-
THE
JOURNAL 0}
O F
C L A S S I C A L
4THE
AMERICAN SCHOOL
STUDIES
AT
ATH
EN
I 7I: NUMBER VOLUME 2002 JANUARY-MARCH
AmericanSchoolof StudiesatAthens Classical 2001
S
and TheAmericanSchoolof ClassicalStudiesatAthensis a research studyof thearchaeology, art, teachinginstitution dedicated to advanced of GreeceandtheGreek history,philosophy, language, andliterature of nineAmerican universiworld.Established in 1881bya consortium ties,the Schoolnowservesgraduate studentsandscholarsfrommore actingasa baseforresearch than150affiliated collegesanduniversities, andstudyin Greece.The mainbuildingsof the Schoolandits library andpublications officesin arelocatedin Athens,withadministrative Princeton, NewJersey.As partof its mission,the Schooldirectsongoingexcavations in theAthenianAgoraandat Corinthandsponsors allotherAmerican-led excavations andsurveyson Greeksoil. It is the andclassicists andthe officiallinkbetweenAmericanarchaeologists Archaeological Serviceof the GreekMinistryof Cultureand,as such,is resources andto the dedicated to thewisemanagement of cultural world.Inquiries about dissemination of knowledge of theclassical in the SummerSessions membership in the Schoolorparticipation shouldbe sentto theAmericanSchoolof ClassicalStudiesatAthens, 6-8 CharltonStreet,Princeton, NewJersey08540-5232. Hesperiais published quarterly bytheAmericanSchoolof Classical
to the StudiesatAthens.Foundedin 1932anddevotedprimarily andSchool-directed timelypublication of reportson School-sponsored fromallscholars workingin the projects, Hesperiawelcomessubmissions history,andliterature, from fieldsof Greekarchaeology, art,epigraphy, journal. earliestprehistoric timesonward. Hesperiais a refereed
VOLUME
7I: NUMBER
JANUARY-MARCH
I
2002
-
I _m
-
I
THEJOURNAL OFTHEAMERICAN SCHOOL OFCLASSICAL STUDIESATATHENS
PUBLICATIONS STAFF EDITOR-IN-CHIEF KerriCox EDITOR, Hesperia TraceyCullen EDITOR, MONOGRAPHS Michael Fitzgerald PRODUCTION MANAGER SarahGeorge Figueira CREATIVE COORDINATOR JordanPeled
PUBLICATIONS C OMMITTEE Carol C. Mattusch (Chairman) George Mason University Darice Birge Loyola Universityof Chicago Jenifer Neils Case Western ReserveUniversity Thomas G. Palaima Universityof Texas at Austin James P. Sickinger Florida State University Kathleen W. Slane Universityof Missouri-Columbia Stephen V.Tracy (ex offrio) Ohio State University
JEFFREY M. HURWIT Reading the ChigiVase PETERKRENTZ Fighting bythe Rules:The Inventionof the Hoplite Agon BRICE L. ERICKSON Aphrati andKatoSyme:Pottery,Continuity, andCultin Late ArchaicandClassicalCrete ALAN S. HENRY The AthenianStateSecretariat andProvisions for Publishing andErectingDecrees
23
41
91
Submissions:Manuscriptsand communicationsshouldbe addressedto Dr.TraceyCullen,Editor,Hesperia,AmericanSchoolof ClassicalStudiesat NewJersey08540;tel.609-683-0800; Athens,6-8 CharltonStreet,Princeton, andphotocopiesof ilManuscripts fax609-924-0578;
[email protected]. originalartworkandphotographs mustbe submittedin triplicate; lustrations aremadewith the Editor.A shouldnot be sent unlesspriorarrangements of the articleshouldalsobe the majorconclusions shortabstractsummarizing reviewprocessandauthors included.Articlesaresubmittedto a double-blind withouttheirnameor accordingly, arerequestedto preparetheirmanuscripts notes,bibliograpreparation, The stylefor manuscript affiliationappearing. canbe foundin the Guidelinesfor on submissions phy,andotherinformation Authorspublishedin Hesperia62,1993, pp. i-xvi; on the School'sWeb site at the aboveaddress. orbywritingto ASCSAPublications (www.ascsa.org); The AmericanSchoolof ClassicalStudiesatAthenswillnotknowinglyprint orinitialscholthe announcement in Hesperiaoranyof its otherpublications afterDecember30,1970, byanymeans of anyobjectacquired arlypresentation otherthanthroughan officiallysanctionedexcavationor survey,unlessthe existingcollectionorwaslegallyexportedfrom objectwaspartof a previously the countryof origin.
Copyright(C)2002 The AmericanSchoolof ClassicalStudiesat Athens Producedat EdwardsBrothers, Inc.,AnnArbor,Michigan. Designby EllenMcKie. The Chigivase. Coverillustration: Rome,VillaGiulia22679. PhotoJeffreyM. Hurwit. postagepaidat Periodicals Princeton,NewJersey,andat additionalmailingoffices. Sendaddresschanges Postmaster: to Hesperia,P.O.Box529, Canton, MA 02021-0529, U.S.A.
ISSN 0018-098X price ISBN 87661-500-0 The annualsubscription Hesperiais publishedquarterly. Subscriptions: $70 forinstitutions($80,interna($70, international), is $60 forindividuals Paymentmust tional),and$33 forstudents(proofof studentstatusrequired). beinU.S.dollars,drawnon aU.S.bankorbymoneyorder,andsentto Hesperia, SubscriberServices,P.O. Box 529, Canton,MA 02021-0529; tel. (U.S.) 800-821-7823; (outsideU.S.) 781-828-8450; fax 781-828-8915; e-mail of Singleissues(currentandbacknumberswhenavailable)
[email protected]. for $15 eachpluspostagefromthe DavidBrownBook Hesperiaareavailable Company,P.O.Box 511, OakviXe,CT 06779; tel. 800-791-9354,860-9459329; fax 860-945-9468; or (outsideNorth America)OxbowBooks,Park End Place,OxfordOX1 1HN, U.K.;tel. +44 (0) 1865-241249; fax +44 (0) 13-17 and19-30 arealsoavailable 1865-794449. IndexII andSupplements fromDavidBrownBooks.Reprintsof IndexI, Supplements1-12 and 18, b.v., andearlyissuesof Hesperiashouldbe orderedfromSwetsandZeitlinger,
[email protected]. P.O.Box 810,2160 SZ Lisse,Netherlands;
Stiitit;itEiashEESiS2< ;00-atl)i; 0A}1
H -
ESPERIA
OR
|int.ti& tt
ff:
NOW/
t1
I N
ELECTRONIC ARCHIVE
E
<
We areproudto announce thatHesperiais now available JSTOR, a nonprofit online organization whosemissionis to createa through cessible widelyacelectronicarchiveof scholarly journalsin manyareas, the arts,humanities, including andsocialsciences. JSTORpermitsa detailed ofthousands of issuesofjournals search extending backasfarasthe17th Hesperia cannowbeaddedto thelist century. of over200journals byJSTOR. Theentirebackrunof currentlyarchived Hesperia,withthe three years,is fullysearchable of thepast in digitizedform.Not exception find onlycanonequickly a specificarticlebyauthor ortitle,butit is alsoeasier to research term intellectual longtrendsby tracingthe development of or specific approaches to a givenmonument. questions CurrentlyJSTOR 1,000 participating libraries at colleges,universities, servesmorethan andresearch tions in 60countries, instituincluding theBlegenLibrary at the of Classical American School StudiesatAthens. VisitJSTOR's Websiteatverwwjstor.org formoreinformation. SEARCHJSTORBasicSearch
1. ENTER SEARCHTERMS: .
4000000000 0 7;|1 Vn; 0
:7-tE11
. t;
7 A;:f
| ln i11 ttf iSS :
|in>i1ltext
i
i
:
.;
,,,
0;00 400f;; 00jf -00;0s ;;
0
;t
| in adorit 0000000 0>^|
2.SELECT DISCIPLINES/JOURNALS TO SEARCH(required): _=fii to searchspecificjournals i
Hesperia (1932-1998) t2002JSTOR
HESPERIA
71 (2002)
Pages 1-22
READING THE
CHIGI
VASE
ABSTRACT Long consideredone of the technicalmasterpiecesof ArchaicGreekvase Chigivase(ca.640 B.C.)hasdefiedattemptsat painting,the Protocorinthian Its imageryhasmostoftenbeenunderstoodas a randomasinterpretation. sortmentof exquisitebutunrelatedscenes-hunts, horsemanship, theJudgmentof Paris,anda hoplitebattle.It is arguedherethatthereis in factalogic behindthechoiceof scenes,andthatthevasedisplaysapliablethematicunity, of the Corinthianmaleandtheinterfocusinguponthe stagesof maturation of the the everyday, exotic,the heroic,andthe divinein the lives penetration of mortals. Therewas a time,not verylong ago,when no one botheredthinkingmuch aboutwhy particularsubjectswerepaintedon particularGreekvases,or why specificscenesarefound togetheron the samevase.1The harddistinctionbetweenmyth and genrewas the only distinctionthat mattered, and since a scene on a pot had to be one or the other,the choicewas inherentlyuncontroversial: mythwas alwaysappropriatebecause,well, the Greekslikedmyth,andgenresceneswerenatural,too,becausethe Greeks had dailylives like everyoneelse. Consequently,the searchfor programmatic or thematicrelationshipsbetweentwo or more sceneson a single vase was rarelyundertaken:the iconographyof Greekvase paintingwas virtuallya randomthing. That time has passed.We now recognizecategoriesof imageryin which the distinctionbetweenthe genericandthe mythological,between the mortalandthe heroicor divine,is not as strict.Considercertainscenes on Attic LateGeometricandArchaicvases,forexample,wherethe every1. In completing this article,I have benefited greatlyfrom the assistanceand adviceof many people, including Alfred Acres,Judith Barringer,LarissaBonfante, Anna Rastrelli(Museo archeologico, Florence), Rex Wallace,the
very helpful anonymousrefereesfor Hesperia,and, above all, Anna Maria Moretti (Villa Giulia),who graciously allowed me to removethe Chigi vase from its vitrine for study and photography.My researchwas also greatly
facilitatedby residenceas a visiting scholarat the AmericanAcademy in Rome, by a Universityof Oregon Summer ResearchAward,and by a University of Oregon Humanities Center Fellowship. I am very gratefulto and for all.
2
JEFFREY
M.
HURWIT
day life is given a heroic characterthrough the depiction of Dipylon shields or battle chariots,2 or where (on a few works by the Amasis Painter) Dionysos makes his epiphany among mortal men who areon routine hunting expeditions.3 Over the last three decades, the choice of subject has also attracted intense attention, from a variety of perspectives.In the 1970s, for example, John Boardman began to interpret Attic vases painted during the Peisistratid era as political, even subversive,documents. Exekias'sfamous scene of Ajax and Achilles amusing themselves with a board game when they should be out looking for Trojans to kill4 is, in Boardman'sview, really a thinly veiled allusion to lax behavior at the Battle of Pallene, ca. 546, when Athenians allegedly played dice as Peisistratos attacked and won his final tyranny (Hdt. 1.63). The presence of the Lakonian cult heroes Kastor and Polydeukes on the back of the same vase supposedly indicates Exekias's pro-Spartan sympathies as well: taken together, the scenes on the Vatican amphora comprise an antityrannicalmanifesto cloaked in myth. Problematic as Boardman's"currentaffairs"approachsometimes is (and important as it is to remember that a privately owned pot is not the same as a work of public propaganda), it has had more than its fair share of proponents, and it has helped clarify the ideological dimension-the political reflections-of many Greek images.5 The syntagmatic relationship between scenes on many other nonpolitical pots is also clearernow.We may not know why the Protoattic Nessos Painter chose to paint the myth of Herakles and Nessos on the body of his name-vase in New York(ca. 675-650) or Exekias,on his fragmentarykrater from the north slope of the Acropolis (ca. 530), the combat over Patroklos's corpse. But the odd-looking lion attacking a deer on the neck panel of the Nessos amphora and the lion fights on the Exekian kratersurely function like Homeric similes: the heroes fight centaurs or each other the way lions maul deer or cattle.6 More broadly, recent structuralist,anthropological, semiotic, and narratologicalstudies have firmly established not only that black- and red-figurevase painting is a "construct"encoding culturalthemes and social attitudes, but also that Archaic and Classical vase painters could approach their task with specific programs and messages in mind, that there is often a correlation between subject and vase shape, and that the particularcombination of scenes on a vase could have paradigmaticvalue (by pairing heroic and mortal behaviors, for instance).7 All in all, the search for thematic unity on a vase is now an orthodox enterprise.8One Archaic vessel has been especially fortunate in the
2. See, e.g., Snodgrass1980; Hurwit 1985b and 1993, esp. pp. 34-36; and Sinos 1998. 3. See von Bothmer 1985, pp. 46-47; Stewart 1987, pp. 36-38. 4. Vatican344; Beazley 1986, pls. 64-65. 5. Major documentsin the debate include Boardman1972,1978a, 1984, 1989; Williams 1980, p. 144, n. 55; and Cook 1987; see also Sparkes1991a,
pp. 69-71. For a recentinvestigationof the use of the Dioskouroi in Athens,see Shapiro1999; and for an uncompromising attackon those who would find political content beneathArchaicimagery,see Neer 2001, esp. pp. 292-294. 6. New YorkNessos amphora: Hurwit 1985a, p. 174 and fig. 72. Exekias'sNorth Slope krater:Beazley 1986, pl. 73; Markoe 1989, esp. pp. 9495, pl. 5:a-b.
7. The literatureis now vast, but see, for example,the variousessaysin Berard1989, Hoffmann 1977 and 1988, Lissarrague1990, Scheibler1987, Steiner 1993, and Shapiro1997. Generally,also Stansbury-O'Donnell 1999, pp. 118-157. 8. This is not to say that the scenes on a pot arealwaysthematicallyrelated; even for Bron and Lissarrague1989, p. 21, "thereis very often no directlink,
READING
THE
CHIGI
VASE
3
devotion it has attracted: the Francois vase (ca. 570), by Kleitias and Ergotimos, which (despite disagreement over details and possible poetic inspiration) has emerged as an anthology of myths chosen to narratethe heroic pedigree, career, and death of Achilles, with a countercurrent of scenes relating to the broader theme of marriage-unhappy marriage,on the whole, but marriagenonetheless. With the battle of pygmies and cranes on the foot to supply comic relief, the Fran9oisvase is perhaps the closest approximation to a "paintedepic"in the 6th century.9 I explorebelow the extent to which some organizing principle or principles may be at work on an even earlier masterpiece of the Greek vase painter's craft: a small polychrome pot whose pieces were found in 1881 during the excavation of a huge Etruscan tumulus accidentally discovered on the property of Prince Mario Chigi, atop Monte Aguzzo, above the village of Formello, about 3.5 km north of Veii. The vessel is now on display in the Villa Giulia.l1
THE VASE The Chigivase(Fig. 1) is perhapsthe earliest-knownexampleof a kindof wine jug conventionallyknown as an olpe-an ovoid or saggingpitcher with a flaringmouth and a verticalribbedhandlethat is fixed to the rim with a pronglikefeatureending in circulardisks (rotelles).l1It is usually assignedeitherto the secondphaseof the Middle Protocorinthian period (MPC II) or to the Late Protocorinthian(LPC) period,but it is at any ratealmostalwaysgiven a dateof around650-640.12 other than proximity,between the differentimages decoratinga vase." And there are still a few scholarswho insist that the searchfor iconographic coherenceon a vase (or,for that matter, in the sculptureof a temple) is a waste of time, the anachronisticexerciseof a modern,literatetemperamentthat (conditionedby fixed texts) seeks programmaticlogic and thematic unity where the ancient mind (conditioned by a predominantlyoral culture)did not. See Small 1999, p. 573, n. 24, who believes such attemptsare doomed to failure"because[the problemof iconographicunity] is solely a modern one."Cf. Ridgway 1999, pp. 8294, who believes that the sculptural programsof ancient temples did indeed bear messages,but that they may not have been as logically or carefullyconstructedas the modern mind (long shapedby written texts and the "controlledmessagesof Christianart")would like or expect. See also Stansbury-O'Donnell1999, pp. 124-129, on problemsof what
he calls "paradigmaticextension." 9. Stewart 1983; Schaus 1986; Carpenter1986, pp. 1-11; Haslam 1991; Isler-Kerenyi1997. 10. Villa Giulia 22679; Amyx 1988, p. 32, no. 3. 11. The modern use of the word olpe, restrictedto such ovoid wine jugs, does not correspondwith ancientuse, when "olpe"could indicate the small perfumed-oilflaskwe know as the aryballos;see Amyx 1988, pp. 488-489, 560-561; Sparkes1991b, p. 63. The Etruscansloved the shape and faithfullycopied it in vast numbers; Amyx 1988, pp. 488, 686. 12. Benson (1986, pp. 105-106) places the beginning of the Chigi Painter'scareerin the MPC II period (660-650 B.c.);Boardman(1998, p. 87) dates the vase "latein MPC, near 650 or later";Payne (1933, p. 23), Simon (1981, p. 50), and Amyx (1988, p. 369) date it to LPC, ca. 640. Ducati (1927, p. 70) dated the Chigi vase and the tomb to the beginning of the 6th century;Karo(1899-1901, p. 8) dated
the vase similarly.Mingazinni (1976) has attemptedto reviseradicallyArchaic pottery chronologiesand dates the Chigi vase to ca. 570; his argumentshave not been widely accepted. Salmon (1984, p. 106) notes that although Corinthianvases had found their way to Etruriafrom the mid-8th centuryon, high-qualityCorinthian importsbegan to arrivein significant numbersaround650 (his date for the Chigi vase).This is preciselythe time when Etruscansociety experienced "greatersocial stratificationand centralisationof power... accompanied by the developmentof an increasingly elaborateand variedelite material culture"(Arafatand Morgan 1994, p. 112). The importationof Corinthian pottery appearsto be a symptomof these phenomena.But, as Small (1994) argues in the case of Attic paintedvases,the importationof foreignvases may tell us less about the generalcourseof Etruscan cultureand fortunesthan is often thought.
JEFFREY
4
a
c
M. HURWIT
b
d
READING
THE
CHIGI
VASE
5
Figure 1 (opposite).The Chigi vase. Rome, Villa Giulia 22679. Photos author
Figure 2. Chamber tomb from Monte Aguzzo. Museo archeologico, Florence. Photoauthor
The vase was deposited in a monumental tomb that, judging from its rough ashlar and quasi-polygonal masonry, was built before the end of the 7th century-perhaps even as early as 630.13The tomb consisted of a 5-m-long corridorlike dromos, two narrow,corbel-vaulted side chambers (one of which, 3.35 m long and 1.90 m wide, has been reconstructedin the courtyard of the Museo archeologico, Florence; Fig. 2), and a large main chamber (7.4 m long and 2.55 m wide) at the back. It was in this main chamber that the pieces of the Chigi vase were found. The relativelyclose dating of the vase and tomb means that although the tomb might have remained in use for more than a single generation, the Chigi vase could have been made and painted at Corinth, exported to Etruria, and buried on Monte Aguzzo all within the course of a few decades, and perhaps a lot less. And that, together with the vase's exceptionally rich figured decoration, raises the possibility in turn that the Chigi vase (like, perhaps, the Fran9ois vase two generations later) was a commissioned piece, specifically made for an Etruscan in the market for items that would, with their foreign cachet, display the owner's good taste, offer him paradigms of Greekness to emulate, or both.14This possibility admittedly remains small, 13. For the date of the Monte Aguzzo tomb, see Akerstr6m1934, pp. 17-18; de Agostino 1968, pp. 109, 111; Steingraber1981, p. 492; and A. De Santis,in Bartoloniet al. 1994, p. 35 (whereit is suggestedthat "unadatazioneall'orientalizzante
medio"-that is, a date even before 630-is possible for the tomb). 14. For Greek vases and the Etruscan market,see generallyRasmussen 1991 (Corinthianpottery);and Arafat and Morgan 1994 and Small 1994 (Attic pottery).For the Francoisvase
(which was given specialtreatment when it was shipped to Etruria,with strutsadded to preventits handles from breaking),see Cristofani et al. 1981, p. 101; Isler-Kerenyi 1997.
6
JEFFREY
M.
HURWIT
but the likelihood that this unique vase arrivedin Etruria as "saleableballast" is minuscule. It was surely a special purchase.15 The tomb on Monte Aguzzo was apparently ransacked twice: first sometime in antiquity, and then in late 1880 or early 1881, when the inhabitantsof Formello,who had been given the right to dig on Prince Chigi's properties, rediscovered and entered the tomb before Rodolfo Lanciani could be entrusted with its more systematic (but still poorly published) excavation.16Some 500-600 impasto, high-quality bucchero, Italo-Geometric, and Corinthian potsherds were found in the same chamber as the pieces of the Chigi vase (about three-quartersof the vessel is preserved). The finds, though plentiful, were otherwise modest, with the exception of a bucchero vessel (datable to the last quarterof the 7th century) inscribed in five lines with two of the earliest-known Etruscan alphabets, some almost incantationlike, nonsensical syllables (for example, azaruazaruazaruas),and a dedicatory inscription that, though open to interpretation, seems to indicate that the vase belonged, or had at some point belonged, to someone named Atianai.17If Atianai was the principal occupant of the main chamber of the tomb on Monte Aguzzo, we in all probability also know the name of the Etruscan owner of the Chigi vase. Although incision is abundantlyused for outlines and details, the vase is notable for a refined polychrome technique in black, reddish-purple, and various shades of yellowish-brown that is usuallythought to owe much to contemporary wall or panel painting.18We are, in fact, told that a Corinthian (Ekphantos) invented the art of painting in color (though we are not told precisely when) and we hear of panel paintings produced as early as the 8th century.l9But there is nothing to indicate that such production was extensive in the 7th century, and it may be doubted whether painters were particularly specialized this early.The few scraps of preserved 7th-century free painting, from the walls of the Temple of Poseidon at Isthmia (ca. 650) and the metopes of the Temple of Apollo at Thermon (ca. 630), might well have been the work of one of the few Corinthian
15. For the controversialtheory that Greek paintedvases had virtuallyno intrinsicor monetaryvalue,see Vickers and Gill 1994. Its many criticsinclude Small (1994), who points out that in later centuriesEtruscanconsumers purchasedAttic paintedvases for a varietyof reasons-to sit as decorative objectson a shelf, to serve as storytelling objetsd'art(like the Francoisvase) or as souvenirs(like Panathenaicamphoras), and, above all, to be deposited expressly in tombs. Small arguesthat althoughthe Etruscansdo not seem to have actually drunkfrom or dined using Attic painted vases, no single explanationfor the
importationof Attic potteryis sufficient. A similarvarietyof uses, and a similarselectivityof productionand consumption,can be assumedfor earlierProtocorinthianimports as well. 16. Ghirardini1882, p. 292; WardPerkins1961, p. 47. 17. The vase itself belongs to Rasmussen'sclassificationld; see Rasmussen 1979, p. 72. For the "Formello alphabet"and the dedicatoryinscription (mi atianaiaachaprialiceveneliSi/ velthurzinace),see Ghirardini1882; Buonamici 1932, pp. 107-108; von Vacano1965, pp. 76-77; Boitani, Cataldi,and Pasquinucci1975, p. 229;
The MacMillan Figure3 (opposite). H. ca. cm. British 6.5 aryballos. Museum1889.4-18.1.Courtesy
Trusteesof theBritishMuseum
Pallottino 1978, pl. 94; Agostiniani 1982, p. 76, n. 127; Cristofani1985, p. 87 (who translatesthe dedicatory inscriptionas "I am [the vase] of Atianai.Achapri [?] dedicated[gave?] me to Venel.Velthurmade me."); and Pandolfiniand Prosdocimi1990, pp. 24-26. We do not know what the word achaprimeans. 18. Payne 1931, pp. 92-98; Robertson 1975, p. 53. 19. See Plin. HN35.16 (Ekphantos) and 35.55-56 (Boularchos'spainting of the "Battleof the Magnetes,"dated by Pliny before the 18th Olympiad,or 708 B.c.); see also Schaus 1988.
READING
THE
CHIGI
VASE
7
craftsmen who primarily painted polychrome vases and who might have easily adjusted their techniques for different media upon commission; the Isthmia fragments bear decorative patterns and a horse's mane comparable to those found on Protocorinthian pots, and the Thermon metopes are even made of baked clay.20 The origin of the painter of the Chigi vase, if not the vase itself, is also at issue. The person who labeled a few figures on the back (see below, Fig. 8) did not write Greek like a Corinthian: he wrote like an Aiginetan, in the opinion of some, or a Syracusan, in the opinion of others.21 It seems likely that either the writer was not the same man as the painter (a possibility we should not dismiss too hastily) or the painter was not a native of Corinth.22 But in any case the warm, creamy, buff-colored fabric of the vase is recognizably Corinthian and its provenience is not in doubt.23 Though the Chigi vase is far larger than an aryballos (the tiny perfume flask that is the quintessential Protocorinthian product), it is still not very large (H. 26 cm) in comparison with other Greek vases and so demanded the skills of a consummate miniaturist. This artist-properly known as the Chigi Painter-is generally credited with at least three other works as well: a fragmentary olpe from Aigina, an aryballos in Berlin, and the British Museum's MacMillan aryballos (where, in its lower friezes, less than a centimeter high, the artist worked on a nearly microscopic scale; Fig. 3).24 The Chigi Painter is also considered the central figure in a small group of polychrome vase painters working in the middle of the 7th century (the Chigi Group),25 and a recently published
20. Cf. Hurwit 1985a, pp. 161-162. It is true that the smallwooden pinakes from Pitsa differ from even the most colorfulof Corinthianvases in technique and style (red and blue predominate), but since they date to the second half of the 6th centurythey cannot fairlybe used to suggest great differences between free painting and works like the Chigi vase a centuryearlier.For the generalproblem(though focusing on the 6th century),see Amyx 1983. It is also useful to keep in mind that even such latervase paintersas the Athenian Euthymides(ca. 510) could execute independentpanel paintings (such as his chocolate-brownwarrioron a plaque from the Acropolis)with a different rangeof color from that seen on his pots; see Boardman1975, p. 54, fig. 53. 21. Payne 1931, pp. 38-39 (Aiginetan,following Rumpf);Jeffery 1990, pp. 125, n. 3,264 (Syracusan). See also Lorber1979, pp. 14-15, no. 13;Wachter2001, p. 31.
22. See Amyx 1988, p. 602, where he points out the dangerin assuming"that the writing on [a Corinthian]vase was in every case providedby the vase painter himself";and Payne 1931, p. 39, where it is suggestedthat "theinscriptionson Protocorinthianvases show us foreign artistsworking at Corinth, in the Protocorinthianstyle." 23. Some have doubted a Corinthian origin for the Chigi vase;for example, Rambo (1918, p. 13, n. 1) believesthat the vase is Etruscan,giving as her reasons its use of landscapein the lowest zone, its findspot nearVulci [sic],and the supcostume of the posedly "sub-Mycenaean" flute playerin the battle scene. I assume that Payne (1931, p. 182, n. 1) is being ironic when he calls Rambo'sconclusion "areal contribution." In the opinion of Karo(1899-1901, 7), p. "derThon ist warmgelbund nicht sehr fein, also von dem hellen, griinlichen, feingeschlammtenThone der gewohnlichen protokorinthischenVasen
verschieden."But even he considersit "einBeispeil der h6chsten Bliite des protokorinthischenStils"(p. 8). No works incontrovertiblyby the painterof the Chigi vase have been discoveredat Corinth itself, though fragments of an aryballosand alabastron relatedto his style have been found at Perachora(Amyx 1988, p. 32, C.1 and C.2) and a fragmentaryalabastronwith a hoplite battle scene found at the socalled Potter'sQuarterin western Corinth was attributedto the artist (the MacMillan Painter)by Dunbabin and Robertson1953, p. 179; Amyx (1988, p. 38, no. 7) prefersan assignment to the broader"Chigi Group."For other sherdsfrom the Potter'sQuarter with affinitieswith the Chigi Group, see CorinthXV, iii, nos. 285,288, 289, 304, and 341. 24. Amyx 1988, pp. 31-32, 369-370. 25. The group also includes the Boston Painterand the SacrificePainter; Amyx 1988, pp. 33-37.
8
JEFFREY
M.
HURWIT
Figure4. OinochoefromErythrai, fragmentaryoinochoe from Erythrai in Asia Minor, with similar scenes of warfare,hunting, and horsemanship,is likely to come from his circle if not from his own hand (Fig.
drawing. Scale1:2.AfterAkurgal1992, p. 84, fig.1:a
4).26
THE SCENES The inside of the rim of the Chigi vase is decorated with hatching and white pinwheel rosettes, while fine lotus palmettes adorn the pronged handle and rotelles (Fig. 5). On the exterior,between the neck and shoulder (again coveredwith lotus palmette chains painted in white over a dark background) and the base (with two abstract zones, one with black rays that lead the eye upward, the other of reddish-purple horizontal stripes against a dark ground), there are four figured bands or friezes (I-IV). I. In the lowest frieze (2.2 cm high), three nude short-haired hunters use a pack of long-tailed dogs to ambush long-eared hares (and, in one case, a vixen) from behind four or five bushes that have the fluidity of aquariumplants (Fig. 6). These are the only elements of landscape found in any zone. One kneeling hunter carries a lagobolon,or throwing stick, as he signals a companion carrying a brace of dead hares on his back to stay low behind a bush. There is no clear indication in the preservedfragments of the sort of trap or net found in other representationsof hare hunting.27 Filling ornaments (hooks, crosses, pinwheel rosettes, S-spirals, zigzags,
26. E. Akurgal,cited in Cook and Blackman1970-1971, p. 41 (attribution to the Chigi Painter);Neeft 1991, p. 15 (E-1); Akurgal1992 (attribution to the "ErythraiPainter");Schnapp 1997, p. 478 (5 bis); Boardman1998, p.278. 27. See, e.g., Schnapp1989, figs. 99-100. A few archinglines preserved below the front hooves of the chariot team in the zone abovemight belong to such a trap.
READING
THE
CHIGI
VASE
9
Figure5. Chigivase,rimandmouth. Photoauthor
Figure6. Chigivase,detail. Photo author
28. For the harehunt, see Anderson 1985, pp. 32-34, and Schnapp1997, pp. 180-181.
and lozenges formed of opposing triangles) are lightly scattered in the spaces between the figures.The direction of the pursuit is mostly right to left.28 II. The next frieze (4.6 cm high) appearsat first glance to be a collection of four or five formally discrete elements, with more abstract ornaments-S-spirals, lozenges, zigzags-tastefully strewn about. First, there is a parade of long-haired horsemen, wearing tunics, riding bareback,using
IO
JEFFREY
M.
HURWIT
only reins and halters, and moving fairly stiffly from left to right (Fig. 1:a). Each rider also leads a riderless horse, and so these youths are probably not jockeys themselves (like the racers on the MacMillan aryballos; Fig. 3)29but squires (hippostrophoi) leading mounts for absent companions or warriors (hippobateis),as we know them from other vases of the period and afterward,at Corinth and elsewhere.30Perhaps, as some think, those missing warriors are to be found dismounted and fighting on foot in the zone above.31Alternatively, the four squires could be holding the horses for other youths in the same zone (as we shall see) or for use in a team. For next comes a light four-horse chariot, driven by a lone youth but led by another (this time nude) youth on foot, who looks back upon his fellows over his shoulder (Fig. l:b).32Although the S-spiral hovering between the lead rider and the chariot has the character of a punctuation mark, the horsemen and chariot are probably part of the same procession. The paradeis broughtto an abrupthalt by a static bicorporatesphinxa monster with two bodies but a single face, who wears an elaborate floral crown (or else the ornament grows out of her head) and who, like a good Archaic figure, smiles a little smile (Fig. 6). Sphinxes are usually innocuous members of the Protocorinthian menagerie and take their place in frieze after frieze beside lions, panthers, boars, goats, and birds, singly or in pairs. They are never found in a Protocorinthian narrative context,33 and there is no hint that they could also be, in Greek art and imagination, destroyersof men and posers of existential questions. Even so, it is worth recalling the sphinx or sphinxes who carry off a fallen warrior from the battlefield on a roughly contemporaryrelief from the acropolisof Mycenae (sometimes thought to be Corinthian in origin).34The Chigi sphinx may not be as sinister a creatureas those but, given the brutal scene to follow (in which a youth is savaged),it may nevertheless introduce intimations of mortality or (as I shall suggest) liminality. At all events, while creatures such as double-lions are known in Near Eastern and Mycenaean art,35the double-sphinx seems to be a specifically Corinthian invention, and this example may be the first of her strange breed. Next to the sphinx, a nearly symmetrical and self-contained lion hunt takes place (Fig. 7). Four youths (one nude but belted, the others wearing cuirasses)spear a magnificent beast that has caught a fallen comrade in its jaws-he is the only human casualty found on the vase. Purplish blood pours out of all (apparentlymortal)wounds. Whether lions actuallyroamed 29. It is possible that they are leading the horses to a startingline for others to race,and it is interesting that, accordingto Pausanias(5.8.8), horse racingwas introducedto the Olympic Games in the 33rd Olympiad, or 648B.C.-close to the date of the Chigi vase and MacMillan aryballos. 30. Greenhalgh1973, pp. 84-88, 96-146; Simon 1981, pl. 67 (Lako-
nian hydriaby the Hunt Painter, ca. 555). 31. Cf. Greenhalgh1973, pp. 8586. 32. Five chariotsrace aroundthe second frieze of the Chigi Painter's aryballosin Berlin (Amyx 1988, p. 32, no. 2), and the chariothere may be such a racerbeing led to the starting line by the youth on foot. 33. Amyx 1988, p. 661.
34. Payne 1931, pp. 89-90; Boardman1978b, p. 39, fig. 35; Fuchs and Floren 1987, pp. 192, 205. 35. For an 8th-centuryplaquein New Yorkwith a double-bodiedwinged lion from Ziwiyeh in Iran,see Osborne 1996, fig. 42 (MetropolitanMuseum 51.131.1). For a double-bodiedlion on a Mycenaeanlentoid gem (Athens, National Museum 2316), see Mylonas 1983, p. 192, fig. 148.
READING
THE
CHIGI
VASE
II
Figure 7. Chigi vase, lion hunt. Photoauthor
the 7th-century Peloponnese is impossible to say, given the present state of evidence.36But it is also beside the point. For even if the Chigi Painter had seen one in the wild (or in a cage) or had seen an imported lion skin, the lion he painted and incised here, with its flamelike mane, is usually thought to be based on Neo-Assyrian models: the Chigi Painter was a rough contemporary of Assurbanipal (669-626
B.C.).37
Horsemen participate directly in the lion attack depicted on the oinochoe from Erythrai (Fig. 4), and so we may wonder whether the whole passage on the Chigi vase from the horsemen and chariot to the lion hunt is a Protocorinthian revision of imagery found in the palace reliefs of Nineveh or Nimrud, where kings and their entourage, riding chariots and horses, slaughter animals by the dozen. Seventh-century Corinthians like the Chigi Painter might have seen such images on imported Assyrian 36. Lions are so common in Minoan, Theran, and Mycenaeanart, and their representationis at times so detailed,that Aegean artistsand their audiencesarelikely to have seen them in the wild; see Morgan 1988, pp. 4445. Lion bones and teeth have actually been found in Late Bronze Age contextsat Kea, Kalapodi,and Tiryns (see, e.g., Boessneck and von den Driesch 1979 and 1981), though that evidenceis equivocal(the teeth may have been importedas amulets).Herodotos (7.125-126) says that lions were presentin northernGreece as late as 480 (when they attackedthe camels
of Xerxes'invasionforce).Though it is sometimeswonderedwhat kind of lions these were (mountainlions?),there is anotherstory that at the end of the 5th centurythe great pankratiastPoulydamas (an Olympic victor in the year 408) killed a lion with his barehands in emulationof Herakles.This beast was said to have come from the region of Mount Olympos, and that it was a "real"lion is suggestedby Lysippos's later representationof the renowned feat in relief on a statuebase at Olympia (ca. 337); see Paus.6.5.4-6; Moreno 1995, pp. 91-93. Nevertheless,by the beginning of the 4th century,according
to Xenophon (Cyn. 11.1-4), one had to leave Greece for foreignlands (such as the mountainsof Mysia) in orderto hunt lions. Cf. Arist. Hist. an. 579b7, 606b15; also Anderson 1985, pp. 4, 55-56. As far as Corinthianvase painting is concerned,lions virtually disappearfrom animalfriezes around 575-550; Payne 1931, p. 67; Amyx 1988,pp.664-665. 37. Payne 1931, pp. 67-69; Brown 1960, pp. 170-176; Amyx 1988, p. 663 (who, citing representationsof female lions with manes,doubts Corinthian vase painterswere directlyfamiliarwith realones).
12
JEFFREY
M.
HURWIT
ivories, metalwork, or textiles.3 The lion on the Chigi vase is the earliestknown example of the Assyrian type in Protocorinthian vase paintingthe normal Protocorinthian lion had previously been based on Hittite models-which suggests a sudden exposure to strong Assyrian influence, somehow precipitated by Assurbanipal'sconquests and fostered, perhaps, by the policies of Kypselos, who overthrew the aristocraticBacchiads and established a tyranny in Corinth around 657, toward the beginning of the Chigi Painter'scareer. Finally, below the handle, in a spot that would have been obscured by the forearmof anyone actuallypouring from the vase, is the only explicitly mythological scene on the olpe (and, with the exception of the scene with the frontal and non-narrative double-sphinx, the only one with female figures): the Judgment of Paris (Fig. 8). This is the earliest extant representation of the myth, but the storywas presumablyfamiliar(to the Greeks, anyway) from popular folktale as well as from the cyclic epic Kypria.39Set between the lion hunt and the last rider of the procession, this scene, too, is formally self-contained: to the left, a long-haired Paris (who here goes by his usual Homeric alias, Alexandros), then the missing Hermes (identified by the tip of his kerykeion),who presents the divine beauty contestants Hera (who is all but lost), Athena (who is helmetless but labeled Athanaia and who carriesin her hand a floral ornament reminiscent of the lotus palmette chains on the rim and neck of the vase),40and, last,Aphrodite (in appearance she is nearly identical to Athena but Aphrod[ita] is inscribed vertically beside her). The discovery of the bucchero vessel with the Etruscanabecedariain the same tomb suggeststhat its occupant,whether or not he was named Atianai, knew the myth, or knew Greek, could at least have sounded out the labels.41 Now, this inconspicuously placed scene seems to announcethe themes of beauty,decision, and ultimatelymarriage(if we loosely regardthe subse38. It is, of course,unlikelythata Protocorinthianvase painterwould have visited Assyriancapitalshimself. See Frankfort1970, figs. 211-214; also Gunter 1986; Barnett 1956, pp. 232233, fig. 2. Anderson (1961, p. 15) suggeststhat horsebackridingbecomes more popularin the 7th centurythan it had been beforebecauseof Near Eastern influences,and Payne (1931, p. 71) even suggeststhat the type of horse seen on the Chigi vase is indebted to Assyrianmodels. 39. The Judgmentwould also be depicted seventyor eighty yearslater on the elaborateCorinthianmythological encyclopediaknown as the Chest of Kypselos,made of cedar,ivory,and gold, and copiouslyinscribed(Paus. 5.19.5). While a varietyof Greek myths and mythologicalfiguresinvade Etruscanart in the centuryafter650,
the Judgmentof Parishad no impact until around550, when the myth appearsfor the first time on the socalled Boccanerapanels from Cerveteri and the so-called Pontic amphorain Munich by the ParisPainter;see Spivey and Stoddart1990, p. 100, fig. 51; Brendel 1978, pp. 153-157. For the iconographyof the Judgmentin general,see Clairmont1951; and LIMC VII, 1994, pp. 176-188, s.v.ParidisIudicium(A. Kossatz-Deissmann). 40. In latervase paintingAthena is sometimes shown holding a branchor flowerin her hand;see, e.g., LIMC II, 1984, pp. 960, 1005, 1011, nos. 31, 523b, 583,584, pls. 706, 758, 761, s.v. Athena (P. Demargne).On the Chigi vase the device may be an attemptto feminize this most masculineof goddesses.At the same time, in many early representationsof the Judgmentthere
is little to differentiatethe three contestants,either in appearanceor attributes;see, e.g., LIMC II, 1984, p. 958, no. 10, pl. 703, s.v.Athena (P.Demargne);LIMC V, 1990, pp. 324-325, no. 455b, pl. 238, s.v. Hermes (G. Siebert);LIMC VII, 1994, p. 178, nos. 9, 12, 13, pl. 107, s.v.ParidisIudicium (A. Kossatz-Deissmann). 41. Cf. Boardman2001, who suggests these labeled figures"arethe only figuresthat might have puzzled the Etruscanbuyer"(p. 31). One might note here the tradition that the Greek alphabetwas introduced to Etruriaby the Corinthianmerchant Demaratus,a Bacchiadrefugeewho settled in Tarquiniaafter 657; see Spiveyand Stoddart1990, p. 96. In fact, the alphabetwas probablyintroduced by Euboeansby 700 or so.
READING
THE
CHIGI
VASE
I3
Figure8. Chigivase,Judgmentof Paris. Photoauthor
42. For sacredprostitutionat Corinth, see Salmon 1984, pp. 398400; and Williams 1986, p. 21, who arguesthat the sexualactivityitself would have taken place in the city below Acrocorinth(a difficulthike), and that only the proceedswould have been dedicatedabove. 43. On many latervases Paris himself realizesthe dangerand attemptsto flee; see, e.g., Beazley 1986, pls. 21.2 (C Painter),35.3, 35.5 (Lydos);also Gantz 1993, p. 569. 44. LIMC II, 1984, p. 47, no. 359, s.v.Aphrodite (A. Delivorriaset al.) is Corinth Museum 4039, a terracotta statuettein pudicapose datableto the mid-7th century.The goddess also appears(with Pegasos)on an early-7thcenturyplaquefrom Perachora;Williams 1986, p. 14. Aphrodite appears lateron the Chest of Kypselosnot only in its Judgmentscene, but also in one panel with Medea andJason and in anotherwith Ares (Paus.5.18.3, 5). For Aphrodite and earlyCorinth, see also Blomberg 1996, pp. 82-84. 45. See Salmon 1984, p. 398; Pindar,fr. 107 Bowra (Ath. Deipnosophistai13.573c-574c). See also Kurke 1996.
quent union of Paris with Helen of Sparta as a marriage),which might at first suggest that the Chigi vase was commissioned as a wedding present. But there may be more to it than that. The contest, after all, ultimately led to war, the subject of the zone above. This display of females and Paris's imminent choice had disastrousconsequences, sending the strong souls of many heroes to Hades as surely as did the epic anger of Achilles. On the vase, however,judgment has not yet been rendered.Moreover, the contest's winner was not just any goddess: she was Aphrodite, with Apollo the most important deity of Corinth, the city's patron goddess and protector. She was probablyintroduced to Corinth from the Near East by the end of the 8th century-her analogue is the Phoenician Astarte-at the time of the unification of the Corinthian polls by the aristocratic Bacchiads (they may have promoted her as a unifying force in the synoikismos).At any rate, when Aphrodite arrived she brought with her the phenomenon (unique in Greece) of sacredprostitution, an activitycentered in a poorly preservedtemple of Aphrodite atop Acrocorinth as early as the 7th century, around the time of the Chigi Painter.42It is possible, then, that the depiction of the Judgment, which of course resulted in the selection of Aphrodite, Paris'sillicit relationship with Helen, and the Trojan War, operates on more than one level. First, it acts as a warning to the (male) symposiast or banqueter to avoid such decisions himself: the female of the species (divine or mortal) is dangerous.43Second, it may reflect the relatively recent selection and establishment of Aphrodite in the city (the goddess may make her first appearancein Corinthian sculpturearound the same time as the creation of the Chigi vase).44The scene, with its erotic overtones, furthermorehints not at marriagebut at the sacredprostitution-and in the words of Pindar, the charms of "youngwomen, hostesses to many,handmaidens of Peitho"-for which lascivious Corinth was so famous, and which, for Strabo,was even the principalsourceof Corinth's proverbialwealth.45
I4
JEFFREY
M.
HURWIT
III. The narrowthird frieze (2.5 cm high) representsanother hunting scene, with badly faded white hounds chasing four white mountain goats, three stags, and one hare over a darkbackground markedby an occasional white pinwheel rosette (Figs. 6-7). The mostly clockwise chase echoes the predominant direction of the hare hunt on the lowest frieze, but here (as in most Protocorinthian hare hunts) the dogs are on their own: there are no human figures lying in ambush or directing the dogs in their pursuit.46 IV. The battle scene of the fourth zone (5.2 cm high), which is not technically a frieze since it is interrupted by the handle, has always received most of the scholarly attention given to the vase (Fig. 9). The reason is that it is usually considered "the earliest unequivocal representation of what is known as 'hoplite warfare,"'thought to have been developed just a generation or two earlier.47Its representation of hoplite warfare, however, is not so unequivocal.The Chigi warriorsdo not carry short swords like standard hoplites and some of them (like Geometric warriors and Homeric heroes) carrytwo spears-one for overhand thrusting, the other a reserve or even a throwing spear.A soldier with two spearsbut no sword is not the sort of hoplite Tyrtaios-the Chigi Painter'srough contemporary-had in mind when he advised:"let [our man] close hard and fight it out with his opposing foeman, holding tight to the hilt of his sword, or to his long spear."48
The Chigi warriors are certainly heavily armed foot-soldiers fighting in close array,with hoplonoverlappinghoplon.But either hoplite side sideide as the tactics, Chigi vase (and a few other Protocorinthian vases) depict them, had not yet uniformly reached their "classic"stage of development or the Chigi Painter did not intend an exact documentation of military tactics;he may instead have used all those spearsto create pleasantly intricate linear patterns, for example, or to give an impression of sheer numbers and the claustrophobia of battle, or even to elevate his warriors to heroic rank (or all of the above).49His goal, after all, was to decorate a vase and convey certain ideas with its imagery, not to produce a tactical training film.50 In any case, two armies, each aligned in two unequal ranks with perhaps a little more space between them than a classic hoplite phalanx ought to have, meet just to the right of center (Fig. 9). It is the instant when the lines first collide (the othismos,or "push"),and no one has yet fallen or died. 46. Schnapp1997, p. 180. 47. Osborne 1996, p. 164; also Cartledge1977, p. 19; and Murray 1980, pp. 125-126, who describesthe battle as "themost successfulportrayal of hoplite tacticswhich has survived." Salmon (1977, p. 87) concedes that the Chigi battle is an inaccuraterepresentation of hoplite warfarebut still "depicts very effectivelythe essentialnatureof hoplite tactics ... [successfully]representing massedformationin a pleasing
manner."But perhapsour conception of hoplite warfareand the hoplite reformin the 7th centuryis not as accurateor complete as is often thought;see, e.g., Krentz1985. [See also P. Krentz'sarticle"Fightingby the Rules:The Invention of the Hoplite Agon"in this issue of Hesperia.-Ed.] See also Shanks 1999, pp. 107-119, 126-130. Generally,see Hanson 1991. 48. Tyrtaios8.33-34 (trans.Lattimore).The next lines (35-38), inci-
dentally,indicate that light-armed fighterscould dartout from the ranks of the hoplites to throwjavelinsand even rocksat the enemy,and then returnto the protectionof the hoplites' shields. 49. Cf. Anderson 1991, pp. 18-20; Salmon 1984, pp. 73-74; for Homeric heroeswith two throwingspears,see, e.g., Patroklosat II. 16.139-141. 50. Cf. Shanks 1999, p. 129.
READING
THE
CHIGI
VASE
I5
Figure 9. Chigi vase, battle scene. Photoauthor
The army on the left, in fact, has been caught off guard:its front rank sets only four men against five while, at the far left (Fig. l:a), two soldiers are still arming-spears of unequal length, fitted with throwing loops, lean beside them-and cohorts carrying only one spear have to run to join the fray.Like the army he painted, the Chigi Painter has seemingly nodded, too, since there is one head too many for the nine shields of the second rank and the four soldiers in the front rank have five pairs of legs (Fig. 9).51 But what the army on the left lacks in preparationand arithmetic it gains in lyricism, as the self-absorbed, pompadoured auletes-spatially isolated and additionally set off by the dark color of his tunic-sets the rhythm for the advance with his double-flute (strapped tightly around his head).52
51. This has been taken as evidence that the Chigi Painterhas compressed and transferredto the small surfaceof his vase a largerbattle painting,with many more figures,found on a wall or panel,the numericaldiscrepancies arisingduringthe processof translation. On the one hand, Robertson (1975, p. 53) doubts that the meticulous Chigi Paintercould have been so clumsy,suggestinginstead that the Chigi Paintermerely"felthe had not spacedthe legs quite right and that the composition needed thickeningin that
area."On the other hand, Robertson finds it difficultto believe that the vase paintercould have conceivedof such a battle "unaided"by the inspirationof a wall painting;cf. Payne 1933, p. 14. For Shanks (1999, p. 115), the problem apparentlydoes not exist, since in art as in realitythe individualhoplite had no identity apartfrom the massed formation:the "body"of the phalanxis all that mattersand the loss of individualitymakesthe numbersirrelevant. But the individualityof the hoplites within the armyattackingfrom the
right is clearlyemphasizedby their differentshield blazons, and on the MacMillan aryballos,in any case, the Chigi Paintersurelynodded once again, since he painted the fifth warriorfrom the left, moving right,with his shield blazon visible,when we should be looking at its emblemlessinterior. 52. Another flute playerappearson a Protocorinthianaryballosfrom Perachora;Amyx 1988, p. 25 (D.1). The Spartansused flute playersto help keep their formationseven and tight as they attacked(Thuc. 5.70).
i6
JEFFREY
M.
HURWIT
Another reason the battle scene has been the focus of most discussions of the Chigi vase is its suggestion of pictorial depth, with its layersof overlappingshields simply but effectively indicating spatialrecession (such Classical works as the Nereid Monument, 250 years later,do not represent depth any better).53The shields of the hoplites advancingfrom the rightthe only shields whose blazons we can see-bear the expected emblems of power,prowess, and ferocity:birds of prey,bull'shead, growling lion'shead, and a boar (Figs. l:c, 6). One shield is, however, unusual for its gorgoneion. Within the imagined scene Medusa'sfrontal-faced, tongue- and tusk-baring scowl is, of course, intended to frighten the enemy away.In reality its function was also apotropaic,meant to fend off evil spirits from those enjoying the wine poured from the olpe itself. It is worth noting that bronze shields with gorgoneiaare known from 7th-century Olympia and also from Carchemish, where one was probablylost by a Greek mercenaryfighting in defense of the city against the Babylonians.54
THE READING What, if anything,do thesevariousfriezesandsceneshaveto do with one another?How shouldwe readthe imageryon this vase?Is this vaseabout anything?The answerhas most often been "no." The usualway of looking at the Chigi vase has been as a random assortmentof exquisitebut disconnectedimages.So, for example,John Boardmanhas suggestedthat the Judgmentof Pariswas "presumably and Tom Rasmussenhas conpaintedsomehowas an afterthought,"55 cludedthat it is unlikelythat anyonewill be able to find "aconnecting threadrunningthroughall the majorscenes.... Many Greekvasesof all periodsshowquiteunrelatedscenesat differentlevelsor on oppositesides, andthereis no needto searchforunityof themeat this earlydateevenon ForRasmussen,then,the Chigi Painter sucha rigorouslyplannedwork."56 knewwhat he was going to paint on the vase beforehe sat down to the task-how else couldit be "rigorously planned"?-buthe had no point to make.This view hasbeen the scholarlyconsensus. There havein the pastbeen a few minorityopinions;for example,I arguedonce that the Chigi vase "forthe most partdisplaysthe kindsof activitya Corinthianyouth of about640 couldbe expectedto engagein The hunting,equestrian,and battle scenes,in and show off his arete."57 otherwords,displaythe skill,courage,andelitismof the idealCorinthian cannotquiteaccommodate thelionhuntmale,thoughthisinterpretation not to mentionthe divinebeautycontest-unless we positthe existenceof lionsanddivinitiesin the 7th-centuryCorinthia.RobinOsbornehasmore recentlyagreedthat"itseemsunlikelythatthe combinationof sceneshere is accidental," but he sees"nosinglewayto 'read'these images"anddrops 53. Hurwit 1985a, pp. 160-161; the subjectaftervaguelysuggestingthat"thethemesof display,decision, Robertson1975, p. 431. 54. Boardman1980, p. 51 and andpursuitthatrunthroughthe figureddecorationheresuggestivelyopen 20. fig. up criticalpathsfor anyviewer."58 55. Boardman1993, pp. 31-32. If the vasehas a singleoverarchingtheme,it is surelythe ag6n(com56. Rasmussen1991, p. 62. 57. Hurwit 1985a, p. 158. petition,struggle,contest)-a conceptthat subsumesthe hare and lion 58. Osborne 1996, p. 164. hunts,the battle,the Judgmentof Paris,andpossiblythe cavalcade,if the
READING
59. Stansbury-O'Donnell 1999, pp. 71-74, fig. 29, finds a similar viewing axis on the Chigi Painter's aryballosin Berlin,where a "nucleus"of two groupsof opposing warriorsin the main frieze is alignedwith the lion's head spout above,two racingchariots in the zone below, and a confronting lion and bull in the zone below that. In fact, the alignmentis not precise:the space between the two chariotsis just to the left of the axis establishedby the opposition of warriorsin the zone above,so that there is a slight deflection from the purelyvertical.This asymmetry is characteristicof the Chigi vase, as we shall see below. 60. Vidal-Naquet 1986, pp. 118119. 61. Forrest1968, pp. 51-54. The origins of the ag6og are notoriously murky.It is possible that the regimen was instituted or more rigorously codified in the aftermathof the battle of Hysiae, which Spartalost to Argos in 669, but it could be much later,the productof a lengthy evolution rather than a single reform.See, for example, Kennell 1995, p. 146, who dates the foundationsof the agoge to the early 6th century. 62. For which see Vidal-Naquet 1986, pp. 117-122; Schnapp1997, pp.123-144.
THE
CHIGI
VASE
I7
horsemen and charioteer are to be considered potential racers. But the idea of the ag6n is too broad to be of much use: it is hard to think of many Greek works of art that do not concern conflict or competition in some way. Beyond this it is possible to read the imagery of the vase more tightly along two dominant axes: 1) the vertical,up and across the stack of figured zones; and 2) the horizontal, around the second zone (the main one). That the imagery was not randomly selected and deployed-and that the Chigi Painter engaged in some degree of advance planning-seems likely from a number of considerations. The squires of the middle frieze, again, might be holding horses for hoplites in the battle zone above (unlikely,but not impossible) and the inconspicuous position of the Judgment of Paris on the back of the vase (Fig. l:d) seems an appropriatechoice for a painter whose interest in mythological narrativewas on the whole minimal. (Alternatively,it is possible to argue that the handle functions as a pointer, leading the eye down to the scene, and thus emphasizing it. But from the perspective of a reclining banqueter having his cup filled by a slave or attendant pouring from the olpe, the scene would have been virtually unnoticeable.) As significant,perhaps,is the direct and surelynot coincidental alignment of the grinning frontal faces of the double-sphinx and the gorgoneion of the shield in the zone above (Fig. 6)-a short axis of (female) apotropaismthat would have been completely visible to the putative (male) symposiast. So, too, it may not be accidentalthat the flute player sounding the notes of the attack in the battle scene is placed almost precisely above the boy gesturing to his companion to stay down in the hare hunt two zones below (Fig. l:b)-a short axis of signaling and signalers.59 But there is also a longer vertical axis and it delineates a process of maturation across the three principal zones: the boys hunting hares in the lowest zone are, with their short hair and nudity, in fact mere boys (harehunting, relying upon trickery,is especially associated with adolescents);60 the horsemen, charioteer,and lion hunters (and even the figure of Paris) in the second zone, with their long hair and tunics, are more properlyyouths; and the heavily armed foot soldiers of the top zone are presumably men (the small auletesis short-haired, like the boys in the lowest zone, though his coiffure is different).The vertical axis, in other words, marksa progression of the Corinthian male from boyhood, to youth, to full manhoodtransitions all made in the context of various agones,a Corinthian paideia loosely comparableto the three-stage agogethat marked the public education and military training of males at Sparta.61There is, as far as I know, no evidence for an analogoussystem at work in 7th-century Corinth (Bacchiad or Kypselid) and it would be unwise to argue for such an institution on the basis of a few vases. Nevertheless, the same progression appears on the Chigi Painter'saryballosin London (Fig. 3) and, to a lesser degree, on the Erythrai oinochoe (Fig. 4), where boys are missing from the hare hunt below. There can be no question of the role hunting played in the education, initiations, and ethos of a hoplite society.62Indeed, the hunting engaged in by the boys and youths in the lower zones on all three vases can be seen as preparation for the warfare of the men above. The various notions that hunting is a rehearsal for battle, that man is an animal who exists to be hunted like any other animal, and that war is a subcategory of hunting (or
I8
JEFFREY
M.
HURWIT
hunting a subcategoryof war) arewell documented later."The exercise [of hunting] itself is the best possible training for the needs of war,"writes Xenophon in the Cyropaedia(1.2.10), and for Aristotle "the art of war is from one point of view a naturalmode of acquisition. Hunting is a part of that art; and hunting ought to be practiced both against wild beasts and against men who, though intended by natureto be ruled by others, will not submit, for that kind of war is by naturejust" (Pol. 1256b, 20-26).63 The vertical progression from hunt to battle on the Chigi vase (as well as on the MacMillan aryballosand the oinochoe from Erythrai) seems to be an early expression of this ingrained Greek attitude, and it may suggest the sort of initiatory practices expected of youths in Archaic Corinth. From this point of view, the Chigi vase is a programmatic piece, designed to inform its buyerand audience-Greek symposiastor Etruscanbanqueterof what makes a man a man.64 This vertical axis, with its paradigm of Greek maleness, is grounded in the generic 7th-century present: Corinthian boys really hunted hares and Corinthian youths reallyrode horses and chariots and Corinthian men really fought other men (even if a few of those shown fighting on the vase wield two spears,like Homeric heroes). The horizontal axis, on the other hand-that is, the course of the second zone-moves from concrete reality to fantasy and myth. Genre fades away when the parade of horsemen and chariot-itself a heroizing vehicle, often used to dissolve the boundaries between mortals and heroes65-reaches the double-bodied sphinx (Fig. 6). In later myth and art, again, the (single) sphinx can be both a dangerous and erotic interlocutor of youths, "posingthem riddles of what life and manhood may be when they are still too inexperienced to understand," combining "the clawed body of a man-eater with the wings of a raptorand a face made for love,"a female destroyerof males.66But she can be a faithful guardianas well as a predator:in sculptureshe is by the end of the 7th century the markerof tombs, squatting atop grave stelai, protecting the dead as the "dog of Hades," as she is known in one funerary inscription.67 The Orientalizing creatureon the Chigi vase may function as a similar kind of sign, a boundary markersignaling a new and different level of being. For on the other side of her is the lion hunt and the only human casualty on the vase (Fig. 7). Even if lion hunts did take place in the 7thcentury Peloponnese, they must still have been considered rareand exotic occasions. This example is still more likely to be a reference to Eastern hunts. It is surely quasi-heroic as well: these Corinthians are killing (and 63. See also Xen. Cyn.1.18,12.1-8; Anth. Pal. 14.17, quoted in Rihll 1993, p. 84, from Burges 1876: "huntingis a practicefor war;and hunting teaches one to catch a thing concealed;to wait for those coming on; to pursuethe fleeing."Cf. Isoc. Panath.163, who states that next to the universalhuman war againstsavagebeasts, the most righteousand necessarywar is the one
that the Greeksperpetuallyfight againsttheir naturalenemies, the Persian barbarians.See also Lissarrague 1989, p. 43, who notes the fusing of the usually"separatespheres"of the hunter and hoplite on some vases;Rihll 1993, pp. 83-84; Schnapp1997, pp. 150-156. 64. It is possible that even the polysemousJudgmentof Parisscene plays a role in this outline of maturation,if
(insteadof warningmen about the danger of Woman or indicatingcivic pride in the city goddess,Aphrodite) it refers to the kind of choice men must make when they take a bride and so embark upon a new stage of life. 65. See Sinos 1998, pp. 75-78. 66. See Vermeule1979, p. 171. 67. See Richter 1961, p. 6.
READING
68. Cf. Schnapp1997, pp. 181, 192, who also notes the contrastbetween the lion-hunting "heroes"of this zone with the simple "jeuneshommes"of the harehunt. 69. The Iliad (24.28-30) sets the location of the Judgmentonly in The Kypriaand "Paris'scourtyard." Euripideantragediesset the scene specificallyon Mt. Ida;see Gantz 1993, pp. 567-571.
THE
CHIGI
VASE
I9
in one casedying)like the heroesof theirown legendsandepic similesas andthe otherGreatKingsof NimrudorNineveh. well aslikeAssurbanipal It is as if these five youthshave dismountedthe four horsesand chariot heldby the squireson the otherside of the sphinxandhavesteppedacross or behind it into anotherontologicalrealm,one very far from that of the hare-huntingboys in the zone below.68An associationbetweenthe equestriansand the lion huntersseems to be confirmednot only by the numbers-five hunterscorrespondto the four riderlesshorses and the chariot-but alsoby the Erythraioinochoe(Fig. 4), wherehorsemenactuallyparticipatein the hunt. This hunt,probablyto be thoughtof as takingplacein somevaguely imaginedEasternlandscapeor mountainside,is followedby the onlyscene of pure myth on the vase, the Judgmentof Paris (Fig. 8), managedby Hermes,god of transitions.The Judgmentsceneis supposedto haveoccurredon Mt. Idain the Troad,closeto areasthat stillboastedlions in the Classicalperiod;for a Corinthianof the 7th century,this settingwas,like The sphereof heroesanddiviniAphroditeherself,sufficientlyEastern.69 ties andthe sphereof the exoticEast,in otherwords,havemerged,andso perhapshasthe sphereof everydaylife.As we progressaroundthis middle to divine zone, we seem to proceedfrom realityto Orientalizing/heroic realms.But what appearto us as differentlevels of being may not have seemedso to the Archaictemperament,just as in the supposedlydocumentaryhoplitebattleof the upperzone,the presenceof two spearsin the handsof manywarriorsmay be an attemptnot so much to fill spaceor a heroic or epic tinge. Taken activatethe scene as to give that "reality" these imagesmay suggest,instinctivelyor by design,the intertogether, of the everyday,the heroic,andthe divinein the livesof men. penetration This axis,perhaps,showswhat makesa man a hero:leoninecourageand the companyandfavorof the gods.But it hintsaswell at the permeability of the boundariesbetweenthe mortalanddivineand,with the ambiguous doublenessof the double-sphinx,the maulingof the youthby the lion, and the imminent,fatefuldecisionof Paris,the dangersof suchan existence. We can only wonderwhetherthe Etruscanownerof the Chigi vase would havegraspedits logic. But he might well havebeen struckby the formalasymmetryof its imagery,seen in the inequalityof the armiesof the battle frieze-their collision takes placejust to the right of center (Fig. l:b)-or in the displacementof the heraldicsphinxfromthe center of the vase,wherewe might haveexpectedit, or even in the Judgmentof Paris,which is not set with perfectsymmetryalongthe line of the handle abovebut is shifteda little to the right (Fig. l:d). This off-centeredness encouragedthe turningof the vase in one'shands,and that very action would haveencourageda processof associationand obliquelyreinforced the kindsof transitionsarticulatedalongthe axesof the vase.Like Paris, who has not yet madehis choice,the vieweris offeredoptions-different coursesto follow,one vertical,one horizontal-ratherthana single,rigid, controllingthematicstructure.In thisway,the unityor thematicarmature of the vaseis pliable.And this maybe whatthe Chigi Painterreliedupon fromthe start-the virtuesof displacement,the intricaciesoficonographic association,and the dynamicpleasureof the tangent.
20
JEFFREY
M.
HURWIT
REFERENCES Agostiniani, L. 1982. Le "iscrizioni parlanti"dell'Italiaantica,Florence. Akerstr6m,A. 1934. Studieniiberdie etruskischen Graber,Lund. M. 1992. "EineprotoAkurgal, korinthischeOinochoe aus Erythrai,"IstMitt 42, pp. 83-96. Amyx, D. A. 1983. "ArchaicVasePaintingvis-a-vis 'Free'Painting at Corinth,"in AncientGreekArt and W. G. Moon, ed., Iconography, Madison, pp. 37-52. .1988. CorinthianVase-Painting the of ArchaicPeriod,Berkeley. Anderson,J. K. 1961. AncientGreek Berkeley. Horsemanship, .1985. Hunting in theAncient World,Berkeley. 1991. "HopliteWeapons and OffensiveArms,"in Hanson 1991, pp. 15-37. Arafat,K., and C. Morgan. 1994. "Athens,Etruria,and the Heuneberg:Mutual Misconceptionsin the Study of Greek-BarbarianRelations,"in ClassicalGreece:Ancient HistoriesandModernArchaeologies, I. Morris,ed., Cambridge,pp. 108134. Barnett,R. D. 1956. "AncientOriental Influenceson ArchaicGreece,"in TheAegeanand theNearEast: StudiesPresentedto Hetty Goldman, Locust Valley,pp. 212-238. Bartoloni,G., A. Berardinetti, L. Drago, and A. De Santis. 1994. "Veiotra IX e VI sec. a.C.: Primi risultatisull'analisicomparatadelle necropolisveienti,"ArchC 46, pp. 1-46. Beazley,J. D. 1986. TheDevelopment ofAtticBlack-Figure,rev.ed., Berkeley. Benson,J. L. 1986. "MiddleProtocorinthianPeriodization,"in Del Chiaro 1986, pp. 97-106. Berard,C., et al. 1989.A City ofImages: and Societyin Ancient Iconography Greece,Princeton. Blomberg,P. E. 1996. On Corinthian TheBridledWinged Iconography: Horseand theHelmetedFemale Head in the Sixth Century B.C.,
Uppsala. Boardman,J.1972. "Herakles,
Peisistratos,and Sons,"RA 1972, pp. 57-72. . 1975. AthenianRedFigure Vases: TheArchaicPeriod,London. .1978a. "Exekias," AJA82, pp.11-25. 1978b. GreekSculpture: The ArchaicPeriod,London. .1980. TheGreeksOverseas: TheirEarly Coloniesand Trade,rev. ed., New York. . 1984. "Imageand Politics in Sixth-CenturyAthens,"in Ancient GreekandRelatedPottery:Proceedingsof theInternationalVase H. A. G. Brijder,ed., Symposium, Amsterdam,pp. 239-247. 1989. "Herakles,Peisistratos, and the Unconvinced,"JHS109, pp.158-159. . 1993. OxfordHistoryof ClassicalArt,Oxford. . 1998. Early GreekVase Painting: 11th-6th Centuries B.C.,
New York. .2001. TheHistoryof Greek Vases: Potters,Painters,andPictures, New York. Boessneck,J., and A. von den Driesch. 1979. "Ein L6wenknochenfundaus Tiryns,"AA1979, pp. 447-449. .1981. "Ein Beleg fur das Vorkommendes L6wen auf der Peloponnesin'Herakleischer'Zeit," AA 1981, pp. 257-258. Boitani, E, M. Cataldi,and M. Pasquinucci.1975. EtruscanCities, London. Brendel,0. 1978. EtruscanArt,Harmondsworth. Bron, C., and F. Lissarrague.1989. "Lookingat the Vase,"in Berardet al. 1989, pp. 11-21. Brown,W. L. 1960. TheEtruscanLion, Oxford. Buonamici,G. 1932. Epigrafiaetrusca, Florence. Burges,G., ed. 1876. TheGreek Anthology,London. Carpenter,T. 1986. DionysianImagery in ArchaicGreekArt:Its Development in Black-FigureVasePainting, Oxford. Cartledge,P. 1977. "Hoplitesand Heroes,"JHS97, pp. 11-27.
Clairmont,C. 1951. Das Parisurteilin derantikenKunst,Zurich. Cook,J. M., and D.J. Blackman.19701971. "Archaeologyin Western Asia Minor, 1965-1970,"AR 1970-1971, pp. 33-62. Cook, R. M. 1987. "Potsand PeisistrateanPropaganda,"JHS107, pp.167-169. CorinthXV, iii = A. N. Stillwell and J. L. Benson, ThePotters'Quarter: ThePottery,Princeton1984. Cristofani,M., ed. 1985. Civilta degli Etruschi,Milan. Cristofani,M., et al. 1981. Materiali per servirealla storiadel Vaso Franfois,Rome. de Agostino, A. 1968. Museo di Firenze,Florence. archeologico Del Chiaro,M. A., ed. 1986. Corinthiaca: Studiesin Honorof DarrellA.Amyx,Columbia,Mo. Ducati, P. 1927. Storiadell'arteetrusca, Florence. Dunbabin,T. J., and M. Robertson. 1953. "SomeProtocorinthianVasePainters,"BSA 48, pp. 172-181. Forrest,W. G. 1968. A Historyof Sparta, 950-192 B.C.,New York.
Frankfort,H. 1970. TheArt and Architecture oftheAncientOrient, Harmondsworth. Fuchs,W., andJ. Floren. 1987. Die PlastikI: Die geometrische griechische undarchaische Plastik,Munich. Gantz,T. 1993. Early GreekMyth: A Guideto LiteraryandArtistic Sources,Baltimore. Ghirardini,G. 1882. "Notizie degli Scavi:VIII, Formello,"MemLinc 1882, pp. 291-300. Greenhalgh,P.A. L. 1973. Early Greek Horsemenand Chariotsin Warfare: theHomericandArchaicAges, Cambridge. Gunter,A. 1986. "EarlyGreek Vase Painting and the Chronologyof PhalanxWarfare,"AJA90, p. 186 (abstract). Hanson, V. D., ed. 1991. Hoplites:The ClassicalGreekBattleExperience, London. Haslam, M. W. 1991. "Kleitias, Stesichoros,and the Jarof Dionysos,"TAPA121, pp. 35-45.
READING
THE
CHIGI
VASE
Hoffmann, H. 1977. Sexualand AsexualPursuit:A Structuralist Approachto GreekVase-Painting, London. . 1988. "Why Did the Greeks Need Imagery?An Anthropological Approachto the Study of Greek Vase-Painting,"Hephaistos9, pp. 143-162. Hurwit,J. M. 1985a. TheArtand CultureofEarly Greece,1100480 B.C.,Ithaca.
. 1985b. "The Dipylon Shield Once More,"CAnt 4, pp. 121-126. . 1993. "Art,Poetry,and the Polis in the Age of Homer,"in From Pastureto Polis:Artin theAge of Homer,S. Langdon,ed., Columbia, Mo., pp. 14-42. Isler-Kerenyi,C. 1997. "Der Fran9oisKraterzwischen Athen und Chiusi," in AthenianPottersand Painters, J. H. Oakley,W. D. E. Coulson, and O. Palagia,eds., Oxford, pp. 523539. Jeffery,L. H. 1990. TheLocalScripts ofArchaicGreece,rev.ed., with A. W. Johnston, Oxford. Karo,G. 1899-1901. "Vaseder SammlungChigi,"AntDenk11.4, pp. 7-8, pls. 44-45. Kennell,N. M. 1995. TheGymnasium of Virtue:Educationand Culturein AncientSparta,Chapel Hill. Krentz,P. 1985. "The Nature of the Hoplite Battle,"ClAnt4, pp. 50-61. Kurke,L. 1996. "Pindarand the Prostitutes,or ReadingAncient Arion 4, pp. 9-75. 'Pornography,"' F. 1989. "The World of the Lissarrague, Warrior,"in Berardet al., pp. 39-51. .1990. TheAesthetics of the Greek Banquet:Imagesof Wineand Ritual, A. Szegedy-Maszak,trans., Princeton. Lorber,F. 1979. InschriftenaufkorinthischenVasen,Berlin. Markoe,G. E. 1989. "The 'Lion Attack'in Archaic GreekArt: Heroic Triumph,"ClAnt8, pp. 86115. Mingazinni, P. 1976. "Ladatazione della ceramicaprotocorinziae di altreceramichearchaiche," MemLinc19, series 8, pp. 491-531. Moreno, P., ed. 1995. Lisippo:L'artee la fortuna,Milan.
21
Morgan, L. 1988. TheMiniature Wall-PaintingsofThera,Cambridge. Murray,0. 1980. Early Greece, Brighton. Mylonas, G. E. 1983. Mycenae,Richin Gold,Athens. Neeft, C. W. 1991. Addendaet Corrigendato D. A. Amyx, Corinthian Vase-Paintingin theArchaicPeriod, Amsterdam. Neer, R. T. 2001. "Framingthe Gift: The Politics of the Siphnian Treasuryat Delphi," ClAnt20, pp.273-336. Osborne,R. 1996. Greecein theMaking, 1200-479 B.., London.
Pallottino,M. 1978. TheEtruscans, New York. Pandolfini,M., and A. L. Prosdocimi. 1990. Alfabetarie insegnamento della scritturain Etruriae nell'talia antica,Florence. Payne,H. 1931. Necrocorinthia:A Study of CorinthianArt in theArchaic Period,Oxford. 1933. Protokorinthische Vasenmalerei,Berlin. Rambo,E. F. 1918. "Lionsin Greek Art" (diss. Bryn Mawr College). Rasmussen,T. 1979. Bucchero Pottery from SouthernEtruria,Cambridge. . 1991. "Corinthand the OrientalisingPhenomenon,"in Lookingat GreekVases,T. Rasmussen and N. Spivey,eds., Cambridge, pp. 57-78. Richter,G. M. A. 1961. TheArchaic Gravestones ofAttica,London. B. S. 1999. Ridgway, Prayersin Stone: GreekArchitectural Sculpture,ca. 600100 B.C.E., Berkeley.
Rihll, T. 1993. "War,Slavery,and Settlement in Early Greece,"in War and Societyin the GreekWorld, J. Rich and G. Shipley,eds., London, pp. 77-107. Robertson,M. 1975. A Historyof Greek Art, London. Salmon,J. B. 1977. "PoliticalHoplites," JHS 97, pp. 84-101. . 1984. WealthyCorinth: A History of the City to 338 B.C.,
Oxford. Schaus,G. 1986. "Gold or Clay? Dionysos'Amphora on the Francois Vase,"EchCl5, pp. 119-128.
22
JEFFREY
Schaus,G. 1988. "The Beginning of Greek PolychromePainting,"JHS 108,pp.107-117. Scheibler,I. 1987. "Bildund Gefass: Zur ikonographischenund funktionalenBedeutungder attischenBildfeldamphoren,"JdI 102, pp. 57-118. Schnapp,A. 1989. "Erosthe Hunter," in Berardet al. 1989, pp. 71-87. . 1997. Le chasseur et la cite: Chasseet erotiqueen Greceancienne, Paris. Shanks,M. 1999. Art and theEarly GreekState:AnInterpretive Cambridge. Archaeology, Shapiro,H. A. 1997. "Correlating Shape and Subject:The Case of the ArchaicPelike,"in Athenian Pottersand Painters,J. H. Oakley, W. D. E. Coulson, and 0. Palagia, eds., Oxford,pp. 63-70. 1999. "CultWarfare:The Dioskouroibetween Athens and Sparta,"in AncientGreekHeroCult, R. Hagg, ed., Stockholm,pp. 99107. Vasen, Simon, E. 1981. Die griechischen Munich. Sinos, R. H. 1998. "Divine Selection: Epiphanyand Politics in Archaic Greece,"in CulturalPoeticsin
JeffreyM. Hurwit OF OREGON
UNIVERSITY DEPARTMENT EUGENE,
OF ART
OREGON
HISTORY
97403
[email protected]
M.
HURWIT
ArchaicGreece:Cult,Performance, Politics,C. Dougherty and L. Kurke, eds., Oxford,pp. 73-91. Small,J. P. 1994. "Scholars,Etruscans, and Attic PaintedVases,"JRA7, pp. 34-58. 1999. "Time in Space: Narrativein ClassicalArt,"ArtB 81, pp.562-575. Snodgrass,A. M. 1980. "Towardthe Interpretationof Geometric FigureScenes,"AM 95, 51-58. Sparkes,B. A. 1991a. GreekArt, Oxford. .1991b. GreekPottery:An Manchester. Introduction, Spivey,N., and S. Stoddart.1990. EtruscanItaly, London. Stansbury-O'Donnell,M. D. 1999. PictorialNarrativein AncientGreek Art, Cambridge. Steiner,A. 1993. "The Meaning of Reception:Visual Redundancyon ArchaicAthenian Vases,"Jd 108, pp.197-219. Steingraber,S. 1981. Etrurien:Stadte, Munich. Heiligtiimer,Nekropolen, Stewart,A. F. 1983. "Stesichoros and the Fran9oisVase,"in AncientGreekArt andIconography, W. Moon, ed., Madison, pp. 5374.
. 1987. "Narrative,Genre, and Realismin the Work of the Amasis Painter,"in Paperson theAmasis PainterandHis World,Malibu, pp. 29-42. Vermeule,E. 1979. Aspectsof Death in Early GreekArt and Poetry,Berkeley. Vickers,M., and D. Gill. 1994. Artful Crafts:AncientGreekSilverwareand Pottery,Oxford. Vidal-Naquet,P. 1986. TheBlack Hunter:Formsof Thoughtand Formsof Societyin the GreekWorld, A. Szegedy-Maszak,trans., Baltimore. von Bothmer,D. 1985. TheAmasis PainterandHis World:Vase-Painting in Sixth-CenturyB.C. Athens,Malibu. von Vacano,O.-W. 1965. TheEtruscans in theAncientWorld,S. A. Ogilvie, trans.,Bloomington. Wachter,R. 2001. Non-AtticGreekVase Inscriptions,Oxford. Ward-Perkins,J.B. 1961. Veii:The HistoricalTopography of theAncient City (BSR n.s. 16), London. Williams, C. K., II. 1986. "Corinthand the Cult of Aphrodite,"in Del Chiaro 1986, pp. 12-24. Williams, D. 1980. "Ajax,Odysseus, and the Arms of Achilles,"AntK23, pp.137-145.
HESPERIA
7I
(2002)
Pages 23-39
FICHTINC
THE
BY
INVENTION
THE
OF THE
RIULE$
HOPLITE
AGON
ABSTRACT This examination of the unwritten rules of Greek warfaresuggests that the ideology of hoplite warfareas a ritualized contest developed not in the 7th century,but only after480, when nonhoplite armsbegan to be excludedfrom the phalanx.Regularclaims ofvictory, in the form of battlefieldtrophies,and concessions of defeat, in the form of requestsfor the retrievalof corpses, appeared in the 460s. Other 5th-century changes in military practice fit the theory that victoriesoverthe Persiansled to the idealizationof massedhandto-hand combat. Archaic Greeks probablyfought accordingto the limited protocols found in Homer. In a collection of essays published in 1968, Jean-Pierre Vernant, Marcel Detienne, and Jacqueline de Romilly spoke of Greek warfareas an agon, a contest, conceived like a tournament with ceremonies and rules.' Though it was not altogether new, this idea soon spreadto other influential French scholars such as Yvon Garlan and Raoul Lonis.2 Pierre Ducrey and W. Kendrick Pritchett have put the subject on a much firmer foundation by meticulously collecting the evidence for many Greek military practices, and Victor Davis Hanson has describedthe misery of Greek battle in gritty detail, even while popularizing the idea that Archaic warfarefollowed unwritten protocols.3This view of Greek warfaredominates the field.4 1. Vernant1968, with Vernant("la guerregrecqueclassiqueest un agon," p. 21); Detienne ("Leheurt des phalangesest soumis a rigles, il a des aspectsludiques:c'est un agon, a la fois concourset combat,epreuveet jeu," p. 123); and de Romilly ("Laguerre entre cites etait, en effet, un etat latent mais non pas incontr6e. Congue comme un tournoi,elle comportaitses rites et ses limites,"p. 211). I am gratefulto the National Endowment for the Humanities for a generousfellowship,and to the
American School of ClassicalStudies at Athens for a wonderfulyear as a visiting professorin 2000-2001. 1 also want to thank audiencesat the College Yearin Athens and the American College ofThessaloniki for their good questionsafteroral presentationsof an earlierversion of this paper.Edgar Krentz,Marion Krentz,M. B. Richardson, Hans van Wees, and Hesperia's refereesmade many helpful comments on written drafts. 2. See Brelich 1961; Garlan 1974, 1975; and Lonis 1979.
3. Ducrey 1999; Pritchett 1971-1991; Hanson 1995, 2000b. 4. In his Warfarein AncientGreece: A Sourcebook, a textbook intended for undergraduates,Sage describesthe way of war in Archaic Greece as "shortsharp clashes that were the productof mutual agreementand had some aspectsof an arrangedcontest"(1996, p. xvii). Connor describesthe "extensivecodificationand thoroughritualization"of Archaicland warfare(1988, p. 18), and phrasessuch as "rulesof combat [battle,conflict, conduct]"run throughoutMitchell 1996.
24
PETER
KRENTZ
Josiah Ober has made the most explicit attempt to set out the unwritten conventions of hoplite warfare. In his article "The Rules of War in Classical Greece,"sOber lists a dozen "common customs (koina nomima) of the Greeks"that governed interstate conflict. He maintains that these rules of war developed after the Homeric epics were put into writing about 700, and that they broke down after about 450, especially during the Peloponnesian War. During the Archaic period, the rules of hoplite warfare"helpedto maintain the long-term practicalworkabilityof the hoplitedominated socio-military system"by making frequent wars possible without risking "demographiccatastrophe."6Hanson also believes that hoplite ideology dominated Archaic warfare,as farmersagreed to decide disputes through pitched battles. "After the creation of the hoplite panoply,"he writes, "for nearly two and a half centuries (700-480 B.c.) hoplite battle was Greek warfare."7Hanson attributes the breakdown of this admirable system to the Persian Wars and the growth of the Athenian empire, a generation before the Peloponnesian War. The earliest referencesto Greek military protocols come in Euripides and in speeches in the historians Herodotos, Thucydides, and Xenophon. In the HerakleidaiEuripides mentions "the customs of the Greeks"(-ootvL 'EXXYlvov v6Vjot;,1010) regarding the killing of prisoners. In the Suppliants he refers to customs regardingburial of enemy soldiers:the "customs of the gods"(v6ot,Ljia Ocov, 19), the "customsof all Greece"(v6ot,i[a aorqS v6oJov,526, 'EXX&rog,311), "the custom of all Greeks"(-cv nIavXXovlvov In refers "the Xerxes to customs of all 671). Herodotos, people"(-c&TrcV'cov the about vOpT7r0ov v6o,u[ja, 7.136.2) inviolability of heralds, and Mardonios describes the way in which he heard the Greeks were accustomed (0aac(i, 7.9f3.1)to wage war.In Thucydides, Archidamos says it is not "customary"(v6ojittov, 1.85.2) to attack someone prepared to make restitution, the Mytilenians refer to "the custom established among the Greeks"(r6 xoa0occo -Tog"EXXYlot 3.9.1) about those who revolt v6oJttzov, the cite the "common customs among the Greeks" a Plataians during war, v (r&oxotv&TE'EXXijvov v6otL[ca,3.59.1) regarding treatment of enemies who surrender,and a Theban herald (in Thucydides' summary) refers to "the customs of the Greeks"(-ca v6ptLt[ja-&Cv 'EXXvwovv, 4.97.2) regarding invaders and sanctuaries.In Xenophon's summary of the Eleians' refusal to let Agis pray for victory in war, the Eleians cite "the old custom" (-6 o&pxaiovv6oittov, Hell. 3.2.22) that Greeks not consult an oracle about a war against other Greeks. Claims made in the second half of the 5th century, however, do not prove that the customs were really old. By the time of the Peloponnesian War, for example, Thucydides could describe the annual public burial of Athenian war dead as an "ancestralcustom"(7rd'prtog v6too, 2.34.2). Kimon probably began this practice when he brought back the ashes of the men who died at Eurymedon (Paus. 1.43.3), and the law requiringpublic burial at home probablygoes back no further than the mid-460s. So this "ancestral custom"started only one generation before the Peloponnesian War.8 Some customs-the ones in which the gods took an interest-certainly go back to Homer: oaths, including oaths sworn as part of a negotiated surrender, were respected; heralds, priests, and suppliants in sanctuarieswere inviolable; the dead were buried.9What about the other
5. Ober1996. 6. Ober 1996, pp. 60-61. 7. Hanson 1995, p. 241. 8. Pritchett 1985, IV, pp. 112-124. The 460s date fits the earliestinscriptional evidencefor the burialgames (IG I3523-525, the first of which, however, Lewis andJefferyincline to put ca. 479) and the earliestinscribedcasualty list (IGI3 1144). 9. On oaths, see Karavites1992. On heralds,"messengersof Zeus and men," see II. 7.274-276 and elsewhere;in the Odyssey,the Laistrygoniansrevealtheir inhumanityby eating Odysseus'sherald (10.110-117). On priests,note that Odysseussparesa sacredgrove of Apollo, along with the priest,Maron, and his child and wife (Od.9.197-201; the troublesin the Iliad begin when Agamemnon mistreatsChryses, anotherpriest of Apollo, by refusingto acceptransomfor his daughter,1.9100). On suppliantsin sanctuaries,see Parker1983, pp. 181-182, and Karavites 1992, pp. 150-155. On burying the dead, see II. 7.394-432, where the Greeks acceptthe TrojanheraldIdaios's requestfor a truce to burythe bodies. Achilles'attempt to mutilateHektor's body is the exceptionthat provesthe rule.Apollo protectsthe body,and in the end Zeus has Achilles grant Priam's requestfor a truce to hold Hektor's funeral.
FIGHTING
BY THE
RULES
25
alleged protocols? Do the rules apply to the fighting in the Iliad? If not, when do they first appear?I will argue that some practices go back to Homer, that others are matters of tactics ratherthan conventions, and that several important new rules and practices appear only in the 5th century. I will propose an alternativemodel below for the developmentof Greekwarfare, agreeing with Hans van Wees' recent suggestion that the hoplite phalanx did not reach its Classical form until after the PersianWars.?1A new, nostalgic ideology of war developed as fighting became more destructive.
A REVIEW OF MILITARY PROTOCOLS Ober assumesratherthan defends the existence of his informal rules,which he draws from the works of other scholars. Nevertheless, Ober's formulation of the rules (indicated by italics) will serve as a convenient foil for discussion. I consider them not in descending order of formality, as Ober lists them, but in the order in which they would arise during a campaign. hostilitiesagainst Thestateofwar shouldbeofficiallydeclaredbeforecommencing an appropriatefoe. In their study of Greek diplomacy,FrankAdcock and D.J. Mosley say that "althoughsurpriseattackswere made it was the habit of the Greeks to make a formal declaration of war."1lSo it would certainly seem from Herodotos and Polybios. Herodotos has the PersianMardonios say (7.9p.1, Waterfield trans.): Besides, from all I hear, the Greeks usually wage war in an extremely stupid fashion, because they are ignorant and incompetent. When they declare war on one another they seek out the best, most level piece of land, and that is where they go to fight. The upshot is that the victors leave the battlefield with massive losses, not to mention the losers, who are completely wiped out. And Polybios, comparing the practices of his own day to those of an earlier era he admired, says (13.3.2-6, Patton trans.): The ancients would not even consent to get the better of their enemies by fraud, [3] regarding no success as brilliant or secure unless they crushed the spirit of their adversariesin open battle. [4] For this reason they entered into a convention among themselves to use against each other neither secret missiles nor those discharged from a distance, and considered that it was only a hand to hand battle at close quarterswhich was truly decisive. [5] Hence they preceded war by a declaration [rob; 7coX?[0ouo 10. See van Wees 2000, pp. 155156. 11. Adcock and Mosley 1975, p. 202.
aXX]iXo; rTcpo6Xeyov],and when they intended to do battle gave
notice of the fact and of the spot to which they would proceed and arraytheir army.[6] But at the present they say it is a sign of poor generalship to do anything openly in war.
26
PETER
KRENTZ
The earliest attested instance of a herald declaring war, however, is the Corinthian herald sent to Corfu before the Corinthian fleet set sail in 435 (Thuc. 1.29.1).12 In his detailed description of the outbreak of the Peloponnesian War a few years later,Thucydides has the Lakedaimonian king Archidamos say that it is not "customary"(v6otL,tov,1.85.2) to attack someone preparedto make restitution. Rather than declaring war on Athens, the Lakedaimonianssent delegations making various demands, which the Athenians rejected, though they said they were willing to go to arbitration as required by treaty (Thuc. 1.145.1). Once the Peloponnesians were on the march,a final ambassadorwas refusedadmission (Thuc. 2.12.12). These delegations, sent to negotiate before an invasion, sound Homeric: a young Odysseus went to Messene to seek reparationsfor 300 sheep and their shepherds that the Messenians had taken (Od. 21.16-21), and Menelaos and Odysseus went to Troy to demand the return of Helen (II. 3.205-224, 11.138-142). Greek practice, therefore, remained fundamentally the same from Homer to Thucydides: Greek states normally sought reparations for injuries before invading enemy territory,but did not declare war in the formal Roman manner. Hostilities are sometimesinappropriate:sacredtruces,especiallythosedeclared for the celebrationof the Olympicgames,shouldbe observed. literAbundant evidence shows that Greeks observed an xseXseptia, for a the Eleusinian and the festiMysteries ally "hands-off," panhellenic vals at Olympia, Delphi, Isthmia, and Nemea.13These truces protected pilgrims and contestants going to and from the festivals, as well as the state sponsoring the festival.They did not prohibit all warfare.Moreover, to be valid the truces had to be declared and accepted. People celebrating a festival were not necessarily immune from attack. In fact, as Aineias Taktikos noted in the 4th century, an ideal time to attack an enemy was during a festival (4.8).14 It is not certain when the custom began. Homer does not mention of any these festivals or the sacredtruces for them. The Olympic truce was alleged to have begun in the 8th century,but the earliest historical reference appearsin an inscription from Selinous dated ca. 460, about the same 12.Thereis somepossiblenegative evidence:earlyin the 5th centuryAigina and Athens fought a "warwithout a herald"(n6oXsioSa&xYpux-o;, Hdt. 5.81.2), and the Lakedaimoniansand Messenians are said,by much later writers,to have done the same earlier still (Paus.4.5.8, Plut. Pyrrh.26.11). If this expressionmeans"unannounced," as Myres (1943) argued,it implies that a war would normallybe announced.But it may carrythe sense,well attestedin
the 4th century,of a war so bitter that the two sides did not communicate throughheralds. 13. See Baltrusch1994, pp. 117122, with literaturecited. Herodotos (6.106.3-107) mentions a rule (v6Loq) forbiddingthe Lakedaimoniansfrom marchingout before the full moon, but as Pritchettnotes (1971, I, p. 120), the ban appliedonly to marching,not fighting. 14. For example,Kolophonianexiles
shut the gates againstthe men of Smyrnawhen they were outside celebratinga festivalof Dionysos (Hdt. 1.150); the Aiginetans attackedthe Athenians duringa festivalat Sounion (Hdt. 6.87); the Athenians hoped to surprisethe Mytilenianswhile they were outside the city celebratinga festivalof Apollo (Thuc. 3.3.3); and Phoebidas seized the Theban Kadmeia duringthe Thesmophoria(Xen. Hell. 5.2.29).
FIGHTING
BY THE
RULES
27
time the Eleusinian truce first appears in an inscription.15The truces for the festivals at Delphi, Isthmia, and Nemea could not predate the founding of the festivals in 582,581, and 563, respectively.Given the state of our evidence, it would be rash to assert that these truces began only in the 5th century,but nothing puts them in the 7th century either. Waris an affair of warriors,thus noncombatantsshouldnot beprimarytargets of attack. No one ever claims that civilians should not be attacked. In practice invadersdid not attack noncombatants because defenders got them out of the way, either behind city walls or off to a friendly state or into the hills.16 For example, when Agesilaos invaded Akarnania in 389, the Akarnanians fled into walled cities and sent their cattle to the mountains (Xen. Hell. 4.6.4). Similarly, each time the Peloponnesians invaded during the ArchidamianWar,the Athenians brought their children,women, and possessions into the city, and sent their sheep and cattle to Euboia and other islands (Thuc. 2.14.1). This alleged protocol is therefore no protocol at all, but rather a matter of military tactics. Battlesshouldbefoughtduring the usual (summer)campaigningseason.
15. For the referenceto the Olympic truce,see Jameson,Jordan, and Kotansky1993: side A, line 7. For the Eleusiniantruce,see IG I3 6 B 847. The story of the Olympic truce's origin appearsin Phlegon, FGrHist257 F 1, Plut. Lyc.1.1-2, and Paus.5.4.5-6, 20.1. The "discusof Iphitos,"apparently seen by Aristotle, contained an inscribedversion of the truce,but was probablymade to justify Eleian control of the games (Lammer1982-1983, pp. 49-50). 16. Miiller (1975) cataloguesthe occasionson which Greeks sent children,women, and possessionsawayfor safety.Pritchett (1991, V, pp. 350-351) notes some additions,mostly from after
350.
17. Pritchett 1974, II, pp. 147-155. 18. Pritchett 1974, II, p. 148.
In the 4th century,Demosthenes remarkedthat the Lakedaimonians, like everyone else, used to campaign only during the four or five summer months (9.48). But no source turns this fact into a should,into a rule of proper conduct. Battles in the Archaic period were fought during the summer because for Greece's farmer-soldiers, fighting at other times of the yearwas impracticalif not impossible.When increasedeconomic resources in the 5th century made pay for military service the norm, at least in Athens, campaigns occurred at other times of the year too. The timing of campaigns was another matter of military tactics ratherthan military conventions. A battleisproperlyprefaced ofthe challenge. bya ritual challengeand acceptance The title of Pritchett's chapter on this subject, "The Challenge to Battle," does not include the word "ritual,"and even so it suggests something more formalthan what he describes.17 Despite Polybios 13.3.5 (quoted above), there is no known case in Archaic or Classical history of a Greek herald issuing a challenge to battle at a particulartime and place-though a Persianherald challenges the Lakedaimonians at Plataia to a single combat between equal numbers of Persians and Lakedaimonians (Hdt. 9.48). Pritchett takes the deployment of an army in battle formation as a challenge to fight: "phalanxbattles normally began when both sides were ready.They were, to use Polybios's phrase, [wxXpi 6EooXoyoo" (battles by agreement).18In this sense, each of the four days of fighting in the Iliad begins with a "challengeto battle":both sides arm and go out to fight
28
PETER
KRENTZ
(3.1, 8.53-59, 11.15-66, 20.1-4). The formal language mentioned by Pritchett, however, does not appear in Homer. When does it first occur? The phrase"byagreement"(~6otoXo6yoo) is common in Polybios (1.87.9, 2.66.4,3.90.5,4.8.11,11.32.7, F 144), but absent from the Classical historians. Another term for pitched battle,rrap&rcxTa tL, is also common in Polybios and absent from the Classical historians-or rather all but absent, for Thucydides does use it once, when he says that the battle of , 5.11.2).19 Amphipolis in 422 was not a pitched battle ([Ji ex 7apocCTdscoS The verbcapaTadooC occurs frequently in Thucydides and Xenophon, but only three times in Herodotos (8.95.1,9.31.2,32.2). I wonder whether Archaic troops were ever deployed more specifically than they are in the Iliad, where the heroes sometimes arrangetheir men in five sections under five named leaders (4.293-296,16.171-198). Nestor places the chariots in front, the braveinfantryat the back, and the cowardsin the middle (4.297300).20The linguistic evidence, therefore, points to a more formal deployment only after the Persian Wars. The main issue, however, does not revolve around terminology. Homeric warriors happily deceived their enemies, yet Herodotos's Mardonios and Polybios (quoted above) assert that Archaic Greeks fought open battles when both sides were ready.Their assertions do not apply to Classical warfare,which is full of deceptions.21It is true that Greeks deployed in a plain rarely attacked an enemy's camp or sprang an ambush during a battle. Commanders who brought their armies out into a plain believed that their troops were a match for the enemy. Under those circumstances, they generally hesitated to try risky deceptive maneuvers. A desperate, or daring, commander might, like Peisistratos in 546, attack during the afternoon siesta (Hdt. 1.63). And if the risks could be minimized, even a Lakedaimonian king might attack a camp during a meal. When the Argives put off battle in 494, Kleomenes observed that they were obeying his herald's commands, and had his men attack after the herald gave the order for breakfast (Hdt. 6.77-78). These examples from the Archaic period suggest that practicehad not changed from what Homer describes. Greeks did not feel obligated to accept a challenge to battle if they were heavilyoutnumbered.Most battles took place between armiesof about the same size.22If the defending armywas heavily outnumbered-as must often have been the case, given the variations in size among the Greek poleis-the leaders usually declined a battle. Perikles' famous refusal to lead the Athenians out to fight a land battle in the Peloponnesian War was not a strategy devised on the spur of the moment in 431.23 19. For the term rotapacrcaL,see Isoc. 10.53, Dem. 9.49, Aeschin. 3.88, Polyb.2.18.2,2.21.5,2.26.8,2.51.3, 2.70.6, 6.26.11,15.12.3, 30.4.2. Plutarch(Mor.231E) describesthe battle that broke out afterthe "Battleof Champions"(Hdt. 1.82) as a,duXrEx but Herodotos does not. atpocar&zco;, 20. Xenophon'sSokratesalso recommendsputting the best men in the front and the back (Mem.3.1.7-8).
21. See Krentz2000. 22. Hanson 1995, p. 277. 23. In additionto the Greekswho abandonedtheir cities in the face of the Persians(Byzantinesand Chalkedonians,Hdt. 6.33.2; Naxians, Hdt. 6.96; Phokians,Hdt 8.32; and the Athenians themselves,Hdt. 8.41), we know of many invasionsbefore 431 that did not culminatein a battle:Tolmides'capture of Chaironeiain 447 (Thuc. 1.113,
though he was ambushedon his way home); the unsuccessfulAthenian siege of Oiniadai in 454 (Thuc. 1.111.3); the Athenians'failureto take Pharsalosin 454 (Thuc. 1.111.1), when the Thessaliancavalrykept them pinned to their camp,and similarlyin 457 (Diod. Sic. 11.83.3-4); the unopposedLakedaimonian burningof trees in the Megarid in 457 (Thuc. 1.108.2); the Phokian invasionof Doris in 457, when the
FIGHTING
BY THE
RULES
29
Use of nonhoplitearmsshouldbe limited. Despite Polybios'sclaim (quoted above) that "the ancients"agreed not to use unseen missiles or missiles shot from a distance, the only such agreement known is the one Strabo says was inscribed on a column in the sanctuary of Artemis Amarynthia, prohibiting missiles in the Lelantine War (10.1.12). Polybios and Strabo probably drew on the 4th-century historian Ephoros for this pact, and Everett L. Wheeler has arguedthat Ephoros invented it as part of a protest against the catapult, a frightening new distance weapon in his day.24 But even if the agreement is historical, it is the exception rather than the rule. Archaic battles included many projectile weapons, with lightarmed men-javelin- and stone-throwers, slingers, and archers-fighting in the phalanx, not in separate units or behind the hoplites.25The 7thcentury poet Tyrtaios, for example, advises (F 11.35-38, West trans.): You light-armed men, wherever you can aim from the shield-cover, pelt them with great rocks and hurl at them your smooth-shaved javelins, helping the armouredtroops with close support. In the Archaic period, the distinction between "light-armed"and "hoplite" was not always sharp, as a few examples will demonstrate. Athenian redfigure vases sometimes depict archerswith greaves, helmets, and shields, and a mid-6th-century bronze statuette of Herakles as an archer,found near Amphipolis, wears a bronze cuirass. A 6th-century molded pithos found at Sparta shows a slinger with a crested helmet. The north frieze of the 6th-century Siphnian Treasuryat Delphi has two giants, armed with helmets and shields, throwing stones. The interior of a 6th-century cup found in the Athenian Agora shows a running warriorwearing an Oriental leather cap and greaves, carrying a hoplite shield and two spears.26The Chigi vase from Corinth, ca. 640, shows fully armed hoplites with two spears, one a javelin.27Athenian vases continue into the 5th century to show some hoplites with javelins, and burialsexcavatedat Sindos, in northern Greece, regularlyinclude a larger and a smaller spear until the late 5th century.28
Phokianscapturedone citybeforethe Lakedaimonians arrivedandcompelled themto leave(Thuc.1.107.2);Miltiades'unsuccessfulsiege of Parosin 489 (Hdt. 6.133-135); the Phokians'
flightto the mountainsbeforetheir nightassaulton the invadingThessalians,probablyin the 480s (Hdt. 8.27); Histiaios'ssiege ofThasos in 494 (Hdt. 6.27); Hippokrates'sieges of Kallipolis,Leontinoi, Naxos, and Zankle in the 490s (Hdt. 7.154.2); Miltiades'siege of Myrina on Lemnos in the 490s (Hdt. 6.140.2); the Lakedaimoniansiege of Samos in ca. 523
(Hdt. 3.47, 54-56); the siege of Kirrha or Krisain the early6th century(Isoc. 14.33, Aeschin. 3.107-113, Diod. Sic. 9.16, Paus. 10.37.4-8). Campaigns without battles surelygo underreported in our sources.For an argumentthat the Athenians had devised their PeloponnesianWar strategyby the 450s, when the long walls were built, see Krentz 1997. 24. Wheeler 1987. 25. See vanWees 2000, pp. 146-156. 26. For a list of armedarcherson Athenian vases, see Lissarrague1990, p. 129; the statuetteof Heraklesis
Boston 98.657 in the Museum of Fine Arts, Boston; for the molded pithos, see Wace 1905-1906, p. 292 and pl. 9; for the SiphnianTreasurystone-throwers, see FdD IV.2, pp. 84, 89. The Agora cup is publishedin AgoraXXIII, pp. 299300, no. 1678 and pl. 109. 27. Forthe Chigi vase,see vanWees 2000,pp.136-137. [Seealsofigs.1 and9 in J. M. Hurwit's article"Readingthe -Ed.] Chigi Vase"in this issueof Hesperia. 28. Sindos(1985), a museum exhibition catalogue,documentsthe Sindos finds.
30
PETER
KRENTZ
This mix of warriors and weapons sounds very Homeric.29By Thucydides'time, hoplites and light-armed troops normally fought in separate units. Before the hoplites engaged at the battle of Syracuse in 415, for example, the stone-throwers, slingers, and archers routed each other repeatedly,"aswas likely for light-armed troops"(Thuc. 6.69.2). When did the Greeks exclude nonhoplite troops from the phalanx?There may have been a small cavalryforce at Athens in the 6th century,but only after 479 did the Athenians establishwhat I. G. Spence calls a "propercavalrycorps," increased to 1,000 plus 200 mounted archersby 431.30True horsemen, as opposed to mounted hoplites, do not appearin Peloponnesian cities until the late 5th or early 4th century.31The Athenian contingent of infantry archers first appears at Plataia in 479.32 As for light-armed soldiers, Thucydides says that at the time of the battle of Delion (424), Athens had no organized light-armed troops (tLXoli x rpocPaoxEuYg, 4.94.1). Little more than a dozen years later, however, Athens did have its own lightarmed men, and no longer had to depend on Thracians or allies.33The Thebans had their hamippoi,infantry fast enough to serve with cavalry,by 418 (Thuc. 5.57; Xen. Hell. 7.5.23). Pursuit of defeatedand retreatingopponentsshouldbe limited in duration. No such rule is attested for Greeks in general. Thucydides says that the Lakedaimonians fought stubbornly until they routed the enemy, but then pursued them neither far nor for a long time (5.73.4). He does not explain their reasoning, and the fact that he makes the point at all suggests that other Greeks pursued their opponents long and hard. Seven hundred years later Plutarch explained that the Lakedaimonians thought it ignoble for Greeks to kill men who had fled, and added that the policy had the practical benefit of making enemies more inclined to run (Lyc. 22.9-10; Mor. 228F). Perhaps the cautious Lakedaimonians thought more about not exposing their troops to a reverseif they scatteredin pursuit.After the battle of Haliartos in 395, the Thebans pursued the Lakedaimonians into the hills, where they rallied, first with javelins and stones, and killed more than 200 Thebans. Lakedaimonian practice avoided such a reverse.
29. In the Iliad spearsare more often thrown (87 times) than thrust (79 times), arrowsoutnumberswords as weapons (21:19), and warriorsthrow stones 12 times (vanWees 1994, p. 144). Some contingents specializein archery-Paionians (2.848, 10.428) on the Trojanside and Lokrians(13.712718) and Philoktetes'men (2.718-720) on the Greek-but there are also individualslike AlexandrosandTeukros who fight now with a bow (11.369378, 8.266-272), now with a spear (3.330-339, 15.478-483), and even
individualswho seem to have bows and other weapons in the same fight, such as Pandaros(spearand bow, 5.171-216, 238-285), Meriones (spearand bow, 13.159, 650-652), and Helenos (sword and bow, 13.576-595). The Lokrians fight well with bows and slings,without helmets, shields, or spears(13.712718), but other archers,such as Pandaros,do have armor(5.294-295), and the Catalogueof Ships describesthe LokrianleaderAias as an expertwith the spear(2.527-530). All seven of the namedwarriorswho throw stones
(Agamemnon,Aineias, Aias, Diomedes, Hektor,Patroklos,and Peirosa realrosterof champions)fight elsewhere with other weapons (Pritchett 1991, V, pp. 4-5). 30. Thuc. 2.13.8; Spence 1993, pp. 9-19; Bugh 1988, p. 39. 31. Spence 1993, pp. 1-9. 32. Hdt. 9.22, 60; archerson ships appearat the battle of Salamisa year earlier,Aesch. Pers.454-464. See Plassart1913. 33. Thuc. 8.71.2; Xen. Hell. 1.1.3334. See Best 1969.
FIGHTING
BY THE
RULES
31
When they had the opportunityto do so safely,Greeksshowedlittle hesitationin slaughteringtheirenemies.In a fragmentary poem foundon seems to that the Messenians "willkill every papyrus,Tyrtaios imagine that catch the battle" West (F 23a.20-22, trans.).In fleeing Spartan they 510, the Krotoniatesroutedthe invadingSybaritesand did kill everyone they caught(Diod. Sic. 12.10.1).Fleeingfromthe Atheniansaftera defeat in 460, some Corinthiansbecametrappedin a field surroundedby a ditch,with no exit.The Atheniansblockedthe frontwith hoplites,surroundedthe Corinthianswith light-armedtroops,andstonedthem all to death(Thuc. 1.106.1-2). After the battleof Delion in 424, the Boiotians and especiallythe Boiotianand Lokriancavalrychasedand killedAtheniansuntil darknessstoppedthe pursuit(Thuc.4.96.7-8). In 392, at the battlebetweenthe Corinthianlong walls,the Lakedaimonianskilled so manyCorinthians,Xenophonsays,that "menused to seeing moundsof grain,wood, and stones now saw moundsof corpses"(Hell. 4.4.12). In killedmorethan 10,000Arkadianswithoutlos368, the Lakedaimonians a in man what was called-from the Lakedaimonianpoint of ing single view-the "TearlessBattle"(Xen.Hell. 7.2.31;Diod. Sic. 15.72.3). The extentof a pursuitthereforeappearsto be anothermatterof militarytacticsratherthanmilitaryprotocols. Punishmentofsurrendered shouldberestrained. opponents
34. Naiden 2000, pp. 71-72. 35. For similarsentiments,see F 53; DissoiLogoi, Herakleitos,FVorsokr FVorsokr vol. 2, p. 410; P1.Resp.468a, Leg.626b; Arist. Pol. 1255a.
In Euripides'Herakleidai(961-966, 1009-1011), "the customs of the Greeks"forbid later killing a prisoner taken on the battlefield, but do not say that a soldier offering to surrender must be spared. Warriors never spare individuals who try to surrenderduring combat in the Iliad: when Menelaos was once about to show mercy, Agamemnon told him not to spare anyone, not even the unborn child (6.37-65). Other rejected appeals include 11.122-147, where Agamemnon cuts off Hippolochos's arms and sends him spinning away like a log, and 21.64-135, where Achilles feeds Lykaon'scorpse to the fish. For an example of mercy granted during fighting, we have to go to Odysseus's Cretan tale at Od. 14.276-284, where the Egyptian king spareshim as he grasps the king's knees. In his recent study of supplication, Fred S. Naiden suggests that Greek soldiers did not spare battlefield suppliants either because they might fight again, or because, on the principle of reciprocity,they deserve no mercy.34The same rationale applies to cities taken by storm: Homeric heroes kill the men and enslave the women. That's what Achilles did at Lyrnessos (II. 2.691), Lesbos (II. 9.665-666), Skyros(II. 9.667-668), andTenedos (II. 11.623-626). It'swhat Odysseus does to the Kikonians(Od. 9.39-61). And it's what Hektor imagines will be done to Troy (II. 6.447-465). The evidence for laterwarfareshows little change. When a city fell, by siege or assault, the defeated might be killed or sold into slavery. As Xenophon's Cyrus says, "it is a custom established for all time among all that when a city is taken people [v6(oo; yacp EvTae&v Cv0pr7oTLa&i't&o] in war, the persons and the property of the inhabitants belong to the captors" (Cyr. 7.5.73).35
32
PETER
KRENTZ
Erectinga battlefieldtrophyindicatesvictory;suchtrophiesshouldbe respected. This rule appliesverywell to the fighting inThucydides andXenophon, as Pritchett's catalogue of examples shows.36Pritchett finds a trophy in II. 10.465-468, where Odysseus dedicates Dolon's equipment to Athena and puts it on a bush. But he puts it on the bush only in order to find it later, when he recoversit and plans to offer it elsewhere (10.570-571). When did the custom of erecting trophies begin? Herodotos never mentions them, nor does he mention the epinikia,the victory sacrificethat The earliestliteraryreferenceprobaccompaniedthe erection of a trophy.37 comes from the tentatively dated to ably mock-epic Batrachomyomachia, first of the 5th More dated are the referencesto the half century.38 securely trophies in lines 277 and 954 of Aeschylus's Seven against Thebes,produced in 468. The earliest physical remains of trophies belong to monuments from the Persian Wars: the marble columns erected, probably in the 460s, at Marathon and Salamis.39Battlefield trophies of the simple sort mentioned in Thucydides and Xenophon-pieces of captured armor and weapons hung on a post or tree-do not appearin vase painting until the middle of the 5th century, as John Beazley observed.40Trophies appear in other art forms (relief sculpture, coins, and gems) later still.41The later trophies differ from the monuments of the PersianWars in severalways: they were erected immediately,constructed of perishable materials,and placed where the enemy turned to flee, rather than where most of the enemy died.42 After a battle,it is right to returnenemydeadwhen asked;to requestthe return of ones dead is tantamountto admitting defeat. Greeks took no unwritten law more seriously than the obligation to allow defeated enemies to retrieve and bury their dead.43 Pritchett's exhaustive study of the burial of Greek war dead details the conventions. A herald regularlyasks for a truce, and Justin says that the Greeks considered the sending of this herald as a concession of defeat (6.6.10). The last 36. Pritchett 1974, I, pp. 246-275. examples,one Boiotian black-figure, and one Campanianred-figure,in 37. Pritchettnotes that none of the 96 occurrencesof 060 in Herodotos Caskeyand Beazley 1963, pp. 66-67. The Boiotian black-figurefragmentwas refersto a post-battle sacrifice(1979, III, p. 186). originallypublishedas late 6th century bei ThebenI, p. 123 159: o-YJoo.s?V (Kabirenheiligtum 38. Batrachomyomachia O(s -cp6oOCLov. and pl. 19.7), but fits better in the late ?U06tUio ToV[u6x-rovov 5th century(Kabirenheiligtum bei The date depends on the poem's ThebenIV, pp. 5-7). attribution(by Plutarchand the Suda) 41. Woelcke 1911;Janssen1957. to Pigres, the son (Plut. Mor. 873F) or 42. Often, of course,most of the the brother(Suda s.v. Pigres) of the casualtieswould fall nearwhere they CarianArtemisiafamous from her turnedand fled. But the findspot first in Xerxes' invasion of participation of the fragmentsexcavatedby VanderGreece,in which Pigres also participool at Marathondoes not,paceWest pated (Hdt. 7.98.1). 39. For Marathon,see Vanderpool 1969, p. 7, fit "thetopographical 1966. For Salamis,Wallace 1969. requirementsfor the trophy."According to laterpractice,we would expect to 40. Beazley made the observation find a trophynearerthe Athenian burial afterlisting eleven Attic red-figure
mound, or soros.The monument apparentlystood nearthe ancient marsh,where the largestnumberof Persiansdied. 43. The Thebans came closest to violating the rule,firstwhen they refused to returnthe Athenian dead at Delion until they recoveredthe temple the Athenians had fortified,on the groundsthat the Athenians had violated the rules of the Greeksregarding sanctuaries(Thuc. 4.97-101), and againwhen they attachedconditions to the truce at Haliartosin 395 (Xen. Hell. 3.5.24). On the first occasion,the Theban case had some plausibility, since Athenians and other Greeks regularlydenied burialto templerobbers.
FIGHTING
BY THE
RULES
33
sentence ofXenophon's Hellenikacomments on the odd result of the battle of Mantineia in 362, when "both sides returned the dead under a truce as though victorious,and both receivedback their dead under a truce as though defeated"(7.5.26). The Lakedaimonians condemned their king Pausanias to death in 395 partlybecause he opted to retrievethe corpses of Lysandros and others under a truce, ratherthan try to recoverthem by fighting (Xen. Hell. 3.5.22-25). The right to bury the dead also appears in Homer (see note 9), but an important change occurs. The Iliad has a burial truce after the first day of battle, but not after the second, third, and fourth, so it does not appear to be standard practice. When the Trojan herald Idaios makes the request that first day, not only does he not concede the victory, he promises to fight again (7.396).44 By Thucydides' time, the losers' request for permission to retrieve their dead has become as regular as the erection of trophies-in fact, Pritchett notes that "the context in which historians refer to the avaxpcQEoG TS'V Vxpcv
(retrieval of corpses) is re-
"Historians"here peatedly that of justificationfor erection of the trophy."45 not include who no more mentions burial truces than he does Herodotos, does trophies. Prisonersof war shouldbe offeredforransom,not summarilyexecutedor mutilated
44. Similarly, whenPriamasks Achillesto returnHektor'sbodyfor burial,Priam too promisesto fight again (II.24.666). 45. Pritchett 1985, IV, p. 247. For examples,see Thuc. 1.63.3; 2.79.7, 82, 92.4; 4.72.4, 101.4; 5.10.12-11.2, 74.2; 6.70.3-71.1, 97.5, 103.1; 7.5.3, 72.1-2; 8.106.4; Xen. Hell. 4.3.21; 6.2.24, 4.1415; 7.1.19, 4.25, 5.13, 5.26. 46. See Ducrey 1999. 47. Pritchett 1991, V, pp. 245-312.
As discussed above, Greeks never felt a moral obligation to take prisoner an opponent begging for mercy during a fight. But both Homeric and Classical ethics dictated that prisonerstaken in another context should be spared.46In the IliadAchilles is said to have releasedfor ransom Priam's sons, Isos and Antiphos, whom he caught on Mt. Ida (11.104-106), and to have sold at least one other, Lykaon, whom he caught in Priam'sgarden one night (21.35-41), and probably more, given the plural references at 21.102 and 22.45. Achilles' sacrifice of twelve Trojans on Patroklos'sfuneral pyre (23.175-176) indicates not that killing prisonerswas acceptable behavior, but that his anger still raged out of control. In Thucydides, as well as in Euripides'Herakleidai(quoted above), we find the claim that by Greek custom captors should not execute prisoners who had surrendered (3.58, 3.66.1). An incident early in the Peloponnesian War shows that other captives ought not to be killed either. In 430 the Athenians executed a half-dozen Peloponnesian ambassadorsintercepted on their way to the Persian king. They threw the bodies into a pit, thinking they were justified since the Lakedaimonians had begun acting this way at the beginning of the war, when they had killed and thrown into pits all the Athenian and allied tradersthey caught on merchantships aroundthe Peloponnese (Thuc. 2.67.4). This phrasing suggests that the Athenians were retaliating against what they perceived as improper behavior. Sparing prisoners did not necessarily mean releasing them for ransom. They might be sold into slavery or forced into exile. After an extensive survey of captives'fate, Pritchett concludes that ransoming was relatively infrequent and done for financial, not humanitarian,reasons.47The only real evidence for better treatment of prisoners in the Archaic period is the 4th-century oratorAischines' referenceto an oath supposedly sworn
PETER
34
KRENTZ
by the Amphiktyons at the time of the First Sacred War (early 6th century?) not to depopulate any Amphiktyonic polis (2.115). The authenticity of this oath is tied up with the knotty question of the historicity of the First Sacred War itself, which John Davies pronounced "a plausible hypothesis, but no more."48Because the Amphiktyonic oath inscribed at Athens in 380 makes no reference to this clause (CID I 10, lines 4-10), it seems safer to conclude with Pritchett that Greeks treated captives similarly from Homer's time through the Classical period.
FIFTH-CENTURY
CHANGES
The preceding analysis suggests that the list of customary practices traceable to Homer includes demands for satisfactionpriorto an invasion.Treatment of surrenderedopponents and prisoners of war does not change in the Archaic period, and does not deserve the term "convention."Other alleged military protocols turn out to be a matter of military tactics rather than of formal conventions designed to ameliorate warfare:noncombatants were not attacked because cities got them out of the way; campaigns usually took place in the summer because the weather was most reliable then and farmerscould afford to leave their fields, trees, and vines; victors sometimes refrainedfrom vigorous pursuitof a defeated armybecause scattering in pursuit might make them susceptible to a counterattack. Some important changes remain. Nonhoplite arms began to be excluded from the phalanx about the time of the PersianWars, when cavalry and light-armed troops started to fight in their own distinct units. Only for a pitched battle. Regular thereafter do we find the word Tocp6dratoc claims of victory, in the form of battlefield trophies, and concessions of defeat, in the form of requests for the retrievalof corpses, appearedin the 460s.49Trophies placed at the "turning"only make sense when the first turn generally became a rout, as it did in Classical hoplite battles, but not in Homeric fighting.o5 Hoplite warfare, therefore, did not break down gradually in the 5th century,but quite the opposite. New military protocols developed not in the 7th century,but only after 480. Other relevant military practices also developed in the 5th century rather than the 7th. The paean sung before joining battle first appears in Aeschylus's account of the battle of Salamis in 480 (Pers. 393).51 In the Iliad the Greeks sing paeans to appease the god who sent a plague (1.472474) and to celebrate a victory (22.391-394), but they advance into battle in silence (3.8-9, 4.427-432). The Dorians by the time of Thucydides, and the Athenians by the time of Xenophon, also advanced to fight singing the song that brought courage to friends and dispelled fear of the enemy (Aesch. Sept. 270). As Pritchett points out, Greeks who advanced singing the paean gave up any attempt to achieve surprise.52 Second, Herodotos mentions aristeia, the awards for individual braveryin battle, for sixteen individuals,while Thucydides and Xenophon mention no such awards,since individual exploits mattered less in Classical battles than they had in Archaic warfare.53A change in shield devices also emphasizes the egalitarianism of the Classical phalanx. Apparently a
48. Davies 1994, p. 206.
49. I passoverherethe trucesfor festivalsthat appearabout 460, because they may have startedearlier.In seeking to explainthe Olympic officials'change in name from &LaTacmTop to EXXovoxrxqa ca. 480, Siewert (1992, p. 115) suggests that, becauseof the Olympic truce,the Eleianswere named the guardiansof the panhellenictruce agreedupon in 481 for the defense againstthe Persians.Alternatively,the truce of 481 may have promptedthe Olympic truce. 50. The battle of Solygeiain 425, where the Corinthianleft wing retreatedto a hill, regrouped,and chargedagain (Thuc. 4.43.2-3), resembles the fighting in the Iliad more than it does other Classicalbattles. 51. Pritchett (1971, I, pp. 105-108) has collected the evidence. 52. Pritchett 1971, I, p. 105. For instance,at the Nemea Riverin 394 the Lakedaimoniansdid not realizethe allied troopswere advancinguntil they heardthe paean,whereuponthey immediatelydeployedfor battle (Xen. Hell. 4.2.19). 53. Pritchett (1974, II, pp. 276-290) has assembledthe evidencefor aristeia.
FIGHTING
54. Anderson 1970, pp. 17-20. 55. Jackson1991, pp. 246-247, and 1992. His volume on the arms and armorfound at Isthmia is in preparation. 56. Hanson (1995, p. 344 note) dismisses the decline in panoply dedications by suggestingthat hoplite battles became less common. But Connor (1988, pp. 6-8) arguesthat scholarshave overestimatedthe frequencyof Archaicwars.For a plausible argumentthat a specific dedicationat Olympia led to a ruthlessreprisalat Sepeia in 494, see Jackson2000. 57. Connor (1988, p. 9) includes the consultationof an oraclein his description of a typicalArchaic militarycampaign. 58. On fighting in Homer, see Snodgrass1993 and van Wees 1994 againstPritchett (1985, IV, pp. 7-44), who acceptsLatacz's1977 argument that Homer describesphalanxwarfare.
BY THE
35
RULES
matter of individual choice in the Archaic period, shield devices tended to become standardized in the 5th and 4th centuries: a lambda for the Lakedaimonians, a sigma for the Sikyonians, a mu for the Messenians, the club of Herakles for the Thebans, the trident of Poseidon for the Mantineians.54This stress on the similarityof hoplites goes ideologically hand in hand with the exclusion of other troops from the phalanx. Third, dedications of armorat panhellenic sanctuaries,extremelycommon in the Archaic period, decline in the 5th century.At Isthmia excavators have found none after the destruction of ca. 470-450, at Olympia few after the 430s, at Delphi few after the spoils sent byTegea and Mantineia in 423.55This change in offering patterns anticipates Plato's advice that Greeks should not dedicate armor and weapons if they want to preserve good relations among Greeks (Resp. 469e-470a).56 We have here a new military protocol designed to ameliorate warfare. Finally, in Xenophon's summaryof the Eleians' refusalto let Agis pray for victory in war, the Eleians cite an "oldcustom"(opxocLov vo6ttl[ov, Hell. not an oracle about a war that Greeks consult 3.2.22) against other Greeks. This alleged protocol does not appear on Ober's list, presumablybecause the assertion is so flagrantly untrue for Archaic warfare. Herodotos, for instance, relates that the Lakedaimonians once misinterpreted a Delphic oracle and ended up confined in the chains they had brought for their intended Tegean prisoners (1.66).57The Eleians are not necessarily inventing the prohibition they cite. It fits with the previous decline in panoply offerings at panhellenic sites. But "old"cannot mean more than one generation, since the Lakedaimonians consulted Delphi in 432 about going to war with Athens (Thuc. 1.118.3). It is of course true that we have less-much, much less-information about Archaic than about Classical warfare.The fact that extant literary sources do not attest a practice until the 5th century does not disprove its existence earlier.But the material evidence supports the conclusion that some military protocols came late ratherthan early.If we can find a coherent theory to explain this late development, we ought to accept it.
THE DEVELOPMENT
OF GREEK WARFARE
I suggest the following alternative model for the development of hoplite warfare.Although they improved their equipment, Archaic Greeks continued to fight in the way Homer describes.The fighting was "mass"fighting, but not "massed"fighting.58That is, battles were not simply fought by champions in front of nameless, and unimportant, followers. The mass of men mattered. But they did not deploy in a tight formation, massed together. Rather, they advanced, retreated, and advanced again in a formation loose enough to allow horses, perhaps even chariots, to approach the killing zone and withdraw again. Brave men moved forward; tired men, frightened men, wounded men moved back. In exceptional circumstances, such as a struggle over a fallen warrioror a break in a wall, a group of men might bunch together. Stones, javelins, and arrows flew, thrown and shot by some of the same men who dared to advance and fight hand-to-hand,
36
PETER
KRENTZ
as well as by less courageous men who hung back.This kind of fluid battle had no single turning point. Then the Mede came. At Marathon in 490, the Athenians confronted a larger Persian force.59They knew how the Persians fought, from their experience during the Ionian Revolt a few years earlier when they were routed in a battle at Ephesos (Hdt. 5.102). They knew the Persians liked to soften the enemy with a barrageof arrowsbefore closing. AtThermopylai ten years later,the Greeks were told that the Persians shot so many arrows that the shafts hid the sun-leading to Deinekes' famous remarkthat this news was good because the battle would take place in the shade (Hdt. 7.226.2).The Athenians decided to arm entirelywith hand-to-hand weapons (Herodotos says they had no cavalryor archers,6.112.2) and charge at a run (Herodotos also reportsthat they were the first Greeks to run against the enemy, 6.112.3). The effect must have been something like what happened in Cilicia, when Cyrus the Younger arrangedhis Greek mercenaries four deep for a paradeand had them charge,wearing their bronze helmets, crimson chitons, greaves, and shields. They went faster and faster, finally breaking into a run toward the camp, and frightened away the Cilician queen and the people in the market, who left everything behind as they fled (Xen. Cyr. 1.2.16-18). So, at Marathon, the Athenian charge routed the Persian wings, once and for all. The experience encouraged the Athenians to behave more like foxes than hedgehogs. They realized the physical and psychological power of a massed infantry charge, and continued to exclude nonhoplite forces from their phalanx. But they also appreciated the value of organized contingents of horsemen and archers.They soon established a largercavalryforce and started an archerycontingent, with some success at Plataia, where an archer killed the Persian cavalry commander Masistios (Hdt. 9.22.1). By 431 they even had Persian-style mounted archers,a remarkableinstance of"Perserie."60 Athenian hoplites, on the other hand, especially after Salamis and the growth of the Athenian navy, remembered Marathon as the essence, the model, of what warfare should be.61"We know,"writes Nicole Loraux, "that in the 5th century a whole ideological structurewas built up around Marathon at the expense of Salamis, the victory of the oarsmen people."62 In his Laws Plato argued not only that the infantry battle of Marathon began the salvation of Greece and that of Plataia finished it, but also that these battles made the Greeks better,while the navalbattles of Artemision and Salamis made them worse (707c). We can trace this idealization of Marathon back to the 420s, when Aristophanes spoke of the Marathonomachai(Ach. 182, Nub. 689), to the 450s, when Aeschylus's epitaph mentioned his courage at Marathon, and to the 460s, when the Athenians erected a marble monument at the battle site.63Perhaps it was also in the 460s, at the instigation of Miltiades' son Kimon, that a few lines honoring the Marathon fighters were added to an inscribed epigram honoring the men who fought at Salamis.64Hoplites made sure that Salamis did not eclipse the memory of Marathon. As for the Lakedaimonians, they visited the battlefield at Marathon and, no doubt, askedthe Athenians how they did it. To judge by Herodotos's
59. Marathoncontinuesto produce controversy.For two good recentstudies, see Evans 1993 and (even better) Lazenby1993, pp. 45-80. 60. See Miller 1997 for a study of how profoundlyPersiancultureinfluencedthe Greeks,who professedto despise orientalluxury. 61. On the paradigmaticfunction of Marathonin Athenian ideology,see Prost 1999 with literaturecited. On the hoplite ideal, see also Hanson'sexcellent brief summary(2000a, pp. 219222). Of the sixteenpassagesHanson cites (p. 229, n. 32) denigratingcavalry and light-armedtroops,only one, II. 11.385-387, was written before the second half of the 5th century. 62. Loraux1986, p. 161. 63. Aeschylus,the greatAthenian poet who fought in the battle of Salamisand describedit in his Persians (472), wrote an epitaphfor himself that mentioned neithertragedynor Salamis, but said that the grove at Marathon and the Persianswho landed there witnessed his courage(Paus. 1.14.5). 64. ML no. 26, with Amandry 1960.
FIGHTING
the Greekfor65. In describing Herodotos mation, saysthat35,000 helots light-armed protectedthe 5,000 sevenstationedwitheach Spartiates, man (Trrepavv80pcExac-rov ircTa 9.28.2), and that the other TrTc-ay[ulvoL,
Greekshadonelight-armedsoldierfor eachhoplite.Thesetroopsallfoughtin the battle (9.61.2), and since Herodotos says lack of armorhurt the Persiansbecausethey were fighting againsthoplites (9.63.2), Hunt (1997) has plausiblysuggestedthat the Spartiatesformed only the front line, with the less heavilyequippedhelots behind them, making the phalanx eight-deep. 66. Pritchettassertsthat "the extension of the word [ag6n]from 'agora'(Homer) to either an athletic contest or a militaryone seems to have developedmore or less simultaneously" (1974, II, p. 284, n. 28), and finds the oracle'sambiguityin the numeralfive. But the earliestevidencehe can find of agonin a militarysense is Aesch. Pers. 405, dated to 472. 67. Harrison(1971) arguespersuasivelythat the word in the "Kallimachos epigram"(IG I3 784 = ML no. 18), inscribedbefore 480, refersto an agonisticvictory ratherthan to the battle of Marathon.Korres(1994, p. 174) gives a reconstructiondrawing of this monument,including the base, the Ionic capital,and the Nike statue (Acr.690) that stood atop the column.
BY THE
RULES
37
account of Thermopylai in 480, the Lakedaimonians continued to use a fluid fighting style effectively.They repeatedly pretended to flee, only to turn on the barbarianswhen they pursuedin disorder(7.211.3). The struggle over Leonidas's body sounds truly Homeric, with the Greeks routing the Persiansfour times before they were able to drag the corpse away (7.225.1). But at Plataia in 479 it is clear that the Lakedaimonians admired Athens' way of fighting, for the Lakedaimonian king Pausanias asked the Athenians to switch places with the Lakedaimonians and oppose the Persians, since the Athenians had the experienceof Marathon (Hdt. 9.46.2-3). Later he asked for the aid of the Athenian archers(Hdt. 9.60). In the battle, the Persian archersshot from behind a fence of wicker shields until the Greeks charged. Unable to get away-probably there were too many of them too close together-the Persiansthrew awaytheir bows and tried to fight handto-hand, rushingout individuallyor in small groups (Hdt. 9.62). The Greeks won, then, by charging en masseto close quarters,just as the Athenians had done at Marathon.65These great victories, Marathon and Plataia, won by hard hand-to-hand fighting, loomed large in the collective memory of hoplites. After Xerxes' retreat,the Greeks soon launched counterattacks as far away as Cyprus and Byzantion (Thuc. 1.94.2). With the growth of the Athenian empire, war became much more destructive than it had been in the Archaic period. Earlier, as Thucydides points out (1.15), neighbors fought most wars without coalitions of allied forces, except in the shadowy Lelantine War. As Classical Athens accumulated a monetary surplus through campaigns against enemies and taxes from allies, the nature of warfare changed. Athens sent out expeditionary forces that remained in the field much longer than Archaic campaigns had lasted. As early as the 460s, the Athenians sustained the siege ofThasos for more than two years (Thuc. 1.101.3). By the time Herodotos wrote Mardonios's speech, hoplites had idealized the Archaic way of war as a ritualized agon, or contest. The story of the mantis Tisamenos makes it unlikely that the term agon was in general use for "battle"before the Persian invasions. When Tisamenos asked the Delphic oracle about a child, the Pythia predicted that he would win five ag6nes.So he trained for the pentathlon, and almost won at the Olympics (he lost in wrestling). The Lakedaimonians then realized that the oracle meant five battles, and persuaded Tisamenos to become their seer. He then helped the Lakedaimonians win five victories, beginning with the battle of Plataia in 479.66 The earliest text to use agon in the sense of "battle,"Aeschylus's Eumenides(914), dates to 458.67 The Archaic way of war was not a single, head-on collision of hoplite phalanxes, excluding cavalryand projectile weapons, commemorated by a battlefield trophy.The idea of agonistic warfare matters-it helps to explain why the Classical Greeks fought big battles such as Mantineia, Koroneia, and Leuktra. But we should not treat the Classical ag6n as a debased form of an Archaic way of war that never existed. Despite changes in armor and weapons, Archaic Greeks fought according to the conventions found in Homer. Greeks invented the hoplite ag6n in the mid-5th century.
38
PETER
KRENTZ
REFERENCES Adcock, F. E., and D.J. Mosley. 1975. Diplomacyin AncientGreece,New York. AgoraXXIII = M. B. Moore and M. Z. P. Philippides,Attic BlackFiguredPottery,Princeton1986. Amandry,P. 1960. "Surles epigrammes de Marathon,"in Theoria:Festschrift F. Echstein, fiir W H. Schuchhardt, ed., Baden-Baden,pp. 1-8. Anderson,J. K. 1970. Military Theory and Practicein theAge ofXenophon, Berkeley. Baltrusch,E. 1994. Symmachieund Spondai,Berlin. Best,J. G. P. 1969. ThracianPeltastsand TheirInfluenceon GreekWarfare, Groningen. Brelich,A. 1961. Guerre,agoni,e culti nellaGreciaarcaica,Bonn. Bugh, G. R. 1988. TheHorsemenof Athens,Princeton. Caskey,L., andJ. D. Beazley.1963. Attic VasePaintingin theMuseumof FineArts,Boston,Boston. Connor,W. R. 1988. "EarlyGreek Land Warfareas Symbolic Expression,"PastandPresent119, pp. 3-28. Trans.by J. Odin as "La guerreterrestreen Grece archaique comme expressionsymbolique,"in La guerreen Grecea l'epoqueclassique, P. Brule andJ. Oulhen, eds., Rennes 1999,pp.289-314. Davies,J. 1994. "TheTraditionabout the First SacredWar,"in Greek S. Hornblower,ed., Historiography, Oxford,pp. 193-212. de Romilly,J. 1968. "Guerreet paix entre cites,"in Vernant1968, pp. 207-229. Detienne, M. 1968. "Laphalange: Problemeset controverses,"in Vernant1968, pp. 119-142. Ducrey,P. 1999. Le traitementdes prisonniersdeguerredansla Grece antique,2nd ed., Paris. Evans,J. A. S. 1993. "Herodotosand the Battle of Marathon,"Historia 42, pp.279-307. FdD IV.2 = C. Picardand P. de la Coste-Messeliire, Les Tresors Paris 1928. '"oniques," FVorsokr= H. Diels andW. Kranz,Die
10th ed., FragmentederVorsokratiker, Berlin 1960. de Garlan,Y. 1974. Recherches poliorcetique grecque,Athens. . 1975. Warin theAncientWorld, London. Hanson, V. D. 1995. The OtherGreeks, New York. .2000a. "Hoplite Battle as Ancient GreekWarfare:When, Where, and Why?"in Warand Violencein ClassicalGreece,H. van Wees, ed., London, pp. 201-232. .2000b. The WesternWayof War: InfantryBattlein ClassicalGreece, 2nd ed., Berkeley. Harrison,E. 1971. "The Victory of Kallimachos,"GRBS12, pp. 5-24. Hunt, P. 1997. "Helots at the Battle of Plataea,"Historia46, pp. 129-144. Jackson,A. H. 1991. "Hoplitesand the Gods:The Dedication of Captured Arms and Armour,"in Hoplites: The ClassicalGreekBattleExperience, V. D. Hanson, ed., London, pp.228-249. . 1992. "Armsand Armourat the PanhellenicSanctuaryof Poseidon at Isthmia,"in Proceedings of an InternationalSymposiumon the OlympicGames,W. Coulson and H. Kyrieleis,eds., Athens, pp. 141144. .2000. "Argos'Victory over Corinth,"ZPE 132, pp. 295-311. Jameson,M. H., D. R. Jordan,and R. D. Kotansky.1993. A lex sacra from Selinous, Durham.
Janssen,A. J. 1957. Het antieketropaion, Ghent. bei ThebenI = Kabirenheiligtum P.Wolters and G. Bruns,Das bei ThebenI, Berlin Kabirenheiligtum 1940. bei ThebenIV = Kabirenheiligtum K. BraunandT. E. Haevernick, BemalteKeramikund Glasausdem beiTheben,Berlin Kabirenheiligtum 1981. Karavites,P. 1992. Promise-Givingand Homerand theNear Treaty-Making: East, Leiden. Korres,M. 1994. "RecentDiscoveries on the Acropolis,"in Acropolis
Restoration: The CCAMInterventions,R. Economakis,ed., London, pp.174-179. Krentz,P. 1997. "The StrategicCulture of PericleanAthens,"in Polisand Polemos,C. D. Hamilton and P. Krentz,eds., Claremont,pp. 5572. .2000. "Deceptionin Archaic and ClassicalGreekWarfare,"in Warand Violencein ClassicalGreece, H. van Wees, ed., London, pp. 167200. Lammer,M. 1982-1983. "Der sogenannteOlympischeFriedein der griechischenAntike,"Stadion 8-9, pp. 47-83. Latacz,J. 1977. Kampfparanese, undKampfwirkKampfdarstellung, lichkeitin derIlias, beiKallinosund Tyrtaios,Munich. Lazenby,J. F. 1993. TheDefenceof Greece,490-479 B.C.,Warminster.
Lissarrague,F. 1990. L'autreguerrier: Archers, peltastes,cavaliers.Sur duguerrier,Paris. l'iconographie Lonis, R. 1979. Guerreet religionen Grecea l'epoqueclassique,Paris. Loraux,N. 1986. TheInventionof Athens,A. Sheridan,trans., Cambridge,Mass. Miller,M. C. 1997. Athensand Persiain the5th CenturyB.C.:AStudyin Cambridge. CulturalReceptivity, Mitchell, S. 1996. "HopliteWarfarein Ancient Greece,"in Battlein Antiquity,A. B. Lloyd, ed., London, pp. 87-105. ML = R. Meiggs and D. Lewis, eds., A Selectionof GreekHistorical Inscriptionsto theEnd of theFifth Century B.C.,2nd ed., Oxford 1988.
Muller,H. 1975. "O)uyr;Evexv," Chiron5, pp. 129-156. Myers,J. L. 1943. "axlpuxtoogiO6Xseo; (Herodotus,V.81),"CR 57, pp. 6667. Naiden, F. S. 2000. "GreekSupplication prior to 300 B.C.E." (diss.
HarvardUniversity). Ober,J. 1996. "The Rules of War in ClassicalGreece,"in TheAthenian Revolution: Essays on Ancient Greek
and PoliticalTheory, Democracy
FIGHTING
BY THE RULES
Princeton,pp. 53-71. Originally publishedin TheLaws of War. Constraintson Warfarein the Western World,M. Howard,G. J. Andreopoulos, and M. R. Shulman,eds., New Haven 1994, pp. 12-26,227230. Trans.byJ. Odin as "Lesregles de guerreen Grece classique,"in La guerreen Grecea l'epoqueclassique, P. Brule andJ. Oulhen, eds., Rennes 1999,pp.219-239. Parker,R. 1983. Miasma,Pollution,and Purificationin Early GreekReligion, Oxford. Plassart,A. 1913. "Les archers d'Athenes,"REG 26, pp. 151-213. Pritchett,W. K. 1971-1991. TheGreek Stateat WarI-V, Berkeley. Prost, F. 1999. "Les combattantsde Marathon:Id6ologie et societe hoplitiquesa Athenes au Ve s.,"in Armeeset societesde la Grececlassique, F. Prost, ed., Paris,pp. 69-88. Sage, M. M. 1996. Warfarein Ancient London. Greece.ASourcebook, Siewert,P. 1992. "The Olympic Rules," in Proceedings ofan International Symposiumon the OlympicGames, W. Coulson and H. Kyrieleis,eds., Athens, pp. 113-117. Sindos= )ZNv8o0: KaroiAoyoqriS ExOear;,Thessaloniki 1985. Snodgrass,A. M. 1993. "The 'Hoplite
39 Reform'Revisited,"Dialogues d'histoireancienne19, pp. 47-61. Spence, I. G. 1993. The Cavalryof ClassicalGreece,Oxford. Vanderpool,E. 1966. "AMonument to the Battle of Marathon,"Hesperia 35, pp. 93-106. van Wees, H. 1994. "The Homeric Way of War:The Iliad and the Hoplite Phalanx,"GaR41, pp. 118,131-155. .2000. "The Development of the Hoplite Phalanx:Iconography and Realityin the 7th Century,"in Warand Violencein ClassicalGreece, H. van Wees, ed., London, pp. 125166. Vernant,J.-P., ed. 1968. Problemesde la guerreen Greceancienne,Paris. Wace, A. J. B. 1905-1906. "Lakonia. Sparta:The Heroon,"BSA 12, pp.288-294. Wallace,P. 1969. "Psyttaleiaand the Trophiesof Salamis,"AJA73, pp.293-303. West, W. C. 1969. "TheTrophiesof the PersianWars,"CP 64, pp. 7-19. Wheeler, E. 1987. "Ephorusand the Prohibitionof Missiles,"TAPA117, pp. 157-182. Woelcke, K. 1911. "Beitragezur Geschichte des Tropaions,"BJb120, pp. 127-235.
Peter Krentz DAVIDSON COLLEGE DEPARTMENT
OF CLASSICS
P.O. BOX 6973 DAVIDSON,
NORTH
[email protected]
CAROLINA
28035-6973
HESPERIA
7I
(2002)
Pages 41-90
ATI
APHR
SYME
KATO POTTERY, CULT AND
AN CONTINUITY,
IN LATE
ARCHAIC
CLASSICAL
CRETE
AND
ABSTRACT The analysis of ceramicsfrom Aphrati sheds valuablenew light on the history of this Cretansettlementand on its relationshipwith a nearbyruralsanctuary at Kato Syme in the Late Archaic and Classical periods. It has long been held that Aphrati was deserted from ca. 600 to 400 B.C.A pottery deposit from the domestic quarter,however, now supports occupation of the city during this period. A ceramicclassificationsystem is presented and the morphological development and absolute chronology of severalkey shapes at Aphrati and Kato Syme areplotted. Historical implicationsof the ceramic evidence are also explored. To date, the most impressive finds of the historical Greek periods from Aphrati, the site of a polls in East Crete (Fig. 1), consist of grave goods from the Geometric and Orientalizing burialground excavatedby the Italians in the 1920s under the direction of Doro Levi.' On the basis of magnificent tomb furnishings, Levi inferred the existence of an important 7th-century settlement at Aphrati.With few exceptions,none of the material from Levi's excavations commands a date later than the 7th century. For this reason, Levi concluded that Aphrati suffered a major reversalafter ca. 600 B.c.2 Based on this conclusion, historians have presupposed a majordisruption of cult at the nearbyruralsanctuaryat Kato Syme commensurate with the presumed break in occupation at Aphrati. Didier Viviers, following the preliminary conclusions of the sanctuary'sexcavator, Angeliki Lebessi, notes a progressive diminution of offerings at the sanctuary in the course of the 6th century. Viviers interprets this phenomenon as the effect of a widespread economic decline of the area,with 1. Levi 1927-1929, p. 528. 2. Levi 1927-1929, p. 528, followed by Boardman1982, p. 227; Watrous 1982, p. 23; Huxley 1994, p. 128. The epigraphicalrecordat Aphratigives little indication of subsequentactivity in the 6th or following century,with
the exceptionof a 5th-centuryinscription, IC I v 4, which recordsa dedication made to Athena. Levi (1927-1929, pp. 22, 37) reportsfinding a stray Argive silvercoin in the courseof his excavations,dated by him to the third quarterof the 5th century.According
to Hoffmann (1972, pp. 42-44), the latest examplesof inscribedbronze armorfrom the Aphrati hoard,discussedbelow,date to ca. 600-575. Viviers (1994, pp. 241-244) briefly summarizesthe archaeologicaldiscoveriesat Aphrati.
0 F 1)
5 i
__ Figure 1. Map of Crete. After Talbert2000, map 60
I 5
15 10
10 : 15
20
APHRATI
AND
KATO
SYME
43
particularlysevere repercussionsfor the sanctuary.3From Viviers'perspective, the root cause lies in a reorganizationof the territoriesof Biennos and Dattalla in response to the growth of their powerful neighbor, Lyktos. Viviers attributes the supposed abandonment of Aphrati to Lyktian aggression. His reconstructionfurtherholds that the sanctuaryat Kato Syme fell out of use as a result of the political subjugation of Biennos and the absorption of its territoryby Lyktos ca. 600 B.C. When and how this process of territorial expansion came about, if indeed a "policy"of Lyktian aggression existed in the first place, remain matters of dispute.4The acknowledged difficulty of identifying and accurately dating the post-Minoan ceramics from Kato Syme has impeded previous efforts to assess the development of the sanctuaryin the Archaic and Classical periods, and to clarify the history of settlement in this part of Crete. Whether the lacunae in the archaeologicalrecord reflect historical reality or a problem of archaeological perception is open to question. Specialists working at individual sites on Crete have noted breaks in the ceramic sequence between ca. 600 and 400 B.C., but have made no effort to assemble the ceramic evidence from across the island to form a comprehensive picture of local or regional developments.5 Without established ceramic sequences, all other categories of archaeological evidence lack a chronological anchor.6 The present article is intended to establish a sequence of shapes and an absolute chronology for pottery at the sanctuary.The catalogue entries include all diagnostic examples from Kato Syme, and can therefore be regardedas a comprehensive publication of its Late Archaic and Classical pottery. My presentation of the finds from Kato Syme and Aphrati treats the pottery from both sites together in a chronological fashion. I propose a regional model of consumption according to which potters from Aphrati supplied the sanctuaryat Kato Syme with a continuous series of drinking cups of a characteristicshape and pale brown fabric between ca. 600 and 400. A domestic deposit from Aphrati provides an invaluable chronological point of reference for this shared local ceramic tradition. Contrary to the accepted picture of Cretan decline, this collection of pottery demonstrates continuity of activity at both the sanctuary and settlement. I further argue that around 400, Lyktos, a rival polis, replaced Aphrati as the 3. Regardingthe fate of the sanctuary in the 6th century,Viviers (1994, p. 256) concludes:"Le sanctuairen'estpourtant pas laisse a l'abandon."The traditional identificationof the site at Aphrati as ancientArkadeshas recentlybeen challenged by Viviers (1994, p. 257), who offers the alternativeidentificationof Dattalla. 4. Chaniotis (1996, pp. 3-5, 13-16) regardsthe formationof powerfulCretan city-stateswith sizable territoriesas a creationof the Hellenistic age. Van Effenterreand Bougrat(1969, pp. 36-37) propose that the expansionof Lyktian
territoryto the south coast of Crete took place at the end of the 3rd century.Viviers (1994, pp. 252-259), Watrous(1982, pp. 22-24), and Haggis (1996, pp. 415, 419, n. 117) arguethat the processbegan in the 6th century.Van Effenterreand Gondicas (1999, pp. 136-137) examine the evidencefor political dependencies of Lyktos. 5. Kanta's(1991, p. 500) summaryof the currentstate of knowledgeof the Cretan ceramicrecorddeservesrepeating: "the6th and 5th centuryB.C. pottery from Crete has not been publishedor studied and is virtuallyunknown."
Knossos is something of an exception;see below,n. 127. Morris (1998, pp. 66-68) discussesthe problemof pottery identification. 6. The resultspresentedhere stem from a more comprehensivestudy (Erickson 2000) of the island pottery styles ca. 600-400 B.C. I limit my discussion here to Aphrati and neighboringsites. While coins might be expectedto provide a reliablechronologicalanchor,most Cretan poleis did not begin minting until the second half of the 5th century,if not later; see Le Rider 1966, pp. 173-174.
44
BRICE
L. ERICKSON
chief supplier of pottery at the sanctuary.In the final section I address the historical implications of my study by focusing upon Cretan cult practice as a reflection of territorial conflict, arguing for an expropriation of the frontier sanctuaryby Lyktos ca. 400 B.C.,a full two centuries later than the date advocated by Viviers.
EXCAVATION
AND
CONTEXTS
APHRATI
Evidence in support of continued occupation at Aphrati comes in the form of a pottery deposit from a house in the domestic quarter of the ancient city excavatedby the Greek Archaeological Service (Fig. 1). Lebessi provides a plan of the building in the preliminary report of her excavations conducted in 1969.7 The house's destruction sealed a layer of debris over the remains of one of its floors. Among the debris was recovereda host of intact high-necked cups, representativesof a previously unknown Classical Cretan ceramic tradition. The discovery of these cups in association with datable lamps (85-88), discussed below, places the destruction of the house within the narrowchronological limits of ca. 425 and 400 B.C.The size of this deposit, its remarkablepreservation, and the presence of unusually good evidence for its date combine to make it a potential cornerstone of Cretan chronology in the 5th century. KATO SYME
Excavations conducted by the Greek Archaeological Service under the direction of Lebessi in 1972 discovered on the remote southern flanks of Mt. Dikte near the modern village of Kato Syme a small open-air sanctuary dedicated jointly in the Hellenistic period to Aphrodite and Hermes (Fig. 1).8This identification is based upon secure epigraphical sources in conjunction with the characterof the finds. The earliest signs of cult activity at Kato Syme date to the Middle Minoan IB period.9The sanctuary experienced its greatest popularity during the Late Minoan period, but the site retained its sacred characterin later historical times. Indeed, Kato Syme housed one of the longest-lived cults known in the ancient Aegean world. The finds document a period of uninterrupted activity across the normally turbulent period of transition from the Late Bronze to the Early Iron Age. The Late Archaic and Classicalperiodspresentthe last remaining obstacle to documenting full continuity at this Cretan cult center. The physical layout of the sanctuary in the historical Greek periods consisted of a series of terraces that followed the natural contours of the hill.These terraces,in turn,supportedan extensiveopen-air platformhigher up the hill, equipped with a central altarand a large hearth.The sanctuary also included subsidiary buildings of uncertain function. The main focus of cult concerned sacrificial offerings and the ritual consumption of food and drink.These activities have left the most substantialtraces in the
7. Lebessi (1973, p. 459, fig. 2, 402:a-b) pl. brieflydescribesthe contents of the deposit and illustrates two lamps and a proto-Hadrahydria. Preliminarystudy of the finds suggested to her (p. 460) a 5th-century
dateforthe deposit.Priorevidencefor betweenca.600 andthe occupation
late-5th-centurydomestic deposit is discussedbelow,n. 17. 8. The successivearchaeological campaigns,and their preliminary results,are summarizedby Lebessi (1985). 9. Watrous(2001, pp. 217-218) surveysthe earlyhistory of the sanctuary.
APHRATI
AND
KATO
SYME
45
material record in the form of enormous deposits of burnt earth mixed with broken pottery and sacrificialdebris. Stratified Late Minoan deposits were found at the deepest levels. The thinner strata above consisted of disturbed fill, an indiscriminate mixture of Minoan pottery with an assortment of Geometric, Archaic, and later material. A lamentable lack of "clean"deposits of homogeneous fill from the historical periods poses a special problem for the study of the post-Minoan phases of the sanctuary. Among the most spectacularfinds from the sanctuaryin the historical periods are a large collection of Early Iron Age zoomorphic bronze figurines and a rich series of Orientalizing bronze cutout plaques.The bronze cutout plaques have generated considerable interest among art historians because of their fine execution, lively style, and range of subject matter. They depict the principal deities of the sanctuaryor human worshippers bearing animals intended for sacrifice. A 7th-century date has been assigned to the majority of these Daedalic plaques on stylistic grounds. Production is thought to have subsided in the first quarterof the 6th century. For the remainder of the 6th century, the archaeological record indicates an almost total absence of metal artifacts.Fifth-century bronzes from the sanctuaryremain equally elusive.10
THE POTTERY The magnificence of these metal objects from the sanctuaryhas overshadowed study and publication of the pottery. This is unfortunate, since the ceramic evidence has the potential to correct the impression of declining activity or even abandonment of the sanctuarybetween 600 and 400 B.c., an impression generated principally by the ebb and flow of metal offerings. My study of the black-gloss pottery from Kato Syme is the first effort of its kind to recognize local ceramics of Late Archaic and Classical date, either from the sanctuaryor from this region of Crete.1'The pottery from Kato Syme, if the equation with a production center at Aphrati is correct, affords a substantially fuller glimpse of these periods of ceramic development than that provided by any discoveries made until now at the presumed center of production at Aphrati. 10. For the figurinesfrom the sanctuary,see Schiirmann1996. For the cutout plaques,see Lebessi 1985. PublishedLate Archaic and Classical votives from the sanctuaryinclude a 6th-centuryinscribedbronze handle (Lebessi 1975b, p. 191, pl. 193:c);a 6th-centurybronze figurineof a goat (Lebessi 1977, p. 325, pl. 256:b); a bronze figurineof a youth with an inscribedbase dated to the second quarterof the 5th century(Lebessi 1990, p. 276, pls. 133:b, 134:a);a 5th-
centurybronzeplaque(Lebessi1974, p. 197, pl. 185:a);and a 5th-century terracottaplaque(Lebessi 1977, p. 325, pl. 257:c). 11. The pottery from Kato Syme is housed at the HerakleionArchaeological Museum. My study concentrates upon a representativesample.The excavationsof 1972 and 1973 provedto be the most fruitfulyearsfor the study of the post-Minoan periods,since these initial campaignsdealt with the removal of the upperlevels of sacrificial
debris.These upperlevels containeda mixtureof Minoan and laterpottery.I examinedall savedfine ware recovered duringthese two yearsof excavation.In addition,I selectivelysurveyedpottery from subsequentyearsof excavation, concentratingon boxes of previously sortedpost-Minoan pottery.I was fortunate to have AthanasiaKantaas my guide in the storeroomsof the Herakleion Museum, increasingmy debt to her immeasurably.
BRICE
46 FABRIC
ANALYSIS
AND
L. ERICKSON
PROVENIENCE
Pottery specialists working on Crete have noted that ceramics vary considerably even between neighboring sites.l2 Consideration of fabric texture, hardness, inclusions, slip, gloss, and other variablesassists archaeologists in determining fabricand ware groups.The consistency of fabriccolor in an assemblage is also relevant in attempting to differentiate fabrics and attribute undecorated pottery to its source.13The color of the fabric may vary as a result of the utilization of different clay sources, differences in the preparationand the firing of the clay, or a combination of these factors.l4 Furthermore,it is widely recognized that variations in fabric color may be due to postdepositional conditions unrelated to differences in the original material or variation in firing. For these reasons, the value of color as a means of characterizing fabrics has occasionally been questioned. The emphasis here upon color as a means of distinguishing Cretan fabrics and proposed production zones therefore requires explanation. As part of my program of dissertation research,I had the opportunity to examine large quantities of unpublished Late Archaic and Classical pottery from sites all over Crete, enabling me to identify ten regional production centers. In my estimation, fabric color varies considerablyfrom one site to the next, even, somewhat surprisingly,in the case of neighboring sites. Personal experience has led me to consider color, alongside fabric texture, hardness, slip, and gloss, as an important factor in differentiating Cretan fabrics on a macroscopic level. My hypothesized ware groups and source attributions, however, will need to be tested against evidence from future fieldwork, ceramic petrography,and surveys of clay sources. The characteristicfabric of the Late Archaic and Classical pottery at Kato Syme consists of a fine clay of a somewhat chalky consistency fired to a very pale brown color (Munsell 10YR 8/3).15The resemblance of this fabric to Corinthian fabric is often close. A poor dull gloss, black or occasionally streakybrown and prone to flaking, is another distinguishing feature of this group. The pottery exhibits a remarkabledegree of internal consistency with respect to fabric texture,hardness, color, and gloss. There arereasons for thinking that the pottery used at Kato Syme was not manufactured at the sanctuary.Local manufactureof pottery seems somewhat unlikely, given the remote location of the sanctuary and the absence of excavated kilns, kiln wasters, or other signs of ceramic production such as pottery wheels. While negative evidence of this kind is hardly decisiveand thus the sanctuary remains a possible candidate for productionstylistic considerations discussed below suggest another possibility. A more likely scenario posits a production center at a nearby settlement as the supplier of pottery to the sanctuary.Aphrati and Pyrgos stand out as the two most promising candidates for the place of manufacture on the basis of reported fabric color and consistency.l6 Both sites lie in close proximity to Kato Syme (Fig. 1), which suggests that a pale brown fabric was characteristicof the region. Fabric color thus has its limits in revealing exact provenance, even under apparently favorable conditions. Of the two suggested places of manufacture,Aphrati seems the more likely. While the Graeco-Roman settlement at Pyrgos remains unexcavated and its ceramic tradition poorly understood, at Aphrati a discernible local tradition exists which, although once thought to have consisted solely of
12. See Coldstream1973b, pp. 4647.
arisein tryingto 13. Complications determinethe exactplaceof manufactureon thebasisof fabricalone. fabricneednot Recognitionof a "local" imply the exclusivemanufactureof pottery at a single site. Arnold (1985, pp. 58-59) argueson the basis of ethnographicevidence that neighboring communitiesas much as 12 to 18 km apartmight exploit overlappingresourceareas,including common clay beds, therebyincreasingthe likelihood of a single regionalfabric. 14. Sinopoli 1991, pp. 12-13; Orton, Tyers,and Vince 1993, pp. 132138. 15. This fabriccontainsvery few inclusionsvisible to the nakedeye. It is relativelysoft and powdery.The color occasionallyborderson reddishyellow (Munsell 7.5YR 7/6). 16. Eiring (2000, p. 54) characterizes the Hellenistic fabricof Pyrgos as "soft,powdery,and very pale."Another obvious candidateon the basis of geographicproximityto the sanctuaryis the ancientpolis Biennos.The standard pale brown fabricat Kato Syme seems to be identicalin everyrespectto the predominatefabricof a seriesof remarkablepolychromevessels from Late Orientalizingtombs at Aphrati;see Levi 1927-1929, pp. 530-532.
APHRATI
AND
KATO SYME
47
the material from the Geometric and Orientalizing tombs, is now understood to have included a thin representation of 6th-century residual material.This material derives from a building southeast of the acropolis excavatedby Lebessi.7 In addition, the later domestic deposit from Aphrati, mentioned above, documents production of high-necked drinking cups in the final quarter of the 5th century.These cups exhibit an identical form and fabric to those of contemporary pottery from Kato Syme. Peculiarities of this style, most prominently the penannularring underfoot, arefound in the local fabric at virtually no other Cretan site. The availableevidence points to Aphrati as the source of the pale brown pottery at Kato Syme, although naturally Pyrgos cannot be wholly excluded as an alternativesource ofvotives. Moreover, the possibility cannot be ruled out that Biennos or another as yet unexcavated site will produce pottery of closely similar style to that of the sanctuary.A decisive conclusion must await further exploration of the area and greater understanding of the local pottery styles. Aphrati must have been the site of an important Classical polis, alternatively identified either as Arkades or Dattalla.l8I suggest, on the grounds of a common fabric and the exact correspondenceof forms, that Aphrati is likely to have been the supplierof pottery for the nearbysanctuarythroughout the Late Archaic and Early Classical periods. A settlement with a long tradition of pottery manufactureis a more probable source of production than a remote mountain sanctuary.If it is the case that pottery production took place at Kato Syme, then the observed similaritybetween the sanctuary style and that of the settlement would raise the possibility that potters either from Aphrati or trained in the local school worked at the sanctuary. The model of consumption advanced here remains tentative pending future efforts to define clay and inclusion sources, kiln sites, and spatial and temporal distinctions. It is even conceivable, however unlikely, that Kato Syme supplied pottery to the settlement. Around 400, or shortly thereafter(the chronology is discussed below), the predominant pale brown fabric at the sanctuary abruptly gave way to a new one, of an entirely different character and presumably a different origin. Replacing the pale brown fabric at the sanctuaryis one with silver mica and a darkreddish-brown hue (Munsell 2.5YR 5/6 to 5YR 5/4). The primarydistinguishing feature of this new fabric is its extreme coarseness, which makes it a rarity among the Classical Cretan fabrics employed in the production of drinking cups and other fine-ware shapes.19As a general rule, gritty fabric is more typical of cooking and coarse wares. Evidently 17. In the preliminaryreportsof her excavations,Lebessi (1970, p. 416, fig. 1, pls. 425-426; 1973, pp. 457,459, figs. 1-2) describesthe discoveryof a large quadrangularbuilding southeastof the acropolishill, a building that she interpretsas a shrine.The reportsmention volumes of Geometric and Orientalizing finds along with a scatterof 6thcenturymaterial.I have not seen the pottery from this building.In addition, a tomb of presumed6th-centurydate
was discoverednearAphrati at Orthi Petra;see Lebessi 1983. 18. See above,n. 3. 19. A multiplicityof fabricsat a single site can be interpretedin various ways.A varietyof fabricsis a conceivable outcome of a productioncenter whose output included a rangeof coarse,cooking, and fine-wareshapes; see Stissi 1999, p. 87. Plog (1980, pp. 86-88) examinesethnographic evidencein supportof the conclusion
that potters choose differentclays or temperingmaterialsin the manufacture of vessels intended for differentpurposes. Yet at Kato Syme there is no indicationthat functionalcategoriesof potteryplayed a determiningrole in the characterof the fabric.Gritty reddishbrownfabricsuddenlybecame the preferredfabricfor the whole range of Cretan shapes at the sanctuary,as typical of large coarsebasins as of the finest drinkingcups.
48
BRICE
L. ERICKSON
this new fabric was incapable of taking a fine colored slip. A fabric of apparently the same characteristics distinguishes Lyktos from all other Cretan production centers (Fig. 1). A coarse reddish-brown fabric with silver mica may well be a signature trait of Lyktian fine-ware production during the Classical and Early Hellenistic periods. The site of Lyktos remains largely unexplored, although a series of campaigns undertaken by the Greek Archaeological Service has brought to light Hellenistic phases of occupation.20The material from these excavations derives from a destruction deposit dated to the 3rd century B.C., a find plausibly connected with a documented destruction of Lyktos at the conclusion of the Lyktian War in 221. After examining this material, I concluded that the 4th-century pottery from Kato Syme issues from the same source as the later Lyktian wares.21Yet this identification cannot be confirmed by exact correspondence between ceramic forms from one site to the next, since the ceramic record at Lyktos remains a blank during much of the 6th, 5th, and 4th centuries. Lyktos provides yet another example of the need to clarify the local fabrics and pottery styles of the Classical Cretan production centers. While the vast majority of the Late Archaic and Classical pottery from Kato Syme falls neatly under the general rubric of either pale brown or gritty reddish-brown fabric, notable exceptions occur, as in the case of a small cup characterizedby a hemispherical bowl and a short upright rim (110), a common Cretan type that can be dated to ca. 500-480.22 Another exception is a cup base with a splayed pedestal stand (109), whose proposed date of ca. 425-400 rests upon general stylistic grounds and specific parallelswith cups from Gortyn. The fine pale red fabric (Munsell 2.5YR 6/8) seen in both examples suggests that they are Gortynian imports.23 The cup base is coated with a lustrous black gloss of uncommonly high quality, another characteristic of Gortynian production in the Classical period. SIXTH-CENTURY OINOCHOE
POTTERY
(1-27)
(1)
My presentation of the finds from Aphrati and Kato Syme follows a chro-
nological arrangement (proceeding from closed to open shapes, large to small, within each period). I postpone a detailed discussion of the contents of the 5th-century domestic deposit at Aphrati until the appropriateplace in the chronological sequence. An oinochoe from this deposit (Fig. 2) is treated here because it suggests a 6th-century date on the basis of style. This high-necked, broad-bottomed oinochoe is decoratedwith a separate tongue pattern on the shoulder and bottom. A 7th-century Corinthian prototype lies behind the production of this unusual Cretan shape. Aphrati seems to have specialized in copies of Corinthian jugs, to judge from numerous examples of local manufacturefrom the 7th-century cemetery.24The oinochoe from the domestic deposit differs from these earlierexamples in two respects. First, the decoration exhibits an apparent simplification of the 7th-century scheme. Most notable in this respect is
20. See Platon 1952, p. 480; 1957, Alexiou 1969, p. 539. Lebessi 336; p. (1975a, pp. 494-496, pls. 512-513) reportsthe discoveryof Archaicpithoi in a Hellenistic destructionlevel.The excavatednecropolisof Lyktos has yielded local and Attic pottery dated to the second half of the 4th century;see Lebessi 1980, p. 886, pl. 661. For imported Hellenistic black-glosspottery from Lyktos,see Englezou 2000. 21. I extend my profoundthanks to Maria Englezou of the Herakleion ArchaeologicalMuseum for permitting me to examinethe Hellenistic material from Lyktos.She agreeswith my identificationof Lyktos as the probable sourceof the 4th-centurypotteryfrom Kato Syme. Here, reassuringly,we have the publishedobservationsof another pair of scholarswho independently reachedthe same conclusionthat gritty red fabricis a hallmarkof Lyktian fine-wareproduction.Regardingthe fabricof the pottery from the Hellenistic context at Lyktos,Callaghanand Jones (1985, pp. 14-15) remark,"the provincialityof its local ceramicindustryis quite clear.The vast majority of its 'black-glazed'shapesare made of a red gritty clay and are coveredwith a thin dull wash. Some shapes,which would have been glazed at Knossos,are left plain." 22. Unpublishedexamplesfrom Knossos,deposit RR:H (see Coldstream1973b), providea date of ca. 500-480 for the type. 23. My reasonsfor ascribingthis light red fabricto a separateCretan productioncenter at Gortyn arediscussed more fully in Erickson,in press. 24. Levi (1927-1929, p. 220, fig. 247) illustratesan examplefrom his excavationsof the Orientalizing cemeteryat Aphrati.
APHRATI
AND
KATO
49
SYME
Figure 2. Oinochoe. Scaleca. 1:2
the tongue pattern, a secondary motif of 7th-century potters, which now occupies the entire shoulder of the oinochoe. In contrast, 7th-century pot paintersfrom Aphrati favornarrowzones of decorationcomposed of smaller elements. Second, the base or stand indicated by the broken protrusion at the bottom of the jug probably reflects a 6th-century development, since this feature is absent from 7th-century jugs. The lack of anything resembling a seriesof 7th- or 6th-centuryjugs of this type complicatesany attempt to date this vessel. All in all, the differences both in form and decoration between this oinochoe and its 7th-century predecessors are relatively minor, so I am reluctant to accept a date for it later than the first half of the 6th century. 1
Oinochoe,body Fig. 2 Aphrati.1969.A5.High-necked, broad-bottomedoinochoe.Fragmentaryneck andbase.Diam. (max.) CUPS,
25. SeeBoardman andHayes1973, pp. 37-38, no. 2104, fig. 16, pl. 20, ca. 575-550 B.C.
26. See Boardmanand Hayes 1966, 78. p. 27. I owe a debt of gratitudeto Nikolaos Stampolidis,the directorof the excavationsof the Universityof Crete at Orthi Petra,for inviting me to study the unpublishedLate Archaic and Classicalfinds from his excavations (Eleutherna,Tomeas III). The 6thcenturydate for the parallelis based upon securestratifieddepositswhose contents include datableimported Laconianpottery.
DEEP
AND
SHALLOW
9.4 cm. Palebrown(10YR8/3). Decoratedwith a separatetongue patternalongthe shoulderandbottom. Firsthalf of the 6th century(?). VARIETIES
(2-8)
The 6th-century ceramic repertoire at the sanctuary consists chiefly of cups, of which three types predominate.The first two types (Fig. 3) signal the strength of an independent local tradition, while the third category manifests direct borrowing from a mainland Greek source.The first type of cup is composed of a broad bowl with a wide opening at the mouth, whose short inset rim either flaresor is outturned (2-6). The convex shoulder of preserved examples recommends reconstructing the cup as a deep shape. A second variety of cup, of equal ancestry at the sanctuary,exhibits a different profile defined by a narrowershoulder, depressed bodily proportions, and a hemispherical bowl (7-8). This cup is likewise equipped with a short outturned rim. The dating of these two types of cup, both the deep and the shallow varieties, depends upon parallelswith material from better-dated contexts elsewhere on Crete, at Knossos and Eleutherna, or from the Greek overseas colony atTocra, Libya. One-handled Cretan mugs from Tocra exhibit a similar profile to the deeper of the two sanctuary shapes.25Since a reportedly pale brown fabric is characteristic of the exports from Tocra, a shared source with the supplier of the sanctuaryat Kato Syme is a distinct possibility.26A cup of similar conception to the deeper variety from Kato Syme was produced locally at Eleutherna in the 6th century,27and a
BRICE
50
L. ERICKSON
similar type of cup appearsat Knossos by the end of the 6th century.28The evidence suggests a progression in the 6th century from a cup with a high flaring rim to a short one, either everted or upright, with hemispherical bodily proportions eventually giving way to a deeper, more voluminous form. It seems clear,judging from the streamlined proportions of a descendant of the cup at Kato Syme (6), that this class felt a later influence from the repertoireof the Classical Knossian low-necked cup or another similarsourceon its development.29The second type of cup at Kato Symethe shallow variety with a hemispherical bowl (7, 8)-is datable by reference to cups in late-7th- or early-6th-century deposits from Eleutherna.
(
2 3
4
5 )
)
6
)
( I
(
7
8
i
)
Figure3. Cup rims.Scale1:3
2
Deep cup,rim and shoulder
Fig. 3
KatoSyme(1977). Level 1, OM87. Diam. (rim)13.1 cm. Pale brown(10YR8/3). Blackgloss.6th century. 3
Deep cup,rim and shoulder
Fig. 3
6
Deep cup,rim and shoulder
Fig. 3
KatoSyme(1972).Level2, OM5. Diam. (rim)11.0 cm. Pale brown(10YR8/3). Blackgloss.5th century. 7
Hemisphericalcup,rim andshoulder
Fig. 3
KatoSyme(1972).Level2, OM5. Diam. (rim)13.0 cm. Pale brown(10YR8/3). Blackgloss.6th century.
KatoSyme(1972).Level4, OM67. Diam. (rim)11.4 cm. Pale brown(10YR8/3). Blackgloss.6th century.
4
8
Deep cup,rim and shoulder
Fig. 3
KatoSyme(1972).Level 11, OM21. Diam. (rim)11.6 cm. Pale brown(10YR8/3). Blackgloss.6th century. 5
Deep cup,rim and shoulder
Fig. 3
KatoSyme(1977). Level2, OM35. Diam. (rim)11.2 cm. Pale brown(10YR8/3). Blackgloss.6th century.
Hemisphericalcup,rim and shoulder
Fig. 3
KatoSyme(1972).Level6, OM17. Diam 11.3 cm, of rim 11.1 cm. Palebrown(10YR8/3). Black gloss.6th century.
28.The Knossiancupcomesfrom anunpublished Classicalwelldeposit fromthe Stratigraphical Museum Excavations(J/JN7.4987). On typological grounds,it ranksamong the earliestpottery from the well. I am gratefulto PeterWarrenfor permission to mention unpublishedfinds from his recentexcavationsand for giving so generouslyof his time. 29. The "Shrineof Glaukos"at Knossosprovidesthe closest parallels for this shape;see Callaghan1978, pp. 6-7, nos. 10-12, fig. 5, pl. 2.
APHRATI
AND
KATO
5I
SYME
CUPS, KOTYLAI (9-18) The third majorvariety of drinking cup at the sanctuaryin the 6th century is a deep shape whose walls define a simple continuous curve that runs from base to rim (9). This local cup is closely modeled upon the form of a Corinthian drinking cup, the kotyle. An apparentMiddle Corinthian prototype helps pinpoint the date of an intact example to ca. 575-550.30 The base of this local cup has a distinctive profile underfoot formed by a vertical inset underneath where the flaring stand meets the bottom wall of the base. This unusual treatment of the foot results in a mirror image of the standard 6th-century "stepped"profile underfoot (see below). A large group of identical bases (Figs. 4-5) indicates extensive production of the cup in the middle and second half of the 6th century, provided that the date of the fragments lies near that of the intact cup (10-18). These bases show little uniformity in size, a factor that suggests a field of production encompassing cups, large and small, and jugs. This type of base is encountered elsewhere on Crete in the local fabric of Praisos.31 One of the bases from Kato Syme (18) bears painted decoration in the form of a row of dots in a reserved band on the outside edge of the foot. The rarity of painted decoration, figural or otherwise, in the Cretan repertoire of 600-400 attracts attention to this particular example. Its decoration represents an apparent simplification of the pictorial tradition of the Orientalizing Cretan pot painters, whose other 6th-century works include Cretan exports at Tocra and Cyrene decorated with simplified spiraland floral ornaments.32Since Aphrati was the seat of a major Orientalizing vase-painting tradition, the existence of 6th-century painted decoration at Kato Syme, even in a debased form, strengthens the proposed connection between Aphrati and the sanctuary. Figs.4-5 Cup KatoSyme(1974). Level 11, orth.Delta, no. 21107. H. 8.2 cm; Diam. 13.0 cm, of rim 12.8 cm, of base6.7 cm. Palebrown(7.5YR5/4 to 10YR6/3). Blackgloss.Close copy of Middle Corinthiankotyle. Ca. 575-550. 9
30. Forthe Corinthianshape,see Payne 1931, p. 241. Amyx (1988, pp. 48-67) discussesabsolutedating
of MiddleCorinthian pottery.
31. The base from the Praisos survey(1994, Site 506.3) remains
Whitley,Prent,and unpublished. Thorne(1999)providea preliminary reportof the surveyresults.My thanks go to JamesWhitley for permissionto mention this piece. 32. See Boardmanand Hayes 1966, pp. 77-80, nos. 921-931, pls. 55-56; 1973, pp. 36-38, nos. 2101-2108, pl. 20. The same Cretansourcehas been posited for the exported6thcenturyCretanpottery from Cyrene, Libya;see Schaus 1985, p. 10.
10 Cup,base Fig. 4 KatoSyme(1972). Level 15, OM19. Diam. (base)7.1 cm. Pale brown(10YR8/3). Blackgloss. Middle or secondhalf of the 6th century. 11 Cup,base Fig. 4 KatoSyme(1973). Level5, OM30D. Diam. (base)7.1 cm. Pale brown(10YR8/3). Blackgloss. Middle or secondhalf of the 6th century.
Fig. 4 Cup,base KatoSyme(1973). Level 1, OM167. Diam. (base)6.1 cm. Pale brown(10YR8/3). Blackgloss. Middle or secondhalf of the 6th century. 12
13 Cup,base Fig. 4 KatoSyme(1974). Level5, OM33. Diam. (base) 6.8 cm. Pale
brown(10YR8/3). Blackgloss. Middle or second half of the 6th century.
14 Cup,base Fig. 4 KatoSyme(1975). Level 12, OM50. Diam. (base)7.5 cm. Pale brown(10YR8/3). Blackgloss. Middle or secondhalf of the 6th century.
L. ERICKSON
BRICE
52
I
10w
11
b
I(I
(
13
b(
14
'
(
g15
18
17 Figure 4. Kotylai. Scale 1:3
18
15 Figure 5. Kotylai
15 Cup, base
Figs. 4-5
Kato Syme (1973). Level 7, orth. Epsilon. Diam. (base) 6.0 cm. Pale brown (10YR 8/3). Black gloss. Middle or second half of the 6th century. 16 Cup, base Kato Syme (1974). Level 11, orth. Delta. Diam. (base) 7.1 cm. Pale brown (10YR 8/3). Black gloss. Middle or second half of the 6th century. 17 Jug, base
Fig. 4
Kato Syme (1973). Level 6, OM110. Diam. (base) 12.5 cm. Pale
brown (10YR 8/3). Black gloss. Middle or second half of the 6th century. 18 Jug, base and lower body
Figs. 4-5
Kato Syme (1975). Level 3, OM-AB. Diam. (base) 12.1 cm. Pale brown (10YR 8/3). Black gloss. Decorated with dots along reserved outer edge. Middle or second half of the 6th century.
APHRATI
CUPS,
AND
BASE
KATO
53
SYME
WITH
"STEPPED"
PROFILE
UNDERFOOT (19-27) A related type of 6th-century base, mentioned in passing above, consists of a disk foot with a "stepped"profile underfoot formed by a vertical inset where the root of the flaring foot meets a projecting concentric disk, or "medallion,"below (Figs. 6-7). While the stepped foot boasts of a long tradition on Crete beginning in the 7th century, an articulated disk foot with a pronounced edge and a stepped profile underfoot is a combination of features that suggests a more advanced stage in the development of the Cretan cup base. At Knossos, P. J. Callaghan has tentatively dated this type to ca. 525-500 B.C.33On present evidence, it is not possible to distinguish Orientalizing cup bases of this type from possible 6th-century descendants manufacturedbefore ca. 525, assuming that production of the base continued during the first half of the 6th century. While itself of uncertain date, the Knossian parallel provides an approximate guide for the dating of the local base at Kato Syme. Unfortunately, no intact example of the cup or different types of cup to which these bases belonged survives, either at Knossos or elsewhere on Crete. Even so, mainland inspiration in the cup's design can be inferred from the similarity of these bases and the related type from Kato Syme of demonstrable Corinthian origin (9), discussed above.
192 20
6
I
NM (I ( _a
I
(
19 Cup,base Fi|g. 6 KatoSyme(1972). Level 8, OM46. Diam. (base)5.0 cm. Pale brown(10YR8/3). Blackgloss. Secondhalf of the 6th century. 20 Cup,base Filg. 6 KatoSyme (1972).Level 15, OM19. Diam. (base)7.0 cm. Pale brown(10YR8/3). Blackgloss. Secondhalf of the 6th century.
33. Callaghan1992, p. 92.
256 26
lp
27
Figure 6. Cup bases. Scale1:3
21 Cup,base Figs. 6-7 KatoSyme (1974). Level 10, OM42. Diam. (base)6.7 cm. Pale
('
1
24
21 22
I
23
I I
brown(10YR8/3). Blackgloss. Secondhalfof the 6th century. 22 Cup,base Fig. 6 KatoSyme (1973).Level3, orth. 1, OM201. Diam. (base)7.4 cm. Pale brown(10YR 8/3). Blackgloss. Secondhalf of the 6th century. 23 Cup,base
Figs. 6-7 KatoSyme(1977). Level 10, OM231. Diam. (base)6.4 cm. Pale brown(10YR8/3). Blackgloss. Reservedbandon outerface.Second half of the 6th century.
54
BRICE
L. ERICKSON
25
24 26
Figure7. Cupbases
24 Cup,base Figs. 6-7 KatoSyme(1972).Level 18, OM107. Diam. (base)7.7 cm. Pale brown(10YR8/3). Blackgloss. Secondhalfof the 6th century.
26 Cup,base Figs. 6-7 KatoSyme (1973).Level2, OM16. Diam. (base)6.5 cm. Pale brown(10YR8/3). Blackgloss. Secondhalf of the 6th century.
25 Cup,base Figs. 6-7 KatoSyme(1973). Levrel5, OM129. Diam. (base)6.3 cm. Pale brown(10YR8/3). Blackgloss. Secondhalf of the 6th centuiry.
27 Cup,base Fitg. 6 KatoSyme(1973). Level 8, OM13. Diam. (base)7.2 cm. Pale brown(10YR8/3). Blackgloss.
FIFTH-CENTURY
POTTERY
Second half of the 6th century.
(28-81)
Knowledge of the 5th-century ceramic record at Kato Syme relies to a great extent upon the fragmentaryevidence of cup andjug bases.With the exception of a single group of intact Classical high-necked cups from the domestic deposit at Aphrati, fragmentarybases from Kato Syme form the sole basis for plotting the morphological development of a shared local sequence during the 5th century. Fortunately,connections with Knossos express themselves in clear ways in the local repertoire,a factor that facilitates a chronological arrangementof the material and helps in establishing absolute dates for the sequence. This method of dating assumes that Crete possessed the means of communication necessary to ensure the swift and steady promulgation of new styles, at least among major production centers. Whether innovations were disseminated first from Knossos, Aphrati, or some other prominent center is an open question. The influence of new styles may have taken longer to reach peripheralareas.Distance and geography presumably impeded the diffusion of new ceramic styles into remote areas, or in extreme cases developments may have completely bypassed conservativecommunities.Against these uncertainties,the domestic deposit from Aphrati provides independent chronological evidence for the dating of the local sequence around 400. JUGS (28-38) Some local bases of a distinctly different form and uniformly larger size than the remaining examples of cup bases from the sanctuary deserve
APHRATI
AND
KATO
SYME
55
special comment (Figs. 8-9). This category of base ranks among the most distinctive creationsof the indigenous potters.No exact parallelsareknown to me, although a comparableseries of large bases has been recoveredfrom surface survey in the neighboring Vrokastroand Gournia regions, east of Kato Syme. The fabricis intrinsicallydifferent from that of the Kato Syme bases, however, suggesting a separateproduction center.34 The large bases from Kato Syme have a uniform diameter of approximately 9 cm, representing an increase in size of roughly 50 percent over the standard cup varieties. A flaring foot with a pedestal stand forms the usual means of support. These larger bases probably belong to a series of generously proportioned cups or jugs, although the absence of intact examples at Kato Syme makes an exact determination of the shape impossible.35If their identification as the supports of large cups is valid, a simple deep shape, along the lines of an Attic skyphos or Corinthian kotyle, is a preferablereconstruction given the preserved curvatureof the lower body. It should be noted, however, that an intact trefoil-mouthed jug from a 5th-century deposit at Aphrati (37) possesses a base of similar size and profile to one of the fragmentaryexamples from the sanctuary(34).36This comparison suggests that at least one of the bases from Kato Syme belonged to a jug. The difficulty of determining a clear line of development for this series of bases is matched by the problem of ascertainingthe absolute chronology of the sequence. With no independently dated archaeologicalcontexts to shed light on the matter,and few exact parallelswith other Cretan bases to serve as a guide, the dating remains tentative and subject to revision pending future discoveries.What may be the earliest examples of this type (e.g., 28) share a feature in common with a group of standard cup bases manufacturedin the middle and second half of the 6th century under heavy Corinthian influence (9-18). Both types exhibit a distinctive profile underfoot formed by a sharp vertical inset where the root of the flaring foot meets the bottom wall of the base. This peculiar treatment, discussed in greater detail above, creates a reverse impression of the standard 6th-century stepped profile. If this comparison is valid, it suggests that the oversized cup or jug was first conceived in the middle of the 6th century as a larger offshoot of the main group.37 34. Cretanblack-glosspottery from surveysof the Vrokastroand Gournia regions,an areaof Crete encompassing the ancient territoriesof Hierapytna, Istron, and Oleros, remainsunpublished. Thanks to the kindnessof the directorsof these projects,Barbara Hayden and L. V. Watrousrespectively, I was grantedpermissionto examine this body of material.Preliminary reportsof the VrokastroSurveyappear in Hayden, Moody, and Rackham 1992; Hayden 1997. Watrousand Blitzer (1995) discussthe preliminary
resultsof the Gourniasurvey. 35. None of the largebases from Kato Syme preservesgloss on the interior.This would suggest a reconstructionof these bases as the supports ofjugs, althoughmost examplesare probablytoo heavilyworn to preserve gloss. 36. The best parallelsfor this jug come from the Orientalizingcemetery at Aphrati;see Levi 1927-1929, pp. 190, 406, figs. 210, 525. A splaying pedestalbase distinguishesthe 5thcenturyjug from earlierexamples.
Lebessi (1983, p. 342) and Orlandos (1976, pp. 196-197) reportthe discoveryof a trefoil-mouthedjug from a 6th-centurytomb in the areaof Aphrati.It suggestslonger-livedlocal productionof the shape than otherwise indicated. 37. The conclusion that the "oversized"cup or jug began as a variantof the standardgroup is hardlysurprising, since examplesof the "standard" group in realityshow little uniformityin size, the maximumdiameterof the base rangingfrom approximately6 to 13 cm.
BRICE
56
L. ERICKSON
The main developmental tendency of the base in the second half of the 6th and the 5th century involves a proposed shift toward a less broad foot, with a higher stand and greater extension of the outer edge. One jug base (32) bears a resemblance to the bases of skyphoi produced in Attic workshops ca. 500 B.C.; this comparison offers welcome confirmation of a Late Archaic date, inferred on the basis of the vessel's position within an internalstylisticsequence.38Around the middle of the 5th century,or shortly thereafter,the local potters achieved a fuller integration of the base and stand by dispensing with the "stepped"profile underfoot and equipping the base with a new wedge-shaped stand that makes a less abrupt transition with the belly (33-35). Influence from the local repertoireof standard cup bases may be responsible for these changes. Finally, toward the end of the 5th century, a base less broad with a substantially higher foot, fully pedestal in form, appears (38).
K(
2829
A
I
30
33)34
<
\
5
35
32. 38
Figure8.Jug bases.Scale1:3
-
28 Jug,base Fig..8 KatoSyme(1977).Level 10, OM231. Diam. (base)9.1 cm. Pale brown(10YR8/3). Blackgloss. Ca. 550-500.
32 Jug,base Fig. 8 KatoSyme(1974).Level2, OM50. Diam. (base)8.5 cm. Pale brown(10YR8/3). Blackgloss. Ca. 500-475.
Fig. 8 29 Jug,base KatoSyme(1974). Level3, OM51. Diam. (base)9.7 cm. Pale brown(10YR8/3). Blackgloss. Ca. 525-500.
33 Jug,base Fig. 8 KatoSyme(1974).Level5, orth. Omicron,OM4. Diam. (base)9.3 cm. Palebrown(10YR8/3). Blackgloss. Ca. 500-475.
30 Jug, base
Fig. 8
34 Jug,base
KatoSyme(1974).Level5, OM4. Diam. (base)9.5 cm. Pale brown(10YR8/3). Blackgloss. Ca.525-500.
Kato Syme (1973). Level 7, orth. Epsilon. Diam. (base) 10.0 cm. Pale brown (10YR 8/3). Black gloss. Ca. 475-450.
31 Jug,base KatoSyme(1975). Level3, OM124. Diam. (base)10.2 cm. Pale
35 Jug, base
brown (10YR 8/3). Black gloss. Ca. 500-475.
Fig. 8
KatoSyme(1974). Level 8, orth. Omicron,OM12. Diam. (base)9.0 cm. Palebrown(10YR8/3). Black gloss. Ca. 475-425.
38. As AgoraXII, p. 257, no. 311, fig. 4, pl. 14. An exportfound at Itanos (1996.6052.17) stands at this point in the sequence.It comes from the recentlyexcavatedcemetery,from a level particularlyrich in Late Archaic and Early Classicalfinds. Greco et al. (1996, p. 950; 1997, pp. 820-822; 1999, pp. 525-526) presenta preliminaryreportof the excavationsof the Archaic-Hellenistic necropolisat Itanos. I am gratefulto Didier Viviers for permissionto examineunpublished pottery from these recentexcavations.
APHRATI
AND
KATO
SYME
57
r:
I. A
Figure 9. Jug. Scale 1:3 36 Jug, base Kato Syme (1973). Orth. Delta, OM228. Diam. (base) 9.1 cm. Pale brown (10YR 8/3). Black gloss. Ca. 425-400. Fig. 9
37 Jug
Aphrati. 1969.18777. H. 19.4 cm; Diam. 16.0 cm, of rim 6.6 cm, of CUPS,
FORERUNNERS
base 9.6 cm. Pale brown (10YR 8/3). Black gloss. Ca. 425-400. 38 Jug, base
Fig. 8
Kato Syme (1975). Level 2, OM7. Diam. (base) 8.5 cm. Pale brown (10YR 8/3). Black gloss. Ca. 425-400.
(39-45)
Several new types of bases appear at the sanctuary at the turn of the 6th and 5th centuries (Fig. 10). One new type of base, comprised of a low disk foot with a concave profile underneath, displays a distinctive articulated foot with a wedge-shaped edge (39-41).39 It leads the way to a rich series of 5th-century cup bases presumablybelonging to a local low-necked cup whose subsequent development is treated in greater detail below. Another contemporary variety of base exhibits a less strongly articulatedfoot (4245), a primitive feature that fosters an initial impression of an early date; this notion is dispelled, however, by the sharp vertical facet of the edge of the foot, an indication of a laterdate. Parallelsfor both types exist at Knossos in a closed deposit dated ca. 500-480.40
39
39. A similartype of base is found in the repertoireof Knossianlownecked cups;see Callaghan1978, pp. 6-7, nos. 10-12, fig. 5, pl. 2. 40. Datable parallelscome in the form of unpublishedcup bases from Knossos,deposit RR:H.
Cup, base
Articulatedverticalfoot with faceted Fig. 10 edge. Ca. 500-475. 4wedge-shaped Level
KatoSyme(1972). OM67. Diam. (base)7.1 cm. IPale brown(10YR8/3). BlackglossS. Wedge-shaped edge. Ca. 500- -475.
Fig. 10 KatoSyme(1972).Level6, OM18. Diam. (base)6.2 cm. IPale brown(10YR8/3). BlackglossS. 40
Cup, base
41 Cup,base
Fig. 10
KatoSyme(1974).Level3, OM51. Diam. (base)6.1 cm. Pale brown(10YR8/3). Blackgloss. Grooveunderfoot.Wedge-shaped edge. Ca. 500-475.
BRICE
58
K
39
40
N
42 434
(I
W
l(b
44
(j
41
42 Cup,base
Fig. 10
KatoSyme(1975). Level 12, OM50. Diam. (base)6.5 cm. Pale brown(10YR8/3). Blackgloss. Beveledouteredge.Ca. 500-475. 43
Cup, base
Fig. 10
Kato Syme (1973). Level 9, OM47. Diam. (base) 6.5 cm. Pale brown (10YR 8/3). Black gloss. Beveled outer edge. Ca. 500-475.
HIGH-NECKED
L. ERICKSON
45
44 Cup,base
( (
Figure 10. Cup bases. Scaleca.1:2
Fig. 10
Kato Syme (1972). Level 25, OM62. Diam. (base) 5.6 cm. Pale brown (10YR 8/3). Black gloss. Beveled outer edge. Ca. 500-475.
45 Cup,base
Fig. 10
Kato Syme (1973). Level 9, OM47. Diam. (base) 6.2 cm. Pale brown (10YR 8/3). Rich red gloss. Beveled outer edge. Ca. 500-475.
CUPS (46-71)
Two fuller series of cup bases become availablefor tracking development in the 5th century (Figs. 11-12). These two main lines of development represent an apparent intensification of local production. They nicely illustrate the major characteristics of local production in the Classical period. Both types of base show a remarkablyconsistent pattern of morphological development, the individual stages of which can be precisely plotted thanks to parallelswith cups from Knossos. Moreover, a chronological anchor exists in the form of a closed deposit of ca. 425-400 from Aphrati that contains intact high-necked cupswhose form and fabricmatch those of the cup bases from Kato Syme. This evidence is of crucial importance, for it strengthens the contention that Aphrati is the likely source of the Cretan pottery at the sanctuaryin the 5th century and, by extension, the source of the 6th-century pottery of the same pale brown fabric. Distinguishing featuresof the high-necked cup include the following: a tall inset vertical rim, a single vertical handle, a sharp transition between the shoulder and rim, and a deep body with a narrowshoulder.Also characteristic of the cup is a base with a low disk foot and a concentric penannulargroovebelow;the articulatedfoot displaysa thick convex"rolled" outer edge and a "dropped floor" creating a hollow concavity inside. A prototype of the base occurs at the sanctuary ca. 500-475 in a broader configuration than later examples and lacking, at this early stage, the later trademark"droppedfloor"and groove below (46). Judging from Knossian parallels, the introduction of these two features in the high-necked cup repertoire took place simultaneously a generation later in the second or third quarter of the 5th century (47).41 Subsequent development of the type in the second half of the 5th century led to a narrowerversion of the
41. A groove underfootis found on earlierbases dated on typological groundsto the 6th century,both at Kato Syme and Aphrati (Aphrati SurveyCollection, Knossos StratigraphicalMuseum), but the groove does not become a regularfeatureof local productionuntil the 5th century.
APHRATI
42. A dateof ca.425-400 forthe latestbasesin thisseriesrestsupon to the conicalbases theirresemblance of Knossianhigh-neckedcupsat a and comparable stageof development control the independent chronological providedby a largedepositfrom Aphratiof 5th-centurycupsin the companyof datablelampforms. 43. Coldstream (1973b,p. 48) forthis coinedthe term"smearing" Knossiandecorative practice. 44. Accordingto Callaghan(1978, p. 6), afterca.475 Knossianpotters beginattachingthe handleinsidethe rim.The localcupsfromAphratibear a closerresemblance to a different at the Knossos, 5th-century shape see Callaghan1978,pp.10kantharos; 11, nos. 24-25, fig. 7. Like the Knos-
siankantharos, the high-neckedcups fromAphratiexhibita beveledshoulder moldinganda footwitha "dropped floor"inside. 45. Fromthe"Shrineof Glaukos," trench42. 46. As Callaghan1992, p. 93, no. 3,
pls.76, 105:c,datedca.400-375. 47.Thisvarianttypeof baseis discussedby Callaghan(1978,p. 8, no. 16),who identifiesit as anearlier featureof Knossiancupproduction. 48.The presenceof foreignartisans at Knossoswouldnotbe surprising. Perlman(1992,p. 202) callsattention to the provisionsof a smallnumber of LateArchaicandEarlyClassical inscriptionsthat referto resident foreignersat Eleutherna,Lyktos, and Gortyn. She concludesthat they probablyemigratedfrom other Cretan cities.
AND
KATO
SYME
59
base with greater extension of the outer edge (48-49). Later examples (ca. 425-400) also have a higher foot that meets the belly at an increasingly narrow point of attachment. Steady attenuation of the base comes to a swift conclusion at the end of the 5th century with the complete replacement of the low disk foot by a high conical support, whose hollow truncated cone adds greater emphasis to the "droppedfloor"inside (51).42 A group of intact high-necked cups from Aphrati (58-71) stands at roughly the same point in time (ca. 425-400) within a shared local tradition (Figs. 14-16). These cups provide invaluableinformation about every aspect of this shape. The similarities between the examples from Aphrati and Kato Syme extend to minor details of potting, including a concentric groove underfoot of the same circumference (within a millimeter tolerance).The two cups shown in Figure 12, one of which comes from Aphrati (60), the other from Kato Syme (51), are surely contemporary and may even be by the same potter.These high-necked cups differ from a contemporary Knossian version of the shape in several key respects: a beveled shoulder marks the juncture with a deeply inset rim, and the profile of the body,whether tall and cylindricalor depressed and "baggy,"is invariablyof greatervolume than that of the Knossian examples.Another minor difference is that the underside of the base receives either a full coat of gloss or is reserved;never is it decorated with irregularstrokes of dilute gloss in the manner favored by Classical Knossian potters.43The penannular groove underfoot also distinguishes the local vessels. Finally,the cups from Aphrati possess a longer strap handle, which, unlike the handle of Knossian cups of this late date, attaches outside the rim.44One cup (58) differs from the others. It exhibits a stout profile with a smooth transition at the junction of the rim and shoulder.This cup possesses a broad low foot that lacks the elaboration of a "droppedfloor" inside. These features suggest an earlier date for the cup, ca. 475-425. Local production of high-necked cups remained largely free of Knossian influence. Yet manifest examples of Knossian influence appear from time to time at Kato Syme, as in the case of a pedestal base (57) in the local pale brown fabric (Fig. 13), which differs from other varieties in having a recessed central medallion underfoot and a fillet marking the junction between the base and belly. This combination of features recalls an unpublished base from Knossos.45This base, in addition to a fragmentary cup from Kato Syme whose cylindrical shoulder is closely similar to a cup from Knossos,46attests a sporadicbut definite influence of the Knossian high-necked cup upon local production (Fig. 13). On this basis, we might postulate a sideline production by Aphrati potters emulating or unintentionally copying Knossian wares while the main output of cups is of distinctly local inspiration.The high-necked cup first appearedat Aphrati at approximately the same point in time (500-475) as at Knossos. Aphrati even exerted a reciprocalinfluence upon the Knossian field of production, to judge from the occasional appearancein the Knossian fabric of a variant high-necked cup distinguished by a base with a "rolled"edge, "dropped floor,"and groove below.47The production of this variant cup at Knossos may reflect the activities of itinerant potters from Aphrati who worked at Knossos or indigenous Knossian potters intimately familiarwith a neighboring style.48
6o
BRICE
46 6
I
-I
47
49
Cup, base
Fig. 11
Kato Syme (1974). Level 4, OM18. Diam. (base) 8.8 cm. Pale brown (10YR 8/3). Black gloss. Prototype, ca. 500-475. Cup, base
Fig. 11
Kato Syme (1972). Level 6, OM15. Diam. (base) 8.0 cm. Pale brown (10YR 8/3). Unglazed. Penannular groove underfoot. Similar to Callaghan 1978, no.16, fig. 5, pl. 2. Ca.450-425. Cup, base
Fig. 11
Kato Syme (1973). Level 2, OM4. Diam. (base) 8.1 cm. Pale brown (10YR 8/3). Black gloss. Penannular groove underfoot. Similar to an unpublished base from Knossos, deposit SEX: J/JN7.4982. Ca. 450-425. 49
Cup, base
Fig. 11
Kato Syme (1973). Level 5, OM26. Diam. (base) 7.5 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca.450-425.
Figure 12. Cup bases
50 Cup,base KatoSyme(1974). Level 7, OM26. Diam. (base)7.4 cm. Pale brown(10YR8/3). Blackgloss. Ca.425-400. 51 Cup,base
Figs. 11-12
Kato Syme (1972). Level 12, OM7. Diam. (base) 7.1 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca.425-400. 52
48
Figure 11. Cup bases. Scaleca.1:2
51
60
47
4
51
48
46
L. ERICKSON
Cup, base
Kato Syme (1975). Level 2, OM146; Diam. (base) 7.9 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca.425-400. 53
Cup, base
Kato Syme (1974). Level 10, OM42. Diam. (base) 7.4 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca. 425-400.
APHRATI
AND
6i
KATO SYME
I 57 Figure13. High-neckedcups. Scale ca. 1:2
56
60 Cup
Figs. 12, 14-15 Aphrati.1969.18760.H. 16.9 Diam. 11.4 cm, of rim 9.2 cm, of cm; base7.9 cm. Palebrown(10YR 8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400.
66 Cup
61 Cup
Figs. 14-15 Aphrati.1969.18761.H. 17.1 Diam. 11.3 cm, of rim 9.9 cm, of cm; base7.9 cm. Palebrown(10YR8/3). Blackgloss. Ca. 425-400.
67 Cup
56 Cup,rim and shoulder Fig. 13 KatoSyme(1972). Level 5, OM172. Diam. 10.2 cm, of rim 10.0 cm. Palebrown(10YR8/3). Black gloss.Cf. Callaghan1992, p. 93, no. 3. Ca. 400-390.
62 Cup
Figs. 14-15 Aphrati.1969.18762.H. 16.5 cm;Diam. 11.0 cm, of rim 10.3 cm, of base 8.1 cm. Palebrown(10YR 8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400.
68 Cup
Figs. 14-15 Aphrati.1969.18768.H. 17.5 cm;Diam. 10.9 cm, of rim 9.9 cm, of base7.3 cm. Palebrown(10YR8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400.
57 Cup,base Fig. 13 KatoSyme(1977). Level 10, OM231. Diam. (base)5.7 cm. Pale brown(10YR8/3). Blackgloss.Pedestal support.Similarto an unpublishedbasefromKnossos,deposit UM.TR12.42.Ca. 425-400.
63 Cup
69 Cup orjug Fig. 16 Aphrati.1969.A6.Diam. 10.9 cm, of base8.3 cm. Palebrown (10YR8/3). Blackgloss.Ca. 425400.
54 Cup,rim KatoSyme(1974). Level 11, orth.Delta. Diam. (rim)11.1 cm. Palebrown(10YR 8/3). Blackgloss. 5th century. 55 Cup,base KatoSyme (1974). Level 10, OM42. Diam. (base)7.1 cm. Pale brown(10YR8/3). Blackgloss. Penannulargrooveunderfoot. Ca. 425-400.
58 Cup
Figs. 14-15 Aphrati.1969.18769.H. 15.7 cm; Diam. 12.0 cm, of rim 9.8 cm, of base 7.6 cm. Palebrown(10YR 8/3). Black gloss.Ca. 475-425. 59 Cup
Figs. 14-15 Aphrati.1969.18759.H. 17.9 cm; Diam. 11.2 cm, of rim 9.9 cm, of base 7.9 cm. Palebrown(10YR 8/3). Black gloss.Penannulargrooveunderfoot. Ca. 425-400.
Aphrati.1969.18763.H. 16.9 cm;Diam. 11.0 cm, of rim 10.2 cm, of base8.0 cm. Palebrown(10YR 8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400. 64 Cup
Figs. 14-15 Aphrati.1969.18764.H. 17.7 cm;Diam. 10.5 cm, of rim 11.2 cm, of base 7.5 cm. Palebrown(10YR 8/3). Blackgloss.Double penannular grooveunderfoot.Ca. 425-400. 65 Cup
Fig. 16 Aphrati.1969.18765.H. 15.8 cm;Diam. 11.2 cm, of rim 9.9 cm, of base7.9 cm. Palebrown(10YR8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400.
Figs. 14-15 Aphrati.1969.18766.H. 17.2 Diam. 10.1 cm, of rim 10.0 cm, cm; of base7.6 cm. Palebrown(10YR 8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400. Figs. 14-15 Aphrati.1969.18767.H. 14.8 cm;Diam. 10.6 cm, of rim 10.3 cm, of base7.8 cm. Palebrown(10YR 8/3). Blackgloss.Ca. 425-400.
70 Cup,base Aphrati.1969.A5.Diam. (base) 7.6 cm. Palebrown(10YR8/3). Blackgloss.Penannulargroove underfoot.Ca. 425-400. 71 Cup,base Aphrati.1969.A5.Diam. (base) 8.2 cm. Palebrown(10YR8/3). Blackgloss.Ca. 425-400.
62
BRICE
62
L. ERICKSON
64
'
Figure 14. High-necked cups. Scale 1:3
APHRATI
AND
KATO
SYME
63
60
Figure 15. High-necked cups. Scale 1:3
BRICE
64
L. ERICKSON
69 Figure 16. High-necked cups. Scale 1:3
LOW-NECKED
CUPS
(72-81)
A parallel process of development occurs in the case of a second major variety of Sth-century cup (Fig. 17). Its distinguishing features include a low disk foot with a wedge-shaped edge and a penannular groove underfoot (72-80). The presence of a groove underfoot links these bases with the previous class, and is a hallmark of 5th-century local production.49A comparison of these bases with similar examples from Knossos encourages reconstructing the local shape as a low-necked cup, but in the absence of intact examples either from the sanctuaryor the presumed seat of production at Aphrati, the proposed identification requires further corroboration. A fragmentaryrim (6) from the sanctuarymay belong to a Classical low-necked cup, but again, this identification remains uncertain. More compelling evidence for the production of the low-necked cup at Aphrati is provided by a find from Knossos, a Cretan low-necked cup distinguishable from Knossian wares on the basis of its reportedlypale brown fabric.50 It may be a product of Aphrati, although other possible sources for pale brown fabric, discussed above, complicate interpretation. Its base displays a wedge-shaped foot and groove underneath, characteristicsof the Kato Syme-Aphrati Group. In addition, the squat shape and wedge-shaped support of a vessel from a 5th-century domestic deposit at Aphrati (81) suggest its reconstructionas a low-necked cup (Fig. 18), although the fragmentarycondition of this pot forbids confident reconstructionof the shape. The development of this local cup follows suit with respect to the high-necked cups, beginning with early "prototypes"that lack the distinguishing features of a groove underfoot and a "droppedfloor"inside (7273). The development of this series also culminates at the end of the 5th century in the formation of a high pedestal foot with a widely extended outer edge and a "droppedfloor"inside (e.g., 80). Close parallelswith datable material from the Knossian Kiln Group (KKG) establish a date of ca. 400 B.C., or shortly thereafter,for the end of this series at the sanctuary.51 72 Cup,base Fig. 17 KatoSyme(1972).Level6, OM18. Diam. (base)7.0 cm. Pale
brown(10YR8/3). Blackgloss. Prototype.Wedge-shapededge. Ca. 500-475.
49.The additionof a groove underfootcreatesa decorative ring to the paintedones comparable adorningthe bottom of Attic and other
mainlandbases.
50. Callaghan(1978, p. 7) describes the fabricof no. 15 as "thepale fabric common in the south and west coast of Crete ratherthan the buff to red of the centralcoast."
51. SeeHomann-Wedeking 1950, p. 171, pl. 13:a(c).Coldstream(1999, p. 323, no. R29, fig. 2, pl. 31) illustrates
a closeparallelfor80 fromKKG,dated to ca. 400-375.
APHRATI
AND
SYME
72 73
KATO
75
_ _ 79
(
74
Figure 17. Cup bases. Scaleca. 1:2
(
J
80
%\
81 Figure 18. Low-necked cup. Scale ca. 1:2
73
Cup, base
Fig. 17
Kato Syme (1972). Level 14, OM23. Diam. (base) 7.3 cm. Pale brown (10YR 8/3). Black gloss. Prototype. Wedge-shaped edge.
Ca.500-475.
74
Cup, base
Fig. 17
Kato Syme (1972). Level 1, OM381. Diam. (base) 8.0 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca. 475-450. 75
Cup, base
Fig. 17
Kato Syme (1972). Level 1, OM381. Diam. (base) 7.6 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca. 475-425. 76
Cup, base
Kato Syme (1974). Level 3, OM39. Pale brown (10YR 8/3). Black gloss. Penannular groove underfoot. Ca. 475-450. 77
Cup, base
Kato Syme (1974). Level 3, OM45. Pale brown (10YR 8/3).
Black gloss. Penannulargroove underfoot. Ca. 475-450. 78
Cup, base
Kato Syme (1973). Level 6, OM110. Diam. (base) 7.1 cm. Pale brown (10YR 8/3). Black gloss. Penannulargroove underfoot. Ca. 450-425.
79 Cup,base Fig. 17 KatoSyme(1975). Level5, OM90. Diam. (base)6.6 cm. Pale brown(10YR8/3). Blackgloss. Penannulargrooveunderfoot. Ca. 450-425. 80 Cup,base
Fig. 17
Kato Syme (1975). Level 3, OM124. Diam. (base) 7.6 cm. Pale brown (10YR 8/3). Black gloss. As Coldstream 1999, p. 323, no. R29, fig. 2, pl. 31. Ca. 400-390.
81 Cup
Fig. 18
Aphrati. 1969.A5. Diam. 13.0 cm, of base 7.6 cm. Pale brown (10YR 8/3). Black gloss. Penannular groove underfoot. Ca. 475-425.
66
BRICE
LATE-5TH-CENTURY BELL
CRATER
POTTERY
L.
ERICKSON
(82-90)
(82)
Notable among the finds from the 5th-century domestic deposit from Aphrati is a small bell crater of curiously conservative appearance (Fig. 19). This shape finds no exact parallel in the Classical Cretan repertoire. Knossos provides the closest comparison in the form of a large bell crater
Figure 19. Bell crater. Scale1:3
of Protogeometric date.52Such craters are absent from the GeometricOrientalizing cemetery at Aphrati, suggesting that the Classical version is either a revivalof an earlierunattested shape or a new creation of Classical potters.53If the latter explanation is correct, the resemblance of the 5thcentury bell crater to the Protogeometric Knossian shape may be fortuitous. The bell craterfrom Aphrati bears painted decoration on the wall in the form of a frieze of stacked triangles and a row of dots consigned to a separate field below, documenting the survival of the Orientalizing pot painting tradition at Aphrati well into the 5th century. The simplified, geometric decorative scheme enhances this vessel's conservative appearance. A date for the pot in the final quarterof the 5th century is suggested both by the profile of the foot-a high splaying stand comparable to the supports of 5th-century jugs-and by the date of the deposit.
Fig. 19 Aphrati.1969.18776.H. 18.8 cm; Diam. 13.2 cm, of rim 15.6 cm, of base 9.2 cm. Palebrown(10YR8/3). Black 82 Bell crater
gloss.Bandeddecorationwith a row of dots and a seriesof stacked trianglesin a separatereservedfield. Finalquarterof the 5th century.
52. Coldstreamand Catling (1996, pp. 368-372) illustratea selection of Protogeometriccratersfrom the North Cemetery comparablein shape to 82. 53. A craterpreservedin a Protogeometricburialfrom Aphrati (see Levi 1927-1929, p. 176, fig. 196) bearsa resemblanceto and may be a distant relativeof the Classicalshape, althoughthe comparisonis not exact. Callaghan(1978, pp. 12-15, nos. 3741, fig. 9, pl. 4:a) publishes5th-century Knossiancratersof a type similarto our examplefrom Aphrati,suggestingmore extensiveCretanproductionof craters in the Classicalperiod.
APHRATI
AND
KATO
PROTO-HADRA
SYME
67
HYDRIA (83)
Also from the 5th-century domestic deposit, a proto-Hadra hydria ranks Aphrati among select sites known to have produced a Classical precursor of the Hellenistic Hadra hydria, a popular export to Egypt in the 2nd and 1st centuries B.C.54 Callaghan andJones attribute these Hellenistic exports to a Cretan source on the basis of fabric, while they trace the ancestry of the type at Knossos back to the Archaic period.55Knossos and Phaistos have yielded evidence for the production of the shape in the 5th century.56 Both sites laterbecame centers for the manufactureof Cretan hadrahydriae in the Hellenistic period. The discovery of a 5th-century example from Aphrati demonstrates comparablelocal ancestry of the type and raises the question of a Hellenistic production center at Aphrati, although this suggestion is not supported by present evidence from the Hellenistic settlement. Moreover, the exact relationship between the Classical and Hellenistic types remains unclear. 83 Proto-Hadrahydria Aphrati.1969.A5.Palebrown (10YR 8/3). Blackgloss.Illustrated by Lebessi1973,p. 460, pl. 402:b.
Bandeddecorationwith curvilinear designsconsignedto reservedfield. Ca. 425-400.
BOWL (84) From the same deposit comes a large bowl equipped with two horizontal cylindrical handles (Fig. 20). Other distinguishing features of the shape include a collared rim and a ring base. Like the bell crater and hydria discussed above, this bowl is the sole known representativeof its type from Classical Aphrati. A 7th-century prototype for the bowl from an earlier deposit at Aphrati exhibits banded decoration, a deep body, and wide strap handles, markinga departurefrom its Classicalcounterpart.57 Despite these minor differences,the proportionsand dimensions of these two deep bowls are remarkablyclose, enabling one to posit a direct line of descent.58
54. Lebessi (1973, p. 460, pl. 402:b) publishesa photographof this hydria. It is not illustratedhere. 55. CallaghanandJones 1985, p. 11, n. 47. 56. For Knossos,see Coldstream 1973b, p. 50, nos. 16-11, fig. 5; Callaghan 1978, p. 15, no. 42, pl. 4:c. In addition,a proto-Hadrahydriahas come to light in a Classicalwell in the areaof the StratigraphicalMuseum, Knossos,from a context dated ca. 450425. I thank PeterWarrenfor permission to mention this unpublishedfind.
ChiaraPortalebrings to my attention a proto-Hadrahydriafrom the areaof ChalaranearPhaistos (Herakleion ArchaeologicalMuseum, no. 4475). It is a surfacefind dated to the 5th centuryon the basis of the Knossian parallelscited above. 57. See Lebessi 1973, p. 458, pl. 400:a. 58. A small shallowbowl from a late-5th- or 4th-century deposit at Knossos falls under the same general category;see Coldstream1973a, p. 26, no. C12, fig. 14.
68
BRICE
L. ERICKSON
I'
84
Figure 20. Bowl. Scale1:3
84 Bowl
Fig. 20 Aphrati.1969.18774.H. 10.0 cm;Diam. 21.6 cm, of rim23.2 cm,
of base 11.6 cm. Palebrown(10YR 8/3). Blackgloss. Ca. 425-400.
LAMPS (85-88) Four lamps from the domestic deposit at Aphrati (Figs. 21-22) provide the securepegs upon which the late-5th-century date for the deposit hangs. An imported Knossian or Gortynian lamp (85) that copies an Attic form datable to the final quarterof the 5th century provides an invaluablechronological point of reference.59This Cretan equivalent of Howland's Type 21 is a lamp whose profile is defined by a continuous curve formed by a side wall joined seamlessly to a curving rim. This type enjoyed widespread popularity among the Cretan production centers in the middle and second half of the 5th century.60The other three lamps from Aphrati resemble Howland's Type 20, although the comparison is not exact. They exhibit a curved side wall and a ridge around the filling hole, features that ally them with an Attic predecessor of the standard 5th-century type. The Cretan version appears to have continued in production for a longer period of time than the Athenian shape, to judge from the evidence of the deposit at Aphrati. The absence of lamps at Kato Syme is somewhat surprisingin light of their popularity at other Classical Cretan cult centers. In the case of the sanctuaryofDemeter at Knossos, Coldstream plausiblyconnects the lamps with the nocturnal ceremonies held in honor of Demeter mentioned by Diodorus Siculus (5.77.3).61 An obvious practical use comes to mind in the case of lamps from a cave consecrated in the Classical period to Pan
59. As AgoraIV, p. 47, no. 169, pls. 6, 34. The Attic parallelcomes from well deposit M18:8. AgoraXII, p. 395, offers a reviseddate of ca. 430-
420 forthisdepositandthusthelamp. My attributionof the Cretanlamp to a productioncenter at Knossosor Gortyn is based upon considerationof the fabric. 60. The earliestdatableexamples on Crete derivefrom closed 5thcenturydeposits at Knossos;see Coldstreamand Macdonald1997, p. 227, no. K77, fig. 18; Callaghan 1992, p. 92, no. 8, pl. 75; Coldstream 1973a, pp. 24-25, nos. B10-11, fig. 14, p. 11. 61. Coldstream1973a, p. 186.
APHRATI
AND
KATO
69
SYME
and the Nymphs at Lera in the territoryof Kydonia.62No single explanation can do justice to the variety of ritual contexts in which lamps appear on Crete. The absence of lamps from an unequivocal cult context at Kato Syme may reflect cult or dedicatory practices out of step with other parts of the island.
v
I
-
88 Figure 21. Lamps. Scaleca.1:2
(I 85
87
88
Figure22. Lamps
Figs. 21-22 Aphrati.1969.18773.H. 1.9 cm; Diam. 5.5 cm, of base4.0 cm. Pale orange(7.5YR7/6). Unglazed.Strap handle.Illustratedby Lebessi1973, p. 460, pl. 402:a.Similarto AgoraIV, p. 47, no. 169, pls. 6, 34. Ca. 430420. 85 Lamp
Fig. 21 Aphrati.1969.A8.18770.H. 3.2 cm;Diam. 9.8 cm, of base5.4 cm. Palebrown(10YR8/3). Blackgloss. Profilesimilarto Coldstream1973a, p. 24, no. B9, fig. 14. Ca. 425-400. 86 Lamp
andLam62. Guest-Papamanoli braki(1980, pp. 221-222, fig. 11,
pl. 46) illustratea selectionof lamps fromLera.
Figs.21-22 Aphrati.1969.A12.18771.H. 3.4 cm;Diam. 8.2 cm, of base5.0 cm. Palebrown(10YR8/3). Blackgloss. Raisedbase.Straphandle.Ca. 425400. 87 Lamp
Figs.21-22 Aphrati.1969.A11.18772.H. 3.3 cm;Diam. 8.1 cm, of base5.0 cm. Palebrown(10YR8/3). Blackgloss. Raisedbase.Straphandle.Illustrated by Lebessi1973,p. 460, pl. 402:a. Ca. 425-400. 88 Lamp
BRICE L. ERICKSON
70 VOTIVE MINIATURES
(89-90)
Two votive miniatures from Kato Syme attest a definite role of pottery as a dedication at the sanctuary (Fig. 23). The first example (89) indicates local production of a jug shape attested at Knossos by a single unpublished example in a Late Archaic deposit (RR:H). This type is distinguished by a deep, rounded body and an inset flaring rim thickened at the lip to form an overhanging projection.The second example (90), a krateriskos,finds a close comparisonin clayvotives from the sanctuaryof Demeter at Knossos.63 89 Votiveminiature, rim and shoulder
Fig. 23
KatoSyme(1974).Level40, OM37. Diam. 5.3 cm, of rim5.1 cm. Palebrown(10YR8/3). Blackgloss. 6th or 5th century.
90 Krateriskos
?I 89
Fig. 23
KatoSyme(1972).Level4, bothros.H. 3.5 cm;Diam. 4.0 cm, of rim4.3 cm, of base2.6 cm. Pale brown(10YR8/3). Unglazed.5th century.
90
Figure23. Votiveminiatures. FOURTH-CENTURY
POTTERY
(91-106)
At the turn of the 5th and 4th centuries a fundamental change in the character of the local ceramics marks an important turning point in the life of the sanctuary.An abrupt and almost complete replacement of the pale brown pottery presumablyfrom Aphrati with pottery of a completely differentcharacter,and arguablyfrom anotherproductioncenter,took place around 400 B.C.,or shortly thereafter.Two strands of interlocking evidence-the timing of the pale brown fabric'svirtual disappearanceat Kato Syme and the swift introduction of new shapes in a gritty, reddish-brown fabric-document the change. Fortunately, this important event in the history of the sanctuarycan be accuratelydated on the basis of the internal typologies of the two production centers. Stylistic analysis of three different series of Classical cup bases of pale brown fabric indicates that pottery from Aphrati comes to an abrupt and uniform end at the sanctuary in the final years of the 5th century.The latest examples of pale brown fabric include a cup base, discussed above (57), whose close approximationof a Knossian form yields a chronological range of ca. 425-400; another cup base (80) with demonstrableparallelsin shape with the Knossian Kiln Group (KKG), ca. 400-390;64 and a fragmentary high-necked cup (56) of identical construction to an example from a closed deposit at Knossos, UM:H5, dated to ca. 400-390.65 It is reasonable to conclude from these parallels that the latest pottery used at the sanctuary before the pale brown fabric disappears dates to ca. 400390. The earliest appearanceat the sanctuaryof cups in the gritty reddishbrown fabric can be dated to the first quarter of the 4th century on the basis of parallels in shape with Knossian material (Figs. 25-26). For example, a new type of cup base at the sanctuary(95), whose distinguishing features include a high pedestal foot, a widely spreading edge, and a narrow point of attachment to the belly, mirrors developments first seen at Knossos in the early-4th-century deposit KKG.66Another type of base in
Scale ca. 1:2
63. Krateriskoipreservedin a late5th-centurycontext from the sanctuary providethe closest parallel;see Coldstream1973a, p. 25, nos. B14-15, pl. 11. 64. See Homann-Wedeking1950, p. 171, pl.13:a(c),republishedin Coldstream1999, p. 323, no. R29, fig. 2, pl.31. 65. Callaghan1992, p. 93, no. 3, pls. 76, 105:c. 66. See Homann-Wedeking1950, p. 171, fig. 4:a.
APHRATI
67.The parallel,notillustrated, comesfromanunpublished early-4thcenturydeposit at Knossos, SEX:J/ JN6. Thanks go to PeterWarrenfor allowing me to examine Classical materialfrom his recent excavations. 68. My identificationof Lyktos as the sourceof this pottery is based upon examinationof the fabricor the correspondencein form with 4thcenturymaterialfrom Kato Syme (or both). 69. Gournia Survey,Site 106. Watrousand Blitzer (1995) give their reasonsfor identifyingthe site of Profitis Ilias with Larisa.Potteryof presumedLyktianorigin from Site 106 dates to the middle of the 4th century. Donald Haggis (pers.comm.) points out that this site lies in a phyllitequartzitezone with rich sourcesof silvermica and biotite (gold mica). Local productionof red micaceous fine-warepottery cannot be ruled out. 70. This cup comes from an unpublishedClassicalwell from the StratigraphicalMuseum Excavations (J/JN7.4982) dated ca. 450-425. 71. Callaghan1978, p. 18, nos. 5657, fig. 10.
AND
KATO
SYME
71
the new fabric (97) is characterized by a high pedestal stand. It finds an equally valid comparison in the early-4th-century ceramic repertoire at Knossos.67A third base, composed of a particularlymassive pedestal foot, occurs in considerable quantities at the sanctuary (99). Its hollow truncated cone is accompanied by an exaggerated"droppedfloor"inside. The ultimate source of inspiration for this type of base is found in the Attic repertoireof Classical kantharoi,but a more immediate influence springs from a Cretan source, given the fact that Attic bases similarly inspired Knossian production from ca. 375 to 350. I suggested abovethat this gritty,reddish-brown,silvermicaceousfabric was produced at Lyktos. While micaceous fabrics appear periodically in Classical Cretan contexts, particularlyin the case of cooking and coarse wares, the use of gritty fabric for the production of fine wares is rare,and probably sets Lyktos apart from its neighbors. This hypothesis requires further support, either through further investigation of the site of Lyktos and delineation of its local pottery style or analysis of nearbyclay beds and mineral sources. If the proposed Lyktian origin of the 4th-century material from Kato Syme is valid, this body of material supplements our knowledge of the ceramic output of ancient Lyktos in an important way.These exportscompensate in partfor a deficit in the recordof settlement at Lyktos itself, where, owing to the limited scope of archaeological investigation, pottery from the 4th century and earlierhistorical periods remains largely unattested. Pottery of possible Lyktian origin68turns up in other parts of Crete, as at an unpublished surveysite in the vicinity of Hierapetra,tentatively identified by L. V. Watrous and Harriet Blitzer as ancient Larisa (Fig. 1).69 In addition, an intact small Cretan cup with a low offset rim and hemispherical bowl, exported to Knossos and subsequently discarded in a 5th-century well, exhibits the same coarse, silver micaceous, reddish-brown fabric that I attribute to Lyktian production.70This cup is clearly an antecedent of Callaghan's"glazed cup with everted rim," a type otherwise known to exist at Lyktos and Knossos only in a Hellenistic manifestation.71From the new evidence of this Classical forerunner,it can be surmised that little or no change occurredin the cup'sdesign from the 5th to the 3rd century. In addition, there is evidence in the domestic deposit at Aphrati (ca. 425400) of three presumed Lyktian imports (Fig. 24): a small jug (91), lekane (92), and cup (93), each in the characteristic Lyktian fabric as defined here. This group ranks as the largest-known assemblage of 5th-century Lyktian pottery from the island. It also serves as an important reminder that scholars'neglect ofpost-Minoan Crete has led to gaps in the archaeological record at major sites, such as Lyktos, where preliminaryexcavation has shown the likelihood of producing rich Classical deposits. On the sole basis of Lyktian exports, then, an otherwise unattested history of local ceramic production in the 5th and 4th centuries can now be appreciated(Figs. 24-26). The voluminous shoulder of the 5th-century Lyktian cup (93) found at Aphrati gives the impression of a conservative local tradition, although naturallyone would wish for a greater number of exampleson which to base such a conclusion.Lyktiancup productionduring
BRICE
72
L. ERICKSON
91 _
92
92
(
o'/
Figure24. Lyktianpottery.Scale1:3
the 4th century,it seems, remained largely derivative from Knossian, save for the one distinguishing feature of an inset notch at the junction of the base and belly in place of a fillet that normally marks the transition at Knossos. The small lekane (92) is a distinctively Lyktian shape. Comparable examples from Kato Syme and the sanctuary of Apollo at Aghia Pelaghia indicate wider distribution of these Lyktian products.72Similar bowls from Knossos in Knossian fabric confirm a chronological range of ca. 425-400 for the Lyktian example.73 72. A total of four lekanai-two
fromKatoSyme,oneof whichis published (Lebessi 1990, p. 268, pl. 127:e, dated erroneouslyto the Late Minoan IIIC or Protogeometricperiod), and two unpublishedspecimensfrom Aghia Pelaghia-have been identified. 73. The Lyktianbowl representsan intermediatestage of development, subsequentboth to Callaghan1992,
p. 92, no. 6, pl. 75, from a deposit dated ca. 475-450 (my revisedchronology), and Coldstreamand MacDonald 1997, p. 224, no. 44, fig. 18, from a deposit dated ca. 475-450 (my revisedchronology), but priorto Callaghan1992, p. 101, no. 28, pl. 81, from a deposit dated to the last quarterof the 4th century.
APHRATI
AND
KATO
SYME
73
(
94
95
'
104
97
Figure 25. Lyktian cup and jug bases. Scale ca. 1:2
105
106
99
91 Jug, reconstructed profile Fig. 24 Aphrati. 1969.A5. Neck is missing. H. 12.7 cm; Diam. 6.8 cm, of rim 4.5 cm, of base 3.3 cm. Silver micaceous, reddish-brown fabric (2.5YR 5/6 to 5/8). Unglazed. Lyktian manufacture. Ca. 425-400. 92
Fig. 24
Aphrati. 1969.18779. H. 5.5 cm; Diam. 13.5 cm, of rim 15.5 cm, of base 5.4 cm. Silver micaceous, reddish-brown fabric (2.5YR 6/6). Unglazed. Lyktian manufacture. Ca. 425-400.
94
93
95
Lekane
l
Figure 26. Lyktian cup bases
Fig. 24
Aphrati. 1969.A5. Diam. 13.9 cm, of base 8.0 cm. Silver micaceous, reddish-brown fabric (2.5YR 5/6 to 6/6). Unglazed. Lyktian manufacture. Ca. 425-400. 94
99
Cup, lower body
Cup, base
Figs. 25-26
Kato Syme (1972). C)M94. Diam. (base) 7.4 cm. Silv er micar 1 -; ceous, reddish-brown, gril#tiyrlaDrlc (2.5YR 5/6 to 6/6). Unglatzed. Lyktian manufacture.Ca. 400-375. 95
Cup, base
Figs. 25-26
,evel 5 Kato Syme (1974). L Silver OM4. Diam. (base) 7.4 cr m. micaceous, reddish-brownl, gnrtty fabric (2.5YR 5/6 to 6/6). Unglazed.
Similarto an Lyktianmanufacture. unpublishedbasefromKnossos,deposit KKG.Ca. 400-375. 96 Cup,base KatoSyme(1973). Level 7, OM50. Diam. (base)7.5 cm. Silver micaceous,reddish-brown, grittyfabric (2.5YR5/6 to 6/6). Unglazed.Lyktian manufacture. Ca. 400-375. 97
Cup, base
Fig. 25
Kato Syme (1972). Level 1, OM381. Diam. (base) 7.2 cm. Silver micaceous, reddish-brown, gritty fabric (2.5YR 5/6 to 6/6). Unglazed. Lyktian manufacture.Cf. Knossos deposit KKG. Ca. 400-375. 98
Cup, base
Kato Syme (1974). Level 7, OM26. Diam. (base) 7.6 cm. Silver micaceous, reddish-brown, gritty fabric (2.5YR 5/6 to 6/6). Unglazed. Lyktian manufacture.Inset notch at junction of base and belly. Ca. 400-375. 99
Cup, base
Figs. 25-26
Kato Syme (1972). Level 10, OM6. Diam. (base) 7.1 cm. Silver micaceous, reddish-brown, gritty fabric (2.5YR 5/6 to 6/6). Unglazed. Lyktian manufacture. Similar to an unpublished base from Knossos, deposit K67:71, trench 13, level 31A. Ca. 375-325.
BRICE
74
fabric(2.5YR5/6 to 6/6). Unglazed. Pedestalbase Lyktianmanufacture. with flat restingsurface.Ca. 375-325.
100 Cup, base
KatoSyme(1972). Level 13, OM1. Diam. (base)7.5 cm. Silver micaceous,reddish-brown, gritty fabric(2.5YR4/2). Unglazed. Ca. 375-325. Lyktianmanufacture.
103 Cup,base KatoSyme(1973).Level 11, OM180. Diam. (base)8.0 cm. Silver micaceous,reddish-brown, grittyfabric (2.5YR5/6 to 6/6). Unglazed.Lyktian manufacture. Ca. 375-325.
101 Cup,base KatoSyme(1972).Level 15, OM19. Diam. (base)7.7 cm. Silver micaceous,reddish-brown, gritty fabric(2.5YR5/6 to 6/6). Unglazed. Ca. 375-325. Lyktianmanufacture.
104 Cup,base
Fig. 25
KatoSyme(1972).Level 10, OM6. Diam. (base)6.7 cm. Silver micaceous,reddish-brown, grittyfabric (2.5YR5/6 to 6/6). Brownwash. Unglazed.Lyktianmanufacture. Pedestalbase.Late 4th or 3rdcentury.
102 Cup,base KatoSyme(1972).Level 15, OM19. Diam. (base)7.7 cm. Silver micaceous,reddish-brown, gritty
IMPORTS
L. ERICKSON
105 Cup,base
Fig. 25
Kato Syme (1974). Level 8, OM12. Diam. (base) 7.9 cm. Silver micaceous, reddish-brown, gritty fabric (2.5YR 5/6 to 6/6). Unglazed. Lyktian manufacture.Ca. 375-325.
106Jug,base
Fig. 25
Kato Syme (1972). Level 36, OM81. Diam. (base) 12.1 cm. Silver micaceous, reddish-brown, gritty fabric (2.5YR 5/6 to 6/6). Unglazed. Lyktian manufacture.4th or 3rd century.
(107-110)
Imported pottery arrivesat the sanctuaryonly in minute quantities during the 6th and 5th centuries (Fig. 27). Two Corinthian aryballoi (107, 108), preserved only in their top sections, constitute the only discernible imports of mainland Greek pottery during the entire period under consideration. The vast majority of the Late Archaic and Classical pottery at the sanctuary comes from a nearby production center presumably located at Aphrati. Gortynian products appear only on occasion. The extreme paucity of imports reinforces the local characterof the sanctuary. m 107
109
108
1 ,
I
107 Corinthianaryballos,
Figure 27. Imports. Scaleca.1:2
110
Fig. 27
rim
KatoSyme(1973).Level2, OM84. Diam. (rim)4.3 cm. Ca.600-550. 108 Corinthianaryballos,
I
I
Fig. 27
rim
KatoSyme(1972).Level 15, OM20. Diam. (rim)4.7 cm. Ca.600-550.
109 Gortynian cup, base
Fig. 27
Kato Syme (1973). Level 9, OM81. Diam. (base) 6.0 cm. Clean, light red fabric (2.5YR 5/6 to 6/8). Lustrous black gloss. Ca. 425-400. 110 Gortynian cup, rim to lower body
Fig. 27
Kato Syme (1974). Levels 5-12, orth. Omicron-Delta. Diam. 8.6 cm, of rim 7.4 cm. Light red fabric (2.5YR 6/8 to 4/8). Black gloss. Ca.500-480.
APHRATI
AND
HISTORICAL
74. Viviers 1994, pp. 254-258. 75. Viviers 1994, p. 259. 76. Watrous 1982, pp. 84-86. 77. Watrous 1982, pp. 22-23. Seneca (QNat. 3.2.5) preserves
accountof the destrucTheophrastus's tionofArkades.Guarducci (1935,p. 6) discussesthe evidence. 78. Raubitschek's (in Hoffmann 1972,pp.15-16) proposedLyktian originforthe armoris followedby Boardman1982,p. 227, andHuxley 1994, p. 129. 79. If the Archaic scriptof Aphrati were shown to employ anotherform of omega, the case of the armor'sorigins would be decided in favorof Lyktos. As it is, there exist no other Archaicinscriptionsfrom Aphrati to form a basis for a decision.The Spensithiosdecree, which recordsa possible decision of the Dattallians,presumablyone of the Cretanpoleis in the area(although other interpretationsarepossible),employs an identicalform of omega. Whether, as Viviers (1994, pp. 240241) argues,Dattalla should be identified with the site of Aphrati,or, if Dattallawas a city at all, it occupied a differentlocation, it stands to reason that the use of a double omega in anotherdocument outside Lyktos favorsan interpretationof the letter form as a regionalcharacteristicof Cretanscript.
KATO
SYME
75
IMPLICATIONS
The sanctuaryat Kato Syme occupies an important place in recent discussions about ancient settlement and the territorialconfigurationof the Cretan poleis in the Late Archaic and Classical periods. On the basis of purportedly negative findings from Aphrati and Kato Syme, Viviers infers a process of Lyktian territorial expansion on a grand scale in the late 7th or early 6th century, involving the incorporation of territory belonging to Lyktos'sneighbors to the south (Fig. 1).74According to this view, Lyktian success came at the expense of the neighboring communities and their cults. Viviers entertains the possibility that the "gaps"in the archaeological record of numerous Cretan poleis and sanctuaries reflect the abandonment or economic decline of smaller sites as stronger neighbors consolidated power by expanding their territory.75This explanation builds upon ideas advanced earlierby Watrous, whose study of settlement patterns on the Lasithi plateau documents an extinction of small hamlets and rural sanctuariesat the end of the 7th century.He interprets this evidence as a potential index of the territorial ambitions of Lyktos.76According to Watrous, the destruction of Arkades (identified with the ancient site at Aphrati and mentioned by a 4th-century author) was part of an earlier offensive by the Lyktians to expand their territory further south.77 Ceramic evidence from Kato Syme and Aphrati now challenges this reconstruction insofar as it is premised upon a supposed remission of activity at both places. Finds of black-gloss pottery, predominately drinking cups, document continuity at Kato Syme during much of the 6th and the whole of the 5th century.They demonstrate that the religious festivities at the site, in particularthe venerable practice of ritual drinking, continued long after the practice of making bronze dedications had waned. This is not to say,however, that an expansion of Lyktian territoryin the direction ofAphrati could not havetakenplace as earlyas the 6th century.Forinstance, a conceivablymore benign policy of Lyktian expansion might have stopped short of the outright destruction or total economic eclipse of its southern neighbors. Aphrati may have become a political dependency of Lyktos in the Archaic period. Not a shred of positive evidence unequivocally supports a 6th-century expansion of Lyktos to the south coast of Crete. While this is not the place for a full discussion of Lyktian territorial expansion, it is worth briefly considering the nature of the evidence upon which previous arguments have been based. A piece of evidence commonly adduced in support of a hostile takeover of Aphrati at the end of the 7th century, the supposedly Lyktian letter forms of the inscribed bronze armor found at Aphrati, has little foundation. Antony Raubitschek first proposed the idea that this cache of bronze armor, allegedly looted from Aphrati, was originally set up as Lyktian spoils of war in newly acquiredterritory.78Yet the sole basis for this intriguing suggestion, an inscribed omega with a double circle, is an orthographic detail that, as Raubitschek himself rightly cautions, may well be a regional characteristicof Archaic Cretan scripts, not a hallmark of Lyktos alone.79Given the remaining uncertainties about Archaic Cretan scripts, it is equally possible that the inscribed bronze armor found at
76
BRICE
L. ERICKSON
Aphrati originated there and commemorated one of the city's own victories over a defeated neighbor.As for the literarytestimony, although Theophrastus in the 4th century refers to an earlier destruction of Arkades, he does not specifically attributethe destruction of the city to Lyktos, nor has the traditionalidentificationofArkades with Aphrati stood unchallenged.80 A stronger case can be made for Lyktian control of the Lasithi plain, tentatively suggested both by the commanding position of Lyktos along one of its natural arteries of communication and by the supposed abandonment of rural settlements on the plateau in the 6th century.81The acknowledged difficulty of recognizing 6th-century Cretan ceramics warrantsextremecaution in the interpretationof these surveyresults.82Lyktian aggression has also been entertained as a possible explanation for the troubles plaguing Knossos in the 6th century, when for some time (ca. 600-525) Knossos seems to have been severely depopulated, if not completely abandoned.83If warfare is a valid explanation for the decline of Knossos, Lyktos is a possible culprit, but given the lack of decisive evidence it is best to refrain from making a final judgment. At any rate, hypothesized instances of Lyktian aggression against Knossos or settlements on the Lasithi plain around 600 need not imply a contemporary defeat of Aphrati and expansion of Lyktian territoryto the south coast. The rise of Lyktian power may have proceeded at different stages in different areas. The evidence from each part of Crete should be allowed to speak for itself. In considering the historical implications of the new ceramic evidence availablefrom my study,I divide the discussion into four parts. In the first section, I trace an evolving dedicatory practice characterized by an increasingemphasis upon pottery and a declining use of bronzes at the sanctuary after ca. 600. Second, I address central questions raised by the study concerning the origin of the visitors, and by extension, the control of the sanctuary.Next, new evidence from Kato Syme is presented that points to a date of about 400 for Lyktian territorialexpansion to the south coast of Crete. In the final section, I propose a historical context for Lyktian expansion in the aftermath of the Peloponnesian War. 80. For the literarytestimony,see above,n. 77. Viviers'proposed identificationof Aphrati as the site of ancientDattalla removesthe conjecturedlink betweenTheophrastus's testimony and the supposedabandonment of Aphrati. 81. See Watrous1982, pp. 22-26. Spratt(1865, vol. 1, p. 112) noticed the advantagesof Lyktos'sposition in terms of maintainingcontrol over the Lasithi plateau. 82. I have not examinedthe materialfrom the surveyof Lasithi. Publishedexamplesinclude two cup bases of suspected6th-centurydate:a base with an articulatedfoot and beveled edge (Watrous1982, p. 82, no. 70, fig. 12, which he dates to the 5th century)and a second base with a
plain foot and beveled edge (Watrous 1982, p. 82, no. 71, fig. 12, pl. 20:b,which he dates to the Classicalperiod).The buff fabricand fine blackgloss of both examplessuggest a non-Lyktiansource. 83. Hood and Smyth (1981, p. 19) and Coldstreamand Huxley (1999, pp. 301-302) addressthe possibilityof a LyktianattackagainstKnossos.This suggestionis supportedby indirect literarytestimony.Pausanias(2.21.3) refersto a war between Spartaand Knossos,set in the time of the Cretan seer Epimenides.In connectionwith this, the employmentof Lyktianarchers by the Spartansin the conflicts of the Second MessenianWar (Paus.3.12.11) has been construedas evidenceof a militaryalliancebetween Spartaand Lyktosin the 7th century.
APHRATI
AND
KATO
SYME
77
CULT PATTERNS
84.As hasbeenrecentlyemphasized by Kanta's(1991) study of the Minoan and Iron Age ceramics. 85. See Schiirmann1996. 86. See Lebessi 1985, p. 222. 87. Morgan (1990, pp. 28-29), in her study of the pottery from Delphi and Olympia, emphasizesthat it is not alwayseasy to distinguishbetween a utilitarianand dedicatoryfunction in the case of potteryleft at ancient Greek sanctuaries. 88. Snodgrass1989-1990. 89. See Schmaltz 1980, pp. 113, 164.
From study of the sanctuary'stwo principal categories of bronze offerings, including the series of zoomorphic figurines and cutout plaques, it is clear that neither type played an appreciable role at the sanctuary after about 575. How, then, do we explain the decline of metal offerings given the persistence of pottery at the sanctuary?I consider two possible explanations for the pattern. First, the ebb and flow of metal offerings at Kato Syme can be interpretedwithin a context of evolving ritual practice.While the basic characterof worship at Kato Syme, with its emphasis upon openair burnt animal sacrifices and the attendant consumption of wine, probably never underwent substantial modification during the long life of the sanctuary,84the precise ways in which religious piety was expressed may have been influenced by changing fashions. An interlocking chronological pattern arises from a consideration of the corresponding decline and rise of the two categories of bronze offerings at the sanctuary.Zoomorphic bronze figurines predominate in the archaeological record from the 10th to the first half of the 7th century,after which they are almost completely absent.85The timing of their disappearancecoincides with the introduction of the bronze cutout plaques, whose steady increase during the 7th century contrasts with the pattern of declining zoomorphic dedications. The bronze plaques, in turn, begin to dwindle in number during the first quarterof the 6th century,precisely when an economic decline is thought to have affected the sanctuary.A mere thirteen plaques from the sanctuary can be dated to the 6th century, as opposed to sixty-one examples in the 7th century.86 Against this backdrop of evolving dedicatory practice at the sanctuary, the special emphasis placed on pottery to the exclusion of all other categories of finds between ca. 575 and 400 takes on greater meaning. It may mark another substantial shift in the customs of the visitors, involving an almost complete substitution of pottery for metal goods and a new preference for utilitarian drinking equipment instead of symbolic offerings, although the lack of statistical data reduces the force of this argument. According to this view, to the extent that pottery served a dedicatory function at Kato Syme rather than a personal or utilitarian role, it substituted for the more costly metal votives of earlierperiods. Votive miniaturesattest unequivocallypottery'soccasional role as a dedication at Kato Syme, while a more practicalfunction best suits the evidence of plain cups presumably employed in ritual drinking.87Thus, the pottery forms a new continuum of material culture whose initial period of emphasis coincides with a decline in metal offerings. Historical causes may lie behind this apparentshift in dedicatorypractice. As Anthony Snodgrass emphasizes, the disappearanceof small metal votive offerings at Greek sanctuariesat the end of the Archaic period is a panhellenic phenomenon. It marks a change in material culture that, in his view, reflects an evolving religious outlook, a new preference for what he calls "converted"offerings ratherthan "raw"ones.88At severalmainland Greek sanctuaries,such as the Sanctuaryof the Kabeiroi in Boeotia, excavators have noticed that metal offerings are largely replaced by terracotta substitutes and vases.89The sanctuary at Kato Syme represents another clear instance of the substitution of pottery for bronze offerings, although
78
BRICE
L. ERICKSON
the discrepancy in date between the shift on the mainland (ca. 500) and the change at Kato Syme (ca. 600) arguably reflects somewhat different historical circumstances. A modification of the cult with a greaterimpact upon the characterof worship and the size of the gatherings may have occurred if Lebessi is correct in inferring from the scanty architecturalremains a contraction in the size of the sanctuary during the periods under consideration.90If so, the architecturalevidence may hint at a genuine downturn in the local economy, although this hypothesis posits an unsubstantiatedlink between the architecturalelaborationof a remote ruralsanctuaryand the economic resourcesof the surroundingarea.It is difficult enough to determine if the sanctuary becomes richer or poorer in absolute terms between 600 and 400; one might question whether it is even possible to connect a putative decline at the sanctuarywith a recession of the "state"economy. A second explanation for the change from bronze to pottery offerings at Kato Syme deserves consideration. Perhaps the decrease in metal offerings is connected with the availabilityof naturalresourcesnot readily obtainable on the island. Crete boasts of few known deposits of copper and none of tin.91Consequently, the island remained heavily dependent upon overseas sources to supply it with the raw materials necessary for bronzeworking. John Boardman, who recounts the achievements of the Orientalizing Cretan bronzeworkers, considers the paucity of 6th-century bronzework to be a reflection of economic and cultural decline.92Ernst Kirsten most fully developed the idea of Cretan isolation from the commercial, political, and military mainstream of ancient Greece.93His work has become the canonical view on the subject as epitomized by Gerald Cadogan's authoritative statement in TheAerialAtlas ofAncient Crete:"in the fifth century Crete seems to have been something of a backwater,on the whole undisturbed by the stirring events in mainland Greece and the rest of the Aegean."94
In an effort to explain the "inevitableCretan terminus"in the archaeological record around 600, SarahMorris revives an earlier commercial explanation for Crete's decline, first entertained by Pierre Demargne (who rejected it in favor of a general culturalfailure);she attributes the troubles to a collapse of the Cretan economy precipitated by military conquests in Syria and the reconfigurationof Near Eastern trade routes.95It is doubtful that the economy was devastated to the extent that Morris claims, given 90. Lebessi 1985, p. 222. 91. In contrast,Crete is comparatively rich in deposits of iron. Morris (1992, p. 151) suggests that iron was centralto Cretancommercialsuccessin the Iron Age. Rackhamand Moody (1996, pp. 14, 18) surveythe mineralresourcesof the island. 92. Boardman1961, pp. 141-146, 159;1982,p.230. 93. Kirsten1942, pp. 10-24, 63-67. 94. Cadogan 1992, p. 39. Morrow (1960, p. 17) expressesa similarsenti-
ment. The argumentfor commercialisolation finds apparentsupportin the archaeological record.A noteworthyabsenceof 6th-centuryimportsis the tentativepicture emergingfrom preliminaryexcavation reports.On the basis of a study of the ceramicrecordof Kommos,Johnston (1993, p. 377) concludes:"Ifthe negative pictureof sixth-centuryCrete remains afterfurthersites on the island have been investigatedand published,we will have to assumeisolation from the new'international'tradingcircuit."
95. Morris 1992, pp. 170-172. See also Demargne 1947, pp. 214-225. Dunbabin (1952, p. 195) takes issue with the contention that Cretantradewas divertedto other channelsafterNebuchadnezzar'sconquestof Syria.He points out (pp. 195-196) that there is little evidencefor connectionsbetween Crete and Syriain the Orientalizing period.Yet Dunbabin fails to consider the possibilitythat developmentsin Syriahad an indirecteffect upon Cretan trade.
APHRATI
AND
KATO
SYME
79
the evidence from local ceramic sequences that settlement continued in the larger Cretan poleis.96Moreover, it is inherently implausible that the local economies of the various Cretan cities, dependent as they all presumably were on an agriculturalbase, would have experienced such a devastating blow from losing access to Eastern markets. Indeed, a recent trend in scholarshipis to doubt whether the gaps in the Cretan archaeologicalrecord directly reflect demographic or historical realities.97 Be that as it may, a reorientation of overseas trade patterns at the end of the 7th century has much to recommend it as an explanation for fluctuations in the availabilityof raw materials not availableon Crete and the removal of Orientalizing exotica from local elite circulation. Prolonged isolation from overseas trade goods and raw materials may have, in turn, influenced local taste-the two explanations are not mutually exclusiveleading to an even greater emphasis upon local pottery at the expense of imported bronze. The interplayof severalfactors ratherthan a single cause might be expected to create a clearerpattern in the archaeologicalrecord. KATO SYME AND THE SETTLEMENT
AT APHRATI
Another important question raised by the operation of cult at Kato Syme involves the identity of the worshippers.The Late Archaic and Classical pottery from the sanctuary is overwhelmingly local Cretan, and the majority of it apparently derives from a single production center. The predominant pale brown fabric at the sanctuarylinks the pottery to a production center at Aphrati, plausibly identified either as the ancient polls of Arkades or Dattalla.98The cumulative weight of this new evidence suggests that Aphrati was home to one of Crete's longest-lived production centers in the historical Greek period (ca. 800-400 B.c.). Moreover, contraryto previous opinion, the tradition of Orientalizing pot painting does not die out at Aphrati at the end of the 7th century.99A glimpse at pottery from the 5th-century deposit at Aphrati reveals a continuation of this tradition, employing a restricted repertoire of Orientalizing motifs (now limited to stacked triangles, rows of dots, tongue patterns, and lobes) accompanied by simplified execution and arrangementof the motifs. 96. See Erickson2000. 97. Rizza (1967-1968, p. 298) anticipatedcurrentthinking about the problemof the 6th-century"gap"; regarding6th-century Cretanterracotta workshops,he concludesthat the apparentdecline in productionmay be due to "lanostraconoscenzadei materiali."Callaghan(1992, p. 133) first entertainedthe possibilitythat the problemof the 6th-centurygap stemmed from the difficultyof distinguishing Late Archaic ceramicstyles from those of their Orientalizingcounterparts.Prent (1997) hypothesizes "lingeringstyles"to explainthe apparentabsenceof 6th-century
potteryboth at Knossos and within the wider Cretancontext. Kontoleon (1970, pp. 86-87) attributesthe apparentdecline of Crete to a conservativeArchaic society which, it is claimed,prevented the emergenceof a true polis founded upon a citizen hoplite army.Cf. Whitley (1997, p. 659), who assertsthat the failureof an aristocraticclass to develop and commemorateits deeds, rather than demographicrealities,lies behind the apparentdecline of 6th-century Crete and its inabilityto leave a "lasting trace that the archaeologistor historian can recover."See also Morris 1998, p. 68. 98. Either identificationwill suit the
proposedreconstructionof events. By virtue of its location, the ancient city at Aphratiwould presumablyhave resistedLyktianexpansionto the south coast. 99. In his seminalstudy of the Orientalizingceramicand bronzeworking traditionsat Aphrati,Levi (1945, p. 18) wistfully and poetically concludesthat in the "ceramicproducts, as well as in contemporarybronze objects of the second half of the 7th century
B.C.,
we see the last flight of
imaginationof the old civilizationof Crete before it settles into the darkness of its exhausted,lethargicsleep."
8o
BRICE
L. ERICKSON
The discovery of imported pottery at Tocra and Cyrene, apparently made of the same pale brown fabric as the Aphrati material, provides another indication of the vitality of the local Cretan tradition.?00 An identification of Aphrati as the source of this pottery also makes sense in terms of the island'sgeography,since Aphrati lies a short distance from the southern coast of Crete on a more or less direct line across the sea from Libya. Chemical analysis of the Cretan pottery found at Tocra has confirmed its central Cretan origin.'?l Moreover, the base of one of the exported drinking cups from Tocra matches the profile of fragmentary cup bases from Kato Syme.102The finds from Tocra and Cyrene constitute the only identified instances of Cretan pottery exported overseas in the 6th century; as such, they form a potential basis for assessing the strength of Cretan longdistance commercial ties in the Late Archaic period.103 My hypothesized ware groups and estimates of provenancelead to the conclusion that the vast majority of the pottery left at Kato Syme during the periods under considerationwas manufacturedby potters from Aphrati. Whether local potters set up stalls at festival time and sold their wares directly to visitors, as Catherine Morgan suggests may have been the case at Olympia and Isthmia,104or worshippers procured cult equipment at Aphrati and transported it themselves to the sanctuary,is difficult to say. While it might be arguedthat other Cretansobtained pottery from Aphrati for use at the sanctuary,why would they go to this trouble on such a consistent basis? Pottery from Aphrati did not circulatewidely in Crete. On present evidence, only sporadic exchange of local pottery took place between neighboring communities. Thus, a special explanation would be needed to account for the overwhelming preponderance of pottery from Aphrati at Kato Syme between 600 and 400 if normal mechanisms of trade are assumed. A third possibility-the manufactureof pottery at the sanctuaryitself-finds no support in the archaeologicalrecord, either directly in the form of excavatedkilns or indirectly in the form of standardized equipment at the sanctuary. For a variety of reasons, it seems safe to conclude that the preponderance of pale brown pottery at Kato Syme at this time reflects substantial activity at the sanctuaryby visitors from Aphrati.l05The only other visitors to the sanctuarywhose presence can be detected in the surviving archaeo100. See Boardmanand Hayes 1966, p. 78; Coldstream1973b, p. 47, n. 23; Schaus 1985, p. 10. 101. See Coldstream1973b, pp. 4647. 102. Also worth mentioning in this context is the painted decorationof 6th- and 5th-centurypottery from Kato Syme (2,18, 82). Painted decorationis a rarityon pottery from Cretan productioncentersbut commonplace among the Cretanfinds from Tocra and Cyrene.
103. Unless, that is, these Cretan "exports"reflectnot tradebut the belongings of colonial Greek settlerswho broughtpottery and other personal belongingswith them on the voyage. While there is no directevidenceof Cretanparticipationin the colonization ofTocra, Herodotos (4.161) informsus that the Cretanssent a contingent of settlersto Cyrenaica,anotherLibyan colony,in the second generationafterits foundation,an event to be dated in the first half of the 6th century.Cretan
pottery at Cyrene andTocramay well indicatethe presenceof Cretancolonists; see Boardman1980, pp. 122-125; Treister and Shelov-Kovedyayev1989, p. 295. 104. Morgan 1990, p. 124. 105. Viviers (1994, p. 256) argues that the sanctuaryat Kato Syme belonged within the territoryof Biennos in the Archaicperiod.More work needs to be done to define the local ceramictradition of that polis. On currentevidence, potteryfrom Biennos cannot be identified at Kato Syme.
APHRATI
AND
KATO
SYME
logical remains (109, 110) hail from Gortyn, one of the chief poleis of the neighboring Mesara plain (Fig. 1). A later Gortynian pilgrim left an inscribed Hellenistic dedication as a token of his visit.106On the strength of this later testimony it seems reasonableto connect individual finds of Late Archaic and Classical Gortynian pottery with earliervisits, but this interpretation is by no means mandatory. The provisional historical picture presented here, dependent chiefly upon my study of the pottery, indicates a small rural sanctuaryunder the political control of the principal nearby polis, attractingvisitors from further afield rarely,if at all. Epigraphical sources do little either to confirm or reject the hypothesized local origin of the worshippers at Kato Syme. The only contemporary epigraphic evidence for a specific presence at the sanctuaryis an inscribedbronze handle of a 6th-centurybowl, which records the signature of a craftsman from Dattalla.107If Viviers' proposed identification of Aphrati as the site of ancient Dattalla is correct, this piece of evidence might substantiate a link between Kato Syme and Aphrati, but since the signaturebelongs to an artisan,not a visitor, the case is considerably weakened. On the other hand, an intact Late Orientalizing ring vase from Kato Syme with demonstrable parallels in shape and decoration to funeraryequipment at Aphrati provides compelling evidence for a visitor from Aphrati in the 7th century.l08Moreover, excavations at Aphrati have yielded bronze cutout plaques identical in type to those left in large numbers at the sanctuary.109 These finds raise the possibility that Aphrati, long of the rich hoard of 7th-century bronze aras the source acknowledged was the seat of an extensive mor, bronzeworking industry, whose other products may have included the cutout plaques in vogue at the sanctuOn the whole, this evidence strengthens my hypothesis of 6th-cenary.110 visitors from Aphrati by providing a possible precedent for the protury posed later activity.
106.The Hellenisticdedications fromKatoSymeawaitfinalpublication.Theseinscriptions, the majorityof which were scratchedon the sides of pots or on pieces of tile, frequently recordthe ethnic and name of the dedicant,includingvisitorsfrom Lyktos,Knossos,Tylissos,Hierapytna, Priansos,and Arkades.I expressmy thanks to Angeliki Lebessi for allowing me to examine this body of evidence. 107. For the inscription,see Lebessi 1975b, p. 191, pl. 193:g;Viviers 1994, p. 240. 108. See Kanta1991, p. 501, fig. 38. 109. The bronze cutout plaques from Aphrati are illustratedby Levi 1927-1929, pp. 28, 30, figs. 8-9. Boardman(1961, pp. 46-49, 142) and
Hoffmann (1972, pp. 32-33) discuss finds of similarbronze cutout plaques found elsewhereon Crete. 110. Hoffmann (1972, p. 30) notes that Knossos,Gortyn, and Aphrati are "clearlyof centralimportanceas major schools"for bronzeworking.Boardman (1961, p. 142; 1982, pp. 230-232) discussesthe cache of bronze armorfrom Aphrati.Hoffmann (1972, pp. 32-33) noticed a stylistic affinitybetween the smallerbody of plaquesthen in existence and the bronze armorfrom Aphrati.Kato Syme exhibitsby far the greatestconcentrationof these plaques, but they appearsporadicallyat other Cretansanctuaries,as for instance at the Diktaian Cave;see Boardman1961, p. 142.
82
BRICE
TERRITORIAL
EXPANSION
L. ERICKSON
OF LYKTOS
One expected outcome of the conflicting territorialclaims of rivalingGreek poleis is that their cults, and particularlythe rural sanctuariessituated on their borders, might become contested areas and symbolic battlegrounds in the settlement of territorialdisputes. Francois de Polignac has amassed an extensive body of evidence in support of his argument that rural cults played an important political function by serving as territorialmarkersin While the context of the emerging Greek poleis of the 8th century B.C.111 de Polignac's reconstruction of ancient Greek cult practice has not gone unchallenged, it has proved to be a highly fruitful way of exploring the relationship between cult activity and the construction of civic identity in the earlyGreek poleis.ll2Angelos Chaniotis arguesthat Cretan cults served a similar purpose in the historical Greek period. Indeed, the capacity of Cretan cults to serve as markersof territorywas not lost even upon Hellenistic participants,whose inscribed arbitrationsof border disputes in the 3rd and 2nd centuries B.C. frequently mention sanctuaries at contested Chaniotis proposesthat the sanctuaryof Hermes points along the frontier.113 and Aphrodite at Kato Syme itself became a territorialmarkeron the frontier between two powerful Hellenistic rivals,the poleis of Hierapytna and Lyktos.114
What role might the sanctuaryat Kato Syme have played during earlier territorial disputes for which we have no epigraphic documentation? My study of the two Cretan production centers whose products are attested at Kato Syme reveals a fundamental change in the composition of Cretan pottery at the sanctuary ca. 400-390, when Lyktian shapes replaced pottery of the distinctive pale brown fabric of Aphrati. The change is abrupt and complete. One way of interpreting this evidence, given the proposed association of local pottery at the sanctuarywith visitors from Aphrati, is to infer a cessation of pilgrimages from Aphrati after about 400 B.C., when the sanctuary may have been taken over by the Lyktians. A notice preservedby the ancient geographer Pseudo-Skylax furnishes a further piece of evidence pertaining to Lyktian expansion. In his geographic itinerary of the Cretan cities, Pseudo-Skylax describes Lyktos in the following manner:?v [ayooyEia 8' A6xzog, xczi8V6xrt accX aipoT?potico0Ev.115 Pseudo-Skylax's testimony suggests that Lyktos gained territory on the south coast of Crete by the middle of the 4th century.ll6 111. See de Polignac 1995, pp. 3341. 112. Critics point to a greater degree of regionalvariationin ancient Greek cult practicethan admittedby de Polignac;see Carter1994, pp. 180-183. 113. Chaniotis (1988) furnishes examples of Cretan cults that served as
territorialmarkersas indicatedby epigraphic,archaeological,and literary evidence. 114. Chaniotis (1988, p. 33) argues on the basis of the location of the
sanctuaryand the existenceof ethnics that recordthe names of Hellenistic visitorsfrom Lyktos and Hierapytna that Kato Syme became a contested frontiersanctuaryin the Hellenistic period.Viviers (1994, p. 256, n. 157) reachesthe same conclusion.The ceramicevidencesupportsthe conclusion that the sanctuaryat Kato Syme grew in popularityto become a regionalcult center duringthe Hellenistic period.The Hellenistic pottery from Kato Syme has been assigned
to Jonas Eiring for publication. 115. Pseudo-Skylax47. Viviers (1994, p. 253) drawsthe same conclusion from this passage,namelythat Lyktoswon control over territoryon the south coast of Crete. But he dates the event ca. 600, long before the earliestevidence,literaryor otherwise, for Lyktianaggression. 116. Viviers (1994, p. 253, n. 138) discussesthe dating of Pseudo-Skylax's itinerary.
APHRATI
AND
KATO
SYME
83
If Lyktian encroachments precipitated a conflict with Aphrati, it is reasonableto infer that the important ruralcult center at Kato Syme would have come under pressure.On the other hand, had Lyktos taken an interest in the sanctuary as early as 600 B.c., the date suggested by Viviers, its pottery might be expected to turn up at Kato Syme, if only in minute quantities.'17As it is, there is no indication whatsoever of Lyktian visitors or Lyktian products reaching the sanctuarybetween 600 and 400. When Lyktian pottery suddenly appears at Kato Syme shortly after 400, it is completely without precedent. Moreover, it totally replaces the earlierpale brown pottery of Aphrati. The observed absence of Lyktian ceramics at Kato Syme before this date raises the possibility of the exclusion of Lyktos from the rites of Hermes and Aphrodite. Another possibility warrants attention. The sudden appearance of Lyktian pottery at the sanctuary arguably represents a response to the growth of a rival polis, Hierapytna, whose conjectured expansion of territory into the hills above Biennos would have presented a threat to Lyktos. If Aphrati did fall under the political sway of Lyktos during the 6th century, as Viviers suggests, but continued to send visitors to the sanctuary until the end of the 5th century,then the change from pale brown to gritty red fabric would require a different interpretation. Rather than marking the point at which Lyktos established political authority over Aphrati, the change in ceramic composition at Kato Syme might imply a more advanced stage of Lyktian territorial expansion. Kato Syme may have first attractedLyktianvisitorsas a counterto Hierapytnianexpansion.The earlier absence of Lyktian pottery at Kato Syme, according to this view, signifies Lyktos's disinterest in its holdings on the south coast of Crete, a malaise broken by Hierapytnian pressure. Lyktian pottery found at Kato Syme after ca. 400 is thus seen as a valuable source for historical reconstruction, while the pottery from Aphrati dating between 600 and 400 is assumed to be unrelated to the political control of the sanctuary. Both interpretations are based on assumptions about religious practice as a valid expression of political control, and they take for granted a correlationbetween the origin of the pottery and the people who left it at the sanctuary.It would bolster either case to be able to cite parallelsfor the relationship between pottery and the political control of sanctuaries,but unfortunately little comparative evidence has been collected from elsewhere in the Greek world. What little work has been done has focused primarilyupon panhellenic sanctuarieswhere the presenceof ceramicsfrom a wide variety of sources complicates interpretation.18 Morgan and Whitelaw identify Argive pottery at the Heraion during the 8th and 7th centuries B.C.as one manifestation of the emergence of Argive hegemony 117. The observedabsenceof Lyktianpotterybefore ca. 400 B.c. does not, of course,arguedecisivelyagainst an earlierLyktianpresenceat the sanctuary.Too little is still known about the social, cultural,and economic dimensions of pottery consumptionin this case. For instance,if it could be
shown that Lyktianwares did not possess social or symbolicvalue in the venue of public displaybefore 400, the potterywould not be expectedto turn up at Kato Syme. This issue requires furtherconsideration. 118. A particularproblemis how to interpretthe absenceof pottery at
panhellenicsanctuariesfrom production centerswhose citizens are known to have participatedin the festival.For instance,Morgan (1990, p. 53) puzzles over the absenceof Corinthianpottery at Olympia before ca. 675 B.c., a date long afterthe earliestattested Corinthian participationin the festival.
84
BRICE
L. ERICKSON
in the plain.19 Insofar as the Argive Heraion was a ruralsanctuaryassociated with a specific rite of passage, its position may be analogous to that of the sanctuaryof Hermes and Aphrodite at Kato Syme.'20 Given the paucity of examples of alleged political control of Greek sanctuariesand the lack of a theoretical foundation from which to assess the relationship between pottery and political control, a decision between the two proposed explanations for the change in pottery composition at Kato Syme requiresfurther deliberation. A central question is whether or not the preponderanceof pale brown pottery at Kato Syme connotes ownership of the sanctuary. Both hypotheses posit a political consequence to the appearanceof Lyktian pottery, although the first interprets it within the compass of a territorial dispute between Aphrati and Lyktos, the second as part of a wider political confrontation between Lyktos and Hierapytna. The first explanation seems preferablefor severalreasons.The second hypothesis fails to explain why the community of Aphrati stopped participating in the cult after 400 B.C. Cult practice provides a means of demonstrating a united front to an external foe. Lyktos would presumably have had an interest in garnering the support of its allies in the face of growing Hierapytnian pressure. In addition, the hypothesis of Hierapytnian expansion implies increased importance for Kato Syme in the 4th century, on the assumption that it became a regional cult center on the border between two powerful rivals. My impression of the evidence from Kato Syme, however, is that the 4th century marks a period of decline, characterized by fewer offerings (bronzes disappearcompletely at this time) and less utilitarian pottery. Judging from the total volume of pottery left at Kato Syme, visits occurred on a more sporadic basis in the 4th century. In support of the alternativeexplanation of a conflict between Lyktos and Aphrati, additional signs of a change in ownership of Kato Syme in the 4th century buttress the idea of an enemy takeover of the sanctuary. The time-honored practice of making burnt offerings in the open air apOn the strength of this evidence, it seems reaparentlyfell out of favor.121 sonable to conclude that a fundamental change in worship took place at Kato Syme in the early years of the 4th century.Another potential sign of discontinuity in this part of Crete is the destroyed house, discussed above, in the settlement record of Aphrati.122While the excavated remains give no indication of direct Lyktianinvolvement in the destructionof this house, the timing of the event in the final years of the 5th century coincides with the cessation of offerings from Aphrati at Kato Syme, thereby suggesting
119. Morgan andWhitelaw 1991, p. 84. There is a wealth of evidenceto suggest that the Argive Heraion fell under the political sway of Argos in the Classicalperiod.Argos instituted and controlledthe cult, and a procession from Argos forged a physicallink between city and sanctuary. 120. Argive expansionprovides anotherpossible exampleof the use of
cult to legitimize territorialconquests. Archaeologicalevidence supportsthe traditionthat Argos destroyedAsine ca. 710 B.C.,althoughthe Temple of Apollo at Asine apparentlysurvivedthe destructionand continuedto attractworshippersto the ruined city.According to Morgan (1990, p. 11), the maintenanceof this principalcult of the defeatedpopulation suggeststhat "Argosreinforcedits
dominanceby taking over the Apollo cult and incorporatingit into the ritual system of the plain, and that the cult at Asine was maintainedas a reminderof the new statusquo."While plausible, this explanationlacks evidencein support of Argive control of the sanctuary. 121. See Lebessi 1985, p. 222. 122. For a preliminaryreportof the excavations,see Lebessi 1973.
APHRATI
AND
KATO
85
SYME
a connection between the two. More work needs to be done to determine the extent of this destruction horizon: was it confined to a single house or the domestic quarters, or did it extend across the entire settlement?l23 Despite these remaininguncertainties,the existing archaeologicalevidence raises the possibility of a destruction of Aphrati by Lyktos accompanied by the replacement of cult activity at a principal shrine in the newly conquered territory with a Lyktian presence. Lyktian activity at Kato Syme perhaps served not only to legitimize territorialconquests and enforce the status quo with Aphrati, but also to stake a claim in the newly created border between Lyktos and Hierapytna. LYKTOS
AND
THE
GREEK
WORLD
The archaeologicalsources identify the end of the 5th century as a turning point in the history of Lyktos and Aphrati, a time of political and military upheaval. It is worth speculating about what might have triggered a potentially more aggressive policy of Lyktian territorial expansion around 400 B.c. The timing of these events, as determined by the study of the ceramic sequences of Aphrati and Lyktos, coincides with a major event in Greek history, the defeat of Athens at the end of the Peloponnesian War in 404 B.C.and the loss of its overseas empire. The eclipse of Athenian hegemony in the Aegean and the ascendancy of Sparta profoundly upset the balance of power in many parts of the Greek world, in manifold ways. The clearestdocumented instance of Spartaflexing its power on the Mediterraneanstage in the immediate aftermath of the Peloponnesian War involved the toppling of the democraticconstitutions of inimical Greek poleis in favor of oligarchic rule headed by pro-Spartan executive committees. Literary sources, chief among them Xenophon, emphasize Spartan involvement in the affairs of the Greek cities of Asia Minor upon the conclusion of the Peloponnesian War. In the case of Lyktian expansion in 400-390, a combination of indirect evidence urges consideration of tacit Spartan approvalor even direct military assistanceon behalf of the Lyktians as a contributing factor.Greek tradition held that Lyktos was one of Sparta'scolonial foundations, and regardless of the historical merits of the claim, there seems to have been definite substance to the relationship between Lyktos and Sparta in the Archaic and Classical periods.l24Moreover, a later documented instance of Spartan military intervention on Crete on behalf of Lyktos in 343/2, the year in which Archidamos of Sparta led a force to Crete in support of evidencefrom 123.The epigraphic the site remainsequivocal.After a long hiatus of approximately200 years,inscriptionsagain appearat Aphrati in the middle of the 3rd centuryB.C.;see Guarducci1935, pp. 6-28. Yet such a gap, even of this duration,is not unparalleledamong the epigraphic recordsof Classicaland Hellenistic Cretanpoleis. Alexiou (1968, p. 406, pl. 435:a) illustratesan alabasterpyxis
from a pithos burialfrom Aphrati.It is the sole evidencefor 4th-centuryoccupation at the site. If my proposeddate for the pyxis in the final quarterof the 4th centuryis correct,the tomb may signify a Hellenistic recoveryof the town. 124. Malkin (1994, pp. 78-80) examinesthe traditionof Spartan colonization.Coldstreamand Huxley's (1999, p. 297) proposalthat Spartasided with Lyktos in a war againstKnossos
around600 B.C. tends to supportthe tradition.Malkin (1994, p. 80) expresses reservationsabout the purportedSpartan foundationof Lyktos. Since our main sourcesfor this, Ephoros (FGrHist70 F 147-149) and Aristotle (Pol. 1271b.2829), wrote their accountslong afterthe events had taken place, it is easy to see how the traditioncould have been manufacturedto supportSpartanmilitary activityon Crete in the 4th century.
86
BRICE
L. ERICKSON
the Lyktians against Knossos,125raises the possibility of direct military intervention at an earlier unattested point in the island's history. To be sure,the earliest episodes of recorded Cretan history concern foreign military expeditions to the island. Herodotos (3.59) relates that the Samians and the Aeginetans fought for control over the colony of Kydonia in 519. In addition, Thucydides (2.85.5) reveals that the Athenians returned to Kydonia a century later to meddle in West Cretan affairs.Furtherstudy of the Cretan archaeologicalrecordhas the potential to illuminate other possible instances of foreign military intervention. The fates of Lyktos, Aphrati, and Kato Syme at the end of the 5th century call into question the long-held tenet of Classical Cretan history that prior to the great foreign military expeditions in the 340s B.C. the island was a backwater in the major currents of history, left to pursue, in G. L. Huxley's words, "aseparatedevelopment in secure insularity."l26 The of a territorial of one of Crete's Classical major timing expansion premier poleis suggests otherwise. The major events of mainland Greek history may have helped shape the course of internal Cretan developments, although the precise ways in which these influences were felt on the island remain obscure.
CONCLUSION Ceramic evidence from Aphrati and Kato Syme fills a substantialvoid in the publication record of post-Minoan Cretan archaeology.The Late Archaic and Classical periods have constituted an almost complete blank in the history of the island's ceramic development, owing to the lack of attention to post-Minoan sites and the absence of an established ceramic classification system and typology. Knossos stands apart as the only Classical Cretan site where archaeologists have attempted to formulate a partial chronological sequence, although efforts to narrow,let alone close, the 6th-century gap in the ceramic sequence have thus far met with limited success.127Now Aphrati may also take its place as a Classical Cretan production center,thereby allowing an analysis of ceramic forms within a second region. In addition, a continuous series of cups and jugs from Kato Syme spanning the years between 600 and 400 B.C. provides an important correction to the impression of an island-wide gap in ceramic production in the 6th century. The absence of identifiable archaeological evidence from Knossos between ca. 600 and 525 has previously led historians to conclude that most Cretan poleis suffered a demographic or cultural decline of unprecedented proportions.128New evidence for the continuous manufactureof pottery at a site in East Crete (probablyAphrati) indicates that this presumption of island-wide abandonment is inaccurate. Historical speculation regarding the cause of the supposed settlement gap or recession should await concrete evidence that comparablebreaksexist elsewhere in the Cretan archaeologicalrecord. The above formulation of a ceramic sequence for Aphrati and Kato Syme will serve as a partial framework for incorporating future discoveries, inviting new avenues of historical inquiry.Preliminaryresults suggest that the ruralsanctuaryat Kato Syme was linked at first to the ancient city
125. Diod. Sic. 16.62.3-4. Perlman (1992, p. 200, n. 39), Callaghan(1992, p. 134), and Huxley (1994, p. 132) discussthis passage. 126. Huxley 1994, p. 132. 127. Coldstream's(1973b) publication of a Late Archaicwell deposit (RR:H) sheds valuablelight on Cretan developmentsca. 500-480 B.C. Callaghan(1978) complementshis work by publishinga selection of 5thcenturyshapesfrom an unstratified deposit from the Classical"Shrineof Glaukos."By contrast,6th-century Knossosis devoid of identifiable pottery,the only exceptionbeing a group of cup bases characterizedby a "stepped"profileunderfoot,a type dated by Callaghan(1992, p. 92) to ca. 525-500. 128. Coldstreamand Huxley (1999, pp. 289-301) summarizethe evidence from Knossosand examinepossible explanationsfor the apparent6thcenturyrecession.They limit their discussion to Knossos and sites within its territory.Aphratihas figuredprominently in earlierdiscussionsof 6thcenturyCretandecline. Brock (1957, p. 219), for instance,interpretsthe mortuaryrecordat Knossos and Aphrati as a directreflectionof population levels in his concludingremarks in the publicationof the Fortetsacemetery,Knossos:"How can one account for the sudden eclipse at the end of the seventh centuryof the flourishing communitiesround Knossos?Arkades, furtherinland,remainedprosperousa little longer,but by the sixth century the whole of Crete seems to have become affectedby the same paralysis."
APHRATI
AND
KATO
SYME
87
nearAphrati and then, from 400 B.C.onward,to the rising power of Lyktos. Subsequent Lyktian activity at Kato Syme served to legitimize territorial conquests and enforce the new political status quo with Aphrati.
ACKNOWLEDGMENTS An earlier version of this paper was presented at the Annual Meeting of the Archaeological Institute of America in San Diego in January 2001. The financial support of a Fulbright Fellowship (1997-1998), the Bert Hodge Hill Fellowship of the American School of Classical Studies at Athens (1998-1999), a University of Texas Fellowship (1999-2000), and researchgrants from the 1984 Foundation (summer 1999, fall 2000) made possible my program of research. I owe a debt of gratitude to Angeliki Lebessi, Polly Muhly, and Athanasia Kanta for permission to publish the Late Archaic and Classical pottery from their excavations at Kato Syme and Aphrati. All three gave generously of their time and expertise in the interpretation of the results.This project could not have been completed without their help. Alexandra Karetsou, director of the Herakleion Museum, graciously made the pottery from these sites available for study, drawing, and photography.The author is responsible for the profile drawings and photographs.I extend my warmest thanks to Antonino Di Vita, BarbaraHayden, Nikolaos Stampolidis, Peter Warren, and L. V. Watrous for permission to mention unpublished finds from their respective excavations or surveys at Gortyn, Vrokastro,Eleutherna, Knossos, and Gournia. My debt to Didier Viviers extends not only to his kind invitation to examine unpublished pottery from Itanos, but also to his generosity in discussing this material and the historical problem of 6th-century Crete. Athena Tsingarida and Maria Englezou helped resolve questions concerning the pottery from Itanos and Lyktos. Nicolas Coldstream, Lisa Kallet,JackKroll,and Cynthia Shelmerdine read earlier drafts of this article in the form of dissertation chapters.Their suggestions inspired numerous changes and saved me from countless errors. Paula Perlman, the chair of my dissertation committee, lent a critical eye to early drafts and advised in matters large and small. I am profoundly grateful. My thanks also go to Tracey Cullen, Tracy Dickinson, Donald Haggis,Jerry Rutter,and the anonymous reviewersof Hesperia,whose suggestions improved the final draft beyond measure. The INSTAP Study Center provided an excellent base of operations for my activities in East Crete. I owe thanks to Tom Brogan, the director of the center, for his tireless efforts on my behalf. Above all, I wish to thank the people of Crete who so unfailingly extended hospitality to a strangerand offered constant good cheer. I am pleased to offer something, however inadequate, in return for their generosity.The natural attractions of the island inspired an earlier visitor, J. D. S. Pendlebury, to write (The Archaeologyof Crete, p. xxix): "To have stood on Ida, on Dikte, and on Aphendes-Kavousi in the clear shrill wind and to have toiled through the hot little valleys with that unforgettable smell of herbs is an experience the memory of which nothing can ever take away from you." I wholeheartedly agree.
88
BRICE
L. ERICKSON
REFERENCES AgoraIV = R. H. Howland, Greek Lampsand TheirSurvivals, Princeton 1958. AgoraXII = B. A. Sparkesand L. Talcott,Blackand Plain Pottery of the 6th, 5th, and4th CenturiesB.C., Princeton 1970. xal Alexiou, S. 1968. "'ApXratorL6rc [JV.1[EL.La XevTpLXYSxoctavaroXLtxYJ
Kprn;," ArchDelt21, 1966, B'2 [1968], pp. 405-411. A .1969. "XpovLxaC: t 1967," xaot-a KpTYrS &apXat6orqTs CretChron 21, pp. 533-543. Amyx, D. A. 1988. Corinthian VasePainting of theArchaic Period,
Berkeley. Arnold, D. 1985. Ceramic Theory and Cultural Process,Cambridge.
Boardman,J.1961. TheCretan Collection in Oxford, Oxford. .1980. The GreeksOverseas, 3rd
ed., London. 1982. "Crete,"in CAH 111.3: The Expansion of the Greek World, 8th to 6th Centuries B.C.,J. Boardman
and N. G. L. Hammond, eds., Cambridge,pp. 222-233. Boardman,J., andJ. Hayes. 1966. Excavations at Tocra,1963-1965: The Archaic Deposits I (BSA
Suppl.4), London. . 1973. Excavations at Tocra, 1963-1965: TheArchaic Deposits II and Later Deposits (BSA Suppl. 10),
London. Brock, J. K. 1957. Fortetsa:Early Greek Tombsnear Knossos, Cambridge.
Cadogan,G. 1992. "Ancientand Modern Crete," in The AerialAtlas of Ancient Crete,J. W. Myers,
E. E. Myers, and G. Cadogan,eds., Berkeley,pp. 31-43. Callaghan,P.J. 1978. "KRS1976: Excavationsat a Shrine of Glaukos, Knossos,"BSA 73, pp. 1-33. . 1992. "Archaicto Hellenistic Pottery," in Knossos,from Greek City to Roman Colony:Excavations at the UnexploredMansion 2 (BSA
Suppl.21), L. H. Sackett,ed. London, pp. 89-136. Callaghan,P.J., and R. E. Jones. 1985. "HadraHydriaeandCentralCrete:A FabricAnalysis,"BSA 80, pp. 1-17.
Carter,J. C. 1994. "Sanctuariesin the Chora of Metaponto,"in Placing the Gods:Sanctuariesand SacredSpace in AncientGreece,S. E. Alcock and R. Osborne,eds., Oxford, pp. 161198. Chaniotis,A. 1988. "Habgierige Gotter,habgierigeStadte:Heiligtumsbesitzund Gebietsanspruchin den kretischenStaatsvertragen," Ktema13, pp. 21-39. . 1996. Die Vertrdge zwischen kretischen Poleisin derhellenistischen Zeit, Stuttgart. Coldstream,J. N. 1973a. Knossos:The SanctuaryofDemeter(BSA Suppl. 8), London. 1973b. "Knossos1951-61: Orientalizingand ArchaicPottery from the Town,"BSA 68, pp. 33-63. .1999. "Knossos1951-61: Classicaland Hellenistic Pottery from the Town,"BSA 94, pp. 321351. Coldstream,J. N., and H. W. Catling, eds. 1996. KnossosNorth Cemetery: Early GreekTombs(BSASuppl.28), London. Coldstream,J. N., and G. L. Huxley. 1999. "Knossos:The ArchaicGap," BSA 94, pp. 289-307. Coldstream,J. N., and C. F. Macdonald. 1997. "Knossos,Area of South-west Houses: Early Hellenic Occupation,"BSA 92, pp. 191-245. Demargne,P. 1947. La Cretededalique: Etudessurlesoriginesd'unerenaissance,Paris. de Polignac,F 1995. Cults,Territory, and the Originsof the GreekCityState,Chicago. Dunbabin,T. J. 1952. Rev.of P. Demargne,La Cretededalique,in Gnomon24, pp. 191-197. Eiring,J. 2000. "HellenisticPottery from Pyrgos at Myrtos,"in 2000: E' yta rnv ZouvadvTYnor ErrmtaIovtxY eivtaCICTxr xepacLtx : XpovoAoytxd rpopAtiraTa, xictarc ac6voca, epyacaTzpla, S. Drogou et al., eds.,
Athens, pp. 53-60. Englezou, M. 2000. "EXXYjvLCoLx xEpoaqtxiaott6zTlvoapXaoaA6TTo," in 2000: E' EmrcaTrovtxr aovadvTqvnr ytxaV'PV xspac.ltx7: cr.ivtrTtxr
XpovoAoytxa 7rpoplSfaza, xAeLtorr aovoAa, -pyaonptca, S. Drogou et
al., eds., Athens, pp. 61-68. Erickson,B. 2000. "LateArchaicand ClassicalCrete:Island Pottery Styles in an Age of Historical Transition"(diss. Univ. of Texas, Austin). . In press."HistoricalGreek Potteryfrom the Odeion, Gortyn," ASAtene 74-75.
Greco, E., et al. 1996. "Itanos(Crete orientale),"BCH 120, pp. 941-952. . 1997. "Itanos(Crete orientale),"BCH 121, pp. 809-824. . 1999. "Itanos(Crete orientale),"BCH 123, pp. 515-530. Guarducci,M. 1935. Inscriptiones Creticae,opera et consilio Friderici Halbherr collectae1: Tituli Cretae mediaepraeter Gortynios, Rome.
Guest-Papamanoli,A., and A. Lambraki.1980. "Lesgrottes de Leraet de 1'Arkoudiaen Cr&teoccidentale aux epoquesprehistoriqueset historiques,"ArchDelt 31, 1976, A' [1980], pp. 178-243. Haggis, D. 1996. "Archaeological Surveyat Kavousi,EasternCrete: PreliminaryReport,"Hesperia65, pp.373-472. Hayden, B. J. 1997. "RuralSettlement of the OrientalizingthroughEarly ClassicalPeriod:The Meseleroi Valley,East Crete,"AegeanArchaeology2, pp. 93-144. Hayden, B. J., J. A. Moody, and O. Rackham.1992. "The Vrokastro SurveyProject,1986-1989: ResearchDesign and Preliminary Results,"Hesperia61, pp. 293-353. Hoffmann, H. 1972. Early Cretan Armorers,Mainz.
Homann-Wedeking,B. 1950. "AKiln Site at Knossos,"BSA 45, pp. 165192. Hood, S., and D. Smyth. 1981. ArchaeologicalSurvey of the Knossos
Area (BSA Suppl. 14), London. Huxley,G. L. 1994. "On Knossos and Her Neighbours(7th Centuryto Mid-4th Century B.C.)," in Knossos, a Labyrinth of History: Papers Presented in Honour of Sinclair Hood,
D. Evely,H. Hughes-Brock,and
SYME
89
N. Momigliano, eds., London,
du ye au Iersikcleav. J.-C. (EtCret 15), Paris. Levi, D. 1927-1929. "Arkades:Una cittAcreteseall'albadela civilth ellenica,"AS4tene10-12, pp. 1-710. 1945. EarlyHellenicPottery Princeton. Crete, of Malkin, I. 1994. Myth and Territory in the SpartanMediterranean, Cambridge. Morgan, C. 1990. Athletesand Oracles. TheTransformation of Olympia and Delphi in the 8th CenturyB C., Cambridge. Morgan, C., andT. Whitelaw. 1991. "Potsand Politics:CeramicEvidence for the Rise of the Argive State,"4AJA 95, pp. 79-108. Morris,I. 1998. "Archaeologyand Archaic Greek History,"in Archaic New Approaches Greece. andNew Evidence,N. Fisher and H. van Wees, eds., London, pp. 1-91. Morris, 5. 1992. Daidalosand the Originsof GreekArt,Princeton. Morrow,G. 1960. Plato'sCretanCity: A HistoricalInterpretation of the Laws, Princeton. Orlandos,A. K. 1976. "Kp'7n1:
APHRATI
AND
KATO
pp.123-133. Johnston, A. 1993. "Potteryfrom ArchaicBuilding Qat Kommos," Hesperia62, pp. 339-382. Kanta,A. 1991. "Cult,Continuity,and the Evidence of Potteryat the Sanctuaryof Syme Viannou, Crete," in La transizionedal miceneoall'alto arcaismo,dalpalazzoalla cittt&. Atti del ConvegnoInternazionale,Roma, 14-19 marzo1988, D. Musti et al., eds., Rome, pp. 479-505. KretaI: Kirsten,E. 1942. Das dorische Die InselKretaimfiinftenund vierten Jahrhundert,Wiirzburg. Kontoleon, N. 1970. Aspectsde la Gr&e Paris. pr&classique, xxi Lebessi,A. 1970. "'ApxpL6-Mts; I1vYlEtiaX
XeVrt~X)q
XOCL aCVaXto?ilX)
24, 1969, Kp'Tciq;: AqpOd'Cr,"ArchDeIt B' 2 [1970], pp. 415-418. 1973. "ApcxpL6ure;xxcix XVTq[XELOX XSvzftQLX7 xocX OcVOCToXLXVS
ArchDelt25, Kp'-Mq:A9p&-rL," 1970, B'2 [1973], pp. 455-460. 1974. "'Iep6v'Ep[LoF xocx1 Aqpo&'-Mi e'tEo'lylv
Kp'-u-lS,"
Prakt 1972 [1974], pp. 193-203. 1975a. "ApxaL6no-rs; xOci xE XeVrptqLX XOrL o?vrx-coLX7N [LVThLScc
26, 1971, Kp'-M;:A6'-rto;,"ArchDelt B' 2 [1975], pp. 493-499. .1975b. "'Iep6v'Epko6xcxL f eri; ZU'[qvBu6vvoU," 'Aqpo&rqO
Prakt 1973 [1975], pp. 188-199. 1977. "' Ilp6v 'Ep[iobxcd 'Aqpo8Ln2r;ei; Z6[riqvBt6cvvou,"
Prakt 1975 [1977], pp. 322-329. 1980. `A9XGLo,TqrTg
xxci
XOXL OCVOCXoxLXI; [LVYI[LUXL XEVrPLX)q;
29, Kp'qnl;:A6t-ro;,"ArchDelt 1973-1974, B' 3 [1980], pp. 886887. 1983. "`Ap)XaoL6mto ; xOC [LVYflIS-LO XEVt9LX
Kp'-rrq,
ArchDelt30, 1975, B'2 [1983], pp. 341-342. 1985. T6 ip6 wtro3 'Epjh~ xat' zn; 'AApo&6z3o-r' ?6p,7Btdvvou: XcAx va xpz7-rxctiTope6jOpircca,
Athens. -robi'Epj xcci 1990. "'Irpdo a-ri Y26[nBLtvvou," -S; 'AppVo86nl; Prakt1985 [1990], pp. 263285. Le Rider,G. 1966. Monnaiescr6toises
11SpLOuhXXoy), xx&Cotvsi; AutLxiqv
Kp'Tn-v," Ergon,1975 [1976], pp. 195-202. Orton, C., P.Tyers,and A. Vince. 1993. Potteryin Archaeology, Cambridge. A Payne,H. 1931. Necrocorinthia: Studyof CorinthianArt in theArchaicPeriod,Oxford. Perlman,P. 1992. "One HundredCitied Crete and the 'Cretan CP 87, pp. 1937toxhtsEx,"' 205. Platon, N. 1952. "Xpovx6c:'H v Kp'? ocpXocLoXoyLxx xNv-qcmL;
-coE-co;1952," CretChron6, xcar6c pp. 471-481. 1957. "XpovLx6x:'H
Kp'Th x6vqcn; E'v &yxcpuoXoytxrxq -coito; 1957," CretChron11, xoa-r& pp. 326-340.
Plog, 5. 1980. StylisticVariationin PrehistoricCeramics.DesignAnalysis in theAmericanSouthwest,Cambridge. Prent,M. 1997. "The Sixth Century B.C. in Crete:The Best Candidate for Being a Dark Age?"in Debating
BRICE
90
DarkAges:Paperson Mediterranean (Caeculus3), pp. 35-46. Archaeology Rackham,0., andJ. Moody. 1996. TheMakingof the CretanLandscape, Manchester. Rizza, G. 1967-1968. "Le terrecottedi Axs," ASAtene45-46, pp. 211-302. Schaus,G. P. 1985. TheExtramural SanctuaryofDemeterand Persephone at Cyrene,Libya.Final Reports2: TheEast Greek,Island,andLaconian Pottery(UniversityMuseum, Universityof Pennsylvania,Monograph 56), Philadelphia. Schmaltz,B. 1980. Das Kabirenheiligtumbei ThebenVI: Metallbei figuren ausdemKabirenheiligtum Theben,Berlin. Schiirmann,W. 1996. Das Heiligtum desHermesundderAphroditein Syme Viannou2: Die tierstatuetten aus metall,Athens. to Sinopoli, C. M. 1991. Approaches New York. Ceramics, Archaeological Snodgrass,A. 1989-1990. "The Economics of Dedication at Greek Sanctuaries,"Scienzedell'antichita:
BriceL. Erickson DARTMOUTHCOLLEGE DEPARTMENT
OF CLASSICS
REED
HALL
HANOVER,
NEW
316
HAMPSHIRE
[email protected]
03755
L. ERICKSON
Storia,archeologia, antropologia3-4, pp.287-294. Spratt,T. A. B. 1865. Travelsand in Crete,London. Researches Stissi, V. 1999. "Production,Circulation, and Consumptionof ArchaicGreek Pottery (Sixth and EarlyFifth Centuries B.C.),"in The Complex Past
of Pottery:Production,Circulation,and Consumption of Mycenaeanand Greek to EarlyFifth (Sixteenth Pottery CenturiesB.C.),J. P. Crielaard,V. Stissi, and G. J. van Wijngaarden,eds., Amsterdam,pp. 83-113. Talbert,R. J. A., ed., 2000. Barrington Atlas of the Greekand Roman World, Princeton. Treister,M. J., andT. V. ShelovKovedyayev.1989. "AnInscribed Conical Clay Object from Hermonassa,"Hesperia58, pp. 289-296. van Effenterre,H., and M. Bougrat. 1969. "Lesfrontieresde Lato," CretChron21, pp. 9-53. van Effenterre,H., and D. Gondicas. 1999. "Lyttos,ville fant6me?"in La demographie historiqueantique,
M. Bellancourt-Valdherand J.-N. Corvisier,eds., Arras,pp. 127139. Viviers,D. 1994. "Lacite de Dattalla et l'expansionterritorialede Lyktos en Crete centrale,"BCH 118, pp. 229259. Watrous,L. V. 1982. Lasithi:AHistory of Settlementon a HighlandPlain in Crete(HesperiaSuppl. 18), Princeton. .2001. "Cretefrom Earliest Prehistorythroughthe Protopalatial Period,Addendum:1994-1999," in A Review (AJA AegeanPrehistory: T. Suppl. 1), Cullen, ed., Boston, pp.216-223. Watrous,L. V., and H. Blitzer,1995. "The Gournia SurveyProject:A PreliminaryReport of the 19921994 Field Seasons,"AJA99, p. 313 (abstract). Whitley,J. 1997. "CretanLaws and Cretan Literacy," AJA 101, pp. 635-661. Whitley,J., M. Prent,and S. Thorne. 1999. "PraisosIV: A Preliminary Report on the 1993 and 1994 Survey Seasons,"BSA 94, pp. 215-264.
HESPERIA
7I (2002)
Pages9I-II8
TH
E
EN
ATH
STATE
IAN
S ECRETARIAT AN I S I O
P ROV PU
B Ll
E RECTI
S H
N
S
AN
I NG
NG
F O
DECRE
D R D ES
AB STRACT responsiblefor This articlepresentsa surveyoftheprincipalstatesecretaries thepublicationof decreesandtheirerectionon stonestelai,followedbya filll analysisof the formsof the publicationanderectionprovisionsfiom the 5th that,duringaSl centuryB.C. to the 2nd centuryA.C. The studydemonstrates butothersequenceswerealso periods,one sequencetendedto predominate, employed.Attentionis paidto detailwithinthe constituentelementsof the in severaltexts. aremadeforalteringrestorations andsuggestions formulations,
whose InthispaperI firstexaminethetitlesof thevariousstatesecretaries anderectionof decreespassedby the inscription taskit wasto supervise thebouleanddemos,andthenanalyzeanddiscussin detailtheformof the anderection. suchpublication authorizing provisions THE SECRETARIES Fromperhapsjustbeforethe middleof the 5th centuryB.C. we meeto of theCouncil.Untilsometimein theSecretary Tr5 DovANs, ytoa,u,ua£vs fora termof a singleprytanyonly,so the360s,thisofficialwasappointed of theCouncil;2 thatin anygivenyearthereweretendifferentSecretaries but of thetribesnotin prytany, he waschosenfromamongthe members washimselfa bouleutes. almostcertainly bytheyear363/2,3 At sometimeduringthe360s,anddemonstrably was nowheldofficefortheentireyear,andalmostcertainly the secretary 1. Fordiscussionof the secretaries, see Ferguson1898,pp. 14-27;Rhodes 1985,pp.134-141;1993,pp.599-605. forthis The bulkof the research articlewasdonein the PenroseLibrary in the BritishSchoolatAthensduring mytenureof the 2001VisitingFellowship.I amdeeplygratefulto the Direcandhis colleagues tor,DavidBlackman,
at the School.I amalsopleasedto acknowledgethe helpfulcommentsand suggestionsof Hesperia'stwoanonymous revlewers. TextsfromOsborne1981arecited All datesin this by theirD numbers. paperareB.C. unlessotherwiseindicated. 2. It is forthisreasonthat,in financial recordskepton an annualbasis,asin the
Propylaia, and caseof the Parthenon, Erechtheion, we finddatinggivenin termsof who"wasthe firstto be secretary." See Dinsmoor1931,p. 351, n. 2. 3. In 363/2,fourprytanies(II,Vl, VII,andthatof Aiantis)allhavethe of Pallene: samesecretary, Nikostratos see Dinsmoor1931,p. 351,n. 3.
ALAN
92
S. H EN RY
nolongerabouleutes. In356/5(IGII2128),commencing withVIIKekropis, the annualrotationof the secretaries in officialtribalorderwasinaugurated.4New,andas yet unpublished, evidenceindicatesthatthe old arrangement withtensecretaries eachyearpersisted atleastuntiltheyearof the archonKephisodoros (366/5).5This evidencenot onlybringsdown the previously heldlowerlimitfrom368/7,butalsoexplodesFerguson's hypothesis thatthesecretaries of thedecade366/5-357/6alsocameeach froma differenttribe,notin officialorder,butin a randomsequencedetermined bythelot.6 It is onlyafterthealteration in thetermof theSecretary's officefrom oneprytany to oneyearthatwe firstencounter a newtitle,o ytoa,u,ua£vs o xaTa stovTav£LavX the PrytanySecretary.7 At firstsight,it mayappear incongruous andpuzzlingthat,nowthatthe StateSecretary no longer servesfora singleprytanyalonebutfora wholeyear,he is nevertheless designatedxaTa stovTav£Lav. But this restson a misconception: xaTa stovTav£Lav does not mean"throughout (one) prytany," but rather,as Fergusonsaw,"prytany by prytany," referring to the Secretary's function throughout thesuccession of thetenprytanies of theyear.8 Thisnewtitleis firstclearlymetwithin IG II2120,aninscription of theyear353/2.9At lines15-16,ToyytoayMaT£a ToyxaTa I [sto]vTav£Lav is instructed, alongwithTovs aBBovsytoa,u,ua{X£}£aq Tovs £g16 ToLl[5 8]NyoototsytoayMastv, to makecopiesof theinventory of treasures in the Chalkotheke. It is notuntila littlelater IG II2210 +259 (349/8);IG II2 223A(343/2) thatwe firstfindhimexplicitly charged withthedutyof publication anderectionof a decree.10 It is manifest,however,fromextantepigraphical evidence,thatthe publication anderectionof decreesmaynowbe entrustedeitherto the secretary designated asxata stovTav£Lav orto the(earlier) officialdesignatedN5 DovANs. Forthesetwodesignations nowappearin thepublication provisions of ourtextsdownto the end of the 4th century.1l This circumstance has generatedthe obviousquestion:werethesetwo titles merelydifferentwaysof designating the sameofficial,12 or arewe now
4. ConfirmedbyAgoraI 7495,an unpublished lawof theyear354/3 (see Whitehead1989,p.102). Forwhatis nowtermed"Ferguson's Law,"see Ferguson1898,pp.32-38;and19141915. 5. A newtextof 366/5,to be publishedbyAngelosMatthaiou,will furnishclearevidencethatmorethan one secretary operatedduringthatyear. I ammuchindebtedto Mr.Matthaiou forthisprivileged information. 6. Forthelowerlimitof 368/7,see Rhodes1985,p. 135;1993,pp.601602,withAddendap. 781.Ferguson's hypothesisis mostrecentlyespoused byWhitehead(1989,SEGXXXJX 71), whorefersto thisputativearrange-
mentas"Ferguson's proto-law." 7. AtAth.Pol. 54.3 theytoa,u,ua£vg xaTa stovTav£avis the onlytitleexplicitlymentioned,althoughwe may reasonably assumethatthe secretary who,we aretold,waspreviously elected (1Tt00T£t00V
. . . O VT0g
NV X£ ttOOTOVT0g)
is the officialwhomwe finddesignatedasthe g povMg. Ath.Pol. tellsus that"now" the secretary is chosenby sortition(vvv8£ y£yov£xhtoog). It is likelythatthe changefromelectionto sortition occurredat the sametimeasthe shift froma tenureof a singleprytanyto an annualone. 8. Ferguson1898,p. 36. 9. Forthe date,see Schweigert ytoa,u,uaT£vg
Tr
1938,pp.281-289;cf. Rhodes1985, p.92,withn.4. 10.Note thatIG II2223C,lines12, [ytoa,u,ua]£[v]gxaTa s[ovTa]v£av I KA£oceoaTog
TC,UOC70£VOVG
ALYLAL£VG,
takenin conjunction withthe fragmentaryprescripts in IG II2224 and225, bothalsoof theyear343/2,provesthat thegrammateuskataprytaneianwasa designation forthe principalState Secretary. 11.Andeven,sporadically, beyond: seebelow,p. 93. 12. So,e.g.,Ferguson1898,pp.3536,63-66; Brillant1911,pp.34-49; PritchettandMeritt1940,p.2, with n. 6; Rhodes1985,pp.136-137;1993, p.600.
century,e.g., D75, line 34 (paullo post 286/5);16SEG389, lines 1-2
THE ATHENIAN
STATE
SECRETARIAT
93
dealingwith two distinctofficials,with separatetitlesandseparate or overlapping-functions?13 Onepieceof evidenceis centralto thisissue,namely,theinscription alreadyadducedabove(IGII2120),relatedto the inventoryof the treafollowingthe For,in the wordsimmediately suresin the Chalkotheke. linescitedabove,we read: x DovANsavaytoaQavTa 5 xaBxo0Nx[g].
£g1£taaV 8£ £g£TaO0! t savta
[at] avaytoaxpt, Toy ytoa,u,ua£a 5 £Mt00a0£V t0LVt osat [£V] OXt
The problemis obvious:canwe believethatoneandthe sameofficialis, to by two differenttitles? withinthe spaceof a coupleof lines,referred 13. So, e.g.,Dinsmoor1931, officials. Surely,somewouldurge,thesemustbe twoseparate pp.352-353; Adessandri1982, pp.15wereit not so obvious, appears what to accept difficult not be would It 32; cf.Whitehead1989, p.102, n. 1. the apalleviate help to whichmay forthe existenceof furtherevidence, 14. As Rhodes(1985, p.137, bytheTreapublished In SEGXIX129,adocument parentcontradiction. n. 7) notes,it wasPritchettandMeritt (1940, p.2, n. 6) whowerethe first is madein lines13-14 to the steleset up in surersof Athena,reference to makethelink.Schweigert(1938, £aa£V Nv]<>tBox8Ng in frontof theChalkotheke: 353/2byPhilokedes pp.281-289 [SEG ADG129, This can be none otherthanthe stele the atoXovTog. £g16 eOV8N[OV ca.352/1]) didnot makethe connecprovito set up by thepublication instructed is Trs DovXNs ytoa,u,ua£vs tion,misledby Dow'serroneous of of Dorotheos son Philokedes Now, 120. of IG II2 17-19 of lines sion (Dow 1937, pp.34-36) identification eponymous asthe g pouSg xaL Toi) firmlyestablished of theytoa,u,uaT£ig Palleneis,withminimalrestoration, 8N,uou-aquitedifferentofficial inAgoraXVI 55 (IGII2138)andIG II2139,bothof 353/2;and, secretary see Rhodes1985, p.136altogether; as we alreadyknowfromIG II2223C andII2224, 225,the eponymous qg oURNG. withtheytoa,u,uaT£ig Hence,the stovTav£Lav. xaTa couldbearthe titleytoayMaT£vs secretary 15. So Dinsmoor1931, p.352; areone xaTa stovTav£Lav andthe ytoayMaT£vs m5 DovXNg ytoayMaT£vs cf. Rhodes1985, p.140; 1993, p.600. andthesame.14 in the Corpus 16. The restoration text,IGII2652, lines34-35, tov entitledytoa,u,ua£vs Althoughincidenceof mentionof thesecretary xaL Toi) 8NI,uou, qg p[oUNg ytoa,u,uaT£a of decreesdefor the publication as the officerresponsible TrsDovANs of the Bouleandof the the Secretary of thesecin favor replaced progresses, century 4th as the steadily creases People,is bothimpossibleandunmisleading, it is certainly stovTav£LavX xaTa ytoayMaT£vs entitled retary We shouldread,as Osborne necessary. indeedfalse,to suggestorimplythatwe hearno moreof theformerdes(1981) does(D 75), p[ouSus£V OXt of the Bouleand foundwellintothe 3rd The Secretary AL0LIvC. Forit is unquestionably ignationafter318/17.15 of the Peopleis anofficialwitha totally differentfunction,andis to be identifiedwiththe electedReader(see Rhodes1985, p.136). 17. Meritt(1961, p.215, no. 10) hadrestoredTol[vytoa,u,ua£ag this, L0LVt, caXing £aXt 00Xg formof the "anabbreviated strangely, Tr g oUMs xaL Toi) titleTovytoa,u,uaT£a ,,
^,
oNFoU.
18. Seep. 95 below. l9.AgoraXV,p.89. 20.Dowl937,p.42,andn.2.
21.AgoraXVI 123,lines23-24 (302/1)furnishesanexampleof in a textthat omissionof the secretary also,asit happens,includesthe phrase xaLTa ovoFaTaai)Xiov.Forother examples,seepp.110-113below.
the totallyrestoredexampleatAgoraXV 77, (ca.med.s. III).17However, lines32-34 (28>275), avaytoaf at 8£ T08£] T0 f gatI [a Tovytoaa£a with suspicion, shouldbe regarded XI[t0Lvt, £]V OXt Tov5 DovANg with the repeateddefinite formulation partlybecauseof the anomalous employedto justifythe restoration. andpartlyforthe reasoning article18 to thelatedateformention Dow'sobjection ForMerittandTraillrejected the decree] of thisofficeron the groundsthat"thefundsfor [inscribing a curious wereto comefromthe moneysat the disposalof the Council," omissionof fivelettersfromthe Dow'scautionin assuming justification.19 he alsonotedthatthe space is preferable; title of the prytanysecretary Thisis not couldbe exactlyfilledwithzat Ta ovoyaTav stovTav£v.20 to the whatsoever so fancifill:instancesof the omissionof anyreference do occur.21 secretary and Trs ,BovANs designated of thesecretaries Apartfromtheactivities we findthat,duringthe twobriefperiodsof political zaTa stovTav£cavX
94
ALAN
S. HENRY
turbulence atAthensin theyears321/20to 319/18and294/3to 292/1,it wastheavaytoaxp£vs whoassumed thestatusof eponymous secretary and thedutiesof publishing decreesof thebouleanddemos.Prytany secretaries do makean appearance in decreeprescripts of the firstof theseperiods, buttheyhavelost theirannualstatus,beingchosenfor a periodof one prytanyonlyfromamongthemembers of theprytanizing tribe.22 Beforethe 4th centuryis over,we encounteryet anothersecretary entrusted with the publication of decrees,or,at least,anothertitle of a secretarywith this responsibility. This is o ytoa,u,ua£vs Tov8N,uov.23 FergusonandBrillantargueforthe identification of this secretary with theytoayMaT£vs xocTastovTav£cav andtheearlier ytoa,u,ua£vs5 ,BovANs, anidentification alsosupported byPritchettin the editingof a fragmentarytext.24 It is not impossible, however, asWoodheadremindsus, that "thiswasa functionary separatefromthosealreadymentioned."25 Certaintyis unattainable; forourpurposesaneconomical hypothesis willbe to acceptthatthisis simplya newtitlefortheprincipal statesecretary. Theytoa,u,ua£vs Tov8N,uov is firstencountered in publication provisionsin the posthumous honorsproposedforLycurgus by Stratokles in 307/6([Plutarch] Vit.Xorat. 852),wherewereadava0£LvaL 8£ Tov ytoaysuaT£a Tov 8Nsuov £V oBats Al0tvatsxat osat £V aXtOOG10X£t ZNOLOV v ava0Nyav. Epigraphically, he appearsslightlylater,e.g., D 61 (IG II2496 + 507),lines37-38 (303/2);possiblyalsoin D 57 (IG II2576) andD 58 (IG II2696),bothplacedbyOsbornein ca.307-303/2.Forthe
next200 yearshis titleoccursregularly butmuchlessfrequently thano stovTav£Lav. He is stillto be foundmentioned atthe veryendof the2ndcenturyB.C. (IG II21011,line62 [106/5]). Thisbriefsurveyis notcomplete withoutmentionof thefactthatthe relevantsecretary appearsin a few casesto havebeendesignatedas o ytoayMa£vs tout court.26 As already noted,on occasionin thepublication provision thereis no mentionof anyresponsible officialat all:"inallsuch casesthe subjectof the infinitiveis presumed to be the familiarofficial, underwhatever designation."27 ytoayMa£vso xata
THE PROVISION FOR INSCRIBING AND ERECTING In theinstructions includedin decreesandlawsfortheirpublication and erection, thereareessentially sixelements, notallofwhicharealways present in eachinstance, andnotallappearing in a single,unvaried orderof occurrence.Thesesixelements,outof whichthewordingof theinstructions is formulated, arethe designation of the responsible secretary, the instructionto attendto theinscribing of thedocument, thespecification ofwhat textis to beinscribed, thematerial onwhichthetextis to beinscribed, the instruction to erectthemonument, andthelocation.Withineachof these elementswe shallfindvariations of wordingorvocabulary. The presentation of the massof materialavailable for an overview and thoroughanalysisof the variousformulations is no easymatter.
22. See Henry1977,pp.50-57, esp. p.55; cf.Rhodes1985,p. 140;1993, p.600. 23. As Woodhead(AgoraXVI, p. 191)notes,Rhodes1985doesnot specifically discussthissecretary, althoughhis note3 (Rhodes1985, p. 136)mightbe takento implythat the secretary is to be identifiedwitho ypa,u,uaT£ig
g
poUfg
xaL
Toi)
8ffot),
i.e., the Reader.Rhodes(1995,p. 600) statesbaldlythat,apartfromthe two periodsof avaytoap£tg,"thetitle ypaaT£i)c; xaTa spOTav£Lav became standard(untilin the secondcentury A.D. it gavewayto a newtitle1T£pt T0 ,8ffa)." This is somewhatimprecise, at leastin so faras the designation of the secretaries responsible forinscribing anderectingdecreesgoes. 24. SeeFerguson1898,pp.63-66 andBrillant1911,pp.37-49;also Hesperia10,1941,pp.270-271,no.70, withn.10; cf.D 64 andAgoraXVI 121. 25.AgoraXVI, p. 191. 26. Seepp.l lO-113below. 27.Woodhead,inAgoraXVI, p.240.
THE ATHENIAN STATE SECRETARIAT
95
overlapat all Consciousof thefactthatthereis anobviouschronological for theinscribthat instructions given periodsexceptbeforeca.365B.C. (or,at ing anderectionof stelaiareassignedto morethanone secretary thatthe withmorethanonetitle) I haveconcluded least,to secretaries in if it is presented anddigestible28 willbemosteasilyintelligible evidence to thevarious(titlesof) secretaries. dividedaccording categories OF THEBOULE THE SECRETARY OTN5,30VAN5) (OYTOaXUXUaT£V5 natureof earlychancery giventheidiosyncratic As is fullyto be expected, style,the 5th centuryrevealsitselfas a periodwhenthe wordingof the or is onlyjustbeginningto edgeits wayforwardto a standard provision form.29 predominant IG I310,lines22-26, the Ourearliestpieceof evidenceis perhaps30 Phaselitedecree,dated469-450in theCorpus: stoich.22
X [O 8£ N(pLO] ya T0[8£] avaytoafa
O25 D°XL5 [TZOYTOaCURU]aT£V5 xat xaTaO
t06]Vt
[£aXt [£TO £
Z0X£t
To whatis to be inscribed, byspecifying commence Heretheinstructions to inscribethedecree,herein Qpts,uato8£,3lfollowedbytheinstruction inthenomihimself,ofnecessity Nextcomesthesecretary theimperative.32 nativecase,and,be it noted,withthearticlerepeated;he is o ytoa,u,ua£vs text,a stele to beartheinscribed Thenappearthematerial o Tr5 DokNs.33 to erectthe stele, the instruction by £V + dative;35 expressed of stone,34 28. A certaindegreeof whatthe DavidLewismight late-lamented is wellhavetermed"rebarbativity" in studiesof thisnature. inescapable We are,I fear,backin "theaustere thatWhitehead(1998,p. 493) realms" withmy not too unkindly, associates, workon the languageof Athenian .
.
.
lnscrlphons.
29. I shouldindicateat the outset that,althoughI haveexcludedmany possibleinstancesin whichrestoration a role,I havenot playstoo substantial hesitatedto includeexamplesnot where,in my totallypreserved, judgment,sufficienttracesremainon the stoneto justifyconfidencein the overalloriginalwording.To citeonly textswould completelyunrestored somepotenhaveriskedoverlooking
tiallysignificantevidence.The firstexthis amplecited(IG I3 10) illustrates pointwell. 30. On the strengthof the short dativeq!aqAxats (line5), Harold Mattinglywouldplacethisdocument in 425/4.LikeLewis,I do not find thlsconvlnclng..t 1S not mylntenton for hereto rehearsethe arguments "therightdatingcriteriaforfifthcenturyAttictexts"(cf.Mattingly1999). I addressthistopicin ZPE 137,2001, forthcoming. wordorder,T08£ 31.The alternative To @NCoya, cannotbe accommodated here,norwouldit be toleratedwherethe by the connective provisionis introduced of the sequence 8£. Thereis no instance T08£ 8£ T0 @NCoya. Cf. n. 77 below. will,of course, 32.The imperative .
.
.
.
.
.
.
.
givewayto the infinitive eventually construction. the title 33. In the nominative, mayindeedbe found,butrarely, withoutthe repeatedarticle:so IG I3 156, lines21-22 (440-425); IGII2106, line 16 (368/7); II2141, line 13 (364?). the titleis invariably In the accusative, withno DOVN5, 5 TOV ypaaT£a doublingof the article.Forthe alleged Tov5 instanceof Tov in AgoraXV 77, line33 (280OVN5 275), seep. 93 above. 34. ButnoteIGII2687 (+ 686), line 43 (265/4), £V aT(>(t XaXx[Nt. is almostas 35. £65 + accusative commonas £V + dative.Cf.IG I398 (411), whereat line 13 we find£V ce[qNtA0tv, andat lines27-28 £5 (avaypafat)
ypa,u,uaT£a
vV
alv]v
Xfv.
96
ALAN
S. HENRY
expressed bytheverb andfinally, thelocation,,u soRrt,37 on theAcropolis, totallyrestored here,butequallytotallycertain. Thus,all six constituent elementsarepresent,in a patternthatcan mostsimplybedescribedas O(bject)-V(erb)-S(ubject).38This pattern seemsto havebeenoneto commend itselfinthe5thcentury: compare, for example, themoreelaborate butessentially identicalIGI378,lines48-51 (ca.422?),thedecreeon the Eleusinian Firstfruits: xaTaTL0rlCuL;36
stoich.50 Tas8£Xavvytoa(pag xatTo(por(pts,ua To8ravay toa(psato ho ytoaarus ho z5 DoWzszv ozAatv Avotv At0tvat v xat xata0zto rv rv 'ERzvotvtzv tot htrtoot, rv 8r
hrztoav[z] Z soR
The samesequenceis foundin theaccusative andinfinitiveconstruction, e.g.,IGI365,lines5-7 (ca.427/6): To 8r (por](p[t]a[a To]8r ava[ytoa(psat ] [oy ytoaa]ra [Tr]5[o]Xzs zv (7[T£]Xt [t0] [tvrt xat xa]X[a0]rv[at] rZ groR[z].
stoich.30
Compare IGI3106,lines19-21 (409/8): stoich.50 T
[o 8r (po]z(ptoZa to8r zV
ava[yp]a(psat
Tov ypaara
1rr5 ,(3Oz5
oTz
[Xrt At]0tvrt xat xaTa0z[v]at
zv groRrt
andIGI3110,lines20-24 (408/7): To ar Q
stoich. 23
ptoZa To8r avaytoaQat Tovy toaara 5 DoRNszv orrAN t At0tvt xat xaTa0rvat rZ z oRr.
A moreelaborate wording,butrecognizably the samepattern, , is IGI340, lines57-63 (446/5):
seen
in
36.The regularalternative, aswe shallsee,is ,u. 37.Thereappearsto be a distinct shiftfrom£ W0X£t to £V aXpOZO£t somewhere aroundtheyear386 B.C. (seeHenry1982),probably witha periodof overlapon eithersideof the apparent divide(SEG XXXII50, lines 17-18 [379/8],althoughtotallyrestored,maybe the latestdatedexample of £ W0X£t). As wellas £V + dative,we alsooccasionally find£65 + accusative: SO £5 WOLV, IG I340, line60 (446/5); I3127,line39 (405/4);£65 axposoktv, IG II2238b,lines15-16 (338/7);D 37 (II2391),line 15 (318);II2571, lines78 (fin. s. IV). Instanceswiththe article arerare:£V t IG II2133, line 17 (355/4);£tg mV axposo[Xcv, II2 221,line5 (paullo post 344/3);£65 1 [V axposoRtv, II2725,lines8-9 (s.III, axpoW[o£tX
To 8s (por(ptoZaTo8r xat Tov
hotoxov avaytoa(psat 'A0rvrot rv Tovytoa Z[a]Xza z5 ,BoRzs rozArtAt0tvrt xat x aTa0rvat r5soRtvzArotTotsXaRxt8r ov, zv 8r XaRxt86zv Tot htrtoot To/\tosTo 'OBVM=CO h£ 0X£
a xaTa0£o.
xaXxC8£ov
avaytoasas
stolch. 32
pars prior).
38.The wordingforthe erection willregularly, butnotinvariably (cf., e.g.,IG I3153,lines19-21 [440-425]), followon afterthe principalinstructionsforthe inscribinghavebeen given.
THE ATHENIAN
STATE
SECRETARIAT
97
Here,of course,wehaveprovision forinscribing anderectingbothatAthensandatChalkis, whichaccounts forsomeofthevariation. Wemaynote too the switchfromthe accusative andinfinitiveto the nominative and imperative, whichis paraHeled, withthereverseswitch,in IGI3156,lines 19-26 (440-425): srtot[8] stoich.23 z Arov8oTas(por(pts,urva a[v] aytoa(psaTo ho ytoaarus r 5 5°Xz5zzot 065 ArovtAo sv ozAatv Avotv,xat rv rv
hrztoavozoat ,u soRrt,rv 8r hrztoavzvhaRtxatovasso CzvTothstootToAsoRRovos We shouldobservethat1) the articleis not repeatedin the titleof the Secretary;39 2) the expectedAt0tvatv is omitted;3) the allocationof the coststo Leonidashimselfcomesbeforetheprovision forerection, nodoubt becauseof thecomplication of thewordingof thedoubleerection; 4) the verbemployedto expressthe erectingis aTrl,ut, ratherthanxaTaC0Nt; and5) at line 27, the construction movesbackto the imperative again (av8toastooorAroEo ArovtAlzs xX.).40 To returnbrieflyto IG I340, it shouldalsobe observed how,in lines 62-63,the formulation hasbeenconverted fromavayto&goatxatxaTa0rvatto avaytoasava (participle) xata0zto. ThissyntacticaXy "reduced" formis encountered notinfrequently in the 5th century, aswellas occasionallyin the4th.4l In the5thcenturywe alsofindthesequence V-O-S, withtheinfinitive,forexample, IG I366,lines20-22 (427/6): xatavaytoa[satTavTaToyy] [oa]ZZararsdoRzsrozArtAt0[tvrtxatxaTa0] zMat
stoich.38
r,usoRrt
whereie restoration Tatrca is confirmed bytheresumptive Tav]*a V avaytoa(psaL in lines22-23. Comparethe slightlydifferentIG I398.I,lines2628 (411): stoich.30
Grtoovava] [y]oaQaL [v]rr
39. See n. 33 above. 40.The heavilyrestoredandnot entirelysecureIG I3163,lines3-7 (44s415) appearsto movefromthe imperative to the infinitive 41. Seep. io3 below. lnaeea,aswe snallsee,tnlssequenceis by farthe commonestwithall (titlesof) secretaries.
8 v orr
xaL
XY2V Toy
To8
To
@f
[(pLoZa
ypaZ[aa
rr
r5qv a] 5 D°AY15] .
Thissequence, V-O-S, is theonethatappears to predominate in the4th century andbeyond;42 compare, forexample, IG II2107,lines18-20(368/7): non-stolch. avaypaQaL
avaypa](psaSro
8[
T]oa
xaTa]0£vaL.
an zT.
T
1
1
1
11
1
*
[r to W]N[?t]aZaTovytoaaranmg[ov]X[g]rvorXtALeLv! [t
xato]rnsatzvaxtoosoRrt-
andD 14 (IGII2226+),lines19-23 (ca.342):
98
ALAN
stoich.21
ava
8Z To8Z
ytoaQaL
To
S. HENRY
@f(pLoZa
toy ytoaCuCuarVa rr5 DovAr 5 zvorrArlt At0turlt xaL xa Ta0zTvaL
Zv
aXtooGrozT
43
and,withthe variationof the phrase"ona steleof stone"appearing before, ratherthanafter,mentionof the Secretary, IG II2232,lines20-23 (340/39): avaytoaQa] [C] 8 To8 To @N[(ptoZazv oXt A] [t]0LVY26TOV yto[aara q5 DovA] xat osa[ zv axtoogroRrt
stoich.25
SEGXL 74, lines 24-27 (337/6): xat avaytoa Qat To8r To @N(ptoZa zv orrAt At0tvt To v ytoaara q5 ,BovANsxat oNsat zv axto
stoich.33
osoR
andthemoreelaborate AgoraXVI 73,lines22-27 (337/6): avaytoatvat or Tov ar Tov vo,uov zv oRats At0tvatsAvotvTovy toaara rrs DovANsxat osat qZ rv rst T CooAov ris rts"Atortovflayov N5 rt5 To Do VAzVTrtOLOV tOLOVTt, TY2V Oz zV Tr t zxxtata o
>
,
,
,
\
As
o
,
-
As
,
>
o
stoich.36
,
-
or,with the completeomission of the phrase"ona stele of stone, 44 for example, IGII2351 (+624),lines33-35 (330/29): avaytoaQat [8]r to8r To @N(ptoZa Tovytoaara 5 DovANsxat oNsat zv axto o groRrt -
,
,
stoich.
,
Thesamewordorder, V-O-S,isstandard inproxenygrants, whenthe 43. Cf. alsoIGII2660.I,lines19-22 wordgrtooE,zvtav is employed inlieuofTo8rTo@N(paZainthepublication (s.IV,parsprior);II2204,lines54-57 provision: soIGII280,lines5-7 (ca.380-370?):4s (352/1);II2212,lines44-47 (347/6), withTo TO8£; II2410, lines 37-39 (ca.330),AgoraXV 49, lines5557 (328/7),IGII2343, lines17-19 (323/2?). 44. Although,of course,thevery existenceof the stoneshowsthatthis wasthe intention. 45. SeePecirka1966,p.29, forthe text(cf.SEGXXIV78). Qyi(pecTza
, , , - \ [xat avay]oaQat avXt vrZ grtoo[irvtav] [V OX]Y26 t0[t]VY26 TOV ytoaa[Xza VV] [rrs DoR]s xa[t o]Xsat zv axtoos[oRrt V]. \
andIGII2149, lines17-20
(342):
.
stolch.29
THE ATHENIAN STATE SECRETARIAT
99
a
stoich.37
[vaytoaQat8£ zat q]v stooirvtav, £av zat Xt 8nZ q5 DovANszv orrA X 46 [t0tvt xat orroat] zv axtoosoRrt 8rxa ZoXv [Xt aOXY2t,TOVytoaZ]ara
we alsofindthe following sequences In additionto thesepredominating twosequences: 1) O-S-V, so IGII243,lines63-66 (378/7): to 8[r Q(pt]aZa to8r o ytoa,u,uaTrug o q5 DoR[s avayto]aQa zv oqAt At0t VY26xat xaTa0z[X] gratoatov lta Tov'ERzv
stoich.31
0ztotov
2) S-V,so IG II276,lines17-20 (ca.378/7?): oq o] 8r y[oa]arus 5 DoRNsav[ayto]aQa zv ot zv t0[LVY26] xat 0z YiXY26 axtoogroRr[t
stoich.20
decree,wheretheobjectof avaytoaQa is to be suppliedfrom a proxeny proxenygrantin the formsuat [8]1r avTo[v preceding the immediately stoog]vov
xX.
anderectionprovision Finally,we cometo a formof the publication the so-called producing verbof erectingis expressed, whereno separate This formwarrantsspecialtreatment,given formulation. "telescoped" basedon it by Harold arguments of the chronological the importance 47 Mattingly With the orthodoxdating,ourearliestinstanceof thisphenomenon Egestadecree,IG I3 11,lines11-12 (458/7),wherethe is in thenotorious seemsbeyondchallenge: restoration stoich.48 ava[yto]a[(psa] To 8r (por](paZa To8r xat Tov [hotox]o[v] [ rozzt C0tvrt,u s]oRrttov ytoaRRara z5 Dos
(420/19)in Apartfromthe wordingof the textof"thegrandalliance" Thucydides5.47.11,Tas 8r iov0Nxag . . . avaytoaQat zv rlxn xetvn knowsof onlytwoSth-century 'A0Nvatovs rv rZ groRrt. . ., Mattingly IGI3119,lines6-9 (407),wherethetextis lessthan parallels: epigraphical secure: V-O-S, 46.Forthesamepattern, seebelow, Secretary, withthePrytany pp.104-106. 1984,pp.34447.SeeMattingly
xat ava] ytoaQatToy ytoaa[Xza q5 DoRNszv orrAC] At0tvYlt zv [soRrtTas r iov0Nxas xat To @N]
6e;s - - [spLs,ua
stoich.34
ALAN
S. HENRY
IOO
andtheequallyuncertain IG I3125,lines29-32 (405/4): ]
stoich.29
[o] 8r +NColla [o8r avaytoaQatTovytoa] [Z]ara q5 13OR[Y15 r11sORrt zv '7tYlAYlt] [Xt]0tvC.
Mattinglyis, of course,happyto acceptthe readingsof thesetwo texts sincethepointhe wishesto establishis thatthe"telescoped" formis not paralleled beforeca.420.If he candoso,thisisolatesIG I311 (Egesta)if it is retained in theearly450s. It is forthisreasonthathe seeksto remove thephenomenon whereit hasbeenrestoredin textsearlierthanthe 420s.Thus,in the caseof the Kolophonian decree,IG I337,lines38-40 (447/6),48 givenin theCorpus in thefollowingform: 8r Wr(pa[a To8rxat Tovotoxov avaytoaQaTo o ytoa] [1la]Xzv5 o z5 D[oRss rozArtAt0tvrt ,u soRrtzAro] [t]ots KoRo(po[vtov
[o]
Mattinglywouldprefer,on the basisof the appearance of the participle avaytoaQav[Xzs (line 41) plusthe imperative xaTa[0rvTov(line42), to read,a la Hiller(IGI214/15,lines26-27): stoich.38-42 [o] 8r @z(pa[a To8ravaytoaQasrozArt At0tvrto ytoa] [,ua]Xzvso z5 5[oRzsxaTa0zTorZ woRrt'A0rvrotzAro]
Thisdispenses withthe"telescoped" format the expenseof reference to theinscription of theoath. Likewise, in thedecreeaboutErythrai, IG I315,lines42-44 (ca.450), Mattingly wouldprefernotto followthe Corpusin reading stoich.47 [tO]XOV Z[V] t[0]LVt
OTZzL49
avaytoaQat8r TavTaxat Tovho] [xat Tovhotoxov Tovz5 DoRzsrZ
woR] ,
butinstead,againon the modelof Hiller(IG I212/13),he wouldreadat lines43-44: [o]xov
r[v] A[0]vrt
O1zzt
[xat
xaTa0rvat'A0rvrotvrv
rZ
woR]
rt, 'E[pv0]pa[a]t 8r xX.
As for the fragmentary and uncertainIG I3 70, lines 3-4 (430420),
[c]
,u
To 8r [Q]N[
48. WhichMattinglywouldplace ca.427. 49. Note the anomalous wordorder.
oRzs Ktotoova 'A[ zArot 10 to[ts [ xat ........ ...........To] 1614. . ]]
THE ATHENIAN
STATE
SECRETARIAT
IOI
Mattinglyfavorsa muchlonger wherenoteventheline-lengthis secure,50 formulation thateschewsthe line,5lwhichwouldpermitan alternative telescoped form.He maywellbe correct. Eleusinian Epistataidecree, Mattinglyis stillleftwiththe awkward IG I3 32, lines 32-34 (ca. 449_447),s2in which thereis no escaping thetelescoped format: ytoa(psat Za zv
8r To [(por(pto]
ozArt'ERzvotuc xa[ zv 'ERzvotv[tot.
aszt
stoich.32
xat 42]
aR[z]ooc zv Tot *
-
publication andabbreviated compressed This,manifestly, is anextremely of three provision,53 andMattinglyarguesthatit was the specification groups needto specifythreedifferent separate locations andtheconsequent brevity.54 to taketheeasywayoutin extreme of officialsthatledthedrafter Thus,Mattinglyseesthisasa specialcase,notto be adducedasa normal earlierthanca.420. formappearing instanceof thetelescoped of the of the generalcaseforthe downdating I remainunconvinced feelthe sameneedto removethe Egestadecree,55 andI do nottherefore citedabove.Forourpresent fromalltheexamples telescoped formulation purposes, someof the abovetextswouldnot havebeenincludedin this positionvisto considerMattingly's discussion hadit notbeennecessary a-visthisphenomenon. decrees,wherethe grantis exWe maynoteherethat,in anderectionarecomandwherethepublication pressed with compare, is oftentelescoped: binedwiththegrantitself,the formulation forexample, IGI327,lines5-11 (ca.450/49):57 proxeny
avaypaQaL56
stoich.23 5 a8z?o5 [o5 zx6vo to5 tX] , , - , \ \ \ vo5 xat ToZ saztoa avTov ava z5 ] ytoa(psattov [ytoaara oRr5rz soRr[t rozArt xat zv] Tot
DoRzvz[otot
stooxozvog]
'A0rvatov xX. e v.5 50. "l)e w. paullolongioribus est"(Lewis). fortassecogitandum 51. Cf.Walbank1978,pp.125-127. 52. Mattinglyprefersa datenot before433/2. secretary is not 53.The responsible evenmentioned(seepp.110-113 below),butin the 5th centurywas of the Boule. certainlythe Secretary 54. MattinglynotesthatevenL0LV£t .
.
Agora XVI 11
(IG I3155), lines 4-9 (435-430) 58 stoich.28
avaytoa(psat grto] 8z(P°S xat lrx[.... oXorvo5xat zv[ztoyrTaszv ozArt A] t0tvrt,u soRrt [xat zv Tot DoRzvz] z5 5] totot r5 sav8a [ov ytoaara
fl
1S OmltteC .
55. See,mostrecently,Henry1998. 56. See Henry1983,pp.116-130. 57. Mattinglyinclinesto a datein the 420s. 58. See Henry1983,p. 117,forthe text.
andIG II213a+ 68 +Hesperia40,1971,pp.149-150,no.3 (SEGXL 54), lines7-12 (399/8): 'Apar [av Tov'A]xa[C]ovTovAtyta ava
stoich.21
ALAN
S . H EN RY
I02
[ytoaQat][ov] ytoaara N5 [5oRNszv gro]Xt zv oX X [t0LVY26ZvOO]6zVOV xat zvrto [yrqv avTov] xat zxyovos
Contrastthe non-telescoped formulations in IG I3174,lines5-11 (425410): Avxxua Tov'Axat OV, zt8YI
[5], avaytoaQa
stoich.21
'A0NvaLo
zV Z0zt
xa
stooizvov
Czvrtoyrmv'A0Nvatxv zv a Tr ANt A0tvrt
aaro5 aTa0z
rR soR
o yto
° 715 zotNs xat x rZ groRr.59
\
X
n
>
\
andIG I380, lines 12-18 (421/0): xat av aytoa(psaTostooxozvov xat zvrtoyrrv'A0rvatov xaEa srto floBvooaTov Tov 42Xrt astov rozArt At0tvrto yto aarus ho z5 DoRzsxa t xaTa0zTo zv soA.zt 60 ,
,
stoich.21
\
The phenomenon is commonenoughin the 4thcentury,6l forexample, AgoraXVI 36 (D 8),lines33-36 (394/3):
stoich.37-39 tov 5 avaytoaQat 5 zv oXt
To @N(ptoZa tvasrto
8s [y]oaZ[ara To8r
[Xr]Xr[at
]T5 tot]s
,BoR[N]
E0otov[o]
xX.
.. 59.Notetheunexpected repetlton of£ W0X£t (lines9 and11). 60.Heretheobjectis tobesupplied fromthepreceding provision (lines89) £TraLv£saL A£av Tov AR£IOVX with whichxaL avaypacpsaSro is coordinated. 61.IGII2140,lines31-35(353/2) andII2365b,lines12-16(323/2)arein
Compare AgoraXVI 40 (D 9),lines13-16 (388/7or375/4),a citizenship a slightly category, inthat bothzivedifferent instructions to inscribe the decreein whichavTos takestheplaceof To8rTo@N(ptoZa: textonstelaialready inposition. Thus xa[t] o
at avto5 vt Dofs
toy
\
o
stoich.18
ava[ytoa@] o
rorrfLt
r
ytoaZ[ara
*
rs ]
tst]
,
-
rrs]
zv ax[oogroz-
thereis noneedto specify thelocation separately. 62.SEG XXXIX75b,lines9-12 (353/2)appears tobeanexample of the telescoped form,butthetentatlve restoration ofline12cannotbecorrect: TOV 8£ ypaaT£a
andIGII2238,lines14-17 (338/7): avaytoa]at [a
Tov ytoa]ara
[65axposot]v [vYlv 62
[avaypaQal
o
8r To8r To f(pto
N5 DOVXY]5 z rt5otYiRYlV At0t
* tv
A
stolcll.we o
£V
T]N5 DOV
5 aXpOZO£]l
1
T08£ T0
Ql[N9lOya £aX£l XlolV£]l 8£[X]a N£I[pOV xai zOal? T£X£Ol] TOl5
Nlxa[.].If thelocatlon £V aXpOZO£l 1S correctly restored inline10,wewillnot expectto findtheinstruction to erect, ceal, laterinline12.
STATE
THE ATHENIAN
Io3
SECRETARIAT
ofthevariousformulations offthisdiscussion Wemayconvenientlyround of the Bouleby a employedduringthe tenureof officeof the Secretary to attendto the of the formin whichthe instruction briefexamination to a parof thedecreeis reducedfromtheimperative/infinitive inscribing ticiple.63
The firstinstanceis perhapsto be foundin IG I324, lines 9-14 (ca.450): stoich.15
o ar y
toa[ az]vs o z5 Do Ar[5to (por](ptoZaTo [8r avaytoa(p]sas r5 a [XrArtAt0tvrtx]aT [a0zTo rZ groRr]sacat
CompareAgoraXVI 15, lines 9-10 (426/5?): stoich.50 xat To8rTo[Q](ptoZaavaytoaQ[aso] ytoa,u,uarv[so rrs r] DoArls x]a[X]a0z rZ gro[Xrt t0LVYI[t [v] orrXYIt
andIG I384, lines26-28 (418/17): stoich.52 ,8oRo,urvot, To8r,osos av rt r8rvat To[C] To8r (por(ptoZa avaytoa(psa xaTa0zTo zvozArtAt0tvrt o z5 ,8oRzs 5 o ytoa,u,uarus ,
,
zvTotNrArt on 63. Alreadybrieflyintroduced p. 97 above. formis probably 64 The participial alsoto be foundin the 32-letterversion of IGI3165 (seeSEGXXVI 19,lines 6-11), butthe textis too disputedto citehere.(I hopeto dealwithIG I3165 The formuin somedetailelsewhere.) aslateasthe lationis stillencountered secondhalfof the 4th century:see IG II2276,lines18-20 (ca.342):To8£ T08£ avaypaQaq
@ftoya
yp]atutuaT£05
£V CTX£t
£+f(pLoCu£va 3o]M5 t0LVL
L0LV£t
£Ta
£V aXpOZO£t,
T]oy
ypaCuCu[a£a
rrEs
TOV
| [aTparryOv
£a1Nt
ZaL ZaTa]0£vaL
is foundinIG I340,lines61-63(446/5),where, formulation A parallel of aftera regularavaytoa(psat+ xaTa0rvatprovisionforthepublication fortheequivathedecreeandoathatAthens(lines57-61),theinstructions form: at Chalkisaregivenin the"reduced" lentprocedure
To/vtosTo zv 8r XaRx86 zv Tot hztoot avaytoa(psas XaRx8rov hzDoRr 'OBv,usto
stoich.32
a zaTa0zTo.
[o 0
wherethe is designatedsimplyas o Secretary (seebelow,pp.109-110). ypayyaSr£vs of the 65. Forthe collaboration cf.IG I3127, generalsandSecretary, 8£ Ta lines38-39 (405/4):avaypaQaL sa[^rlO]
ot satoa Ta3yXtota.64
£5 wot[v.
proaswellas in IG I3118,lines33-36 (408),where,in an amendment withtheSecreareto actin conjunction thegenerals posedbyAlcibiades, taryof theBoule: zaL
zaTa0£vaL
[£5]
[80X£5
[V] £665
£V [z0X]£t [T]as
[p]aT£og [
TooT
avaypasavTaq
Cu£Ta To ypaCuCuaT£og
oUv0£[X]aq .....
18
]
£V^T£X£t
T
t06
Io4
ALAN
S. HENRY
THEPRYTANY SECRETARY (o yRoa,u,ua£vs o zaTa
=ovTav£av)66
Althoughwe findoccasional examplesof the sequencesS-V-O, forexample,IG II2354, lines26-29 (328/7): stoich.34
TOV 8£ yRoa[a£a]67
tov zaTa =ovTav£tavavayRoaQat T08[£ T0 @N] ptoa £V Xt t0LVt zat oqaat £[V TOL] t£pOt TOV 'AsxAztov andIG II2653, lines 52-54 (285/4): Tov yRoa,u,ua£a
[xaTa =]ovTav£tav [£V ^T]NNt t0LVt
Tov
stoich.36-38
avayRoaWat T08£ T0 WgtoZa zat rsat
£V aXtOOZO£t-
V-S-O, in a proxeny grant,IGII257, lines1-9 (ante 387/6): stoich.11-14
[- - - avayRoaQa] [t] tov [yRoaa] ,3ovN [5 £]
rr 5
£a
V X!t ,
t0LVt ,^
£
,
ZOA£t =tOOt£V \
,
,
ovs zat £V£R0y£T as avTovs ,
zat £X ,>
yOVOV5 Tu5 ZOA £X5 q5 oA0Nvat v
andO-V-S,in thelatterpartofthe2ndcentury, AgoraXVI 310,lines5052 (ca.135): non-stoich. ca.38-44 T0 8£
[a£a [0£LvaL
WN(pLo[a]a T08£ avayRoaQat tov yRoa
Tov]zaTa =ovTav£Lav saRoa]rrv £Cxova68
[£t]5
^TXNV
t0LVNV
zat ava
thesequence V-O-S is otherwise applied withoutexception, theonlyvariablebeingthepositionof the phraseexpressing inscription "ona steleof stone."Examplesin whichthe phraseappearsbeforethe mentionof the 66.Thereare,in fact,no examples of the nominativeandimperative with thisofficial,onlyof the accusative and infinitive.Invariably, he appearsasTow ypaax£a Tow scaxa wpuxav£xaw, with doubledarticle.It maybe notedhere that,atIGII2463, lines31-32 (307/6), we findava[yp]a[@]ax 8£ T08£ T0
<)ri(p[t]a[F]a
To[w
Z]aT[a]
w()oTaV£Lal[w
a designation of the Secretary thatmayalsobe required at IG II2551,lines13-14 (paulloante 307/6)andatAgoraXV 322,lines2526 (ca.A.D. 120).This rarewordorder, alsointroduced in IG II2564,lines6-7, is rightlyeschewedbyWoodheadat ypa,u,uaT£a,
AgoraXVI l l l, line 19 (307/6-302/1 [sedqvEx post 306/5?]).
67. Forthe intrusiveintervocalic iotaandthe accentuation, seeThreatte 1980,pp.151-152. 68. Cf.the heavilyrestoredIG II2 1019,lines36-37 ( fin.s. II).
THE ATHENIAN
Io5
SECRETARIAT
STATE
thanthosewherethe phrasecomesafter arefarlessnumerous Secretary see,forexample,D 22 (IGII2222),lines Forthe former,69 the Secretary. 26-29 (ca.334): stoich.27
8£ T08£ T0 @N96a[Z]
avaypaWat
t0LVNV TOypaaT[£] [a] £65 ^qV [a] Tov zaTa =ovTav£tav zat orr sat [£V] axRoosoR
Hesperia43, 1974, pp. 322-323, no. 3, lines 17-21 (331-324?):
[t]
£V ^qNt
08£ T0 Wgtoa
stoich.22
8£ [T]
avaypaWat
Tovz[a] 0tvt ToyyRoa,u,ua£a £ Ta =ovTav£avzat sat ,
\
,
-
\
v axRoosoR£
of SEGXM 357, lines6-9 (286the slightlylongerwording70 Compare 262): non-stoich.ca.43 avaypaQat] zat ] v twzaRoXxv 8£ T08£ T0 @Ntoa zat Ta ovoa[a t0LV£t TOV [yRoaa£a TOV X] t)V vOXatoXXv £V ^vNt wovTav£Lav
xat
t
\
\
,
\
aTa
Tots
=oo5
orroat
te_at5.
IGII2339b,lines9-13 ( fin.s. IV): also,with,u =ooi£vtav,71 Compare stoich.25
a] vayRoaQaL ov
£t5
orr
[,u]aT£La72 [t orr
saL]
8£ QV =poi£v[Lav V Tov
t0LVNV zaT[a]
avT] yRoa]
T[ov
za]
=[o]vT[av£Lav
£V [aXtOOZO£t-
andnoteIG II2240, lines19-23 (337/6),in whichthe mentionof the untilafterorr sat: postponed is abnormally Secretary a] [va]y[o]aWat
[NANv] At0tvNv zat sat [a£]a [°=°X]£t-
69. Apartfromthe examplescitedhere range,note to indicatethe chronological also,e.g.,IG II2338, lines24-28 (333/2); AgoraXVI 248, lines22-24 (s. III/II). 70. Forthistypeof wording,wherethe objectincludesthe phrasescax Ta ovoZaxa
stoich.27
8£ QV =tOOi£VLa[V £65 ^T]
[T]ov za[X]a
[ov
yRoa]
=ovTav£t[av
£V axRo]
73
zX., cf.IG II2792,lines11-14 (275/4);AgoraXVI 188,lines45-49 (271/0). 71. Cf. the similarIG II2235,lines 26-29 (340/39),wherethephraseis omittedaltogether.
72. See n. 67 above. is rare;but 73. Suchpostponement of the Boule, cf.,withthe Secretary IG II229, lines7-11 (387/6),in which doesnot appearuntil the Secretary after£\ aXpOZO£t.
>
n
\
ALAN
Io6
S. H EN RY
Forthesequence V-O-S followedby"ona steleof stone,"74 whichis manifestlythesequenceemployedmostcommonlythroughout the entireperiodof activityof thePrytanySecretary, compare IGII2426,lines11-14 (336-334): 8£ T08£ T0
avayRoa@]aL [f(pLoa
Tov
yRoa,u,u]aT£a
stoich.25
+
zaTa
£V ^q]Xt t0LVt £V aXtOO]=O£t-
[zRovTav£Lav [XaL orr
Tov
saL
IGII2360.I,lines21-24 (325/4),withanextendedobject: 8£ T08
avayRoaQaL
£ T0
@ftoa
\
\
TOV yRoaaT£a ,,
zaL
TOV5
TOt
£V ^At
>
S
,
aMov5
TOV zaTa
,
\
£=aLvovs
ALULV£t zaL
TOV5 osaL
stoich.39
=ovTav£Lav ,
,
y£y£VN£VOV5 £V aXRoosoA£t-
av
Hesperia Supplement 17,1978,pp.2-4, lines105-107(270/69):
non-stoich. 42-49 QatT08£
T0
av TovzaTa spvo
[Q]gtoa Tovypaa£a
[£V] ^v£t
t0LV£t
zat osat
sapa v
oay[pa] nav£tav
£tzova
74. There are dozens of examplesof this formulation:e.g., IG II2483, lines 27-31 (304/3); II2500, lines 36-40 (302/1); II2505, lines 59-62 (302/1); non-stc )ich.ca. 36 II2641, lines 25-29 (299/8); D 68, lines 54-57 (295/4); D 74, lines 36-39 avayRoaFvat o£ T0o£ T0 F)N96^ (286/5); IG II2657, lines 68-70 ^TN (283/2); AgoraXVI 181, lines 37-40 £V (282/1); AgoraXVI 182, lines 27-30 (281/0); IG II2665, lines 31-33 and IGII2892,lines15-17 (188/7): (266/5); II2668, lines 33-36 (266/5); AgoraXV 89, lines 19-20 and 38-40 non-stoich. ca.39-40 (259/8)*;IG II2682, lines 87-89 (259/8?); II2780.A, lines 22-24 8£ T08£ T0 (253/2); II2788, lines 26-28 (235/4); AgoraXVI 224, lines 45-47 (226/5); ^q£t t06 IG II2786, lines 32-34 (ca.225?);Agora [V£t £]V aXtOOZO£t XVI 225, lines 18-20 (224/3-222/1); Hesperia47, 1978, pp. 49-50, lines As fora "telescoped" formwiththePrytanySecretary, I canciteonlyone 33 (ca.203); IG II2896, lines 17-1931example, D 88 (IGII2707),lines6-8 (286?): and 53-55 (186/5);AgoraXVI 291, lines 38-42 (169/8); IG II2949, lines 19-21 (165/4); II2 1006.I, lines 47-48 8£ T08£ T0 @fgto,ua [Tov yRoa,uR] stoich.37 and 96-97 (122/1); II2 1008.II, lines =ovTav£Lav £V X[t t0LVt £] 72-73 (118/17); II2 1009.I, lines 24-25 [V] aXtOOZO£tand 54-55 (116/15); II2 1011.I, lines 29-30 and II, lines 51-52 (106/5). The veryfactthatthisappears to betheonlyexample laterthantheendof (*Forthe archonsof the mid-3rd centhe 4thcenturymighthelpto supportthedatingof IGII2707in the ear- tury,I have in the main followed the lier partof the3rd.75 schemes of Osborne 1989 and 2000.) 75. See Henry 1990, pp. 182-183.
Agora XVI 224, lines 45-47 (226/5):
o
o
os
o
N
\
o
a xovyRoaa£aTov zaTa=ovTav£tav £V AtAt0tvt zatsat ayoRoat
ava[yRo]aWat Wgtoa[o] [vyRoaa£a]o[v]zata=ovTav£tav £V zatosat
avayRoaWat [a]X£a Tov zaTa
THE ATHENIAN
STATE
Io7
SECRETARIAT
THE RECORDER (o avaytoa(p£v)
Duringthetwobriefperiodswhentheavaytoa(p£Ls occupiedthepostof chiefsecretary,76 it is clearthatthepredominant, if notthesole,sequence is yet againV-O-S, forexample,D 31 (IGII2392 + 586),lines15-17 (321-318):77 avaytoaWat [£ T08£ T0 Wgaa]
Tov ytoaa£a
stoich.28
8
za
[ orr sat £V aXtOO=]O£t-
D 32(IGII2393),lines9-12(321-318): avaytoaWat To W(paa X£t t0LV£t 0z0X£t
zat osat
stoich.25
8£ T08£
Tov avaytoa(p£a
£V ^T
£V axto
-
SEGXL79 (IGII2407 + SEGXXXII94),lines15-18 (321-318): av] aytoaWat 8£ T08£ [T0 Wgaa 9£a
[£]V ^mNt
[t0LVt
stoich.31
Tov avaytoa]
zat oqaat
£V ax]
t00z0X£t-
D 36 (IG II2398b), lines 6-8 (318): avayto]aWat [+N(paa
Tov avayto]a(p£
[Xt0LV£tzat oqaa]
8£ T08£ T0
stoich.27
£V ^q£t
£V aXtOOZO£t-
SEGXLV 101 (IG II2649+),lines 48-50 (293/2): stoich.39 avaytoaWat8£ T08£ T0] @NCoa Tov avaytoa[£] [a £V orr Bats At0tvatsza]t orr sat rr v £V £V axto[o] [z0X£t, QV 8£ satoa qv £tz]ova
It willbe observedthat,apartfromD 31, whichmakesno reference to inscription "ona steleof stone,"alltheseexamples placethatphraseafter the mentionof the Recorder.78 Contrast IGII2396,lines4-6 (321-318): 76. See Henry1977,pp.50-66. 77. OsbornerightlyrestoresT08£ To fX4CoFa ratherthanthe unnecessary To +nx4CoFa T08£ of Kirchner and Karapa. To+N(paZa T08£ occursonly occasionally: so D 37, line 13 (318); D 48, line 6 (ca.303/2);PIGII2845, line 19 (paullopost 249?);II2810,line5 (ante230);AgoraXVI 310,line 50 (ca.135);IG II21019,line36 (fin. s. II);AgoraXV 264,line 15 (ca.80/79).
It shouldnotbe introduced gratuitously intorestorations, ashasbeendonein IG II2397,line2 (321-318);II2542, line 12 (ante303/2);II2516,line4 (fin. s. IV); II2521,line4 (fin. s. IV); AgoraXV 147,lines14-15 (203/2)notethatat line49 we findthe normalorder;AgoraXVI 276,line22 (190/89?) Osborne(D 100)hasthe correctwordorder;the faultoriginates in IG II2954;AgoraXVI 301,line5
( s. II,parsprior);IG II2984,line22 (ca.med.s. II);AgoraXV 236,line 1 (ca.150). 78. Cf. alsotheveryheavilyrestored D 34 (IG II2395),lines8-10 (321318);IGII2397, lines1-4 (321-318), where,incidentally, thewordorderTo +N(ps,ua T08£ iS unnecessarily introduced.(SeealsoHesperia 58,1989, p.86, no.12, fora possiblenewfragmentof IG II2397.)
Io8
ALAN
avaytoa@]at [V ^v£t
8£ T08£ [T0 Wgtoa
stoich.28
£]
t0]LV£t TOVa[vaytoa£a
S. HENRY
za]
[t orr sat £V aX]tOOZO£t-
andthe lesssecure,butquiteacceptable, D 33(IGII2394), lines16-18 (321-318):
stoich.38
avaytoa] [Wa] 8£ T08£ T0 W[gtoa [ayto]a(p£a
£V ^vNt
zat o[Nsat
.
t0LVt
TOVav]
£V aXtOOZO£t-
t
wherethephrase precedes thementionof the Recorder. One exampleof the sequenceO-V-S maybe noted,D 37 (IG II2 391),lines13-15 (318),a text"execrably executed":79 stoich.29-30 O8£ [avaytoaWat
To W]N(paa
[t At0tv]Nt Tov avayto[a(p£a
zat
£V^T!<X!>]
ava0£tv]
[aL80 £65a]Xtoso<X>tv Althoughheavilyrestored andmiserably cut,sothatcertainty ofreading is impossible, thesequencecannotbe doubted. THE SECRETARY OF THEPEOPLE (o ytoa,u,ua£vs
Tov
8N,uov)
Oncemore,thepredominant formulation is V-O-S,withthephrase"ona steleof stone"occurring apter mention of the secretary. So,forexample,82 D 61 (IGII2496 + 507+),lines36-39 (303/2): ytoaQat8£ ,
\
,
\
T08£ T0
,
\
,ua£a Tov8N[,u]ov £V^TXt zat osat
stoich.28
ava @Ntoa Tovytoa t0LVt
£V aXtOOZO£t-
aVa0£tuaL
D 79 (IG II2712),lines16-18 (ca.273-262): ava]ytoaQat8£ T08£ [T0 @NCoa Tov] [ytoa,u,ua£aTov8],uov £V^v[Nt t0LVt zat ox] sat £V aXtOOZO]£t-
stoich.37
IG II2844.I,lines28-30 (229/8):
non-stoich. 42-50 8£ T0
avayRoaQat
8£ T0 [v]v
@Ntoa
zat
Tov
sat
yRoaa£a
saRoa
Tov
rr
v
8Ntov
79. So Osborneadloc. 80. Forthe lateoccurrence of thansa), cf. IGII2741,line 10 (init.s. III). 81.Thatthereis no connective8£ at the beginningof the provisionis to be explainedby the introductory clauseof "hortatory intention" (seeHenry1996). This hasallowedthe drafterto employ the uncommonwordorderToWr-aZa T08£ (seen. 77 above),unlessthistoo is to be set downto the carelessness of the cutterso rightlydeploredby Osborne. 82.Apartfromthe examplescited here,cf.alsoD 89 (IGII2570),lines 11-13 (s.III;see SEGXL 89);IGII2 651,lines26-28 (286/5);andthe more elaborate IG II2660.II,lines43-45 (281/0).
£65 ^TXNV
t06
£tzova
ContrastAgora XVI 164, lines 17-20 (between300/299 and295/4):
(rather
STATE
THE ATHENIAN
SECRETARIAT
IO9
£V ^q£t
T08£ T0 Wgtoa ov ytoa,u,ua£a
stoich.29
8£]
ava[ytoaWat
[Xt0LV£tT]
Tov 8N,uov za[t orr sat £]
V aXtOOZO£t-
mentionof thesecretary.83 wherethephraseprecedes we appearto havethe In IG II2845, lines19-21 (paullopost249?)84 sequenceO-V-S: non-stoich.41-44 T]0 8£ @N(pt[aa
T08£ avaytoaWat
£V]
8N,uov zat orr sat £V]
[ov
At0t[v]N[t] to[v yto]a,u,ua£a
[aqA]t
[aXtOOZO]£t-
andwe alsofindS-V-O inAgoraXV1213, lines22-24(248/7):85 stoich.38 TOV 8[N]OV
TOV8£ ytOaaT£a zat Ta ovoaTa
[8]£ T0 Wgt[aa]
v
avaytoaWat
To
£zL80VT[X]V £V ^v£t
V \
£t z[a]t oqaat
,
,
ayotoat
£V t
zX.
formin IG II2672,lines It wouldseemalsothatwe havea "telescoped" 74): XEVIII see SEG 280-270; 14-16 (ca. stoich.68 [oaQat8£ T08£ T0 @N96aa86 zat ]as 8Xto£a[g ytoa,u,ua£aTov8N,uov£VaXtOOZO£t [£V
avay] To]v ^T]
[X£t t0LV£t-
THE SECRETARY (o ytoa,u,ua£vs) tout aso ytoa,u,ua£vs designated wefindtheSecretary Ona fewoccasions withoutfurtherqualification. where,for exmaygo backto the 5th century, This phenomenon in IGI3102,lines21-22 (410/9): ample,it is restored
court,
83. So too in the heavilyrestored IG II2809,lines1-3 (ca.300).
84. Forthe date,seeTracy1988, p. 320. alsoin the heavily 85.Andprobably AgoraXVI 178,line 7 (286restored 262?). To 86. I havealteredthe unnecessary f gtoZa T08£ of the Corpus.(At IG II2 542,line 12 [ante303/2] I would likewisereadT0[8£ T0 f gaZa.) See n. XX aDove.
zat avaytoa(psa
[o hoytoaa£vs
stoich.36
Ta £9a£96a]£Va
The referenceis clearlyto the seemsinescapable. Herethe restoration later,withhisfulltitle,in connection of theBoule,whoappears Secretary aseuergetai (lines fellowconspirators ofThrasyboulos's withtherecording 28-30): £V£t0y£[Ta]5 [a]vaytoav sat £ a £5
W0X£[t £V £X£t 50X£5-
]t0LV£t TOVytoa[a]T£
ALAN
IIO
S. H EN RY
Anotherexampleis foundin IGII2276,lines18-20 (ca.342): To 8£
T08£
@f(pLoa
[yto]aiuiuaT£vs
£V
^v£t
avaytoaQag87
t0LV£t
stoich.31
[O]
sa[T]
[fi)] £V aX[tOOZO]X£t-
Thenthereis a smallclusteraroundthe turnof the 4th/3rdcentury:88 IGII2456b,lines28-29 (307/6): stoich.41 avaytoaWat X
8£ T0[8]£ T0 [Qgt]a[a] Tov [ytoaa£a za[t] orr [a]at [£V a]XtOOZO[X£t-
A0tvt
£V ^T]
D 56 (IGII2519),lines3-5 (ca.307-303/2),wherespatial
considerations
maketherestoration virtually inescapable: avaytoaWat To Wgto[a 0tvt
Tov ytoa,u,ua£a
stoich.31
t]
za[t orr sat £V aXtOOZO£t-
166,lines5-8
Agora Xvl
8£ T08£]
£V ^qNt
(295/4
avaytoaWat 8£ T0 W[gtoa
selpaullo
post):
8£ T0]
stoich.22
Tov ytoaa£]
a £V ^q[Xt
t0LVt zat ] sat £V a[XtooZo£t 89
Woodhead alsorestores theytoaaT£v5 toutcourt inAgora XVI214,lines 22-23(244-241),where,witha slightlyshorterlinethanthatenvisaged byMeritt(stoichedon 52, asopposedto 55),he proposes: 8£ T08£ T0
avaytoaQaL
*
[XaL
*
.
@fto[a
zaL
T
Ta
ovoaTa
TOV ytoaaT£a]
£V TOt T£T£V£t TOV tL[05-
oT]foaL .
.
*
t
*
t
t
SECRETARY NOT EXPRESSED Aswillhavebeenclearfromthepreceding analysis, theAtheniansregularly specifiedthedesignated secretary chargedwiththetaskof seeingto the inscribing anderectingof decreesandlaws.Not entirelyunexpectedly, however, we findinstancesearlyin the5thcenturywheretheresponsible official is onlyimplied,ratherthanexplicitlyspecified.Compare, forexample, IGI323,lines5-11 (ca.447),a proxeny grant,wheretheawardand the publication provision arecombined:90 Kop[oa]yt8£v (i)aXvXt8£v aL
zaL
M£V£aTt0aT0V
'A0£vatov To5 (9£a=La5
[a]psaL
z[aL]
=00X^£VO5
zaL
X
avayto
£V£0y£Ta
87. Forthe participleseep. 103 above. 88. Cf.Woodhead,inAgoraXVI, p.240 (onAgoraXVI 166). 89.The samewordingis foundin IGII2567,lines17-18 (fin. s. IV), which Woodhead(AgoraXVI, p. 240) calls, perhapstoo confidently, a"necessary restoration." 90.Walbank(1978,p. 89) draws attention to the factthat"theletterforms areunusual,moreappropriate to a Boiotianthanto anAthenianinscription." This point,togetherwith the absenceof anymentionof the Secretary, suggeststo himthatthe honorands themselvesmayhavelooked after theinscribingof theiraward. However, in viewof the factthatthe poletai areto let outthe contractand the kolakretaiareto providethe funds, this seemsunlikely.
STATE
THE ATHENIAN
III
zat tos satAas tos
[5 'A]0£vatov [£X£VO]V £
SECRETARIAT
W0X[£]t £V OT£X£t t06
[V£t.
Compare IG I392,lines9-13 (422/1): zat avaytoaQat avTov £^ t0LVt WtOOi£VOyzat £V £t0y£V 'A0NvaLOvavToy zat to [5] 7ratAa5tos KaR;Atsozat zata
stoich.25
Xt
[0£VaL £
=0X£t].
In non-proxeny texts,comparethe decreeon the EleusinianEpistatai, IG I332, lines32-34 (ca.449_447):91
a
£V AT£X£t
a[£]pO
ytoa(psat 8£ T0 [9a£96a] ER£votutza[ £V asT£L zat O]
stoich.32
£V T06 ER£V^LV[606.
andthe steleis not describedas "of Herethe provisionis "telescoped," stone,"as likewisein a secondexample,Sgora XVI 7a, lines 17-18 (439/8?): ytoa(psat 8[£ TavTa] stoich.35 [£V ^T£X]£t
zat
zaTa0£vaL
£
=0X£t-
byzat zaTa0£va. is introduced where,however, theerectionprovision In bothof theseinstances(IGI332 andAgora XVI 7) the subjectof the infinitivecouldbe thoughtof astheAtheniandemos,ratherthanthe Secretary of the Bouleas such,just as we find in the Treatywith the to have Bottiaians,Xgora XVI16,lines21-25 (422),wherewe alsoappear ofavaytoa(px forheretheparticiple avariant ontheformwiththeparticiple; is preceded bytheverbof erecting: stoich.42 Tas 8£ XAV zat] Tov [hotozov zaTa]0£vat 'A0£vatog ,U£ £aT£X£t] Xt0LV£tzat Ta ov V £ =0X£[t aVaytOa]9a[aVTag [o]aTa Tov [z0X£0V] TO[V BoTtatov ]ov XAVVTL0£0£VOV zX. T£V tAta[v zat T£V Xavaxtav v0£Xaq
Ta[a8£
bythefollowing: Additional instancesarefurnished formula1) IG II282, lines14-16 (ca.390-378?),another"telescoped" tion: adatenot 91.Matangly prefers theyear432/1. before 433/2;Clinton p.938. SeeIGI3Addenda,
T0 8£ @N96a
[a T08£ avaytoa]@at£V aXt00z0X£t [£V ^qNt
t0]LVt
stoich.25
ALAN
II2
S . H EN RY
2) IG II2125, lines 17-19 (343?): avaytoaWat a,
\
tvt
zat oNsat
,
8£ TO [Qgtoa
,
£aXt
£V axtoow OA£t zat £V t ,>
\
,
t]
stoich.39
,
ayotoat
zat £V TOt t£VL
wherethereseemsto be no roomfor08£.92 3)AgoraXVI 79,lines18-21 (332/1): ava[ytoaWat [£]V ^TXt [0]=O£t
8£ T08£ T0 Wgtoa]
[Xt0LVt zat osat
stoich.29
£V axto]
-
although Woodhead, notingthat"thetitleof theyRoa,u,ua£vs exactlyfills thelineandmayhavebeenomittedby oversight,"93 is inducedto insert asline19 . Woodhead notesfurtherthat"therearein facta fewapparently intentional omissionsof the words, e.g.,IGII2493 +518,508,648(=M.J.Osborne, NaturalizationI, pp.128-130,D 53, D 54 andpp.148-150,D 69, respectively), 123and 141,"94 although howonedistinguishes betweenacarelessly omittedphrase andanintentionally omittedoneis notat allclear. ToWoodhead's listof "apparently intentional omissions" onemayadd: 4)IG II2448.I, lines 26-28 (323/2): stoich.41 avaytoaQat 8£ T0 @NCo[a £V oNkats A] t0tvatszat orr sat rr v £V tav satoa [ov Ata, qv 8£] £T£t0aV £V a[Xtooso£t satoa TovV£O N5 HoRtaAos
where we havealready notedtheomissionof TO8£.95 At AgoraXVI 141(#n. s. IV [ca.304/3?]),too fragmentary to merit listing here,Woodhead's note(p.214) on the omissionof the Secretary could be takento implythatsuchanomissionis mainlyconfinedto the last yearsof the4th century. Thephenomenon, however, continuesto be met forsomeconsiderable timeafterthat.Compare thefollowing: S) SEGXVIII22, lines20-21 (165/4):
non-stoich.36-51 8£ T0 @N96aZa96 £V £V TOt TOV AsxAztov t£pO[t] -
avaypaQaL
ATrXY16 t0LVY16zaL
oTf oaL
6) IG II2lOll.IV,line72 (107/6):97 8£ [T0
avay[oaQaL] ,
avToL5
,
,
N
£zLo£60V
@]ftoa ,
£t5 N
£LVaL00Xt.
XfV
t0LVNV
zaL
oTfoaL
ov
av
92.T08£ iS onlyoccasionaXy omitted in thisphrase:so IG II2448.I,line26 (323/2);SEG XVIII22 (IG II2950), line20 (165/4);IG II2lOll.II, line51, III,line62, IV,line72 (106/5).In IG II2983,line9 (ca.med.s. II), however,thereis sufficientroomto read[T08£ T0 +NtoZa. 93.AgoraXVI,p. 121. 94. I.e.,AgoraXVI123 and141. 95.Seen.92above. 96. Forthe omissionof To8r,see n.92 above. 97. AlsowithoutTo8r.
THE ATHENIAN
STATE
SECRETARIAT
II3
7) IG II21039,lines65-66 (79/8): a[vaytoa]@aL
8£ T08£ T0
£Ta
@[fto]a
TOV
[aOv
£t5 qV
[av]v oq)trlvon2t98 of SypaletFinally,we maymentionthe decreein honourof Sosandros 15, p. 75, no. 16 (IG II21023),lines18-22 tos,99HesperiaSupplement anderectionis given, forinscription (Jin.s. II),wherethe responsibility of the butto "themenelectedforthe zaTasx£vN100 notto the Secretary, Templeof Athena": non-stoich.ca.38 avaytoaQaL zaL
Tov
oT£avov
£t5
oqV
8£ T08£ T0 t0LVNV
av8toa5 £=t QV tOOTOVNN£VOV5
@ftoa
Tovs
X£X£t
xaTasX£vNv
Tov vaov rr 5 'A0Nvas vvzat orr sat
£V axtooso
£t -
CONCLUSION
98. Cf. alsoIG II21043,line58 (37/6?),whichhasanidentical wording. 99.Tracy(1975,p. 76) notesthat"it wouldbe especiallyinterestingto know benethe exactnatureof Sosandros' andhowit ficenceand(ptRorXvta affectedthe Parthenon." See Dinsmoor lOO."Decoration"? 1934,p. 102.
that,althoughon occasion,from It is manifestfromtheabovediscussion for the officialresponsible texts designate the 5th centuryonward,some or (o ytoa,u,ua£vs) anderectingsimplyas"theSecretary" theirinscribing of cases evenneglectto insertmentionof himat all,in thevastmajority specifiedwitha precisetitle. theofficialis carefully of the Council" The earliesttitle we encounteris the "Secretary firstfoundaroundthe middleof the5th cenq5 ,BovANs), (ytoa,u,ua£vs (ytoa,u,ua£vs Secretary" bythe"Prytany supplanted tury.He is gradually Froma 4th century. of the the middle from around zata =ovTav£tav), tov of the People"(ytoa,u,ua£vs the"Secretary pointlatein the century, thatthesethreetitles however, 8N,uov) entersthescene.It is notunlikely, of theholderof oneandthe aresimplythat threedifferentdescriptions sameoffice. (321/20to 319/18and294/3to 292/1)when In thetwobriefperiods it is noneof wassuspended, of government apparatus thefulldemocratic decrees of having the function to perform called upon who is thesethree (avaypa(p£vs). anderectedon stone,butthe"Recorder" inscribed it is hardlyto be expectedthatalltheelementsthatconstiNaturally, tutethe filllestformof the wordingof the provisionfor inscribingand erectingwill all occuron everyoccasion,or all in the samesequence,or Andthis withinthevariouselementsthemselves. withoutminorvariations is borneoutby a detailedanalysisof theextantevidence. orof concerned ofthe (titleofthe) Secretary regardless Nevertheless, it is evidentthatoneprincipal or,later,infinitive, thechoiceof imperative sequencefor the itselfas the predominant essentialpatternestablished followed, regularly V(erb)-O(bject)-S(ubject), provisionfor inscribing:
II4
ALAN
S. HENRY
ratherthanpreceded, by"ona steleof stone"(and,of course,thatin turn followedby the wordingforthe erection). The sequenceO-V-S is also found it is indeedprevalent in the 5th century as is, occasionally, the sequenceS-V-O.In proxeny grants,we haveminorevidenceforO-S-V, V-S-O, andevenS-V withtheobjectimpliedratherthanexpressed. Tworefinements of thesebasicpatterns areto benoted.In thefirstof these,whichI havetermedthe "syntactically reduced" formulation, the verbof inscribing is reducedfroman imperative or infinitiveto a participle,thusleavingas the onlyfiniteformtheverbof erecting. Thistype appears notinfrequently in the5thcenturyandoccasionally in the4th.In thesecondtype,conversely, andagainmostcommonly in the5thcentury andin proxeny grants,we encountered thephenomenon of the so-called "telescoped" formulation, in whichdetailsforthe erectingareappended directlyto the verbof inscribing, with no intervening separate verbof settingup. Whatemergesstrongly fromthisstudyis therealization thatthelanguageof thestockprovisions of Atheniandecrees, whilefirmlyformulaic andtendingtowarda predominant form,neveradoptsa single,standard wording.Variatiadmittedlywithin fairlynarrow parametersis always likelyto beencountered.
T H E AT H E N I A N S TAT E S E C R E TA R I AT
IIS
A PPEN D IX TEXTS SPE{IALLY {ITED OR DIS{USSED
TEXTS WITH IG I3 REFERENCE IG I3
Walbank1978
Agora XVI
95
10
99
ll
100
15
23 24 27
Page(s)
110-111
ll
103
23 13
101 101; 111
32
100
37
96; 97; 103
40
7
46
15
62 65 66
39
70
19
76
14 16
111
103 96 97 100-101 111
96 102 103
78 80 84 92
65
111
98.I
75
97 109 96 96 103 99
102 106
85
110
87
118 119
28A
125 155
156 174
24 22 50
11
100 101
97 102
ALAN
II6
S. HENRY
TEXTS WITH IG II2 REFERENCE IG II2 13a + 68 (SEGXL 54) 17 25
Osborne 1981
D8 D9
Agora XVI
36 40
43 57 76 80 82 107 125 139+ 289 (SEG XXXIX75) 149 222 D 22 226 D 14 232 238 240 242 + 373 (SEGXL 74) 276 339b 351 + 624 354 360.I 391 392+586 393 394 396 398b 407 426 448.I 456b 496+ 507 519 542 649(SEG XLV 101) 653 672 707 712 791(SEGXXAhI 118) 844.I 845 892 983 lOll.IV 1023 1039
Page(s)
101-102 102 102 99 104 99 98 111
97 112 102 n.62 98-99 105
97-98 98 102
-
105
98 110 105
D37 D31 D32 D33 D36 106J
D38
98 104 106 108 107 107 108 107-108 107 107 106 112 110
D61 D56
108 110
D88 D79 213
109 n.86 107 104 109 106 108 109 108 109 106 112 n.92 112 113 113
THE ATHENIAN
STATE
SECRETARIAT
II7
TEXTS WITH NO IG REFERENCE Hesperia
4, pp.525-529,no.39 8, pp.26-27, no.6 13,pp.242-243,no.7 17,p.11 21, pp.355-359,no.5 28, pp.185-186,no.7 30, p.210, no.4 36, pp.59-63, no.6 37, pp.268-269,no.4 43, pp.322-323,no.3 Suppl.17,pp.2-4
SEG Osborne 1981 XXV 106 XXXV 71 XXIV 119 XXIII 67 XII 87 XVIII 22 XXI 342 XXIV 135 XXV 84 XXI 357 XXVIII 60
Agora
XVI 224 79 164 214 73
Page(s) 106 112 108-109 110
98 112
141 310 166 106H
112 104 110 105 105
255D
106
REFEREN C ES AgoraXV = B. D. Merittand J. S.Traill,Inscriptions:TheAthenian Councillors, Princeton1974. AgoraXVI = A. G. Woodhead, Inscriptions: TheDecrees,Princeton
1997. Alessandri, S. 1982."Alcune osservazionisuisegretariateniesinelIV sec.a.C.,"AnnPisa12,pp.7-70. Brillant,M. 1911.Lessecretaires atheniens,Paris. Dinsmoor,W. B.1931. TheArchonsof Athensin theHellenisticAge,
Cambridge, Mass. .1934. "TheRepairof the AthenaParthenos: A Storyof Five Dowels,"AJA 38, pp.93-106. Dow,S. 1937.Prytaneis(Hesperia Suppl.1),Athens. Ferguson,W.S.1898. TheAthenian Secretaries (CornellStudiesin ClassicalPhilology7), NewYork. . 1914-1915."TheIntroduction of the Secretary-Cycle," Klio 14, pp.393-397. Henry,A. S. 1977.ThePrescriptsof AthenianDecrees(Mnemosyne
Suppl.49), Leiden. .1982."Polis/Acropolis, Paymasters, andtheTenTalent Fund,"Chiron12,pp.91-118. . 1983.Honoursand Privilegesin AthenianDecrees(Subsidia Epigraphica 10),Hildesheim. . 1990."BithysSonof Kleonof Lysimacheia: FormalDating
CriteriaandI. G. ii2 808,"in Owls toAthens:Essayson ClassicalSubjects Presentedto SirKennethDoqJer,
E. Craik,ed.,Oxford,pp.179-189. .1996."TheHortatory Intentionin AthenianState Decrees," ZPE 112,pp.105-119. .1998. "TheSigmaEnigma," ZPE 120,pp.45-48. Mattingly,H. B. 1984.Reviewof D. M. Lewis,InscriptionesGraecaei3: InscriptionesAtticae annoEuclidis anteriores,fasc.1,Berlin,1981,in AJP 105,pp.340-357.
. 1999."WhatArethe Right DatingCriteriaforFifth-Century AtticTexts?" ZPE 126,pp.117122. Meritt,B. D.1961. "GreekInscriptions,"Hesperia30, pp.205-292. Osborne,M. J.1981.Naturalizationin Athens1, Brussels. . 1989."TheChronologyof Athensin the MidThirdCentury B.C.," ZPE 78, pp.209-242. .2000. "Philinosandthe AthenianArchonsof the250s B.C.," in Polisand Politics:Studiesin AncientGreekHistoryPresentedto MogensHermanHansenon His SixtiethBirthday,August20, 2000,
P.Flensted-Jensen, T. H. Nielsen, andL.Rubinstein, eds.,Copenhagen,pp.S07-520. Pecirka, J.1966. TheFormulaforthe GrantofEnAtesisin AtticInscriptions
II8
(ActaUniversitatis Carolinae, Philosophica et Historica Monographia 15),Prague. Pritchett,W. K.,andB. D. Meritt 1940.TheChronology of Hellenistic Athens,Cambridge, Mass. Rhodes,P.J. 1985.TheAthenianBoule (1972,reissuedwithadditionsand corrections), Oxford. . 1993.A Commentary on the AristotelianAthenaionPoliteia (1981,reissuedwith Select Addenda),Oxford. Schweigert, E. 1938."Inscriptions from the NorthSlopeof theAcropolis," Hesperia7, pp.264-310. Threatte, L. 1980.TheGrammarofAttic Inscriptions1: Phonology,Berlin. Tracy, S.V. 1975.TheLetteringof an
ALAN
AthenianMason (Hesperia
Suppl.15),Princeton. .1988. "TwoAtticLetter Cuttersof theThirdCentury: 286/5-235/4 B.C.," Hesperia57, pp.303-322. Walbank, M. B.1978.Athenian Proxeniesof theFifth CenturyB.C.,
Toronto. Whitehead,D.1989. "Secretaries, Charidemos, Poteidaia: The Date (andPersonnel) of IG II2118,"The AncientHistoryBulletin3, pp.102106. . 1998.Reviewof C. VeligianniTerzi,Wertbegriffe in denattischen Ehrendekreten derklassischen Zeit, Stuttgart1997,in BrynMawr ClassicalReview 9.5,pp.491-494.
AlanS. Henry UNIVERSITY OFST. ANDREWS SCHOOL OFGREEK, LATIN,ANDANCIENT HISTORY ST.ANDREWS KYI69AL SCOTLAND [email protected]
S. HENRY
(.W. BLEGENAND M. RAWSON
RecentlyPublished
REVISED AND EXPANDED BY J. L. DAVIS AND C.W. SHELMERDINE A
70 pages,39 figures(mostin color) ISBN0-87661-640-6 May2001.Paper$7.50/ £5.75
Guide
to
the
Palace
of
Nestor
The classicguideto theBronzeAge Palaceof Nestorat Pylos,nowillusto inExpanded tratedin fullcolor,includingPietdeJong'swatercolors. as a resultof of nearbysitesaswellas sitesdiscovered cludedescriptions Project.An by the PylosRegionalArchaeological recentinvestigations servesasa guideto theChoraMuseum. appendix
Forthcoming
(HARLES K. WILLIAMS 11 AND NANCY BOO KID IS, ED ITORS Corin f h) f he Cen f enary: I89 6 I99 6
550 pages,500 illustrations, 31 tables E Corinth ISBN 0-87661-020-8 May2002.Clothca. $75/ £57.50 Ca.
attheDecember of thepaperspresented twenty-five Thisbookpublishes anniverone-hundredth the celebrate to Athens in held 1996symposium at Ancient saryof theAmericanSchoolof ClassicalStudiesexcavations therangein subjectmatter Corinth.Thepapersareintendedto illustrate by scholarsof AncientCorinth, beingundertaken currently of research workfor andtheirinclusionin onevolumewillserveasa usefulreference Eachof the topics whichvarywidelyfromCorinthian nonspecialists. by the to Byzantinepottery is presented geologyto religiouspractices expertin thatarea. acknowledged of articlesandvolumes Thebookincludesa fullgeneralbibliography of onehundred at Corinth.As a summary excavated material concerning to come. of scholars it willbe usefulto generations yearsof research,
HESPERIA SUPPLEMENTS 13 MarcusAurelius:Aspects of Civicand CulturalPolicyin theEast,
byJamesH. Oliver(1970).$15.00 14 ThePoliticalOrganizationofAttica,byJohnS.Traill(1975).$15.00
16 A Sanctuaryof Zeuson MountHymettos,by MerleK.Langdon(1976). $15.00 17 Kalliasof Sphettosand theRevoltofAthensin 286 B.C., byT. LeslieShearJr.(1978).$15.00 19 Studiesin Attic Epigraphy,History,and Topography Presentedto Eugene Vanderpool (1982).$15.00 20 Studiesin AthenianArchitecture, Sculpture, and Topography Presentedto HomerA.Thompson(1982).$15.00 21 Excavationsat Pylosin Elis, byJohnE. Coleman(1986).$25.00 22 Attic GraveReliefsThatRepresentWomenin theDressofIsis, by ElizabethJ.Walters(1988).$40.00 23 HellenisticReliefMoldsfAromtheAthenianAgora, byClaireveGrandjouan (1989).$25.00 24 ThePrepalatialCemeteries at Mochlosand Gourniaand theHouseTombsof BronzeAgeCrete,byJeffreyS. Soles(1992).$35.00 25 Debrisproma PublicDiningPlacein theAthenianAgora, by SusanI. RotroffandJohnH. Oakley(1992).$35.00 26 TheSanctuaryofAthenaNike inAthens:Architectural Stagesand Chronology,by IraS. Mark(1993).$50.00 27 Proceedingsof theInternationalConference on GreekArchitectural Terracottas of the Classicaland HellenisticPeriods,December12-15, 1991,
editedby NancyA. Winter(1994).$120.00 28 Studiesin ArchaicCorinthianVasePainting,by D. A. ArnyxandPatricia Lawrence(1996).$65.00 29 TheAthenianGrain-TaxLaw of 374/3 B.C., by RonaldS. Stroud(1998). $35.00 30 ALMIA CeramicKilnin South-CentralCrete:Functionand Pottery Production,byJosephW. Shaw,AleydisVande Moortel,PeterM. Day, andVassilisKilikoglou(2001). $35.00
-
l_l -
T HE J - OURNAL
-
OF
THE
SCHOOL
AMERICAN
O F C L A S S I C A L S T U D I E S AT
AT
H
E N
VOLUME7I: NUMBER2 APRIL-JUNE 2002
AmericanSchoolof ClassicalStudiesaf Athens 2002
S
The AmericanSchoolof ClassicalStudiesatAthensis a research and teachinginstitutiondedicated to advanced studyof thearchaeology, art,history,philosophy, language, andliterature of Greeceandthe Greekworld.Established in 1881bya consortium of nineAmerican universities, the Schoolnowservesgraduate studentsandscholars frommorethan150affiliated collegesanduniversities, actingasa baseforresearch andstudyin Greece.The mainbuildingsof the Schoolandits libraryarelocatedin Athens,withadministrative and publications officesin Princeton, NewJersey.As partof its mission, the Schooldirectsongoingexcavations in theAthenianAgoraandat CorinthandsponsorsallotherAmerican-led excavations andsurveys on Greeksoil. It is theofficiallinkbetweenAmericanarchaeologists andclassicists andtheArchaeological Serviceof the GreekMinistry of Cultureand,assuch,is dedicated to thewisemanagement of cultural resources andto thedissemination of knowledge of the classical world.Inquiries aboutmembership in the Schoolorparticipationin the SummerSessionsshouldbe sentto theAmerican Schoolof ClassicalStudiesatAthens,6-8 CharltonStreet, Princeton, NewJersey08540-5232. Hesperiais published quarterly bytheAmericanSchoolof Classical
StudiesatAthens.Foundedin 1932anddevotedprimarily to the timelypublication of reportson School-sponsored andSchooldirected projects, Hesperiawelcomessubmissions fromallscholars workingin thefieldsof Greekarchaeology, art,epigraphy, history, andliterature, fromearliestprehistoric timesonward.Hesperiais a refereed journal.
VOLUME
7I:
NUMBER
APRIL-JUNE
2
2002
I
X|-El l
THEJOURNAL OFTHEAMERICAN SCHOOL OFCLASSICAL STUDIESATATHENS
PUB LI CATI ON S STAFF EDITOR-IN-CHIEF Kerri
E D I T O R,
Cullen
119
JOHN K. PAPADOPOULOS AND EVELYNLORDSMITHSONt
MONOGRAPHS
Michael
Fitzgerald
PRODUCTION Sarah
MANAGER George
MANUSCRIPT Carol
A NewBronzeAge Sitein the Corinthia: The Orneaiof StraboandHomer?
Cox
Hesperia
Tracey EDITOR,
JEANNETTEMARCHAND
The CulturalBiography of a CycladicGeometricAmphora: Islanders in AthensandthePrehistory of Metics
149
Figueira
EDITOR
A. Stein
YANNISA. LOLOS A PublicColumnDrumfroma Corinthian Quarry
PUBLICATIONS COMMITTEE CarolC. Mattusch(Chairman) GeorgeMasonUniversity DariceBirge LoyolaUniversityof Chicago JeniferNeils CaseWesternReserveUniversity ThomasG. Palaima Universityof TexasatAustin JamesP.Sickinger FloridaStateUniversity KathleenW. Slane Universityof Missouri-Columbia StephenV.Tracy(ex ofScio) Ohio StateUniversity
201
NORADIMITROVA Inscriptions andIconography in theMonuments of theThracianRider
209
Submissions:Manuscriptsand communicationsshouldbe addressedto Dr.TraceyCullen,Editor,Hesperia,AmericanSchoolof ClassicalStudiesat NewJersey08540;tel.609-683-0800; Athens,6-8CharltonStreet,Princeton, andphotocopiesof ilManuscripts fax 609-924-0578;[email protected]. originalartworkandphotographs lustrationsmustbe submittedin triplicate; aremadewith the Editor.A shouldnot be sent unlesspriorarrangements ofthe articleshouldalsobe themajorconclusions shortabstractsummarizing reviewprocessandauthors included.Articlesaresubmittedto a double-blind withouttheirnameor accordingly, arerequestedto preparetheirmanuscripts notes,bibliograpreparation, The stylefor manuscript affiliationappearing. canbe foundin the Guidelinesfor on submissions phy,andotherinformation Authorspublishedin Hesperia62, 1993,pp.i-xvi;on the School'sWeb site at the aboveaddress. orbywritingto ASCSAPublications (www.ascsa.org);
Copyright(C)2002 The AmericanSchoolof ClassicalStudiesat Athens
The AmericanSchoolof ClassicalStudiesatAthenswillnotknowinglyprint orinitialscholthe announcement in Hesperiaor anyof its otherpublications afterDecember30,1970,byanymeans of anyobjectacquired arlypresentation otherthanthroughan officiallysanctionedexcavationor survey,unlessthe existingcollectionorwaslegallyexportedfrom objectwaspartof a previously the countryof origin.
postagepaidat Periodicals Princeton,NewJersey,andat additionalmailingoffices.
Producedat EdwardsBrothers, Inc.,AnnArbor,Michigan. Designby EllenMcKie. Thracianrider Coverillustration: relief.GlavaPanega,Bulgaria. Museum, NationalArchaeological Sofia,inv.3906.Photocourtesy Museum.
Sendaddresschanges Postmaster: to Hesperia,P.O.Box529, Canton,
MA 020214529, U.S.A. price The annualsubscription Hesperiais publishedquarterly. Subscriptions: $70 forinstitutions($80,interna- ISSN 0018498X is $60 forindividuals($70,international), Paymentmust ISBN 87661-500-0 tional),and$33forstudents(proofof studentstatusrequired). beinU.S.dollars,drawnonaU.S.bankorbymoneyorder,andsentto Hesperia, SubscriberServices,P.O. Box 529, Canton,MA 02021-0529;tel. (U.S.) 800-821-7823; (outsideU.S.) 781-828-8450;fax 781-828-8915; e-mail of Singleissues(currentandbacknumberswhenavailable) [email protected]. for$15 eachpluspostagefromthe DavidBrownBook Hesperiaareavailable Company,P.O.Box 511, Oakville,CT 06779;tel. 800-791-9354,860-9459329;fax 860-945-9468;or (outsideNorthAmerica)OxbowBooks,Park End Place,OxfordOX1 1HN, U.K.;tel. +44 (0) 1865-241249;fax +44 (0) 13-17 and19-30 arealsoavailable 1865-794449.IndexII andSupplements fromDavidBrownBooks.Reprintsof IndexI, Supplements1-12 and 18, b.v., andearlyissuesof Hesperiashouldbe orderedfromSwetsandZeitlinger, [email protected]. P.O.Box810,2160 SZ Lisse,Netherlands;
HESPERIA
7I (2002)
A
N
EW
B RONZE
PagesII9-I48
SITE CO
AG
E
INTHE R I NTH
IA
THE ORNEAIOF STRABOAND HOMER?
ABSTRACT A newlydiscoveredBronzeAge site is reportedat Doratiin the Corinthia, overlookingtheNemeaRiver.Surfacematerialindicatesthatthiswasalarge Mycenaeansettlement,with structurespotentiallywell preserved,andthat earlierperiodsof the BronzeAge arealsorepresented. The locationof the sitemakesit possibleto identif;r it tentativelyasthe CorinthianOrneaimentionedby Strabo,whoimplies(contra Pausanias)thatthisis the sitereferred to in the Catalogueof Ships.I suggestthatStrabois correct,andthatDorati mayinfactbeHomer'sOrneai.Acceptingthisidentification helpsclarilithe logicbywhichsitesin Agamemnon's realmarelistedin the Catalogue.
A previously unknownMycenaean siteoverlooking the NemeaRiverin the northeastern Peloponnese has recentlycometo my attention.1 It is largeandthesurfacematerial copious,diverse,andof highquality, yetno mentionof it hasappeared in anyscholarly publication.2 Therefore I offer 1. On November14, 1999,I first visitedthe areaof Dorati,northof the modernvillageof Soulinari,in the companyof P.Panagopoulos of Derveni(nearKiato),who hadlearnedof it fromlocalresidentsas thepossiblesite of a Classicaltemple.On September 20, 2000,I reportedthe siteto the FourthEphoriaof Prehistoric and ClassicalAntiquitiesin Nauplionand fileda draftof thisarticlewiththe ephorat thattime,E. Spathari; in Octoberof 2001 I fileda seconddraft withthe currentephor,A. Mantis. I thankthembothfortheircomments. I wouldalsoliketo thankN. Bookidis, M. Boyd,B. Burke,L. Costaki,J. Herbst,G.J. Lolos,S. G. Miller,G. Sanders,R. S. Stroud,andI. Tzonou
forvisitingthe sitewithme;andLolos, Miller,Stroud,andTzonouforreading earlierdraftsof thisarticle.I owea specialthanksto Herbstforpreparing the accompanying maps.Finally,I am gratefillto the anonymous Hesperia reviewers fortheirsuggestions. 2. The sitehasneverbeenexcavated or systematically surveyed. It is approximately5.5 kmnorthof the summitof Mt. Apesas,the northernlimitof the regionsurveyedbyWrightet al.(1990) forthe NemeaValleyArchaeological Project(NVAP),and14 kmnortheast of the New Nemeavalley,partiallysurveyedin conjunction withNVAP (urbanareaof AncientPhlius:Alcock 1988and1991)andnowbeingstudiedbyJ.MaranandH.-J.Weisshaar
(Liondiandthe adjacentNemean plain).It is ca. 11 kmnorthwestof the areaof modernSolomos,thewestern extentof the EasternKorinthia Archaeological SurveyProject,andit is immediately eastof the areaincluded in Lolos'stopographical surveyof the Sikyonia(Lolos1998).The siteis not mentionedin Blegen1920,Gebauer 1939,Alin 1962,Syriopoulos 1964, Wiseman1978,or Sakellariou and Pharaklas 1971.It is not includedin anygazetteerof Mycenaeansites,such asHope SimpsonandDickinson 1979,Hope Simpson1981,and,more recently, IsthmiaVIII,pp.469-482, andMountjoy1999,pp.197-242 (Corinthia), whichdealsonlywith siteswithpublishedpottery.
I20
JEANNETTE MARCHAND
hereanintroductory description of thesitefollowedbysomeobservations aboutits topographical locationandpossibleidentification, in the belief thatit will figuresignificantly in subsequent discussions of the number, size,andnatureof BronzeAge settlements in theCorinthia.3 LOCATION AND DESCRIPTION
OF THE SITE
Thesiteis locatedontopof abluffontheeasternsideoftheNemeaRiver, overlooking the SilyonianandCorinthiancoastalplains(Fig.1). This bluffisthenorthernmost extension of aridgeofthefoothillsofMt.Apesas thathasbeenerodedawayonthewestbytheNemeaRiverandontheeast byanother largetributary rema, orseasonal riverbed (Fig.2, labeledDaveli), so thatit nowextendsasanisolatedpromontory orfingerof landstretchingtowardthenorth.The ridgesupportsa thicklayerof topsoil,covered withpinetreeswhereit hasnotbeencleared, abovea stratum of conglomeraterock.Belowthisthicklayerof rock,the ridgeconsistsprimarily of soft marlthathassufferedsevereerosionon all sidesexceptthe south. Consequently, the unsupported upperconglomerate shelfhasbrokenup andfallenawayatthenorthern end,andheretheridgenowformsa naturallydefensible "acropolis," notunlikethatatthesimilarly sitedAetopetra farther to theeast.Thesiteis locatedonthisnatural conglomerate stronghold,whichat the northernendformsa relatively flat,double-pronged plateau marked bya GreekArmySurveycolumnat 190masl(Figs.2-4). Themoderntoponymis Dorati.4 A partially pavedroadrunningfromnorthto southgivesaccessto the ridgeandbisectsthe site;it maycorrespond to the routeof an ancient road.5 On the easternsideof thisroad,twovineyards havebeenplanted, onewithinthe lastdecades,the otherwithinthe lasttwoyears(Fig.2:1 and2, respectively). Rubbleandhabitation debris(grinding stones,tripod legs,fragments of largestorage jars)fromanancientsettlement havebeen cleared fromthevineyards andpiledhighalongtheroadandthenorthern edgeof the cultivated area,andthevineyards themselves aredensewith pottery fragments. To thewestof the road,a relatively flatfield(Fig.2:3),clearedand cultivated atsometimein thepast,is overgrown withweeds,butpotteryis stillvisibleonthesurfaceandcontinues to theedgeof theplateau. In late July 2000,whena longsectiona meterwidewasplowedalongthenorthwestern edgeof the plateau,a largeamountof high-quality Mycenaean pottery inverygoodcondition wasturnedup.Belowthesteepwallsofthe conglomerate shelf(Fig.2:4),surfacepotterycontinuesdowntheeroded slopes of the ridgeto boththe eastandwestasfarasonecannowsafely descend, but all of this materialcanhavebeenwashedor pusheddown from aboveduringclearingof theland:thesurfacematerial andthenatural topography suggestthatin thenorththesitewasprimarily confinedto thetop of the two-pronged plateau.Localresidentsreportthatpottery can be foundin quantity allthewayto thechurchof AgiaParaskevi to the
3. This paperis basedon a number ofvisitsto the site,aloneandin the company of colleagues. No surveywas conducted. The presentdiscussionis in nowaymeantto represent the results ofa systematic fieldproject;its aimis tobringattentionto the siteandto recommend thatsucha projectbe conductedin futurebeforemoredestructioncanoccur. 4. Dorati(Nxopaq)is the toponym listedon the GreekArmymapof the Corinthia (1:50,000).The localsknow theareaby the nameDarani(Nxapav), a toponymthatappearson the Greek Armymapof 1:5000andrefers specifically to the promontory at the northwestern endof the sitewhere thesurveycolumnshownon Fig.2 is located. 5. An ancientroadin thislocation andcontinuingsouthalongthe crest ofthe ridgetowardApesaswould parallel the routeon thewesternside ofthe NemeaRiverbetweenTarsina and KoutsiviaStimanga,documented inLolos1998,pp.14() 142.A further indication of thelikelihoodthatthis route canbe tracedbackto antiquity (although not necessarily alltheway back to the BronzeAge)is thepresence ofpreserved wheelrutsalongthe ridge, further southabovethe modernvillage ofSoulinari.
A NEW BRONZE
Figure1. Dorati andthe Corinthian Gulf plain.J.Herbst
Figure2. Sketchplanof Dorati J.Herbst
AGE SITE
I2I
JEANNETTE
MARCHAND
I22
Figure3. Dorati fromthe north. The site occupiesthe flat top of the highestplateauin middledistance. Mt. Apesasis visiblebeyond. Photoauthor
road.I havenotbeenable (Fig.2:5),on theSoulinari-Vrachati northwest thereis a butatthechurchof AgiaParaskevi thispossibility, to investigate Thisspringmayhave wateryear-round. naturalspringthatstillprovides watersourceforthe settlementduringthesummer.6 beentheprimary At the southernendof the site,wherea narrowmarlledgeconnects plateauto the restof theApesasridge,the situationis the conglomerate Hereatthehighestpointof thesitetheplateaunarrows slightlydifferent. shelfcomesto anend,so thaton theeastandwest andtheconglomerate therearesteepslopesratherthana sheerrockface.Mostof thisareahas not beencultivatedandthereis a goodpossibilitythatportionsof the depth. hereoveralargeareaandto a considerable arepreserved settlement To the eastof the road,a largestandof thickpournaria, or holmoak, coversa knollconsistingof a largequantityof rubblethatcannothave treesarependeposited(Fig.2:6).Wheretheseshrublike beennaturally denselyamongtherubble. BronzeAgepotteryisvisibleinterspersed etrable, debris loosematerialon the surface(presumably Muchof thisrelatively into its piled pushedand wasprobably of the settlement) fromstructures presentlocationduringthe creationof the adjacentroadandvineyard, at the centerof the standof treeswherea moundhasbeen particularly formed. thattherewasa goodreasonwhytheareawasnotcleared: It appears sizeanddeptharevisiblein of considerable sectionsof in situconstruction the adjacentscarpto the east,beginningat andjust belowthe current to surfacelevel.Fourlargewalls,parallelto eachotherandperpendicular fromthepresentscarp. theslopeof thehill,canbe clearlyseenprotruding plantedwitholivetreeshasbeencreatedhere A seriesof narrowterraces (Fig.2:7),descendingdownthe slopebelowthe standof oaksandthe thewallshavebeenexposedbetweenthelevelof vineyard; southernmost the top of the slope(elev.211 m) andthe floorof the firstterrace(elev. 203 m).Fromnorthto south,thefirstwallis a solidrubbleconstruction 1.90m inwidth;it is exposedfora 5-m stretchfromjustbelowthetopof its the slopedownto thelevelof theterrace(thefloorof whichobscures
6. Localsalsotellme thatthereused to be otherspringsnearthe churchthat havegonedryonlyin recentmemory, andso it is possiblethattheremayhave beenotherscloserto the ridgeitselfat one time.
A NEW BRONZE
AGE SITE
I23
Figure4. Conglomerateshelfof the westernpromontoryof the site, viewedfromthe east.Photoauthor
Figure5. Conglomerateblocksof a largewallprotrudingfromthe southeasternslope,viewedfromthe east.Photoauthor
7. Someof the rubblefromthe secondwallhasbeendugout andlitters illicit the terraceat its base;apparently the rubbleasposdiggersinterpreted siblybelongingto a tomb,butabandonedthe projectafterproceedinga meteror so intothe scarpandfinding thatthewallstillcontinuedas a solid Therearealsounmistakconstruction. ablesignsof illicitdiggingat the top of the hill,withinthe standof holmoaks.
A secondrubblewallof similardimenthereafter). possiblecontinuation in thescarp3 m to thesouth,followedbya appears sionsandorientation to the south,a wallof heavier Immediately thirdat a similarinterval.7 to therubblewallsbutata fromthescarp,parallel protrudes construction (Fig.5).Thiswallasexposedconsistsof fivelarge, slightlyhigherelevation blocksin twocoursesandisjustunder2 m roughlyshapedconglomerate inwidth.Thetopof theupperblocksis levelwiththemoderngroundsurto a depth onlycontinues face,butunliketheotherwalls,thisconstruction is thatthethickrubble givenbythesefeatures of 1.5-2 m.Theimpression of the constructed for a structure wallsmayhaveservedas foundations largerblocksabove.The presentrateof erosionat this end of the site suggeststhatthe useof stratumis not apparent wherethe conglomerate wasa prudentdecision. suchlargefoundations In thescarpbelowthe firstterraceof olivetrees,roughlya meterbelowthepointto whichtherubblewallsareexposed,anotherconstruction
I24
JEANNETTE
MARCHAND
wall.Unliketheupperwalls, orfoundation possiblyaretaining ispreserved, of medium-sized, the scarpandis constructed parallels thisconstruction froma pointdirectly stones.It is preserved roughlyformedconglomerate north rubblewallandcontinues belowandto theeastof thenorthernmost over stonesscattered 6.5 m;severallargeconglomerate forapproximately originatefromthiswallorthewallof larger nearbyprobably theterraces belowthese stonesabove.In the scarpsandon the ledgesof the terraces thereis although of pottery,and wallsis a highconcentration preserved offinewareanditsqualityareveryhigh. muchcoarseware,theproportion potteryonthese Figure6 givesanideaof thedensityandnatureof surface Thus,on the easternslope, andFigure7 an ideaof its quality.8 terraces, to atleast6 m belowthe currentsurpreserved arepotentially structures to such is preserved likelihoodthatarchitecture facelevel.Thetantalizing bythethicknessof thewalls,the sizeof someof the a depthis increased fineware, oftheaccompanying buildingstones,thequantity conglomerate andalsoby the discoveryof smallfragmentsof paintedplasteron the slopesbelow. fromthescarpsbeandprotruding Thepotterylitteringtheterraces to havewasheddownfromabovepriorto appears lowtheseconstructions wall, Roughly70 m belowthelowestpreserved thecuttingof theterraces. onlybya large, endandtheslopeof thehillsideis interrupted theterraces (Fig.2:8).In thisarea,concurvingaccesspathcutbymodernmachinery finedbytwolargeerosiongulliesontheeastandwest,areseveralaccumudepositedandcannotbe explained lationsof stonesthatarenotnaturally disareovergrown, Mostof theseaccumulations bytheforcesof erosion.9 byconbysoil,butin a fewcasestheyaresurrounded turbed,orobscured to be stillin situin of potteryandsomeof therubbleappears centrations accompanying 8.The photographs thisarticleweretakento illustratethe siteforthe Ephoria.It wasnot possible to movesherdsfromtheirpositionsor to collectandgroupsherdsby typeor at date.All materialwasphotographed its findspotandleft at the site;it was to orientsherdsto take necessary naturallight. of the available advantage The resultis thatalthoughthe photographscontainsomewhatmiscellaneous
reflect theyalsoaccurately groupings, the presentstateof the site.These the full do not represent photographs rangeof materialat Dorati;full of anyof the analysisandpublication potteryfromthe siteat thispoint wouldbe premature. 9.The areais roughly67 m from eastto west,30 m fromnorthto south, and150-164maslin elevation.
Figure6. Artifactdensityat Dorati: surfacematerialfromtwo areason the easternslope (sherdshavebeen slightlyarrangedbut not gathered). Photosauthor.
Figure7. A largefragmentof a stirrupjar.Scale1:2.Photoauthor.
A NEW BRONZE
Figure8. Examplesof LH III paintedpottery:designsincludea crestedbird(a: center) andwhorl shells(b:lower le:i). Scale1:2.Photos author.
AGE SITE
I25
Thesecontinueto thepresentedge construction. somesortof purposeful theDavelirema andindicatethatactivityassociof theravineoverlooking in thislowerarea. alsooccurred probably atedwiththesettlement canalsobeseenonthewesternslopeatthesouthInsituconstruction ernendof thesite.Adjacentto thestandof oaktreesbutwestof theroad a smallolivegrovehasbeenplanted(Fig.2:9).In the scarpa fewmeters isvisibleovera horizonbelowthisgrovemorein siturubbleconstruction appears in genconstruction 2:12). This at least 22 m (Fig. taldistanceof eralto be on a smallerscalethanthaton theeastandto consistof stepped stonesrunningparallelto the rubblewallsof smallandmedium-sized bythecois indicated scarp.ThatthesearealsoBronzeAgeconstructions potteryin the soilerodingfromaroundthe stonesand piousMycenaean coveringthe slopesbelow. The nearlycompleteabsenceof potterylaterthantheBronzeAge is Geometric,and striking.A few sherdsmaydateto the Protogeometric, perhapsevenEarlyArchaicperiod,butI haveobservednothingon the of thereadilyidentifisurfacethatis Classicalorlater.Thevastmajority thelaterphases(through andof thismaterial ablematerial is Mycenaean, Materialis alsopresentfromearlierprehistoric LH IIIB)predominate. phases,for example,muchEH handmadepottery(andsomepossibly sherds, Neolithic),MH yellowMinyan,andMH or LH I matt-painted of finewarearefromLH III fragments of theidentifiable butthemajority Mycenaean kylikesanddeepbowls,manyof whichshowwell-preserved whorlshells,rosettes, bands,runningspirals, designs,includingdecorative figures(Fig.8). andfragmentary siteis clearfromthelargeamountofcoarse Thatthiswasa settlement pottery, suchascups,ladles,andcookingpots,andfromthesheeramount
.
d
n
I26
JEANNETTE
MARCHAND
of looserubbleoriginating fromroughwallsandstructures. Tripodcookingpotlegsandfragments ofgrindingstonescanalsobefoundinparticularlylargequantities, andI noticedat leastten completegrindingstones fromallareasof thesitein additionto ubiquitous fragments. Alsoindicativeofdomestic activities orindustry arethenumerous murexandgaidouropusshells,andlargepithosfragments (Figs.9-10).Otherartifacts include numerous piecesof chert,figurines(Fig.11),oneobsidianblade,andone steatitespindlewhorl.10 At thenortheastern endof theplateau,attheedgeof oneof thevineyards,alargepileof clearedmaterial includesa number of carefilllyworked blocks(Fig.2:10).To the north,belowthe upperplateau,manysimilar large,well-worked conglomerate blockslie low on the sideof the ravine on the east(Fig.2:11andFig.12).Theseprobably cameoriginally from thetop of theplateau,andmostof themdo notdiffersignificantly from thosestillin situatthesouthernendof thesite(Fig.5).Amongtheconglomerate blocksareoneortwoporosblocks.Noneoftheseblockscanbe datedandthereis no immediately evidentsourcefortheporosstone. Withoutexcavation, theexactsize,nature,andperiodsof occupation ofthesiteatDoraticannotbe definitely determined, butthesurfaceindicationssuggestthatit is potentially amongthelargestMycenaean sitesyet locatedin the Corinthia.1l I estimatethatthe top of the plateaualone coversapproximately 46,595m2.12 This is not the totalareacoveredby surface material andit doesnotincludeanyof theareason theslopes,but itis likelyto represent thecoreof thesite.Extending thisestimateto the areacoveredwithsurfacepotteryandartifacts datingto the BronzeAge yieldsaminimum figureof 106,000m2.Tomyknowledge, onlyatKorakou hassurfacematerial coveringalargerareabeenreported fora BronzeAge s1te1ntil1S reglOn.lJ 10.Identicalsteatitewhorlsfrom theAidoniatombshavebeententativelyinterpreted asbeadweightsfor clothing: seeDemakopoulou 1996, pp.66-67, n. 59, fig.59. 11.It is virtuallyimpossibleto compare thissiteaccurately with othersthathavebeenpartiallyexcavatedor identifiedin surveyby the fullextentof surfacescatters. The followingfiguresareonlyroughestimates meantto providesomebasisfordiscussion andarenot to be takenas a definitive quantification of the materialat the site. 12.The surfaceareawasmeasured witha handheldGPS unit. 13. Hope Simpson(1981,p. 33) estimates thatthe settlementat Korakoucoveredan areaof 225,000m2.For mostof the largestsitesin the Corinthia, no extensivesurveyshavebeen
conducted to establishthe parameters ofsurfacematerial.One exceptionis Tsoungiza in theArchaiaNemea valley, wherethe sizeof the Mycenaean occupationhasbeenestimated at75,000m2by the NVAPteam(see thediscussionof Mycenaean Nemea athttp://river.blg.uc.edu/nvap/ MycNVAP.html). The onlypublished estimates of sizeforthe majorBronze Age siteslongknownandin somecases partially excavated in the regionare those in Hope Simpson1981.Manyof his figuresarebasedon observations made in the late1950s(thesamenumbers canbe foundin Hope Simpson and Dickinson1979andto a certain extent in Hope Simpson1965),and the criteriausedforthe estimatesare not alwaysmadeexplicit(in somecases they appearto be basedon the fullextent of surfacematerialandin others
onthe dimensionsof the natural topographical featureson whichthe sitesarelocated);in allcasesthe figures givenareonlyroughestimates. These figures mustthereforebe usedwith extreme caution,butforthe sakeof illustration only,basedon Hope Simpson1981,the relativeextentof surface material at majorMycenaean sitesin thewiderregioncanbe listedas follows: Gonia87,500m2;Kleonai 75,000 m2;Melissi40,000m2;Perdikaria 25,500m2;Vasiliko/Ancient Sikyon 24,050m2;Aetopetra22,500m2; Zygouries 15,300m2;AgiaIrini 15,000 m2;MylosCheliotou6,375m2. See Hope Simpson1981,pp.34 (A57, Gonia; A59, Perdikaria; A56, Aetopetra; A55, MylosCheliotou),35 (A62, Kleonai; A63, Zygouries), 36 (A70, Vasiliko; A66,AgiaIrini),and37 (A73, Melissi).
A NEW BRONZE
Figure9. A typicalassemblage, includingmurexshell and coarse waretripodcookingpot leg (lower left).Scale3:8.Photoauthor.
Figure10. Andesitegrindingstone andpithossherds.Photoauthor
Figure11. Figurinefragments: a) quadruped;b) Phi, backview (on right).Scale1:2.Photosauthor.
a
AGE SITE
I27
I28
JEANNETTE
MARCHAND
Figure12. One of the largeconglomerateblocksat the bottom of the Daveli rema to the east of the site. Photo author
Surfaceareais,however, notnecessarily anaccurate indication of this site'spotentialsizeorsignificance; moreindicative is theparticularly large amountanddensityof identifiable potterydatingto theMycenaean period.A roughideaof the amountof surfacematerialcanbe givenbythe following observation: onthesoutheastern terraces alone,I stoppedcountingthereadily recognizable Mycenaean sherds whenI reached 2,000,having onlyprogressed partofthewaydowntheslope(thisareais notincludedin theestimateof the sizeof the sitegivenabove).In surveysof the northeastern Peloponnese, a totalof morethan200 sherdsrepresents a significantMycenaean component ata site.l4In manyareasof Dorati,morethan 200Mycenaean sherdscanbefoundwithinanareaof afewsquaremeters, andthe densityof sherdsoverthe entiretop of the plateauwherethe surface is visibleis consistently high.As notedabove,thesurfacematerial clearly indicatesthatthisis a settlement site,andlarge-scale construction isatleastpartially preserved. No Mycenaean settlementon a scaleto compare withthe citadelsof theArgolidhasyet beenlocatedin the Corinthia,andonlya fewlargescaleMycenaean settlements havebeenlocatedor excavated. Although sitesin the regionwith LH surfacepotteryarenumerousand several 14.Mountjoy(1995,p. 52) reports that only1,241Mycenaean sherds(all phases) wererecognizedin the southern Argolidsurvey. The largestsite (F5) produced 257 sherds.NVAPreports twenty-five siteswithMycenaean material fromtheirsurvey(notincluding the excavated areas,Tsoungizaandthe Nemean sanctuary), morethanhalf consisting of fewerthanfivesherds;the few largestsitesarecategorized as consisting of"over200"sherds.See"Myce-
naean Nemea"on the NVAPWebsite (above, n. 13).Wellsandhercolleagues report fromthe Berbati-Limnes survey (Wells 1996,pp.123-175)one findspot (no.428,p. 126)consistingof 900 Mycenaean sherdscontainedwithinan area of ca.30 x 30 m, andanother(no. 14, esp.pp.133,166)extendingover 60,000 m2atwhich269 artifactswere collected; theynotethatthisspreadis much largerthananyof the othersand that the coreof the siteshouldprobably
be estimatedas muchsmallerthanthe total scatterarea.The remainingsites listed aresmaller.Foranoverviewof surveys in the region,see Rutter1993, table 1, pp.748-749,ns. 8-15. The caveat statedin n. 11 aboveis particularly relevant here,sinceonlya veryrough sense of the relativesizeof Doratican be obtainedby comparing limitedonsite observations withthe findingsof formal surveysemployingvariedmethods of collectionandquantification.
A NEW B RONZE AGE SITE
Figure13. Dorati andothersitesin the northeasternPeloponnese. J. Herbst
15.Note thatnoneof the siteshave beenextensively excavated. At Gonia, despitethelargesurfacescatter(see above,n. 13),trialtrenchesturnedup onlyerodedtracesof the Mycenaean settlementanda fortification wall: Blegen1920,p.6; Blegen1930-1931; Sakellariou andPharaklas 1971,appendixII, p. 16;Rutter1974;Hope SimpsonandDickinson1979,p.63, no.A57;Hope Simpson1981,p.34, no.A57;IsthmiaVIII,p. 474,no.13. ForAetopetra,seeHatzepouliouKallire1984(mostrecentexcavations);
I29
cemeteries andtombsof theMycenaean periodhavebeenexcavated, most oftheBronzeAgesitesthathavebeenexcavated (suchasGonia,Aetopetra, MylosCheliotou, andAgiosGerasimos) havesofarproduced moreextensiveEH orMH components (Fig.13).15 Thepredominance ofMycenaean potteryatDoratiis therefore of particular interest.In thewiderarea,only Korakou hasso farproduced evidenceforanextensive Mycenaean settlement:foundation wallsfornumerous smallLH III houses,a fortification wall,indications of largerstructures (amassivethreshold blockandpartof Blegen1920,pp.3-4; Corinth I.1, p. 108;Syriopoulos 1964,p. 86;SakellariouandPharaklas 1971,appendixII, p.21;Wiseman1978,p. 99;Hope SimpsonandDickinson1979,p.62, no.A54;Hope Simpson1981,pp.8 (mapA),10 (fig.1),34, no.A56;(but notethatit is incorrectly placedon the mapsas overlooking the LongopotamosRiver:it is actuallylocatedfurther east.Forthe correctlocationseeBlegen 1920,pp.2-3, fig. l); IsthmiaVIII, p. 470, no. 8. ForMylosCheliotou,see Blegen1920,p.3; Blegen1921,p.116;
CorinthI.1, p. 108;CorinthXIII,p. 6; Sakellariou andPharaklas 1971,appendixII, p. 19;Hope Simpsonand Dickinson1979,p. 62, no.A53;Hope Simpson1981,p. 34, no.A55;Isthmia VIII,p. 470, no. 10, andp. 471.For AgiosGerasimos, see CorinthI.1, p. 109;Alin 1962,p. 57;Sakellariou andPharaklas 1971,appendixII, pp.20-21;Protonotariou-Deilake 1974;Wiseman1978,p. 99;Hope SimpsonandDickinson1979,p. 63, no.A55;Hope Simpson1981,p.33, no.A53;IsthmiaVIII,p. 469,no.5.
JEANNETTE MARCHAND
I30
a fresco),aswellasburials.l6 At Zygouries in theAncientKleonaivalley, Blegenexcavated animpressive two-levelLHIIIB:1-2structure withfrescoesandanextensivestoreof potteryaswellas a numberof burialsat a locationnearby; onlya portionof the hillwasexposed.l7 In the adjacent AncientNemeavalley,theMycenaean settlementatTsoungiza hasbeen characterized as a smallhamletin the EarlyMycenaean period,growing to moresubstantial proportions byLH IIIB.18 Thesearetheonlysitesin the Corinthiaat whicha substantial Mycenaean phasehas so farbeen confirmed byexcavation. In thePhliusvalley,therichtombsexcavated atAidoniasuggestthat animportant LH settlementshouldbe locatednearby, butit hasnotyet beendiscovered.l9 Moreover, the numberof siteswheresurfacematerial indicates alargeLH settlement is relatively small.Amongthese,Perdikaria (withanimpressive stretchof cyclopean wall),Kleonai,andAgiaIriniin thePhliusvalleystandoutasthelargest.20 Tracesof Mycenaean occupationhavebeenfoundinAncientCorinthandontheslopesofAcrocorinth, butasyetnolargeLH settlement sitehasbeenuncovered.2l The situation 16.Blegen1921;Dickinson1972; Rutter1974;Hope Simpsonand Dickinson1979,p. 61, no.A50;Hope Simpson1981,p.33, no.A51;Isthmia VIII,p. 469, no.6. 17. Blegen1928;Alin 1962,p. 58; Hope SimpsonandDickinson1979, p.66, no.A67;HopeSimpson1981, p.35, no.A63;Thomas1988;1992; IsthmiaVIII,pp.358-361,469, no.2. Shear(1986)hasinterpreted the remainsasbelongingto a numberof separatesubstantial houses,butit is not certainthatmorethanone structure is represented andThomas(1988;1992, esp.337)hascharacterized the pottery asunusualfora domesticcontext. 18. Hope SimpsonandDickinson 1979,p. 67, no.A70;Hope Simpson 1981,p.36, no.A65;Wrightet al. 1990,pp.631-638;Wright1990, p.353; Rutter1989;1990a;1990b; 1993;Pullen1990,p.333;Thomas 1992;IsthmiaVIII,pp.358-361. 19.Touchais1979;1980;1987; Rutter1993,p. 788,n. 179;KrystalleVotse1989;1996. 20. Forsizeestimatesseeabove, n.13. Forunpublished newobservationsconcerning the siteof Perdikaria, originallylocatedby Blegen(1920),see the fieldreportsforthe 1999seasonon theWebsiteof the EasternKorinthia Archaeological SurveyProject(http:// eleftheria.stcloudstate.edu/eks). See alsoAlin 1962,p. 57;Wiseman1978, o
pp.64-65; Sakellariou andPharaklas 1971,appendixII, p.17; Hope SimpsonandDickinson1979,p.64, no.A59;Hope Simpson1981,pp.3435, no.A59;IsthmiaVIII,p. 474, no.14. ForKleonaiseeGebauer1939, col.271;Alin 1962,p.58; Syriopoulos 1964,p. 87;Sakellariou andPharaklas 1971,appendixII, pp.33-34;Hope SimpsonandDickinson1979,p.67, no.A69;HopeSimpson1981,p.35, no.A62;IsthmiaVIII,p. 469,no. 1. A. Frickenhaus (in Karo1913, cols.114-116)reportedfindingover 100Mycenaean figurinesnearthe churchof AgiaTriadain the mountainsabovethevaSley; theyhaverecentlybeenpublishedin Kilian1990. ForAgiaIrini,seeHope Simpsonand Dickinson1979,pp.67-68, no.A71; Hope Simpson1981,p.36, no.A66; Rutter1993,p. 781. 21. Blegen1920,p.3; Dunbabin 1948;Weinberg1949,p. 157;Broneer 1951,p. 293;Robinson1976,p.211, n.25 (onesherd,whichhe identifiedas LH);andHopeSimpson1981,pp.3334, no.A54. ForAcrocorinth, see Corinth III.1,p. 28. Forthe LH IIIC structure, grave,anddepositfromthe Sanctuary of DemeterandKore,see BookidisandFisher1972,pp.291292;1974,pp.286-289;Rutter1979. Foranextensivebibliography, see IsthmiaVIII,pp.47s471.
A NEW BRONZE
Figure14. View fromthe northern end of the site overthe Silyonian plaintowardthe CorinthianGulf and Perachora.The courseof the Nemea Riveris discernibleto the east (right)of the pavedroad.Silyon is locatedfarleft in the distance. Photoauthor
22.The sheernumberof sitesin the CorinthiaatwhichMycenaean materialhasbeenfoundprecludes listing or discussingthemallhere.Some notableomissionsfromthe above discussionaretheIsthmiansanctuary, Kenchreai, Galataki,KatoAlmyri, Athikia,Phlius,Krines,andMoulki nearSikyon.Forcompletegazetteersof Mycenaean sitesin the Corinthia,see theworkslistedabovein n. 2. Fortwo additionalsmallsitesreportedin the Kleonaivalleynot mentionedin gazetteers,see GauvinandMorin1997.I havenot includedanysitesnorthof the Isthmus;forthe Perachora peninsula,in additionto the generalsitegazetteers, see Fossey1990.Someasyet unpub-
AGE SITE
I3I
is the samein andaroundthe Sikyonian plain;surfacesherdsandburials suggesta Mycenaean presenceat numerous sites(e.g.,Vasiliko/Ancient Sikyon,Melissi),butno substantial settlementsiteon the orderof those listedaboveorcomparable in surface material to Doratihasbeenlocated.22 Thesizeof thesiteatDorati,evenconsidering onlythetopof theplateau, andthe densityof the specifically Mycenaean materialthereonsuggest thatthe Mycenaean phaseof the site is potentiallyas largeas or larger thanmostyetinvestigated oridentifiedin thearea.23 The site'stopographical locationis alsosignificant forestimating its potentialimportance, in thatit commandsan impressive viewoverthe juncture betweenthe Sikyonian andCorinthian plains(Fig.14).It is in a positionto controlsignificant partsof thecoastalplainandaccessintothe NemeavalleyviaatheNemeaRiver.24 Thissetting,on a naturally defensibleplateauabovea riverandalongthecoastalbluffs,similarto thelocation of a numberof otherBronzeAge sitesalongthe Corinthian Gulf, conforms to a preferred typeof locationforsettlements in thisareaatthis time,andfurthersuggeststhatothersimilarareasbetweentheNemeaand the Longopotamos Riversdeservecloserinvestigation.25 lishedinformation fromNVAPandthe EasternKorinthiasurveycanbe obtainedfromtheirWebsites. 23.The precedingdiscussionis not intendedto ignorethe probability that earliersettlementphasesarepreserved. I estimatethatat least90 percentof the materialthatI haveobservedat the site is LH, butthe tendencyto overestimate periodsmoreeasilyidentifiedin the fieldandthe largeamountof totalsurfacematerial(notto mentionthe natureof otherBronzeAge sitesin the region)suggestthatthe earlierprehistoricphasescouldalsoproveto be substantial. 24. Mountjoy(1999,p. 197)suggeststhatthe routealongthe Nemea
River(andthereforedirectlypastDorati),animportantroadin the historical period,wasalreadyin usein theMycenaeanperiodto provideaccessbetween Mycenae,Tsoungiza,Aidonia,andthe CorinthianGulf. 25. Morgan(IsthmiaVIII,p. 354) hasalreadymadea similarobservation concerning Aetopetra: "Thesitelocationtypifiesa localpreference for bluffs,removingsettlementfromagriculturallandwhileallowingeasyaccess to it, facilitating defense,andlying closeto animportantrouteof communication." The siteat AgiosPandeleimonaboveKamariin Achaiais similarlysituated:seeAndersonand Anderson1975,p. 5, pl. 1, fig. 1.
v
,
ton
,
\
,
\
,
I32
JEANNETTE
DORATI AS STRABO'S CORINTHIAN
MARCHAND
ORNEAI
The specificlocationof Doratiis particularly significant becauseit allows fora tentative identification ofthe site.Thegeographer Strabo, in a numberof controversial passages, mentionsthatthereexisteda secondtown calledOrneai,distinctfromthetownin theArgolidof thesamename.He describesthissecondOrneaias abandoned in his time,locatedbetween CorinthandSikyonandnextto ariveronaheightoverlooking theSikyonianplain,butin the neighborhood of Corinth.I proposethatDoratifits exactlywiththistopographical description byStrabo. Accepting Doratias a viablecandidate forthissecond,"Corinthian," Orneainotonlyclarifies thedescription ofthattownbyStrabo, butalsosolvessomeofthedifficultiesin locatingtheArgivetownof the samenameandin understanding theinternal logicofthearrangement ofAgamemnon's realmintheHomeric Catalogue of Ships,in whichanOrneaiis mentioned. It is notmyintentionto reviewallof thearguments already advanced by otherscholarsconcerning the locationof theArgiveOrneai,butit is necessary to prefaceanydiscussion of Corinthian Orneaiwitha relatively detailedreviewof thesourcesforandproblemof thetwosimilarly named towns.An Orneiaiis listedin the Catalogue of Shipsasbelongingto the "realm of Agamemnon": 0' 8e MvxNvag stXov,euxCCuevov sxoRC0toov, a(pvetov xe KototvOov euxCCuevag xe KAsxuaq, 'Otovetag ' eveCuovTo 'Atoa0vtosNv ' stoaxetvNv xat Stxvxv',o0'ato"'A8toaxog stox' sCu,BastAgvgv, 0t U l=CtONatNV TE xat ax=stvNv 1ovosovav IlrEnvNv ' stXov8' A'cytov aCuxptveCuovTo AtytaBov ' ava wavTaxat aCu(p' 'ERtxNv evtostav, xxv exaxovvv t°XextostxvAyaCusCuvxv -
,
.
e
\
?,
-
,
.
.
.
cs
ATtOCtOG.
AndtheythatheldMycenae,thewell-builtcitadel,andwealthy Corinth,andwell-builtKleonai,anddweltin Orneiaiandlovely Araithyrea andSikyon,whereinatthefirstAdrastus wasking;and theythatheldHyperesia andsteepGonoessaandPellene,andthat dweltaboutAigionandthroughout allAigialus,andaboutbroad Helike of thesewasthesonof Atreus,lordAgamemnon, captain, witha hundredships.26 Therearea fewbriefreferences to anOrneaithatindicatethataplace of thisnamewasinvolvedin hostilitieswithSikyonat sometimeduring the EarlyArchaicperiod.27 The firstclearreference to anArgiveOrneai 26. II.2.569-577,A. T. Murray, trans.,Cambridge, Mass.,1924.On the variantspellingOrneiai,see the passage of Eustathiusquotedin n. 34 below. 27. Plut.De Pyth.or.15 (Mor.401d); Paus.10.18.5.Botharepassingreferencesto a dedicationbythe Orneataito the sanctuary at Delphicommemorat-
ing a victoryoverthe Sikyonians. There is nothingin eitherpassageto indicate the dateof thevictory,buta reasonable settingforthe conflicthasbeenfound in a warbetweenSikyonandArgos mentionedby Herodotos(5.67-68) duringthe tyrannyof Kleisthenes. Plutarch(Desera.7, Mor.553a-b)also
appearsto indicatea borderdispute betweenKleonaiandSikyonat this time.McGregor(1941,pp.277-278, n. 49), Kelly(1976,p. 124),Griffin (1982p. 38, n. 20, andp. 51), andLolos (1998,p. 38, n. 107,andp. 49) haveall hintedthatbothKleonaiandOrneai mayhaveformedanassociation with
,
,
-
,>
A NEW BRONZE
,
.
-
\
v
,
,
,
AGE SITE
I33
appears in Herodotos8.73.3,whereherefersto thepeopleof anOrneaiat the timeof the PersianWaras in somewaysubjectto the Argivesand amongthosewhosatapartfromthewar.28 TheOrneatai nextappear alongsidethe Kleonaians as alliesof theArgivesat thebattleof Mantineiain 418 B.C. (Thuc.5.67.2, 5.72.4, 5.74.3-4). According to Thucydides and Diodoros,the Lakedaimonians invadedthe Argolidthe followingyear (Thuc.6.7.1-2, Diod.Sic.12.81.4-5).Afterravaging thecountryside, they settledfilgitivesfromArgosat Orneai,fortifiedthecity,andleftit witha stronggarrison. Soonthereafter, theAtheniansarrived andtogetherwith theArgivesbesiegedOrneai.According to Diodoros,aftertakingthecity, the ArgivesandAtheniansputsomeof the garrisonandexilesto death andexpelledothers.InThucydides' account,theArgivesrazedOrneai.29 Theplaceapparently continued toexist,however, sinceDiodoros(16.39.16) mentions it againinreference tohostilities betweentheLakedaimonians andtheMegalopolitans in 352/1 B.C.: theLakedaimonians advance from Mantineia to the"Argive cityof Orneai" andcapture it beforetheMegalopolitansandtheirallies,includingtheArgives,canadvancefromtheir positionattheheadwaters of theAlpheiusRiver.WhentheArgivessubsequentlytakethe fieldagainstthe Lakedaimonians at Orneai,theyare defeated.The Lakedaimonians eventuallymakean armisticewith the Megalopolitans, butwe do notlearnthefateof Orneai,andits capture by theLakedaimonians is thelasteventin itshistorypreserved in oursources. Fromthesefewreferences it is already indicatedthatArgiveOrneai laysomewhere withintheArgolidin the directionof Mantineia, butthe mostimportantsourcefor the morepreciselocationof this Orneaiis Pausanias 2.25.4-6. He beginswitha description of a road(whichhelater callstheKlimax)leadingfromtheDeirasridgeof ArgosandpastLyrkeia to Mantineia.He givesan aition for a beaconfestivalheldat Argos,in whichhe makesit clearthatLyrkeia andtheLarissa areintervisible. Havinggiventhesecluesto theroughlocationof Lyrkeia, he thencontinues: Cuev 8! xavTrv sv si 'AtoyovgsiNxovxa CuaRa sov a8ta, 8e AvtoxetageVxetoa xovavxa sg 'Otoveag.AvtoxetagCuev 8! ZOACUg, axe oNpUpevNg NoNxaxa zv tvxv oTtoaxetav sst "IAtov,oux CsoLsaTo VOCuoog Ev xaTaXoyX CuvuNv 'OtovEag 8Esg
ex
,>
Argosalreadyat thistimeto wagea territorial waragainsttheircommon andaggressive Sikyonianenemy.See alson. 67 below. 28. Amongthe racesof the Peloponnese,HerodotosincludesOl 8C KoVOOplOl aDXoX0oMGS COWgS aOXCOUOl yOUVOl ClVal''I@VGS, oso
Ts 'Apysixv
CX8g80plCUWal
apxoyevol
Xp°MoO, CoWGS'0pVNTal
xai xai
8g
xou
[Ol] WC0l-
On the strengthof Strab.8.6.7 (C 370),Andrewes(1970,p. 109)explainsthe Kynouria alludedto hereas an areaof the Inachosvalleyandnot the morefamiliarborderareabetween theArgolidandLakonia. 29. Ar.Jv. 399 andscholiaand Paus.8.27.1alsoprobably referto these events. OlXOl.
v
,,
\
,
,
,
tt
CTCyato UxoUvXo
, UssEto
Ev ToLg C=goL stooTEtoag ExaXoUvXo
,,
8E aso
b
\
\
o
>,
,
,
,
v
,
,,
,
\
v
\
TZ TO=U Tt g AtoyELag ExCTvXo, ovTZ xaL
N 4>XCoUvXa TE xaL CxvXva
ootovEXg Tov 'EtoEx0gUg
xaTEgiEv.
Tov 8E ootovEXg Nv
xouxov Ilexexg, xov 8e Meves0eog, 8g'AyaCusCuvovL CusTa 'A0NvaLXv zv Iltota,uovovyxa0CXcv atoXNv.aso Cuev 8! xovxov xo ovo,ua EyEMEToq ZOACt, AtOyELOtOX VgtOOV TOVT@Vutoveaxag averm7sav avaavCg 8e ouvotxot ygyovastv 'Atoyetog. sc= 8e ev xag 'Otoveag'ACCu8og xe Ctoovxat goavov otoOovxat eVxetOog
vaogsog wastvsg xotwov aveuevog.xa os ssexetva utovexvN xe StxvxvtaxatN 4>Xtasta soxcv. \
A
o
\
o
o
\
CS \
o
o
o
tO
The distancefromArgosto Lyrceais aboutsixtystades,andthe distancefromLyrceato Orneaeis thesame.Homerin theCataloguemakesno mentionof thecityLyrcea, becauseat thetimeof theGreekexpedition againstTroyit already laydeserted; Orneae, however, wasinhabited, andin hispoemhe placesit on thelist
JEANNETTE
I34
MARCHAND
beforePhliusandSicyon,whichordercorresponds to theposition of thetownsin theArgiveterritory. The nameis derivedfrom Orneus,thesonof Erechtheus. ThisOrneusbegatPeteos,and PeteosbegatMenestheus, who,witha bodyof Athenians, helped Agamemnon to destroythekingdomof Priam.Fromhimthendid Orneaigetits name,andaftervfards theArgivesremovedallits citizens,whothereupon cameto liveatArgos.At Orneaearea sanctuary andanuprightwoodenimageof Artemis; thereis besides a templedevotedto allthegodsin common.On thefilrthersideof OrneaeareSicyoniaandPhliasia.30 The interpretation of partof thispassageis somewhatproblematic, and Pritchett hasarguedthatthetranslation, quotedabove,byW. H. S.Jones is inaccurate. He arguesthatOrneaiis theimpliedsubjectof exetvxo,and thatthesentencemustbetranslated asfollows: "Butwithrespectto Orneai (forit wastheninhabited), asit is situatedin thecountryoftheArgives,it is mentioned bythepoetbeforeeitherPhleiousorSikyon.''3l Inthistranslationit is particularly clearthatPausanias placesthisOrneaiwithinthe Argolid,andalsothathe identifiesthisOrneainearLyrkeia astheOrneai mentioned in theCatalogue of Ships. Straboalsorefersto thisOrneainearLyrkeia, butin thesamepassage heclearlyrefers to asecondplaceofthesamenamelocatedbetween Sikyon andCorinth: VOCuoog 8 avxagouxol8ev,ov8exo Avtoxetov, ov8''Otoveag xuat 8'sLat qg 'Atoyetaq, N xx sANatov, at 8e xag 'Otoveag xag Cusxaiv KototvOov xat Stxvxvog8tovCuevag. Cuev
oCuxvvCuog
xx
otost
Homerdoesn'tknowthese[ArgiveHysiaiandKenchreai], noryet doeshe knowLyrkeion orOrneai,whicharevillagesin Argeia,the formerbearingthesamenameasthemountainnearit andthelatter thesameastheOrneaiwhichis situatedbetweenCorinthand Sikyon.32
At firstglance,Strabo's comment thatHomerdoesnotknowofOrneai appears strange,sinceanOrneaicertainly appears in theHomericCatalogue.ButStrabosoonturnsto a discussion of therealmof Agamemnon, andit becomesclearthat,unlikePausanias, he believesthatthissecond Orneai,betweenCorinthandSikyonandnotaplacein theArgolid,is the Orneaimentionedin the Iliad. He beginshis discussionof Mycenae, Corinth,Kleonai,Orneai,andso on withthe preface"butlet me speak nextof theplacesthatarenamedin the Catalogue of Shipsassubjectto MycenaeandMenelaus" (Strab.8.6.19[C 377]).AfterquotingHomer andproceeding to describe thesitesin theorderlistedin theCatalogue, he reachesOrneai(Strab.8.6.24[C 382]): 'Otoveax 8' sxotvssxvvCuox xx waoatotosovx soxaCux, vvvCuev stouox, stooxetoov 8' otxovCuevax xaRg, xetoov eXovaax Iltoxasov xuuevov, axp'xv xaxo xa Iltoxasexa soc<sagEvxptoovtog 'Otovearr v xaRex xov 0sov xetvTax 8' v=Cto xovse8xovxovStxvxvxxv, vrv 8e Xoav soxov'Atoyexox. 'AtoaxOvtosa 8'so v N vvv4>Xtasta xaBovCuev.
30.Textandtranslation fromJones 1918, pp.380-383. 31. Pritchett1980, p.22, n. 35. 32. Strab.8.6.17 (C 376). The text givenhereis fromJones[1927] 1954, p.182, butit is problematic. Lyrkeionis
anemendation, no doubtbasedon Pausanias's description (2.25.4-5) of the twoArgivetownsof Orneaiand Lyrkeia. Andrewes(1970, p.108) commentson the passagethus:aou8e ToAuxoupytov (probably Aupxelov), ou8''Opveagxyat 8' elCsi q5 'Ap,
t
yelaq,
fi
t
,
-
,,
-
N Xuev oXuxvoXuog TZ OpCl TZ . . .
(probably Aupxgi), ai 8e xalS'OpvealS xalS ysxaio
KopLvOouxai
ElXU@VOt
i8poXuevalS. Kramer excisedallthe wordsherecitedin Greek,believing thatthissecondOrneaiwasthe inventionof aninterpolator. But6.24,382 is clearenough." Aly (1950, p.249) also advocates the emendation: aou8eTO AuXOupylOu (?)ou8''Opveag.xual 8' t
fi
-
Clot m5
t
*
o
ApystaS,
t
fi
t
o
N CV O@V0yOt
-
TZ
opelxx 5-6 B. al 8e xalS'OpvealS xX. DassLykurgion falschund Lyrkeiongemeintist,siehtjeder."
A NEW BRONZE AGE SITE
I35
Orneaiis namedaftertheriverthatflowspastit. It is desertednow, andhada templeof Priapus it waswell-peopled, althoughformerly thatwasheldin honour;andit wasfromOrneaithattheEuphrotheOrnecallsthegod"Priapus niuswhocomposedthePriapeia butthe atan."Orneaiis situatedabovetheplainof the Sikyonians, is thecountry bytheArgives.Araithyrea countrywaspossessed whichis nowcalledPhliasia.33 Strabo(13.1.12[C 587])mentionsthissecondOrneaiin passingin one Corinth": lastpassage, wherehe callsit merely"near ... xxv setoxKototvOov . . . sE Otovexv fromthe OrneainearCorinth Fromthesethreepassagesit is clearthatStrabobelievesthatthere wasa placecalledOrneaibetweenSikyonandCorinth,nextto a riverof the samenameandon a heightabovethe Silyonianplainbutbetterdebutit wascompletely scribedasnearCorinth.It hada templeof Priapus, B.C. comthe 2nd-century has shown that his day. B. Niese abandoned by underlieStrabo's thatof Apollodoros, mentarieson the Iliad, primarily on thissecond andhisinformation discussion of theHomericCatalogue, Strabois virtually entirelyfromhis sources.34 originates Orneaiprobably using sourceforthis secondOrneai,but Eustathius, the onlypreserved repeatsthe of Byzantium, versionof Stephanus Straboandanunabridged butaddsthefollowing: sameinformation Naso 'Otoveag Etoex0CXgS vTov Kakaat 8e ouxxgNaso 'Otovexg, xx soxaCu. N oCuxvvCug'Otovea vvCufpg, N oxtsfp'vQovgxetvTat, It [theOrneaibetweenSikyonandCorinth]is namedafterOrneus, orthenymphOrnea,orbecauseit is on a thesonof Erechtheus, height,orfromtheriverof thesamename.35 thatcouldhaveultimately information repeating Eustathius is primarily of Orneusto thisOrneai(contra derivedfromStrabo,buthis assignment 33. Jones[1927] 1954, pp. 204-205. 34. Niese 1877; Giovannini1969, p. 8, n. 2. Strabogivesno indication
thathe knowsthe locationof CorinIn his descripthianOrneaifirsthand. tionof Kleonai,he statesthathe saw he the cityhimselffromAcrocorinth; makesno suchclaimforOrneai.Dorati,mycandidateforOrneai,is disandso it is cerniblefromAcrocorinth, possiblethatits generallocationwas pointedoutto himfromthatvantage point,butsincehe doesnot sayso it seemslikelythathe is simplyquoting his sources. 35. Trans.author.Eust.2.291.7-15 (= Strab.8.6.24 [C 382]): 'Oovelai8e N Aixaxoul 'Opveai souxoyap XuaBlaTa
CVXOlv NOCl
XClTal
X@yN CCiV
'Apysia5xaxa xov rexypaxpov.scrcl8e xai sxepa ysxaE, KopivOouxai ElXUxvog. xauTrv 8e o xa 'E0vlxaypaf aS WOXlVgyCl, OU X@yNV. WN09OlX@t
8g
AsyovTal, XS xai ai KAs-
xai aal
vai. ypaxpovTal as Ala AlepOoyyou ai
'Opuelai,xS xai Bpocselaixai Auyelai. XaClTal
8C OUT@tN aso 'Opvexs,
'EpCX0g@5
N
aSo'°09Cat
V0yt,
UlOU N
OTl
P' 0+°0S XCtVTaL,N oXuxvoXuxt'Opve,a TZ, SOTaX. TOUTO8C Xal O rC@ypavOt OTl 0pVgal C=@V0yOl TZ
NOl gyXv, sapappeovxl
soxay
,. og xai Tauxa
vNOlV, 05 V0V yCV C0NyOl, =pOTC0OV8
oixouXueval xaRxq. gTlyaTO IIpiasog,
o0ev xai'OpveaTq
XClVTal 8
D=gp
8' CXCl exaRelTo.
TOU=galOU TOUT@V
zlXU@Vi@V. SeevanderValk1971, pp.448-449.I do notputparticular emphasison the distinctionstressedin thispassagebetweenthe designation Strabodoesnot actuxuN andwoBlg. to Coallyuseeithertermin reference rinthianOrneai,andtheArgiveOrneai wasnot alwayssubjectto Argos.Diocallsthe doros(16.39)specifically ArgiveOrneaia s°AlS-Thereforethe in terminology discrepancy apparent arguefor alonedoesnot persuasively the necessityof a secondOrneaito explainthevariancein the sources.The differenttermscouldrelateto a change in the statusof ArgiveOrneaiafterits reductionbyArgosin 417/16 B.C.
s
I36
JEANNETTE
f
s
v
MARCHAND
Pausanias) andhisreference to a nymphOrnea,bothof whichdo notappearin Strabo's text,maysuggestthathe is usingadditional information derivedfroma differentsource(oroneof Strabo's originalsources).36 Starting withtheearlytravelers to Greece,mostscholarsseekingthe locationof Orneaihaveassumed thatStrabowasconfused, andthatthere wasnosecond,Corinthian, Orneai; alloftheinformation aboutanOrneai thatStrabohadtakenfromhissourcesmustreferto oneplace.Thisview wasfollowed mostinfluentially byFrazer inhiscommentary onPausanias.37 Buta numberof scholars whohavelookedattheproblemin termsof the 5th-century historyof Argosor actually lookedforArgiveOrneaibyautopsyhave begunto arguefortheexistence ofa second,Corinthian, Orneai. Amongtheirarguments, theypointoutthatthereis noplacethatcanfit allthetopographical requirements in the sourcesfora singleOrneai:for instance, it shouldbeimmediately clearthatthereis noplaceroughly120 stadesfromArgos(the60 plus60 of Paus.2.25.4-5:approximately 24 km)thatcanalsobe described asbeingabovethe Sikyonian plainorbetweenSikyonandCorinth(Strabo).38 Themaindifficulty formanyscholars in accepting Strabo's accountof a secondOrneaihasbeenthathisaccounthasseemedconfusedandselfcontradictory: howcoulda siteoverlooking the Sikyonian plainbe both withinCorinthian territory buthavebelongedto the Argives?It is this fact,thatStrabohimselfappears tolocatethesecondOrneaiintheArgolid, thatinitiallyledto theassumption thatStrabowasrelatinggarbledinformationaboutthe sameOrneaias discussedby Pausanias andthe other sources. Meyerhelpedto createwhatis stillperhaps thecommonopinion bystatingflatly"Strabons Unterscheidung ist irrig."39 It is of thegreatest importance, however, to recognize thatStrabois basinghisdescription of thissecondOrneai, andindeedtheentirearea,ontherealmofAgamemnon asdescribed in theCatalogue of Ships.He makesit clearthathe believes thattheCorinthian Orneaiis theonementioned in Homer.Thus,heboth indicates thattheOrneaihehasin mindis aplacethathissources believed to havebeenimportant in theMycenaean period,andthatit is at thattime thattheland(alongwithCorinth,Kleonai,andAraithyrea) belongedto the"Argives." G.J.Loloshasrecently interpreted Strabo's comments about Orneaiin a similarfashion,with "Argives" meaningthe lordsof the Mycenaean citadel,althoughhe doesnot observethatStraboindicates thattheCatalogue refersto thissecond,Corinthian, Orneai: 36.The traditionof the eponymous nymphfindssupportin Diod.Sic. 4.72.1-2,wherehe includesa nymph Ornia,alongwith Kleone,amongthe daughters of the riverAsopos. 37. Frazer([1897]1965,p. 217) placesOrneaiat Liondi,usinginformationfrombothStraboandPausanias.
38. Pritchett1980,pp.22-23; Andrewes1970,p. 108;Aly 1950,
p. 249. Herter(1932,pp.251-252,n. 1) hasalsoarguedthata cultof Priapusas mentionedby Strabowouldfit better witha Corinthian location;thisopinion is secondedby Kruse(REXVIII,1939, col.1124,s.v.Orneates). It is worth notingthatPausanias doesnot mention a cultof Priapusamongthoseat the ArgiveOrneai,andthatStrabomentionsthe cultin relationto the CorinthianOrneaitwice(8.6.24[C 382],
textabove,p. 134,andagainat 13.1.12 (C 587):rIpiasog8' soxisoBlgrsi OaBarm. . . rsxvl)oc, 8' ri xoi) [Iptasol) xZZrvol) sap' ai)xolg,'ix' rE, 'Opvrxv
xxv
sroi
KoplvOov
rxrvNvry-
[lrVOI) T01) lrpOI), rlTz TZ tryroual
ato-
60Z!savv rsl xo xlllaval)xovxxv av sxv. 39. RE XVIII,1939,col.1123,s.v.
vi)ool) X
fi
xai
fi
Vi)llXpg fi
Tov
fi
Orneai(E. Meyer).
0rov fi
s
A NEW BRONZE
AGE SITE
I37
The notionof temporary possession of theterritory at onetime in historybytheArgivesis clearlyimplied.Thishistorical time, I argue,mustbe placedin the almostlegendary eraof Argive domination overAraithyrea (Phlius),Sikyon,andCorinthto whichI brieflyreferred in thefirstchapter[thedomainof Agamemnon asgivenin theIliad Catalogue of Ships].To this periodthegeographer assignstheflourishing of Orneai,which was1nru1ns1nn1scWay. A candidate siteforOrneaialongthe 15 krn.of highland separating SikyonfromCorinthhasyetto be advanced. Judgingfromthe specification srpL KopvOow, onewouldbe temptedto placeit in theeasternpartof theplain,beyondthe NemeaRiver.Wiseman,whosurveyed theCorinthian land,favors thispossibility butsaysthat"noremainsof a suitableancienttown haveyetto be foundbetweenCorinthandSikyon."40 .
.
.
.
.
Suitableremainsof an ancienttowncorresponding to Strabo's accounthavenowbeenfoundatDorati,andin exactlythelocationfavored by LolosandWiseman.Straboandhis sourcesclearlyrelatethatthere was a Mycenaean site calledOrneai,abandoned at leastby the timeof Strabo,neara majorriverandon a heightthatoverlooked theterritory of Sikyon.It waslocatednotwithintheArgolidbutwithinthe areaof the Corinthia. Doratihasa magnificent viewoverthe Corinthian Gulf,but becauseof the relativeheightsof the bluffson eithersideof the site,its primary viewis to thewestovertheSilyonianplain(Figs.14-15).Dorati is, however, withinwhatwasCorinthian territory in thehistorical period, sinceit is justeastof the NemeaRiver,the historicalboundary between CorinthandSikyon.41 The siteis lofty,up on a bluff,andtherearetwo riversthatrunpastit:theNemeaandtheDaveliremato theeast.Ornea(i) 40. Lolos1998,p. 103,quoting Wiseman1978,p. 110.Lackinga candidatecloserto Silyon,Lolos presentedthe possibilitythatAetopetracouldbe identifiedwith Orneai. This site,eastof the Longopotamos River,seemshowevertoo fareastto be describedasoverlooking Silyonian land;the historicalboundary between the Corinthiaandthe Silyoniawas the NemeaRiver.It is worthnoting herethatthereis increasing evidence thatArgoshadgreatinfluence(as wellas periodsof completepolitical control)in the historicalperiodin manyof the areasincludedin the Catalogue.NemeawasArgive,at leastin the 4th century,andallthe extantancientsourcesattributethe altarof Zeus ApesantiosatopMt. Apesasto Nemea or associateit withArgiveheroesas
well.Kleonaiwaspoliticallyintegrated intothe Argivestateat leastfora periodof a littlelessthana hundred years,andevenwhenindependent it wasa verycloseallyof Argos.Finally, Xenophontellsus (Hell.4.4.5-6) that in the 390sthe Argivesremovedthe boundarystonesbetweentheirstate andCorinth,bringingCorinthian territoryintotheirown"citylimits." Althoughthiswasa short-livedphenomenon,it pointsup thatthe boundariesbetweenArgosandCorinthwere closerthanoftensupposed. Thusit is not lmpossl) e to lmaglnea tlmeeven in the historicalperiodwhenthe area of Doratimayhavebelongedto Argive territory, andsimilararguments could be advanced forthe areafurtherwest, includingSilyon andPellene(forinstance,Pausanias [2.7.1]thoughtthat .
.
.
.
.
SilyonwasonceArgive).ForKleonai, seePierartandThalmann1980, pp.261-269,no.3; Miller1982;and Perlman2000,pp.131-149.For Apesas,seeRE I (2),1894,col.2699, s.v.Apesas(G. Hirschfeld); RE I (2), 1894,cols.2698-2699,s.v.Apesantios (O.Jessen);andin particular Hes. Theog.326-332;Paus.2.15.3;[Plut.] deFluviis18.9;Plin.HN4.17; Stat. Theb.3.461-462;Steph.Byz.,s.v. Asroag.ForNemea,seeMiller1994, pp.95-96;NemeaII, pp.100,170,233. 41.The NemeaRiverwasthe boundary by Strabo's time,as he himselfcomments:Strab.8.6.25 (C 382);Livy33.15.Exactlywhen the borderbecamefixedat thispoint is not known,butit wascertainly alreadythe borderbythe Classical period.
I38
JEANNETTE
MARCHAND
Figure15. View fromthe site to the easttowardAcrocorinthandthe Corinthianplain.Photoauthor
couldhavebeenan old nameforeither.42 The surfacematerialindicates thattherewasa majorBronzeAge settlementhere,as Strabosuggests, andalsothatit wasabandoned wellbeforehisday.Thelocationis exactly betweenCorinthand Sikyon,the viewis primarily in the directionof Sikyon,butit lieswithinthe bordersof historicalCorinth.It is hardto imagineanothersitethatcouldnotonlysowellfit allof thetopographical indications asgivenbyStrabo,butalsoexplaintheseemingly self-contradictoryelementsin Strabo's account.43 LOCATION OF ARGIVE ORNEAI With a strongcandidate nowproposedforCorinthian Orneai,thelocationof ArgiveOrneaicanbe clarified. Thereis no longeranyreasonto locateArgiveOrneainearariverorexplainhowit couldhavebeenthought to overlook theSikyonian plain,sincethesedetailsrelateto theCorinthian site.Andrewes, Pritchett,andPikoulashaveallarguedpersuasively that 42. Strabo's i)srtoToi)sr8touat 8.6.24(C 382)impliesthatthe sitewas on a height,as Eustathiusemphasizes. The NemeaRiveris conspicuous from Doratiandpresumably the proximityto the riverwasone of the reasonsforthe locationof the settlement(seeFig. 14). If the sitederivedits namefromone of the tworiversnearit, the Nemea,given its considerable lengthandsize,seemsa muchmorelikelysourcethanthe rema to the east.The nameof the rivermay havesubsequently changedafterOrneai wasabandoned andthe sanctuary of Nemeagainedinfluence.The earliest designation of a riveras Nr,urasxatoa8toashouldin factindicatethatthe nameof the riverderivedfromits origin in the Nemeavalley.However,all the references in whicha contextis preserved forthistitleactuallyreferto
eventsthatoccurrednearthe LongopotamosRiverto the east(satoaTov Nr,uravsoxa,uovin Diod.Sic.14.83.2; Aeschin.2.168,srtotri v Nr,uraAa zaBou,urvnv xatoa8toav, is ambiguous. Ephoros,FGrH 70 F82 [Harpokr.] preserves sAatov ri 5 Nr,urEa8>os Xapa8pa5, alsoin Suid.Phot.,s.v. Nr,urasxatoa8toa, bothwithoutcontext, althoughreference is alsomadeto Aischinesin Harpokration). Presumablythe termis the resultof confused topography derivingfroma misreading of Xenophon,whowhilediscussing eventsin partnearthe Nemearefersto the Longopotamos as simplya xatoa8pa (Hell.4.2.15-23).OnlyStrab. 8.6.25(C 382) andLivy33.15canbe definitelyassociated withthe river flowingfromthe Sanctuary of Zeusout to the CorinthianGulfpastDorati,and
theybothcallit simply"theNemea River." Nevertheless, it shouldbe evidentthatthe namederivesfrom the valleyandthe site.It is therefore not necessary thatthe nameNemeafor the riverpredatesthe periodin which the Nemeansanctuary andgames gainedprominence. SeeREXVI,1935, col.2322,n. 3, s.v.Nemea(E. Meyer) andthe importantobservations of Pritchett(1969,p. 78) andLolos (1998,p. 131). 43. It canalsobe notedthatdirectly uponenteringthe plain,the Nemea Riverturnsto the eastbeforecontinuing straightintothe gulf;therefore, althoughDoratiis on the Corinthian sideof the river,the portionof the plaindirectlyin frontof the siteis "Sikyonian."
A NEW BRONZE
AGE SITE
I39
fortheArgiveOrneaiis foundin theruinson andnear thebestcandidate hasbeenrenamed atKatoBelesi(which,confilsingly, thehillPaliokastraki Lyrkeia), locatednorthwestof Argosin the Inachosvalley(Fig. 13).44 by the circumandtheproblemis compounded Certainty is impossible, stancethatancientLyrkeiahasalsonot beendefinitelylocated.Pausanias,however,indicatesthat Lyrkeiashouldbe locatedapproximately 60 stades(ca.12 krn)fromArgoson the Klimaxroadleadingfromthe Deirasridgeof ArgostowardMantineia,andthatit shouldbe intervisfits a locationnearSchinochori iblewith the Larissa.This description ArgiveOrremainshavebeenfound.45 and Skala,whereappropriate 60 stadesfurtheralongthe sameroadbut neaishouldbe approximately of KatoBelesi, to Paliokastraki stillwithinthe Argolid,corresponding impressive remainsof a Classicaltownand wherePritchettdocumented andwhichPikoulashasshownto havebeenon the Klimax fortification, routeto Mantineia.46 the secure be considered Despitewhatwouldin othercircumstances in the Gymno Orneai, a site with Argive of KatoBelesi identification town.The Gymno identifiedasthe5th-century valleyis stillpersistently valleygivesaccessfromthe Argolidintothe PhliasianplainviaLiondi; of the4thcenturyarelocatedatthe watchtower theremainsof a Classical Thesitewasone entrance to thatvalley,on a heightat KastroKourounas. forArgiveOrneai,and bytheearlytravelers proposed of thoseoriginally no eviThereis absolutely by Meyerwithoutautopsy.47 was supported that fora Classicaltownnearthesite.Pritchettobserved dence,however, notwiththeArgolidbutto thetoweris situatedto permitcommunication thateventhetowerwasArgive.48 thenorthwithPhlius,makingituncertain haspersistedpartlybecauseof Pritchetthasnotedthattheidentification aboutthe second information make all of the scholars to the attemptof accountof theArgivetown,and OrneaifromStrabofit withPausanias's the locationnearGymnowasseento be moresuitedthanKatoBelesi it is neara river(thesource indications: topographical to someof Strabo's of the Inachos)andit lies closerto Sikyon,perhapsleadingto Strabo's 44. Andrewes1970,p. 107;Tomlinson1972,p. 39;Pritchett1980, pp.19-30;Pikoulas1995,pp.267-270. 45. Pritchett1980,pp. 12-17; Pikoulas1995,pp.263-265 (full (1976, Papachatzes bibliographies). p. 186,n. 1) andothersstillfollowthe the rein associating earlytravelers mainsat KatoBelesiwithLyrkeia; however,KatoBelesi,at over18 km fromArgos,is too farawayto fit Pausanias'sdescription(althoughTomlinson [1972,p. 39] hasrightlypointed are measurements out thatPausanias's and Schinochori onlyapproximate). 12 kmfrom Skalaareapproximately shouldbe Argos:thisfirstmeasurement givesa the moreexact,sincePausanias to Lyrfigureof 60 stadesin reference keia;forOrneaihe simplysaysthatit is
aboutthe samedistancebeyond Lyrkeia. 46. Pritchett1980, pp. 19-31. Pikoulas(1995, p. 101, pp. 267-270, thattheArgos290) hasconfirmed routepasseddirectly Orneai-Mantineia of KatoBelesi:oneof by Paliokastraki wheelrutsof the stretchesof preserved northof the the roadis immediately of site.See alsothe earlierobservations Frickenhaus andMuller1911,p. 24. Papachatzes (1976, p. 187) estimates thatKatoBelesiis 18 kmfromArgos; Pritchett(1980, p. 24, n. 41) estimates thatit is farther,andobservesalsothat the ancientroadwouldhavetakena routethanthe modern morecircuitous one.Musti(1986, p. 295) alsoconof cludesthatthe identifications = Lyrkeia,andKato Schinochori/Skala
Belesi= Orneaibestfit the distances givenbyPausanias. 47. REXVIll, 1939, cols. 11234, 1124, s.v. Orneai(E. Meyer);KlPauly 1972, p. 346, s.v. Orneai(E. Meyer). 3 kmto The siteis approximately
the southeastof thevillageof Gymno on the backroadfromPhliusto Sterna andArgos.Pritchett(1980, pp. 2324, 27-31) reviewsthe evidencefrom forremainsnear the earlytravelers thatthey Gymnoanddemonstrates allappearto referto thissame watchtower. 48. Pikoulas(1995, p. 73), while rejectingthe identificacategorically tionof the towerwithOrneai,argues thatdespitethelimitedvisibilitythe towermayhavebeenArgive.
I40
JEANNETTE
MARCHAND
description of Orneaias"above Sikyon."49 Sucha compromise is notsatisfactory, however, andultimately Gymnodoesnotfit eithertheevidenceof StraboorPausanias: alocationnearGymnocannotbecorrectly described as "aboveSikyon," andthe locationdoesnot fit with the evidencethat placesArgiveOrneaiwithintheArgolidandon the routeto Mantineia. With the discovery of a candidate forCorinthian Orneai,allof Strabo's information canbefinallydisassociated fromArgiveOrneai,andtheidentificationby Andrewes, Pritchett,andPikoulasof thattownwith Kato Belesicanbe deemedsecure. Thereis,however, anothersignificant reasonforthestayingpowerof Gymnoas a candidate forArgiveOrneai.Hope SimpsonandLazenby reportedfindingMycenaean sherdsin the vicinityof thewatchtower at Gymno,whichapparently madeit appropriate fora sitementioned in the Iliad.Following FrazerandMeyer,theyidentifiedit astheOrneaiof the HomericCatalogue.5° TherewasnodoubtaMycenaean presence atKastro of Gymno,buttherearenumerous placeswithMycenaean material that arenotmentioned intheIliad,andthesitedoesnotfitPausanias's descriptionor thatof othersources: it maywellhavebeena Mycenaean settlement,butit isveryunlikely to havebeenClassical ArgiveOrneai.Pritchett statesthecasewell:"Unable to findanyotherremains of a sitein theupper Yimnon[Gymno]valley, I cannotaccepttheidentification of aMycenaean settlement, havingafourth-centurywatchtower, oriented toward thenorth, asevidenceforOrneai,anArgivepolisof theClassical period.''5 STRAB O'S CORINTHIAN CATALOGUE OF SHIPS
ORNEAI IN THE
It is of somesignificance thatin discussions of the Catalogueof Ships Gymnois consistently identified withArgiveOrneai,despiteits clearincongruitywith the evidencefor the 5th-century town.Althoughsome prehistoric sherdsandoneworkedobsidianbladehavebeenobservedat KatoBelesi,thereis as yet no publishedevidencethattherewasa LH presencethere.52 Thishascreatedreluctanceevenamongscholars who 49. Frazer([1897]1965,p. 217) evenplacedOrneaifurthernorthin the Phliusvalleynearthe modernvillage of Liondi(andthus"closer" to Sikyon; butevena locationasfarnorthas Liondicannotby anystretchof the imagination be describedasoverlooking the Sikyonianplain).The identificationcanstillbe found(e.g.,Papachatzes1976,pp.189-192)although no moderninvestigator hasyet to reportanyspecificsuitableremainsin the area.In anyevent,thisgeneralareais muchtoo farfromArgosto fit Pausanias'saccountand,asAndrewes (1970,p. 107)pointsout,the areawas certainlyin PhliasianandnotArgive
territory in the 5th century;therefore, anyremainsthatmaycometo lightin the regionof Liondiin futurewouldbe unlikelyto represent ArgiveOrneai. 50. HopeSimpsonandLazenby 1970,pp.66-67:"Around it [thetower on Kourounas hill]andextendingover thewholesummitandthe upperslopes to the southandeast-the northern slopesaremuchsteepewe found a considerable numberof Mycenaean sherdsof a 'provincial' nature,ranging fromLH II to LH IIIB(mainlythe latter),togetherwithfivesherdsof GreyMinyanware,someobsidian,and a fewpiecesof classicalpottery.This site mustnowclearlybe considered in
connectionwiththe locationof HomericOrneai,especially if it shouldin factturnoutthatthereare no prehistoric remainsin the [closer] vicinityof Gymno."Pikoulas(1995, p. 227) reportedmorerecentlyfinding at the siteonlya fewunglazed sherds,one LH kylixbase,andno obsidian. 51. Pritchett1980,pp.30-31. 52. Forthe ceramicsat Paliokastraki of KatoBelesi,seePritchett1980, p. 25;Pikoulas1995,p. 269. Pritchett mentions"prehistoric" sherds;Pikoulas reports,fromthe prehistoric period, onlyNeolithicsherds.
601: "DaPausmit den wortenasrto Towxqg
A NEW BRONZE
AGE SITE
'Atoyaq
rxvTo
I4I
reflectsthe Bronze do not thinkthatthe Catalogueof Shipsaccurately in Homer. of a citymentioned Age to identifyKatoBelesiasthelocation Orneaihasbeenlocated,thereis no forCorinthian Nowthata candidate site at ArgiveOrneai, longeranyreasonto expecta majorMycenaean town a traditionthatit is the Corinthian sinceStraboclearlypreserves Evenwithno otherfactorstakeninto in theCatalogue. thatis mentioned it townis produced, oncea candidateforthe Corinthian consideration, forHomer'sOrneaias becomeaslikelya candidate shouldautomatically forbothclaims. preserved traditions theArgivetown,sincethereareliterary andStrabo,andI It becomesa questionof choosingbetweenPausanias evidenceis to be preferred. proposethatin thiscase,Strabo's Orneai,W. Aly arguedthat Evenwithouta locationforCorinthian Strabo's secondOrneaiis morelikelyto be theonementionedin Homer Apollodor sources, and"Wo Strabo isusing2nd-centurys.c. simplybecause rechnen,dieunsviel vorliegt,kannmanmiteinergenauenOrtskenntnis Indeed,forthisreasonStrabo mehrzwingtzulernenalszu korrigieren."53 of Ships.54 of theCatalogue is a majorsourceformostmoderndiscussions ofArgiveOrneai Pausanias doesnotmentionanysourceforhisdiscussion As hasbeen localinformation. andit is likelythathe is simplyrepeating of places guideto thetopography is a veryreliable oftennoted,Pausanias ofwhathe seesandwhat andafaithfillreporter thathehashimselfvisited, to himis trueornotis anentirelydifhe is told;whetherwhatis reported hasargued thatinPausanias's KaLkmann, following Pritchett, ferentmatter. arguingagainstStraboon description of ArgiveOrneai,he is consciously theissueof theCatalogue: Pausaniasder is thatof A. Kalkrnann, The mostdetailedtreatment byH. Hitzig Perieget(Berlin1886)158-159,whichis summarized Descriptio1.2 (Leipzig1899) PausaniaeGraeciae andH. Bluemner,
159wohl Kalkrnann scheint,vermuthet seineAnsichtzu motivieren derQuelle mitrecht,dassergegendieMeinungApollodors, wolle."Strabo(8.6.17.376),orApollodoros, Strabos, polemisieren wasthecitybetween haddecidedthattheOrneaiof the Catalogue makesit CorinthandSikyon,nottheArgiveOrneai.Pausanias wastheArgiveone.55 clearthatthecityin theCatalogue textin mind.PauhadStrabo's I seeno reasonto concludethatPausanias thathe knowsof anyOrneai hereor elsewhere saniasgivesno indication Orneaihadceasedto exist otherthantheonein theArgolid.Corinthian didnotvisit longbefore.Pausanias day,andprobably already in Strabo's Pausanias's Argive Orneai. at least pass the 53. Aly 1950,p. 249.Andrewesdoes the areaof Dorati,buthe did explanation of ArgiveOrneaiandthe Catalogueappearsto be his own not givea clearopinion.I knowof no difficulty withthe Catalogueof Ships,or otherscholarwho hassupportedStrabo attemptto explaina perceived on thispoint. offeredby localguides,who mayhave evenmorelikely,an explanation 54. Niese 1877,passim;Giovannini beenmoreawareof the "rival" claimant.If theywerenot,it wouldhave 1969,pp.11-17;Visser1997,p. 33 and beenevenmorenatural gives forthemto claimHomericstatus.Pausanias passim.As with anyancientsource, of Argive in the neighborhood when he is ofthe Catalogue his explanation Strabomustbe usedwithsuitable andthusit is logicalthathe wouldgivea localacOrneaiandLyrkeia, caution. 55. Pritchett1980,pp.21-22. it, andnot a placein what Homeras representing countthatinterprets
I42
JEANNETTE MARCHAND
laterbecameCorinthian territory, intheIliad.Hiscomments aboutLyrkeia alsoseemto fit thisapologetic mode.56 EvenifweassumethatPausanias wasawareof Strabo's contrary claim, thereis nogoodreasonto thinkthathehadbetterinformation thanStrabo (orApollodoros) onHomer's meaning. Indeed,hisexplanation oftheCatalogueasgivenabovein Pritchett's translation (p.134)doesnotmakemuch senseperse:he claimsthatOrneaiis mentioned beforePhliusandSikyon becauseit is locatedin the countryof theArgives.Thisexplanation does not explainwhyCorinthandKleonai,not"situated in thecountryof the Argives" in Pausanias's day,arementioned beforeOrneai.Moreover, here "Argives" cannotbeexplained asmeaningthelordsoftheMycenaean citadel,sincethisexplanation wouldalsonotmakeanysense:"Homer" locates all of theplaceson thelistwithinthe countrysubjectto Mycenae, since thatis precisely thepointof theCatalogue. Pausanias seemsto recognize thatthe orderof the Catalogue is topographical, buthe cannotsuccessfullyreconcilethe physicallocationof ArgiveOrneaiwith the list;acceptingJones's translation ofthepassage(above,p. 134)doesnotsolvethe problem. Indeed,thedifficulties in the sourcesforOrneainowappearto arisenotfromStrabomakingtwoplacesoutof one,butfromPausanias or hisguides,unaware of theCorinthian Orneai,conflating thetwoseparate placesintooneArgivetownandassociating allthe mythsof Corinthian OrneaiwiththatArgiveplace. Nowthatasitehasappeared thatvindicates theconsistency of Strabo's sources concerning theexistence, nature, andlocationof Corinthian Orneai, I proposethatthesesourcesalsohadbetterinformation concerning the Catalogue of ShipsthanPausanias's localinformants in theArgolid. DORATI AS HOMERIC ORNEAI? I do notsuggestthattheclaimof Corinthian Orneaito Homericstatusis superior to thatof ArgiveOrneaibecausethe surfacematerial andcommandinglocationof Doraticlearlyindicatethatit wasa muchmoresignificantMycenaean settlement thanKatoBelesi(orKastroof Gymno,for thatmatter).57 I dopropose, however, thataccepting DoratiastheOrneai of the Cataloguecanelucidatethe logicbehindthe arrangement of the nameslistedin the firsthalfof Agamemnon's realm.J. P. Crielaard has recentlyobserved thatthedebateconcerning thehistorical period,if any, whichtheworldof the Homericpoemsmostreflectshasbeenelevated 56. Anderson(1995,p. 181) makesa similarobservation about Pausanias's accountat Donussa: aPausanias wasalsotoldthatDonussa hadplayeda part,thougha humble one,in worldhistory,andthat'lofty Donoessa'hadbeenlistedin theIliad amongthe citiesof Agamemnon's kingdom[insteadof Gonoessa].... His Donussaninformants mayhave sharedhis historiographical outlook,
andfoundin the Catalogueof Ships the bestopportunity to immortalize an otherwiseforgottencornerof an obscuredistrict." See alsohis similar commentson Pausanias's accountof Chaeronea. 57. Dickinson(1986,p. 31) has faultedinvestigators intothe Catalogue forusingthe presenceof Mycenaean materialto argueHomericstatus.This, he says,is to arguewhatis to be proved,
thatthe Catalogueis a Mycenaean document.AlthoughI agreethatthe procedure is prejudiced, one doesnot needto thinkthatthe Catalogueoriginatedin theMycenaean periodor even fromMycenaean sourcesto recognize thata placewithmajorvisibleBronze Age remainswouldbe morelikelyto haveaccrueda heroicpastthanone thatcouldproducelittleor none.
A NEW BRONZE AGE SITE
I43
to the statusof a new HomericQuestion.58 Withinthis largerdebate, questions concerning thesourceanddateof composition of theCatalogue of Shipsandthehistoricity of thepoliticalsituation described thereinpose a numberof problems thatcannotbe addressed here.59 I confinemycommentsto the natureandorderof namesin the list,andindeedit canbe arguedthattheidentification of theplaceslistedin theCatalogue should as muchaspossibleprecedeanyinterpretation of the Catalogue's sources, date,ormeaning. Quiteapartfromthe difficulties of whyAgamemnon wasassigned onlypartof the Argiveplainanda seriesof placesalongthe Corinthian Gulf,with Orneaiidentifiedas the Argivetownit hasbeendifficultto explainwhyin additionto Mycenaehe is assignedonlyoneotherrather unimportant placein the Argolid.Furthermore, withOrneaias the Argivetown,theorderinwhichAgamemnon's holdingsarelistedis curious, as Pausanias clearlynoticed,regardless of whyor whenthoseparticular placesweresingledout to comprisehis realm.C. Morganrecentlydescribedtherealmasaarather strange balanceofplacenames intheArgolid, the Corinthia,andAchaia."60 Althoughby no meanssolvingthe difflcultiesin interpreting Agamemnon's realm,recognizingthe Orneaiof the Catalogueas Doratiremovesthe problemof figuringoutwhyonly ArgiveOrneai,anunimportant placein theBronzeAgeandin allsubsequentperiods,is singledout for mentionalongwithMycenae,whena numberof important BronzeAge sitesin theArgolidarenotmentioned atall. WithOrneaiatDorati,onlyoneplacein theArgolidproperis mentioned:Mycenaeitself.The remaining placesin thispartof therealmare thenlistedin a cleartopographical order,andaswiththe Argolid,only oneplaceis mentionedin eachgeographically discretearea.Theseareas correspond to thevalleysof the threemajorparallelriversthatflowinto the CorinthianGulf (the Longopotamos, the Nemea,andthe Asopos, respectively) andthesectionsintowhichtheseriversdividethegulfplain (Fig.13).The routesalongtheseriverswereimportant foraccessto the gulf andthe Isthmusfromthe Argolidin boththe prehistoric andthe historical periods; it is nothardto imaginewhyoneplacecontrolling each valleywouldbe singledoutformentionin thelist. Thus,the orderof the firstpartof the list is quiteclearandlogical: Mycenaeis mentioned first,sinceit is theseatofAgamemnon's powerand regarded as the controlling cityof the northernhalfof the Argiveplain 58. Crielaard (1995,p. 201, and in the bardictradition). passim)providesextensivebibliography 59.Thereis stilllitde agreement on andsummary of previousandcurrent anyof theseissues.Forexample,in a views.See alsoSherratt1990,pp.822singlerecentvolumeof essaystwo 824.Crielaard arguesthatthe Homeric completelydifferentperiodsforthe poemsreflecta latestagein the Early worldthatthe Cataloguerepresents IronAge (8thor 7th centuryB.C.), werepresented: Hood 1995(LH IIIC) whileSherrattarguesthatthe texts andAnderson1995(late8th century containlayersof information froma B.C.). Fora filllbibliography on all successionof periodsfromthe issuesrelatingto the Catalogue,see prepalatial to the later8th century Visser1997,pp.75>773. (representing activelycreativeperiods 60. IsthmiaVIII,p. 350.
I44
JEANNETTE MARCHAND
The remaining (thesouthernhalfis assignedto DiomedesandArgos).6l orderfromeastto west,startingwith placesarelistedin topographical theeastern ascontrolling hereagainonlyonesiteis represented Corinth: orderto the west is Kleonai, gulf andIsthmus.Next in topographical the Riverandcontrolling accessalongthe Longopotamos representing representing at Dorati, comes Orneai valley. Next modernAgiosVasilios accessto the Nemeavalleyviathe NemeaRiverandcontrolof the cencontrolof the next,represents tralportionof thecoastalplain.Araithyrea, Phliusvalleyandthe sourceof the Asopos;the exactlocationof AraibutbothStrabo(8.6.24[C 382])andPausanias thyreais asyetunknown, Finally,Sikyon, of Phlius.62 the predecessor (2.12.4-5)makeAraithyrea bothcontrolof thecoastal represents standingalonein itsownhexameter, plainwestof theAsoposandthebeginningof a newsectionof therealm in thepoem.63 Strabo(8.6.17[C 378])commentsthatsomeof the sectionsof the order;it is a logicalwayto organize arelistedin topographical Catalogue thatanitinerrequire thehypothesis and it does not place-names a listof doesnot ruleit arylaybehindthe Catalogue(althoughit alsoobviously out).Onlyoneplaceis listedforeachdiscreteregion,andtheseplacesare areas citiesin theirrespective as the mostimportant clearlyrepresented whetherthissituaterritory: andonesthatwerein controlof surrounding to the politicalsituationduringthe Bronze(or Early tion corresponds With Orneaiat Dorati,all of the Iron)Age is anothermatterentirely. AncientCorinthhasproremains: placesin thelist do haveMycenaean town(if located andmuchof theMycenaean material ducedMycenaean bylateroccuin the areaof the Romanagora)mayhavebeendestroyed andI have settlement,65 Mycenaean Kleonaihadan important pation.64 havebeena reality:Kirk(1985,p. 181) 61.This divisionof theplainis the suggeststheperiodof declineat the pointthathascausedthe mostcontroAnderendof the LateBronzeAge.Vermeule versyanddiverseexplanations. son(1995,p. 185)statesthewidespread (1987,p. 133)arguesthatAgamemwithelements opinionthatit is hardto believethat non'srealmcorresponds in the mythof Adrastosat Sikyonand "theMycenaeof the shaftgraves,the vase"didnot liongateor thewarrior's thattakentogetherthe Catalogueand controlthe entireArgiveplain.Jamethe Sikyonianking-list(Paus.2.5.6) a consistentBronzeAge mayrepresent son,Runnels,andvanAndel(1994, of the palatial oraltraditionpredating p. 59) considerthat"thegeography period. has Peloponnesos the northeastern to assignAgabeengerrymandered 62. See alsoSteph.Byz.,s.v.'ApalmemnonandDiomedes,bothimporOI)pra, 'Aparia, and@lOU5; M II (1), of theIliad, 1895,col.374,s.v.Araithyrea tantfiguresin the narrative homebasesin theArgeia,"andFinkel(G. Hirschfeld). berg(1988,p. 39) notesthatthe terri63. It wouldbe possibleto continue further,butthe places toryof DiomedesandArgoshasbeen the argument broughtintoalignmentwiththe lot of listedafterSikyonhavenotbeenpreTemenos,andsuggeststhatthe Cataciselylocated,andit is alsopossiblethat the interests the sectionfollowingSikyonmayemloguein generalrepresents method. Athens,Corinth,Argos, ploya differentorganizational of 7th-century andSparta.Othersseekto finda hisforCorinthabove 64. Seereferences in n. 21. toricalperiodotherthanthe palatialin whichsucha politicalsituationcould 65. Zygouriesis usuallycitedas the
settlement Mycenaean mostimportant in the Kleonaivalley.The site,however, (andas hasbeenpartiallyexcavated by partof animportantexcavation Blegen)andKleonaihasnot.I have confirmedby autopsythe observations in HopeSimpsonandLazenby(1970, p. 66) thatthe acropolisof Kleonaiwas the centerof animportantMycenaean "Indeed,to judgefromthe settlement: sherdswe pickedupwhenwe visited theplacein 1960,the Mycenaean phasewasthe mostimportantin this extendarea,the settlementapparently ing forabout300 metresnorthto south by about250 metreseastto west."It is impossibleto determinethe relationshipbetweenthe sitesat present:only (1971,p. 45) andPharaklas Sakellariou arguedthatthe havesubsequently betweenthe actualstatusrelationship two sitesin thevalleymayhavebeen the reverseof the oneusuallyimagined.
A NEW BRONZE AGE SITE
66.Vermeule1987,pp.134-135; Krystalle-Votse 1996,p.25. 67. Andrewes(1970,p.108) is amongthosewho assignthe references to hostilitiesbetweenan Orneaiand Sikyonin the EarlyArchaicperiodto the CorinthianOrneai(seeabove, n.27): "thereseemsto be a similar confusionin Pausanias X.18.5,a dedicationat Delphiby'Opvraxat ot sv q 'ApyoB8lforthe repulseof Silyonians who hadoppressedthemin war(cf. Plut.De Pyth.or.15, 401d).All this wouldbe easierto understand if there wasin fact,or hadoncebeen,an Orneaiin the neighborhood of Sikyonand Corinth." (Andrewesinterprets the phraseot sv q 'ApyoB86 as an addition byPausanias, not as a directquotefrom the dedication.) In the association of thisconflictwiththe CorinthianOrneai he is followedby Lolos(1998,pp.49, 102);I planto explorethe issueof the possiblecontinuation of Corinthian Orneaiintothe EarlyArchaicperiod morefullyin a futurearticle. 68.Visser1997,p.161. 69. SeeJameson,Runnels,andvan Andel1994,andabove,n. 61. 70. In thislight,seethe recentarticle by Kolonas(1996-1997)in whichhe summarizes the resultsof research in the westernportionof"Agamemnon's realm" overthe pastfifteenyears.He listsmany newsltes,lnc uclng onesln tize reglon of A'cytov andthe AtytaSstaxpa, and concludes(p.490) thatthe Mycenaean presencein the regionof Achaiawas "powerful andpopulous," andthat"this newevidencehasindicatedthe high standard of culturaldevelopment of the Mycenaean inhabitants, thuschallenging theprevailing viewthatAchaeawas onlya peripheral andratherbackward regionof theMycenaean world." .
.
.
.
.
I45
arguedthatDoratirepresents animpressive Mycenaean town.Araithyrea hasnotbeenlocated,butMycenaean material is notlackingin thePhlius valleyandsomehaveassociated thecemetery atAidoniawiththename.66 Furthermore, thesecenterswerethe mostimportant placesin their respective areaseitherin mythor factin the historicalperiodwhenthe poemswerecomposedin roughlytheirpresentform(the8thor7thcentury):Mycenaeis of coursethefocusoftheTrojanWar myth.Corinthwas bythattimethemajorpowerontheeasterngulf,andKleonaibythencertainlycontrolled theAgiosVasiliosvalley.OrneaiatDorati,evenif it had ceasedto existatthistime,wasnodoubtstillvisibleasanimpressive ruin.67 Visserhasarguedthattheinclusionof Orneaiin theCatalogue mayhave to dowithits importance in myththroughassociation withOrneus.This is possiblebut,asseenabove,thismythis aslikelyto belongoriginally to Corinthianas to ArgiveOrneai,anda visibleBronzeAge site is more likelyto haveaccruedsucha mythicpast.68 Araithyrea was seenas the predecessor to Phlius,whichcontrolled itsvalleyin thehistorical period. Thus,evenif theplaceson thelistwereneverthecontrolling Bronze Agesettlements in theirrespective settingsoratthesametime,theycould havebeenperceived in a subsequent periodto havebeenpowerfill citiesof the BronzeAge,or convincingly represented as suchin a mytho-history thatwasafterallsetin theheroicpast.Doratitherefore makesbettersense of the Catalogue's description of therealmof Agamemnon, whetherone wishesto interpret it asrepresenting therealpoliticalgeography of aphase oftheBronzeorEarlyIron Age,orwhether onechoosesto seeit asmerely a logicallyorganized description of a "gerrymandered" realmattempting to reconcilemythsandtraditions aboutvisibleremainswithcontemporarypoliticalconditions.69 The significantpointis thatwith Orneaiat Dorati,therealmis clear,logical,andhasa definitetopographical order; the audienceof theIliadwouldhavehadno difficulty in believingthatit hadexistedasa politicalrealityatsomeunspecified timein theheroicage. The identification of DoratiasOrneaimakessenseof theorderof placenamesin thelist andprovidesanexplanation fortheinclusionof Orneai in thelistthatfitswithvirtually anyinterpretation of thedateandhistoricityof the Catalogue. Evenif onedoesnotacceptthatDoraticanwithsomeconfidence be associated withthe Orneaiof the HomericCatalogue andof Strabo,the site meritsattentionbecauseof its commanding locationandabundant surfacematerial, andbecauseof the strongpossibility thatstructures are preserved. Thepredominance of Mycenaean potterysuggeststhatDorati shouldproveto havea significant LH settlement phase.Giventhedearth of attestedlargeMycenaean settlements in theCorinthia andthecontinuingcontroversies surrounding theinterpretation oftheCatalogue of Ships, futureworkatthissiteis to be encouraged andthefactunderscored that, evenin ourdayin areaswelltraversed, thereis stillmuchto discover in the Greeklandscape.70
I46
JEANNETTE MARCHAND
REFERENCES Alcock,S. 1988.'LargeSite'Survey, Phlius1986,"AJA 92, pp.233-234 (abstract). .1991. aUrbanSurveyand the Polisof Phlius,"Hesperia 60, pp.421-463. Alin,P.1962.DasEndedermykenischen Fundstatten aufgriechischen Festland, Lund. Aly,W.1950. aZumneuenStrabontext,"PP 5, pp.228-263. Anderson, J. G.T., andJ. K.Anderson. 1975.aALostCityDiscovered?" CSCA8, pp.1-6. Anderson,J.K. 1995.aTheGeometric Catalogueof Ships,"in TheAges of Homer, a Tribute toEmilyTownsend Vermeule, J.B. CarterandS.P. Morris,eds.,Austin,pp.181-191. Andrewes,A. 1970.AHistoricalCommentary onThucydides IV,Oxford. Blegen,C. W. 1920.aCorinthin Prehistoric Times,"AJA24, pp.1-13. . 1921.Korakou, a Prehistoric Settlement nearCorinth, Boston. . 1928.Zygouries, a Prehistoric Settlement in theValley ofCleonae, Cambridge, Mass. .1930-1931."Gonia," MMS 3, pp.55-80. Bookidis,N., andJ.E. Fisher.1972. aTheSanctuary of Demeterand Koreon Acrocorinth: Preliminary ReportIV:1969-1970,"Hesperia 41, pp.283-331. .1974. aSanctuary of Demeter andKoreon Acrocorinth: PreliminaryReportV:1971-1973," Hesperia 43, pp.267-307. Broneer,O. 1951.Investigationsat Corinth,1950," Hesperia 20, pp.291-300. Corinth I.1 = H. N. Fowlerand R. Stillwell,Introduction, Topography,Architecture (Corinth I, pt. 1), Cambridge, Mass.,1932. Corinth III.1= C. Blegen,R. Stillwell, O. Broneer,andA. Bellinger, Acrocorinth: Excavations in 1926 (Corinth III,pt.1), Cambridge, Mass.,1930. Corinth XIII = C. Blegen,H. Palmer, andR.Young,TheNorthCemetery (Corinth XIII),Princeton1964.
Crielaard,J. P.1995."Homer,History, andArchaeology," in HomericQuestions:Essaysin Philology,Ancient History,andArchaeology, Including thePapersof a Conference Organized by theNetherlandsInstituteatAthens (lS May 1993), J. P.Crielaard, ed.,
Amsterdam, pp.201-288. Demakopoulou, K.,ed.1996.The AidoniaTreasure: SealsandJewellery of theAegeanLateBronzeAge. NationalArchaeological Museum, Athens,30 May-1 September 1996,
Athens. Dickinson,O.T. P.K.1972.aLate HelladicIIA andIIB:Some EvidencefromKorakou," BSA 67, pp.103-112. . 1986."Homer,thePoetof the DarkAge,"Greeceand Rome33,2nd ser.,pp.2(}37. Dunbabin,T. 1948."TheEarlyHistory of Corinth,"JHS 68, pp.59-69. Finkelberg, M.1988. "Ajax's Entryin the HesiodicCatalogueof Women," CQ38, pp.31-41. Fossey,J.1990."ThePerakhora PeninsulaSurvey," EchCl34, n.s.9, pp.201-211. Frazer, J. G. [1897]1965.Pausanias's Descritfionof GreeceIII,repr.,New York. Frickenhaus, A., andW. Muller.1911. "AusderArgolis," AM 36, pp.2138. Gauvin,G., andJ.Morin.1997. "Quelques sitesprehistoriques a la peripherie desplainesde Cleonees et Phlionte:Une note,"in ArgoloKorinthiaka1, G. Gauvin,J.Morin, andJ. Fossey,eds.,Amsterdam, pp.l-13. Gebauer,K. 1939.Forschungenin der Argolis,"JdI54, cols.268-294. Giovannini, A. 1969.Etudehistorique surlesoriginesduCataloguedes Vaisseaux, Berne. Griffin,A.1982. Sikyon,OXord. Hatzepouliou-Kallire, E. 1984."Arf ava HE xat ME otxtopov o Bovo Arnowrnpa,"ArchDelt 33,1978,A'
[1984],pp.325-336. Herter,H.1932. De Priapo,Tubingen. Hood,S. 1995.aTheBronzeAge Contextof Homer," in TheAgesof
A NEW BRONZE AGE SITE Homer,a Tributefo Emily Townsend J.B. CarterandS. P. Vermeule,
Morris,eds.,Austin,pp.25-32. Hope Simpson,R. 1965.A Gazeffeer andAtlas ofMycenaeanSites(BICS
Suppl.16),London. . 1981.MycenaeanGreece,Park Ridge,N.J. Hope Simpson,R., andO.T. P.K. I)ickinson.1979.A Gazeffeerof AegeanCivilisationin theBronze Age 1: TheMainlandandIslands,
Goteborg. Hope Simpson,R., andJ.F.Lazenby. 1970.TheCatalogueof the Shipsin Homer'sIliad,Odord. IsthmiaVIII= C. Morgan,TheLate BronzeAge SettlementandEarlyIron Age Sansfuary(IsthmiaVIII),
Princeton1999. Jameson,M., C. Runnels,andT. van Andel.1994. A GreekCountryside: TheSouthernArgolidf omPrehistory fo fhePresenfDay, Stanford. Jones,H. L. [1927] 1954. The Geography of StraboIV,London. Jones,W. H. S. 1918. Pausanias Descritfionof GreeceI, London.
Funde Karo,G. 1913."Archaologische imJahre1912,"AA 1913, cols.95132.
Kelly,T. 1976. A HisforyofArgosfo 500 B.C., Minneapolis. in the Cult Kilian,K. 1990."Patterns Argolid: Activityin the Mycenaean HaghiaTriada(Klenies),the Profitis EliasCave(HaghiosHadrianos), andthe Citadelof Tiryns,"in CelebrationsofDeath and Divinify in the of fhe BronzeAgeArgolid:Proceedings SixthInfernafionalSymposiumaf the SwedishInstituteaf Athens,11-13 June 1988, R. HaggandG. Nordquist,eds.,Stockholm,pp.185-197. Kirk,G. 1985. TheIliad:A Commentary 1: Books14, Cambridge. Kolonas,L. 1996-1997. ANZ@TZ! Mvxnvaixn
Tosoypaqxia
rqg
in IlKorcowrlocofxor, oAxaviat," SUPP1.22,II, PP.468-490. K. 1989. "Ta AaxxvKrystalle-Votse, BiAlaaso cic'= aCA 1=
xa
in A8dala Koplv0taq,"
E. Mvicovoss rEXsoycow IOU 60 ETVIOV=v=X=vTXOV
sm1
scf
¢,oyovr, Athens,pp.34-43. of the .1996. "TheExcavation
Cemeteryat Aidonia," Mycenaean 1996, PP.21-30. in Demakopoulou
Lolos,G.J.1998.aStudiesin the (diss.Univ. of Sikyonia" Topography Berkeley). of California, M.1941. aCleisienesof McGregor, Sicyonandthe Panhellenic TAPA72, pp.266-287. Festivals," Miller,S. G.1982.Kleonai, the Nemean Games,andthe LamianWar,"in Studiesin AthenianArchitecture, Presented and Topography Sculpture, fo HomerA.Thompson(Hesperia
Suppl.20),Princeton,pp.100-108. .1994.Sosikles andthe BuildingProgram Fourth-Century of Zeusat Nemea," in the Sanctuary in Proceedingsof theInternational on GreekArchitectural Conference TerracoffasoftheClassicalandHellenisticPeriods,December12-15, 1991, N. A. Winter,ed. (Hesperia
Suppl.27),Princeton,pp.85-98. P.A.1995. aThePotteryof Mountjoy, in the LateHelladicPeriod," TheFinds ArtifactandAssemblage: from a RegionalSurveyof the SouthernArgolid,Greece1: ThePrehistoric andEarlyIronAge PofferyandLithic Artifacts,C. Runnels,l). Pullen,and
S. Langdon,eds.,Stanford,pp.5256. .1999. RegionalMycenaean DecoratedPoffery,Rahden. Musti,D.1986. PausaniaGuidadella GreciaII, Milan. NemeaII = S. G. Miller,ed.,Excavafionsaf NemeaII:TheEarlyHel2001. lenisfic Sfadium) Berkeley ComNiese,B.1877. aApollodors als mentarzumSchifEskataloge RhM 32, pp.267QuelleStrabo's," 307. N.1976. nOwoorvc'ov Papachatzes, Ko,oc>0cofxorERAordosIls,ocriyryoc5:
Aofx>cxor,Aiens. Perlman,P.2000.Cifyand Sansfuaryin in TheTheorodokia AncienfGreece: (Hypomnemata121), fhePeloponnese
Gottingen. Pierart,M., andJ.-P.Thalmann.1980. argiennes aNouvellesinscriptions (1),"in Etudesargiennes(BCH Suppl.6),Paris,pp.256-278. Pikoulas,I. A.1995. 'O8CXO aC'XIVO xac 'Arcor> Ko,oc>dooro ar,llv>
Pritchett,W. K.1969.Studiesin Pt.2: AncienfGreekTopography, Battlefields,Berkeley.
I47
.1980. StudiesinAncienfGreek Pt. 3:Roads,Berkeley. Topography, E.1974. Protonotariou-Deilake, 26, rrpaol,uos,"ArchDelf aVAylog 1971,B' 1 [1974],pp.68-71. Pullen,D. 1990.aTheEarlyBronze Age VillageonTsoungizaHill, egeen AncientNemea,"in L'Habitat dela Tableronde prehistorique:Actes (BCHSuppl.19), internationale P.DarcqueandR.Treuil,eds.,Paris, pp.331-346. Robinson,H.1976.Excavationsat Corinth:TempleHill,1968-1972," 45, pp.203-239. Hesperia Rutter,J.1974."TheLateHelladic IIIBandIIICPeriodsat Korakou (diss. andGoniain the Corinthia" Univ.of Pennsylvania). .1979. aTheLastMycenaeans 48, pp.348Hesperia at Corinth," 392. . 1989."ACeramicllefinition of the LateHelladicI fromTsoungiza,"Hydria6, pp.1-19. Groupsfrom . 1990a."Pottery Tsoungizaof the Endof theMiddle 59, pp.375BronzeAge,"Hesperia 458. .1990b.The'Aegeanization' of anInlandCorinthianVillage: fromMiddleHelTheTransition ladicto LateHelladicI atTsounAJA94, p. 329 giza(Nemea)," (abstract). .1993. aReviewof AegeanPreBronze historyII:The Prepalatial Age of the SouthernandCentral AJA97, pp.745GreekMainland," 797. M., andN. Pharaklas. Sakellariou, (Ancient 1971.Kosoc>8cos-Kisworc GreekCities3), Athens. Shear,I.1986.The PanagiaHousesat Shop'at Mycenaeandthe'Potter's sm1sc5 r@Oin cAc'or Zygouries," rcovE- MVA@v=5ACA1= 60 sg IOV Of>aAX=CXOV IOV ,oyov A',Athens, pp.85-98. E.1990. 'ReadingtheTexts': Sherratt, andthe Homeric Archaeology Anfiquify64, pp.807Question," 824. qs K. 1964.H rcgoocaropc'or Syriopoulos, Athens. nfiO7TOWOV, PerThomas,P.1988.aAMycenaean fumedOilWorkshopat Zygouries?" AJA92, p. 254 (abstract).
I48
. 1992."LHIIIB1Potteryfrom TsoungizaandZygouries" (diss. Univ.of NorthCarolina,Chapel Hin). Tomlinson,R. 1972.Argosand the ArgolidfromtheEnd of theBronze Age fo theRomanOccupafion,
London. Touchais,G. 1979.aChronique des fouillesen 1978[Aidonia]," BCH 103,p. 555. . 1980."Chronique desfouilles en 1979[Aidonia],"BCH104, p.595. . 1987."Chronique desfouilles en 1986:Aidonia," BCH 111, p.530. vanderValk,M. 1971.EustathEiCommentariiad HomeriIliademPerfinentesI, Lyon.
Vermeule, E.1987. "BabyAigisthos andthe BronzeAge,"PCPS, n.s.33, pp.122-152.
JEANNETTE
Visser,E. 1997.HomersKatalogder SchXe,Stuttgart. Weinberg,S. 1949."Investigations at Corinth,1947-1948," Hesperia18, pp.148-157. Wells,B., ed. 1996.TheBerbati-Limnes Archaeological Survey,1988-1990 (SkrAth4.44),Stockholm. Wiseman,J.1978. TheLand of fheAncienfCorinthians,Goteborg.
Wright,J. 1990."AnEarlyMycenaean HamletonTsoungizaat Ancient Nemea,"in L'Habitategeenprehistorique:Actes de la Tableronde internationale(BCHSuppl.19),
P.DarcqueandR.Treuil,eds.,Paris, BCH Suppl.19,pp.347-357.
Wright,J.,J. Cherry, J. Davis,E. Mantzourani,andS. Sutton.1990."The NemeaValleyArchaeological Project: A Preliminary Report," Hesperia59, pp.579-659.
JeannetteMarchand UNIVERSITYOF CALIFORNIA AT BERKELEY GRADUATE GROUPIN ANCIENTHISTORY AND MEDITERRANEAN ARCHAEOLOGY 7303 DWINELLEHALL(2600) BERKELEY, CALIFORNIA 94720-2600 Kleonai@aol. com
MARCHAND
HESPERIA
71 (2002)
Pages I49-199
THE BIOG
CU
RAL
LTU
RAPHY
CYCLADIC
OF
A
GEOMETRIC
AMPHORA ISLANDERS PREHISTORY
IN ATHENS
AND
THE
OF METICS
ABSTRACT This article presents the life history of a large, repairedEarly Iron Age amphora imported to Athens, fragments of which were discoveredin 1939 in and aroundthe Hephaisteion. The context of the vessel suggests that it was used in a tomb. Decorated in an "archaizing"style reminiscent of Protogeometric, the amphoracan be dated to the Late Geometric period. It finds its closest parallelson Syros, an island hitherto little known for its post-Early Cycladicantiquities.How the amphoramade its way to Athens is addressed, and differenttypesof evidencepoint to the existenceof residentaliens(metics) in a period before the reformsof Solon and Kleisthenes. Among the surprises that lay beneath the building that has come to be known as the Hephaisteion (Fig. 1), few were more curious than a huge amphora, Agora P 14819 (Fig. 2), assembled during the 1939 excavation season.l Vincent Desborough saw the fragments at the time, and some 1. The circumstancesleading to the appearanceof this article,publisheda decade afterthe death of Evelyn Smithson, requirea word of explanation.This study grew out of a conversationbetween Evelyn and myself in the spring of 1988, at which time the amphora, Agora P 14819, became the focus of discussion.Over the next few years, Evelyn spent quite a bit of time thinking about P 14819. In her typical fashion, she dissected the originaltrench notebooks of the Agora and scouredevery possible context lot in orderto reconstructthe context of the vessel. Moreover,she acquaintedherselfwith the Early Iron Age Cycladesand, despite her modest assertionthat she was only an "Atticman"(see Papadopoulos1994a, p. 564), she delved into the quagmireof scientific analysisand particularlythe
problemsof distinguishingthe fabrics of variousCycladicislands.After her untimely death in 1992, I came across an unfinished manuscripthidden in one of the manyboxes of her notes. Entitled "OAMDOPEYTTOYANAEIAA"-Anaxilas'samphora-the paperwas in two sections, amounting to some eleven pages of double-spaced text. In the same folderwere many handwrittennotes, primarilyon the phenomenon of an archaizingstyle in later Geometric, and severalphotocopies of photographsof pots from Athens as illustrations.There were, in addition,many handwrittennotes on Naxos and on Cycladicpottery in general,as well as the transcriptionof the funeraryinscriptionof Anaxilasof Naxos-Athenian Kerameikos,inv.I. 388 (see below).The manuscriptitself
compriseda brief introduction,a descriptionof the pot, and a section on "Excavationand Context."The manuscript,little more than an introductory draft,was among the last projectsEvelyn worked on before her death. Enough survivedin the manuscriptand notes to indicate the basic structureof a paper. In the presentarticle,much of the introductionand section on excavation and context, as well as the basic description of the vase, is Evelyn's,with my addendaand corrigenda,often in those sections markedby Evelyn as requiring furtherwork or elaboration.The Cycladic complexitiesof the vase follow some of Evelyn'snotes; so too the discussionof the pot as a burialurn for an infant, and some musings on how it may have made its way to Athens. This said, the manuscriptwas very much
150
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Figure1. The Hephaisteion,east front,May 1936. Courtesy Agora Excavations
years later suggested, by implication at least, that the pot was an import of East Greek or island origin, and Subprotogeometric in date.2As an "isolated find" of uncertain date and without close parallel, it attracted little further attention, and its ungainly size, delicate equilibrium, and heavy plastering relegated it eventually to the virtual oblivion of the back, south storeroom of the Stoa of Attalos, wired to the top shelf for safety'ssake. A chance conversation between the authors in the spring of 1988 recalled the vessel to mind and, following a precarious viewing, it was brought down from its lofty position. Although one of the largest Early Iron Age vessels from the area of the later Athenian Agora, P 14819 received little attention. A problematic context (see below) meant that it could be dated only on the basis of style and, because it was not clearly recognized as Late Geometric, it was not included in AgoraVIII. While the pot and its context do have some bearing on the use of the area before there was ever an agora nearby,particularly with regard to the history of early burials on Kolonos Agoraios, the vessel is later than any intact grave on the plateau or its western slopes. Moreover, in the framework of acceleratedscholarly activity and new discoveries in island and East Greek Geometric over the past few decades, Agora P 14819 is a "problempiece" that should be more widely known to scholars. For these reasons, it seemed desirable to present it at this time and in this form, not least in the hope of finding a more secure provenience and date for it. The piece appears to be Cycladic-rather than East Greek-in origin, perhaps from the island of Syros, and its date 8th century B.C., probunfinished. haveto assume I, therefore, of foranyshortcomings responsibility the finalresultand,indeed,responsibility forthe decisionto publishthispaper in its currentform.Evelynwasa perfectionist,andI hopethe resultdoesnot falltoo farbelowherhighstandards.
2. Desborough1952,p. 34:"Curiof a similar ouslyenough,fragments amphora[likethosefrom8th-century contextsonThera]of coarseredclay, verymicaceous,appearamongthe unpublishedmaterialof theAgoraExcathere vations(P 14819);unfortunately,
He goeson to compare is no context." thevesselto anunpublished amphora on the fromMelosthathassemicircles shoulderandcircleson the belly(Desborough1952,p. 34, n. 2), witha furthermentionon p. 36.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I5I
b
a Figure2. a) Late Geometricbellyhandledamphora,AthenianAgora P 14819;b) detailof handle,as preserved. CourtesyAgoraExcavations
3. At the time of her death Evelyn inclined towardthe view that the amphorawas Naxian on the basis of comparandaavailableat the time; thus the originaltitle of her paper,"Anaxilas'sAmphora,"was appropriatefor an amphorabelievedto come from Naxos, the homelandof the later metic Anaxilas. Now that the Naxian provenience seems less likely,the old title slightly misses the mark.Although I have changed the title, the basic thrustof the originaltitle is as immediate and directas when Evelyn penned it.
ably no earlier than the third quarter,according to the conventional chronology.3As such, the vessel may serve as a point of departurefor any discussion of imported pottery to Early Iron Age Athens.4 It also provides a clearerfocus on the phenomenon of an Early Iron Age "archaizing"style, that is, one alluding to the Protogeometric, but produced in the Late Geometric period. In turning to the particular history of this large pot-its cultural biography5-alternative scenarios are discussed, including the movement of commodities and people. Beyond this, the vessel provides a potential glimpse of the prehistory of a social phenomenon that was to define historic Athens. Various threads of evidence converge to suggest the possibility of resident aliens-metoikoi as they came to be known in historic Athens-not only in the period before Kleisthenes, but arguably generations before the reforms of Solon, in an era before recordedhistory.6 4. It is fairlyclearthat ceramic imports to EarlyIron Age Athens, especiallyduringProtogeometric,but also duringEarly and Middle Geometric, were exceedinglyrare-comparable to the import of the proverbial"owlsto Athens."This situationcontrastswith that at any numberof Aegean sites, such as Lefkandi and Knossos,to mention only two. There, pottery imports,although not abundantin comparisonto the local material,were neverthelesscommon occurrences(see esp. LefkandiI, pp. 347-354, for Proto-
geometric and Geometric imports;Coldstreamand Catling 1996, pp. 393-409; also Coldstream1990). 5. There is a growing literatureon this subject,much of which was either unknown to Smithson, or appearedafter her death. Especiallyimportantare the variouspapersin Appadurai1986, esp. Kopytoff 1986, p. 86; cf. Davis 1997; Gosden and Marshall1999, esp. pp. 169178. 6. For overviewsof metoikoiin Athens, see Clerc 1893; Gauthier 1972, esp. ch. 3; Whitehead 1977; 1984.
I52
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Figure3. Interiorof the cellaof the Hephaisteionfromthe east,showing late burialsunderthe floor. CourtesyAgora Excavations
EXCAVATION
AND CONTEXT
After spending two years clearing the rest of Kolonos Agoraios to bedrock,7the Agora excavatorsreturned to the so-called Theseion in February of 1939, this time to remove the earth filling and Christian burials from within that building (Fig. 3) in order to study its foundations, construction filling, and any remains that might have survived of an earlier sanctuaryon the site. Work began within the cella on 20 February,under the supervision of Dorothy Burr Thompson, with the collaboration of Homer A. Thompson.8 On the very first day "Protogeometric"material was observed in the disturbed levels immediately beneath the tile flooring of the modern "museum"(Fig. 4).9 Two days later more "Protogeometric!" was recorded, including "considerable pieces of a Protogeometric amphora."10All of these "Protogeometric"fragments were from Agora P 14819, and in the course of the next month exploration to bedrock led to the recovery of 165 pieces of the pot, roughly a third of it, the profile complete. So remarkablewas this piece and distinctive its fabric that not a single preserved sherd escaped the excavators'diligence; in 1988 reexamination of the sherd lots from the temple and immediate surroundings yielded no additional fragments. 7. Shear 1936, pp. 14-16,23-24; 1937, pp. 342-352. Limited cleaning was done in both sections KK and IIO in 1937, the latterreferringto Plateia Theseion. 8. The resultswere incorporated into William Bell Dinsmoor'sthorough (1941) study on the Hephaisteion, with specialacknowledgment
to theThompsonson pp.3-4. 9. Forthe so-calledTheseionas the "CentralArchaeologicalMuseum," 1834-1874, see Karouzou1968, pp. ix-xi. The Hephaisteionwas decreed"CentralPublic Museum for Antiquities"on November 13, 1834. The buildingwas first used as an exhibition space,and later,from 1835 until
at least1934,forantiquities storage; see Papageorgiou-Venetas1994, p. 314; also p. 115, fig. 145. See furtherKavvadias 1890-1892, pp. 12-15; Kokkou 1974, p. 104, n. 2; Karo1934, col. 146, for the "Sammlungim Theseion." 10. Referencesin quotationmarks arefrom the excavationnotebooks.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I53
Figure4. Interiorof the Hephaisteion, showingthe sculpturecollection of the then NationalMuseum of Athens.Albumenprint,Pascal Sebah,ca. 1870.The Getty Research Institute,acc.no. 92.R.84 (04.13.04). Institute CourtesyGettyResearch
Some of the fragments of P 14819, all but one of them joining, were found in each of the post-antique layerswithin the cella (Fig. 3), "in general near the south door,"but at least two lay in the loose upper filling of Christian grave 45.11A single sherd clearly derives from the temple construction fill (layerIV) that remainedover bedrockbeneath Christian grave 34.12It is likely that some sherds came, as well, from a pocket of heavily burned red earth (layer VI) over bedrock in the West Peristyle beneath Christian grave 4.13No fragment of P 14819 is known to have derived from the thin layer of gray earth (layerVII) preservedin four small patches over bedrock within the cella.
11. Dinsmoor 1941, p. 5, fig. 1. 12. See below. Dorothy Thompson had suggestedthat this might have been the provenanceof the entirepot. In fact, no cutting preservedwithin the temple is of suitablesize or configurationfor it. 13. This layerhas been relatedto possible Persiandestructionin the area;see Dinsmoor 1941, pp. 126-127. This provenancewould seem the most likely explanationof certainanomalies: in 1939 section KK, lot 476, was packagedfor permanentstoragein a large sturdymanilaenvelope,still intact and firmlyclaspedin 1988. The envelopewas labeled (DBT/HAT [= Dorothy BurrThompson and Homer A. Thompson]) "WestPeri-
style, Under GraveXLI [= grave4 in Dinsmoor 1941]. Protogeometric-VI cent. B.C.,"and so enteredby HAT in the section lot-list; it was the only lot of 46 examinedin 1988 that lacked a field ticket. It containedonly seven small sherds,none conceivablyProtogeometric,two of them black-glaze, one from the handle of a cup,Type C. It seems impossible that pieces describedby the Thompsons as "Protogeometric"would not be readily identifiable,for both had had, by 1939, extensiveexperiencewith EarlyIron Age wells and graves.Tiny lots, such as this, were regularlystoredin smaller envelopes,of two standardsizes. One can only surmisethat the "Protogeometric"materialthey noted had
been removed,and that it was more pieces of P 14819. At least two other lots listed as containing"Protogeometric"now have nothing in them that could have been so described. One may presume,then, that in a final effort to recoverall sherdsof P 14819 before missing portionswere filled out in plasterto stabilizethe mended vessel as preserved(1939), a conservatorhad reviewedall KK lots listed as containing"Protogeometric,"and that additional pieces were in fact found in some of them. Such an interventionmight also accountfor the accidentalloss of the field ticket, normallysecurely fastened.
I54
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
The seven strata within the temple were divided by the Thompsons into three groups; within each the range of sherds in the larger lots is essentially the same: Post-antique(layersI-III): these levels contained pottery that was predominantly Byzantine and Turkish, along with a few Hellenistic and Roman pieces, and a scatteringof Protogeometricthrough Classical sherds. The Christian burialswere either dug down into these layers or the layers had formed over or were placed to cover the tombs. The burials themselves were divided by Dinsmoor into Medieval and Protestant graves.14 Of the former, the great majority of tombs belong to the period between A.D. 1057 and 1453.15Of the latter, the earliest is probably that of John Tweddell, who died on July 25, 1799, while the use of the area as a burial ground probably came to an end immediately after the outbreak of the War of Independence.16 Templeconstruction filling (layers IV-V): these layers contained pre6thand dominantly early-5th-century material, with only a handful of than later 480 B.c. More than half of the lots contained a pieces certainly few pieces of Protogeometric and Early Geometric; Middle Geometric sherdswere rare,Late Geometric more numerous.This filling was dumped within the rectangle of the peristyle as the foundations rose. Foundation trenches for cella walls, porches, and internal colonnades were cut through this filling, mostly to, or deeply into, bedrock; the poros underpinning for the temple pavement cut through it to rest on bedrock or on the firm burned crust of the few surviving deposits that predate the temple.17 Depositspredatingthe temple(layersVI-VII): the red earth of layer VI, often preserving a thick burned crust, contained a fair number of burned sherds, some heavily affected by the fire.The softer gray earth of layer VII contained no clearly burned sherds. Apart from these physical differences, which are significant, the material recovered from layers VI and VII was very similar.The majority of fragments were of the 6th century B.C., with nothing clearly later than ca. 480 B.C.The early component is substantial in nearly all the lots, and about a third of one lot was certainly Protogeometric. Two lots from layer VI from the west end contained a handful of Late Helladic IIIC sherds,18and one along the north peristyle preserved a fine piece of early Mycenaean.l9The LH IIIC material is not surprising in the context of nearby graves of that date,20but this sample is now the earliest pottery recovered from the plateau or the northwest slopes of Kolonos Agoraios. All phases of Geometric are represented, a 14. Dinsmoor 1941, pp. 6-30. 15. Dinsmoor 1941, p. 15. 16. Dinsmoor 1941, pp. 16-17,30. After the War of Independencethe "Theseion"would have been a very unlikelylocation for a burialground becauseof the importanceit had assumedat that time in the creationof the moderncity of Athens. This is perhapsmost clearlyillustratedin Peter von Hess's painting entitledArrival of King Ottoof Greecein AthensonJanuary 12, 1835, now in the Neue Pinakothek
in Munich (see Koch 1955, p. 161, fig. 2; see also p. 163, figs. 7-8; Bastea 2000, pp. 6-7, fig. 1). As a monument, the building itself, along with Plateia Theseiou, came to occupy a prominent place both physicallyand symbolicallyin the newly createdcapitalof the nascent Greek state (Athens became the capital of Greece in 1834, following earlier provisionalseats of governmentat Aigina and Nauplia;see Bastea2000, pp. 6-14). 17. Dinsmoor 1941, pp. 30-31, 65, 126-127.
18. KK lot 466, South Peristyle under grave 16, by the southwest anta; KK lot 479, northwestcornerPeristyle. 19. KK lot 474, North Peristyle, undergrave29. The fragmentis of a verticalstraphandle (W. 0.034 m), well burnished,fired light reddish brown to light red 2.5YR 6/4-6; the piece is LH III A:1 at latest. 20. EspeciallygravesD 6:3 and D 7:1, both of which can be assigned to Final Mycenaean/Submycenaean.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I55
fair quantity of Late Geometric and Protoattic, as well as all stages of black-figure. Many pieces are ambitious and some are of high quality. None of the early material deriving from these deposits is, then, in its original context, and the layers antedating the temple representfilling that had been spread to serve a construction or landscaping project of the 5th century B.C. The source, or sources, of this material is arguable,but some of it came from disturbed graves or cemetery refuse around them. While earth fill elsewhere has been shown to have sometimes been hauled in from a distance,21rarelyis the distance substantial, and in this case a ready source may have lain just to the west and northwest where bedrock falls away, in places abruptly, from the temple area. Moreover, a number of graves were preserved along the west and northwest shoulder of the hill, and dozens of cuttings for what were certainly others.22Surviving burials range in date from the closing stages of LH IIIC into Middle Geometric I,23 but scattered cemetery debris over bedrock around them leaves little doubt that burial in the area continued at least into the late 8th and probably through the 7th century B.C. The latter is suggested by a fragment of a terracottaplastic snake that had surely peeled from a Late Geometric or Early Protoattic funeraryvase.24 Sherds in the temple filling are mostly small, joins few, and even reconstituted fragments small; the same is generally true of the presumed cemetery debris, reinforcing the impression that this is indeed the source of most of the sherds beneath the Hephaisteion. The sherds assembled as P 14819 are an exception. So much of the vessel is preserved, all but one small sherdjoining, that it cannot have been moved far,and its survivalon Kolonos Agoraios is virtually unimaginable outside an original context of a grave deeply sunk into bedrock.25A likely sequence of events, therefore, is as follows: 1. The pot was probably uprooted by builders of the Hephaisteion.26If pieces of it actually derive from layer VI, as seems probable (see above), the grave must have been destroyed at the beginning of the 5th century B.C.The vessel may have 21. A good but ratherrareexample of fill having been carriedsome distanceis that of the dumped road fill of the mid-7th centuryB.C. beside the Tholos Cemetery.The deposit contained sherdsthat joined fragments in the ProtoatticVotive deposit on the Areiopagos,about 100 m to the southeast:deposit H 17:4, Young 1939, p. 10. Another case is that of sherds joining pyre debrisin graveH 16:6, which were found in 4th-century B.C. dumped filling above that deposit, 15 m to the south;see Smithson 1968, pp. 78-79. In both cases the distanceis not substantial. 22. The preservedgravesand numerousgravecuttingswill be fully
presentedin a forthcomingvolume in the AthenianAgoraseries. 23. The latest tomb, C 8:7 (Middle Geometric I), is located about 20 m southwest of the temple. 24. The fragmentderivesfrom lot KK 461 (layerIV), under Christian gravesin the cella.The incidence of such plastic snakes appearsto originatein the Late Geometric IIb workshopof the PhiladelphiaPainter (see Coldstream 1968, p. 57) and continues through the careerof the Analatos Painter,that is, a chronologicalspan of ca. 725-675 B.C. in the conventionalchronology.The work of the Analatos Painter,including hydriaiwith snake-moldedhandles, is fully discussedby Sheedy 1992.
25. While a well is also a possible source,the earliestpre-Archaicwell on Kolonos Agoraios is D 12:3 (Middle Geometric II/Late Geometric Ia), nearly 100 m to the southeast,at the edge of the easternscarp;see Brann 1961, pp. 103114, well I; for the date see Coldstream 1968, p. 22. It is also worth adding that pots found in wells, particularlythose substantiallypreserved(i.e., period-ofuse), are normallysmaller,and rarely exceed a height of 0.40 m. 26. An earlieruprootingof the vessel, though possible, seems unlikely,especially since the surfacebreaksappearsharp and relativelyunworn;reinforcingplaster now coating the entire interiorof the vessel preventscloser inspection.
I56
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
been originally in one of the robbed burial pits just outside the foundations of the west peristyle (see below).27 2. The amphora was deliberatelybroken up into sherds and tossed into or onto the early-Sth-century B.C. fill. 3. The sherds were dispersed through the Hephaisteion construction filling, either entering as a part of the pre-temple fill dumped within the area of the rectangle or churned up from levels disturbed by the Hephaisteion builders. 4. Christian burials and repairsto the church, to the extent that they penetrated and displaced construction filling or pre-temple levels, drew some pieces up into post-antique or modern levels. If the sequence of events presented here is sound, then the relevant contexts in and immediately around the building help to pinpoint the precise date of the construction of the Hephaisteion. They supplement the "new evidence for the dating of the temple" preparedby Lucy Talcott and presented in Dinsmoor's study,28primarily based on the material from the deposit of working chips of Pentelic marble some 33 m to the southwest of the temple, and suggest a Kimonian date for the building. Moreover, the location of Christian burials within the cella (Fig. 3) throws some doubt on the restoration,and even the very existence of the interior colonnades.29 As for the immediate surroundingsof the building, it is worth adding that the bedrock in the area of the Hephaisteion was dug into for numerous Early Iron Age graves. A few of these, such as the "Final Mycenaean/ Submycenaean"tomb D 7:1 (Fig. 5), were dug deep into the natural rock and, being well sealed in some cases, escaped the attention of later builders on the hill.30It was very common, however, to find on Kolonos Agoraios large areasof bedrock with numerous destroyed,looted, or damaged burial pits with a configuration identical to the better-preserved tombs nearby.31 Figure 6 illustrates only a few of these pits, and it is highly likely that P 14819 derives from such a damaged tomb. Assuming the pot to have come from a tomb, as seems most likely, one may well ask of what kind and whose. Before addressing these questions, it is important to describe the vessel more fully and to establish, as far as is possible, its date, since chronology has a direct bearing on the question of preferredburial custom in Athens at any given time in the Early Iron Age. 27. As, for example,one on either side of the third peristylecolumn from the north;gravesD 7:1, D 7:6, and D 7:7 lay within a distanceof 4.0 m. Other pits and an intact grave,D 7:8, dotted the slopes furtherto the west and southwest. 28. Dinsmoor 1941, pp. 128-150. 29. The lack of an interiorcolonnade for the Hephaisteion and the fact that the statuebase that was restoredin the Hephaisteionwas moved to Athens
from Sounion formed part of an unpublished paperpresentedby Frederick Cooper (1985) at the American School of ClassicalStudies on March 27, 1985. I am gratefulto Fred Cooper for sending me a copy of his manuscript. 30. For the later activityaroundthe Hephaisteion,including the establishment of a landscapedgarden,see Thompson 1937. 31. For full discussionand details, see Papadopoulos,forthcoming.
CULTURAL
Figure5. FinalMycenaean/Submycenaeantomb D 7:1,with cover slabpreservedin situ, dug deep into bedrock.Foundationsof the Hephaisteionon the northside in background.CourtesyAgoraExcavations
Figure6. Emptypits in bedrockin sectorIlE (PlateiaTheseiou),in the areaof the Hephaisteion(Kolonos Agoraios),representinglooted or destroyedEarlyIronAge tombs. CourtesyAgoraExcavations
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I57
I58
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
I
J
@
T
l
Ill
aa~~~~~~~~s
:
Figure 7. Agora P 14819. Scaleca.1:7. DrawingA. Hooton
THE AMPHORA AND ITS REPAIR Although never fiullypublished, P 14819 has been briefly mentioned in more than one publication.32It may be described as follows: P 14819: Largebelly-handledamphora Figs. 2, 7 from Imported,Cycladic,perhaps Syros H. 0.730-0.740 m; Diam. (base)0.155 m; Diam. (mouth) 0.203 m Reconstructedfrom manyjoining fragmentspreservinga substantial part of the amphora,including a portion of the base, much of the body, neck, rim, and partsof both handles.About a third of the vessel preserved, the profile complete;missing partsrestoredin plaster,with interiorheavily coated in plasterto stabilizevessel. Surfacevery worn in parts,particularly on upperbody. Tall ring foot; undersideflat.Tall ovoid body,with point of maximum diameterset high. Verticalneck, offset from body on interiorby a substantial thickening,indicatingthat the neck was made separatelyfrom the body and subsequentlyattachedor that the neck was thrown onto the leatherhardbody by attachinga coil to the upperedge of a finished body.The cylindricalneck terminatesin a plain rim, with the top edge smoothed to form a roundedlip. The mouth is not originalto the vessel but has been repaired(see below).Two horizontaldouble handles are set on the body at the point of greatestdiameter.Wheelmarksprominenton neck interior.
32. Dinsmoor1941,p. 126; Desborough1952,pp.34 and36; Oakeshott1966,p. 121,under"Island Protogeometric,"wrongly stated as
P 14189";Smithson1968,p. 85, "Agora underno. 1; Snodgrass1971,p. 101, n. 43; Papadopoulos1998, p. 115, n. 37.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I59
Non-Athenian clay,full of inclusions,predominantlywhite (limestone or quartz?),only a few darkerones, with many individualparticles eruptingonto the surface.Fabricvery micaceous,with mica visible all over surfaceand on clean breaks.Clay body fired close to reddishyellow 5YR 6/6 and light red 2.5YR 6/6, best describedas "brickred."Reserved surfaceon neck interiorand upperpart of the vessel on exteriorfired as the clay body,in places closer to light reddishbrown 5YR 6/4; reserved surfaceson lower partsof vessel fired closer to reddishbrown5YR 5/4, in parts slightly lighter,especiallyon belly zone, closer to light reddish brown 5YR 6/4, in partsgray.Upper half of vessel largelyoxidized,while lower part is reduced. Paint mostly well adheringon lower wall, in partsmuch worn and peeled, especiallyon upperbody.Variouslyfired red throughblack,in accordancewith the fired color of the reservedsurface.Lower exterior edge of foot reserved;remainderof foot, to juncturewith wall, painted. Lower wall reserved,except for a single horizontalband, quite thick. Three broadbands near and slightly below point of maximumdiameter, immediatelybelow the handles, define the lower belly zone. The belly zone is decorated,on either side of the vessel, with sets of mechanically drawnconcentriccircles,each set comprisingnine circles,with a very small dot at center.Alternatingbetween the circlesare two sets of short double squiggles,one set hanging pendant from the band above,the other lower down. On the better-preservedside there arefive sets of circles;the belly zone on the opposite side of the vase is largelylost. Each set is drawn with the same pivoted multiplebrush;individualbrushesof the implement were very thin and the sets of circlesare rathersmall for the size of the vessel.The left set on the better-preservedside is the only one that shows "flooding."33 Judging by the fine pivot point, the circles were executedwhile the clay was ratherdry,probablyleatherto bone hard. Shoulderdecoratedwith a broadhorizontalband framedby two thinnerbands below and two above.From the uppermostof these springupright,mechanicallydrawn,concentricsemicircles.The pivot point for each, mostly not preservedon accountof wear or damage,is set slightly above the lower band,with the resultthat the semicirclesare not perfecthalf circlesof 180?but ratherdefine a slightly greaterarc.They are executedwith the same implement used for the circleson the belly zone, with each set of semicirclescomprisingnine arcs.Only a few sets, or portions of sets, arepreservedon the shoulder,but these would have been continuous aroundthe shoulder.Judgingby the partsof three sets preserved above one side, the semicircleswere set quite far apart,and there is space aroundthe shoulderfor at least seven sets. Thin horizontalband on uppermostshoulder,nearjuncturewith neck. From this band hang pendant sets of double squigglesidenticalto those on the belly zone. These squiggles,judging by what is preserved, on 33.This is a commonoccurrence arefor the most part located in the areabetween the sets of semicircles, setsof concentriccirclesandsemicircles and thereforealternatewith them, but at a higher level. It is possible that paintedwitha pivotedmultiplebrush, there is a second set of double squigglesbelow,between each set of semiforwhichseePapadopoulos, Vedder, andSchreiber1998. circles,but in all cases the relevantareaof the vase is either not preserved
i6o
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
or extremelyworn.The reservedareabetween the semicirclesand the thin horizontalband aboveis ratherlarge and this may be why the semicircles aregreaterthan 180?.Preservedneck, though much worn, is clearlypainted solid;there is no traceof paint on the preservedrim top. Outer faces of handles painted,as is the connecting rib between each double handle, as shown.The paint extends from the lower outer handle attachmentsin a broad sweep, over the three bands below and well down the wall of the vessel;each ends in a raggedtermination.Preservedneck reservedon interior. As noted above, the neck of the vessel now ends in a plain lip that is not original.The fact that the mouth is only partiallypreservedand rather worn does not assist in establishing whether the repairwas executed before or after firing. Nevertheless, a number of factors combine to suggest that the repairwas done after firing, and possibly some time well after the vessel reached Athens. The top surfaceof the rim is uneven and dips down on one side about 0.005 m, with the result that the line of the rim crosses those of the wheelmarks. The trough of a gouge, approximately0.005 m wide, runs diagonally for a distance of 0.040-0.045 m across an area of intact paint on the neck; it terminates about 0.01 m below the top in a rounded "nose,"as if made by a fine chisel that slipped. About 0.015 m to the right a chipped notch interrupts the surface,the bottom of which appears to be what remains of a possible mending-hole that shattered the inner surfaceof the wall and flaked the edges outside. If the chipped notch is indeed part of a mending-hole, as seems likely, then it is clear that the adjustments to the vessel occurred after firing. An attempt to clamp or tie the severed neck and rim seems to have splintered the neck beyond conventional repair;the line of the fracturewas evened up as much as possible, as the irregularitiesat and near the mouth show, and the resulting edge was sanded smooth to form a lip. Salvage repairs,beyond the usual rivet- or tie-holes, are known, especially on large vessels, though they are also known on smaller pots. The ring foot of a very large neck-handled amphora from the Kerameikos,for example, was chipped off and the bottom hollowed, evidently after painting, so that the vessel would stand firmly.34It is not clear from the published description of the vase whether this was done before or after firing, though it seems likely that the repair was executed when the clay was rather hard, perhaps after firing. A classic case of a salvage repair made prior to firing can be seen on a Middle Geometric I vessel from well L 6:2 in the areaof the later Athenian Agora, which was originally designed as a large hydria, but subsequently trimmed down to the base of the shoulder to form a krater.35 Whatever the life history of the amphoraP 14819-a history to which we shall return in the final section of this paper-it is clear that the vessel was prized enough to warrant a repair,which, having failed, necessitated the more drastic step of cutting off the upper part of the neck and all of the rim, and sanding down what remained to form a plain round mouth. The current state of the amphora,with its upper neck and rim gone, gives the appearance of a proportionately dysfunctional vessel, one that does not look quite right in any period, a feature that does not assist the dating of the vessel.
34. Kerameikos I, p. 115, pl. 41, grave mound T 8, inv. 568. 35. The vessel is fully publishedin
1998,alongwitha Papadopoulos similarvessel,alsoof Atticmanufacture,fromthe EarlyIronAge cemetery at Fortetsa,nearKnossos,forwhich see Brock 1957, pp. 43 and 47, pl. 31, no. 454 [13]. A numberof other repairs to smallervases from the Agora are also listed in Papadopoulos1998.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I6I
THE DATE OF THE AMPHORA AND THE PHENOMENON OF ARCHAIZING In her publicationof the burialurn of the celebratedTomb of the Rich Athenian Lady (tomb H 16:6-1 [P 27629]), Smithsonhad occasionto mention Agora P 14819. In discussingthe solidly glazed handles on P 27629, she noted that they were uniquein Athens, adding:"Theyare fromsomeexactlylikethoseon AgoraP 14819,animportedbelly-handler wherein the Cyclades;it mightbe thislate,thoughit is stillProtogeometric in style."36 Althoughin laternotes Smithsoncameto considerP 14819 as Late Geometric,she had alwaysessentiallyviewedthe vesselas "Subprotogeometric"in date,despiteits "Protogeometric-looking" style. Stylistically,a numberof featurescombineto suggestthat P 14819, tradition,is notof Protogeometric althoughfirmlyrootedin Protogeometric date.Perhapsforemost,the mechanicallydrawncirclesandsemicirclesare exceptionallyfine and theirpaintthin. As a result,the circularornament appearsflimsy and is overshadowedby the rathersturdysquigglesand comparativelymassivebands.While the inspirationof the piece is clear, both in compositionand execution,it lacksthe balanceand solid texture of genuine Protogeometricdecoration.For Smithson,this suggestedan adaptation,whetherby meansof survival,revival,or both, detachedfrom an earlierliving traditionand not fiullyunderstood.An analogycan be madewith well-trainedand skilledmodernclassicists:howeverdeft their controloverthe syntaxandgrammarof Greekor Latin,theirpronunciation of the languagesremainsa contentiousissue,stillpracticedbut neverthelessseveredfroma oncethrivingtradition.A few pointsunderscorethe relationshipanddifferencesbetweenP 14819andpiecesof Protogeometric date,especiallyin Athens. belly-handled Protogeometric AgoraP 14819is largerthananysurviving a both and Late Protogeometric,approach amphora,although few, Early when it in size.37Suchlargepots,with a heightabove0.50 m, particularly used as epitymbia,areespeciallyvulnerableto breakage.Fragments,relativelycommonin cemeterydebrisfromFinalMycenaeaninto Late Geometric,show that they werepartof the standardceramicrepertoire,even though the normalsize belly-handledamphoras(H. ca. 0.40 m) become comparativelyrare after the close of Protogeometric.The large versionswerenot chancecuriositiesbutwereproducedroutinelyto meet the ritualneedsand statusrequirementsof wealthierfamilies.The decorative formulasforverylargeProtogeometric belly-handledamphorasdifferfrom thoseof standardsize andarefairlyrigid:circlesacrossthe bellyandsemicircleson the shoulder.38 36. Smithson 1968, p. 85, under no. 1; cf. the solidly painted handles on P 27629, pl. 20 (bottom). 37. An Early Protogeometricexample is Erechtheion Street H-15, Brouskare1980, pl. 4:a (H. 0.64 m); for a Late Protogeometricexamplesee KerameikosIV, pl. 9, grave38, inv. 1089 (H. 0.69 m). 38. There are,as always,a few excep-
tions to the rule.At least one very large belly-handledamphoralacks the circles on the belly zone: Athens, National Museum inv. 18113, from Nea Ionia (H. 0.572 m), Smithson 1961, pl. 24, no. 3. The belly zone of the vase is left clear and the shoulderheight is diminished by extending the paint from the neck and with the additionof three horizontalbands. Somewhat earlier,
two belly-handledamphorasof standard size have circleson the belly zone like their largercounterparts;both are Early Protogeometric,decoratedat a time before conventionswere firmlyfixed:Kerameikosinv. 8808/09 (= Schlorb-Vierneisel 1966, Beil. 11, 4-5, Gr. hS 101; H. 0.446 m); and Agora P 24240 from well J 14:2, with an estimatedheight a little over 0.40 m (unpublished).
I62
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Since the same pivoted multiple brush was regularlyused for both full circles and semicircles,39 the height of the shoulder zone on genuine Protogeometric vases was reduced to proportions commensurate with the height of the semicircles. Two schemes are usual: several bands between the neck and the patterned zone,40or two, rarelythree, superimposed patterned zones; the latter, popular in Final Mycenaean and Early Protogeometric, is abandoned before Late Protogeometric. On P 14819, the single narrow band, closely set beneath the neck, does nothing to reduce the zone height. The semicircles fill less than half of the height and the squiggles dangle high above them. The semicirclesand circlesarevery small for the size of the pot, smaller, in fact, than is usual on Protogeometric belly-handled amphoras of standard size anywhere in Greece.41The heavy shoulder-banding, comprising a wide band framed by two thin bands below and two above, is usual on amphoras found in graves, literally Protogeometric in form, placement, and execution; the lower bands are not. A deep belly zone with an emphatic row of full mechanically drawn concentric circles is one of the hallmarks of Protogeometric grave-amphoras. On true Protogeometric amphoras this belly zone is defined, below the handle attachments, by a repeated scheme of a wide band framed by two thinner bands above and below, or by three narrowbands. In either case, the narrowbands are equal in width and areusually drawnwith the same brush as those on the shoulder above, resulting in a balanced effect. On P 14819 the lower bands are placed comparatively too high, and although they run a little below the base of the outer handle attachments,the attachments themselves areshort and quite unlike the tapered attachments carefully smoothed against the wall of genuine Protogeometric pieces. Furthermore,the handles on P 14819 are too small for this pot. The central rib of the double handle rests squarely on the top band below, a cramped and inelegant juxtaposition that negates the decorative effect of the double handle.42The firm, even mighty, bands below the handle, three times the width of the narrowones on the shoulder,overpower the dainty circles of the belly zone. The circles, placed more or less at the middle of the belly zone, are approximately0.06 m above the point of greatest diameter of the vessel; although it may seem a minor point, such a position is virtually unthinkable on genuine Athenian Protogeometric. If the semicircles on the shoulder have too much room, the circles on the belly have too little. Added to this is the fact that the entire patterned decoration has been squeezed up to the top half of the pot. The solitaryband on the lower 39. Contraryto the argumentsof Eiteljorg(1980), it is clearthat the decorationof Attic Protogeometric and Geometricvessels in the form of concentriccirclesand semicircleswas done with a pivoted multiplebrush implement;see Papadopoulos,Vedder, and Schreiber1998. 40. See Kerameikos I, pl. 55, inv. 561, with a wide band enclosed by a narrow
band above and below, and a pendant zigzag canopyabovethe semicircles; see also ErechtheionStreet,H-15 (Brouskare1980, p. 23, pl. 4:a),where the paint on the neck extendsin a broad band on the shoulder,below which are two narrowbands and a zigzag canopy. Both vessels are EarlyProtogeometric. Cf. furtherKerameikosinv. 1089, which is laterProtogeometric.
41. In additionto Athenian amphorasalreadymentioned, cf. a similardecorativeidiom on Protogeometric amphoras,and other closed vessels,from Boiotia, the Corinthia, the Cyclades,Thessaly,East Greece, Crete, and even Macedonia;see, generally,Desborough 1952; 1972. 42. For such handles generallysee Oakeshott 1966.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
i63
body does little to counterbalance the top-heaviness of the decoration, and although a subsidiary band (or bands) on the lower bodies of very large Protogeometric closed vessels is not uncommon, the one on P 14819 teams up with the heavy triple bands below the handles to detract from the patterned decoration. The manner in which the outer face of the preserved handle is painted contrasts with the normal thinner "archesand bows" of Protogeometric vessels, especially belly-handled amphoras, but its prolongation in tails extending onto the body, below the level of the bands, is an echo of Protogeometric practice.43 A comparison of the size, shape, and decoration of P 14819 with any number of genuine Protogeometric belly-handled amphorasindicates that the vessel cannot be Protogeometric in date.44The distinctive featuresseen in P 14819, particularly the tall ring foot, the ovoid body with point of maximum diameter set quite high, the comparatively broad neck, along with the great size of the vessel, are all consistent with a Late Geometric date. Such a combination of features can be seen in a number of regional workshops. In Athenian Geometric these features are standardin a variety of vessels, including very large amphoras, such as the neck-handled amphora measuring 1.350 m, now in Leiden.45In the Cyclades large bellyhandled amphoras shaped like P 14819, with the characteristic double handles, referred to by Noel Oakeshott as "horned-head,"46are found as early as Middle Geometric. A good example is the amphora said to be from Melos, now in Munich, which is almost certainly Naxian (Fig. 8).47 Standing to a height of 0.730 m, the Munich amphora is of similar proportions to Agora P 14819 (0.730-0.740 m, though originally larger),but is decorated in the standardpatterned style of the time, with much of the lower body covered in black.48A virtually identical amphora, also said to be from Melos and now in Vienna, was published by Elena Walter-Karydi (Fig. 9),49 who went on to list and discuss a number of related vessels and fragments, especially from Naxos.50The stylistic similarity between these and other closely related vessels and contemporary pottery from Athens has been most recently discussed by Photeini Zapheiropoulou, Kenneth 43. Cf. Desborough 1952, pls. 4-5; Papadopoulos1994b, pls. 109:a-b, 119:c. 44. For Attic Protogeometricbellyhandled amphorassee, among others, Kerameikos I, pls. 32 (inv.529), 43-45, 46 (inv.857), 54-56,58; Kerameikos IV, pls. 9-11; Desborough 1952, pp. 20-37, pls. 4-5; Desborough 1972, pp. 35-36, figs. 2-3, pp. 148-149, pl. 27. For amphorasfrom other partsof the Greekworld, see Desborough 1952; 1972. ComparingCycladicProtogeometric amphorasis more difficult, given the rarityof true Protogeometric-especially earlierand developed Protogeometric-pottery in the Cyclades.The existing evidence is
admirablyassembledand discussed by Desborough (1952, pp. 153-163; 1972, pp. 221-224); see also Coldstream1968, pp. 148-157. In its latest phase,Athenian Protogeometricwas widely imitated in variouspartsof the Aegean, particularlyin the Cyclades. 45. Leiden 1.1909/1.1, conveniently illustratedand discussedin Coldstream 1968, p. 55, pl. 11:a. 46. Oakeshott 1966. 47. Munich inv. 6166, Coldstream 1968, p. 167, pl. 34:m; Zapheiropoulou 1983, p. 131, fig. 27; Boardman1998, pp. 47,59, fig. 88. 48. Cf. the relatedfragmentary amphorawith a restoredheight estimated at 0.810 m, DelosXV, pp. 90-91,
to as"Attiques," but pl.XLII:1,referred listed under CycladicMiddle Geometric in Coldstream1968, p. 166. 49. Walter-Karydi1972, p. 390, fig. 4; Zapheiropoulou1983, p. 131, fig. 28; Vienna, Kunsthistorisches Museum inv. 1879. 50. Walter-Karydi1972, esp. pp. 386-390. Foremostof these is the fragmentaryamphorasaid to be from Thera, now in the Louvre (inv.A 266), Walter-Karydi1972, pp. 388,390, n. 12, fig. 5, and a similarfragmentary but smalleramphorafrom Philoti on Naxos (p. 388, fig. 2), first published by Kontoleon (1949, pp. 1-3, figs. 1-3), now fully discussedin Kourou1999a, pp. 183-198, pls. 56-59.
I64
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Sheedy, and Nota Kourou.51The number of recognized Athenian Protogeometric and Geometric imports to the Cycladeshas been steadilygrowing, thanks to the work of Richard Catling and Sheedy.52Large Cycladic amphoras have also been found outside the Cyclades and especially in the various cemeteries at Knossos.53 The Munich and Vienna amphoras shown here (Figs. 8-9) are both assigned to the Middle Geometric period. Similarly large amphoras are common in the Cyclades in Late Geometric, as illustrated by the fragmentary example from Delos (Fig. 10), which stands to a preservedheight of 0.740 m.54Its decoration follows the more standardpatterning of Late Geometric. It is worth adding that earlier Cycladic amphoras of the Protogeometric period are never this large. By the Early Archaic period, largebelly-handled amphorasin the Cycladesbecame proportionatelytaller and more slender,as numerous examples from Delos attest.55On the basis of shape alone, Agora P 14819 can be reasonably assigned to the Middle and Late Geometric period. 51. Zapheiropoulou1983, esp. pp. 130-133; Sheedy 1990; Kourou 1999a, esp. pp. 90-97, 185-187. 52. Catling 1998a, esp. pp. 370-378; Sheedy 1990. 53. Coldstream1990, pp. 26-27, pl. 6, nos. 1 and 3. Of the two amphoras, ColdstreamassignsTekke Q.63 to
Early Geometric I and KMF 283.91 to
MiddleGeometricI. In additionto the imports from the North Cemetery at Knossos,Coldstreamdiscussescomparativematerialfrom the Fortetsaand Tekke cemeteries. 54. DelosXV, p. 37, no. 2, pl. XVIII: a-b, groupAc (B 4.213); neck and rim,
Figure8 (left).NaxianMiddle Geometricbelly-handledamphora, Glyptothekund Museenantiker Kleinkunst,Munich,inv.6166. H. 0.730 m. CourtesyMuseum;photo C. Koppermann
Figure9 (right).Cycladic(Naxian) Middle Geometricbelly-handled amphora,Antikensammlung,KunsthistorischesMuseum,Vienna, inv. 1879. H. 0.695 m. Courtesy Museum(neg.11179)
as well as base, not preserved.For large belly-handledamphorasfrom other islands,see, among many,Coldstream 1968, pl. 39:h ("Melian,"now in Leiden; H. 0.620 m), pl. 58:a (Rhodes;H. 0.560 m, discussedin more detail below). 55. See DelosXVII, pls. I-VI (groupBa).
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
i65
Figure10. FragmentaryCycladic Late Geometricbelly-handled amphorafromDelos, groupAc, no. 2. Pres.H. 0.740 m. CourtesyEcolefranfaised'Athenes
56. Pfuhl 1903, p. 170, Beil. XXII:2 (E 4), from grave29 (72) 6. 57. I took notes on Thera 825 while visiting the island in Septemberof 1997. Reconstructedfrom fragments, with missing partsrestoredin plaster, Thera 825 has no visible breaksfrom which one could betterjudge the clay body of the vessel and I was unable to take a Munsell readingof the surface. The paint is dull, and variesin color from red throughdifferentshadesof reddishbrownto black.
Among the many large amphoras found in the Cyclades, perhaps the closest parallel to Agora P 14819, especially in terms of the painted decoration, is the belly-handled amphora from Thera first published by Ernst Pfuhl almost a century ago (Fig. 11).56Standing to a height of 0.815 m, which would be not too far from the original height of P 14819, the amphora from Thera (inv. 825) is identical in virtuallyall respects to P 14819. The most obvious exception is that it has a thick band framed by thinner bands immediately below its belly zone instead of the three bands of similar thickness found on P 14819. Even the squiggles on P 14819 are echoed in the groups of short vertical strokes found between the full circles on the belly of the amphora from Thera. If Agora P 14819 is restored with an upper neck and rim like that of Thera 825, the two vessels would be very similar to one another, except that the base of P 14819 would be proportionately slightly more narrow,resulting in a lower body fractionallymore piriform, and therefore closer to the shape of amphoras like that from Delos illustrated in Figure 10. Despite these strong similarities, the fabric of the two vessels is not the same. The clay of Thera 825, judging by its reserved surfaces, has prominent dark inclusions and only a light dusting of mica; it is lightcolored, in places reddish yellow, elsewhere approaching a light red with an almost maroon-purple tinge.57In his publication of Thera 825, Pfuhl listed the amphora under his category E "Samisches und Verwandtes," and more specifically under category E II, that is, related to Samian, but not considered Samian. Some 50 years later,Desborough listed Thera 825
i66
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
b
Figure11. a) Late GeometricbellyhandledamphorafromThera,grave 29 (72) 6. H. 0.815 m. b) Detail of a
handle area. a) CourtesyDeutsches Institut,Athens; Archaologisches b) photoauthor
among a number of amphoras from Thera and assumed it to be of local manufacture.58Coldstream, however, like Pfuhl, listed Thera 825 as an East Greek import to the island, and stated: The survivalof the belly-handled amphora, still decorated in the PG tradition, is attested by an example found in the burnt layer at Miletus, and by two others exported to Thera [including inv. 825], both found in LG contexts.59 As for the date, there is no question that the context of Thera 825grave 29 (72) 6-and of a related amphora from grave 89 (109) 3 is Late Geometric.60This, together with the Cycladic Middle and Late Geometric vessels discussed above, helps place Agora P 14819 broadly within the Late Geometric period. Any more precise date would be sheer guesswork, though a date in the third quarterof the 8th century B.C., according to the conventional chronology, seems more likely (on the basis of the overallappearance of the vessel) than one toward the very end of the century. In considering such a late date for Thera 825, Desborough stated that "it comes, therefore, as rather a shock to find that the context of this amphora is undoubtedly eighth century.... The only conclusion seems to be that it is a remarkablecase of the survivalof an Attic Protogeometric form,
58. Desborough 1952, pp. 31, 34, 215. 59. Coldstream1968, p. 269, with referenceto Hommel 1959-1960, p. 39, fig. 1; cf. also the largefragmentary belly-handledamphora,p. 54, pl. 53, no. 4, referredto as "(proto?)geometrischen." 60. Pfuhl 1903, p. 170; Desborough 1952, pp. 34,215; Coldstream1968, p. 269.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
i67
Figure12. Subgeometrichydria, AthenianAgoraP 4614. CourtesyAgoraExcavations
61. Desborough 1952, p. 34. 62. Desborough 1952, p. 34. 63. See, for example,Smithson 1961, p. 166, under no. 43; AgoraVIII, pp. 34-35, underno. 37. 64. Smithson 1961, p. 166, under no. 43. 65. AgoraVIII, p. 19; cf. Brann 1961, p. 125, under L 66: six Mycenaean goblet stems found in the fill of a Late Geometricwell, and perhaps (re)usedas bobbins. 66. Young 1939, p. 27, under no. V 1. Younggoes on to cite as close parallelsnos. X 1 (see below,n. 67) and C 148 (p. 186, fig. 137), as well as a relatedexamplefrom the 7th-century grave10 at Phaleron(Pelekides1917, p. 31, fig. 20). Among many other relatedhydriaisee Charitonides1975, pl. 1O:Cr(FM45, tomb XVI).
at least in Thera."61In this context Desborough made passing reference to Agora P 14819: "Curiously enough, fragments of a similar amphora of coarse red-brick clay, very micaceous, appear among the unpublished material of the Agora excavations (P 14819)."62 Vessels like Agora P 14819 andThera 825 raise the issue of archaizing. A number of scholars looking at the problem from an Athenian perspective have discussed the incidence of a Protogeometric style in later Geometric contexts in terms of survivalsand/or revivals,and even as copies.63 In discussing the phenomenon of specific Protogeometric shapes echoing Mycenaean tradition, Smithson noted that this "maybe, like a number of resurgent types, a Mycenaean revival,i.e. a 'copy,'slightly modernized, of pieces salvaged from disturbed tombs."64A similar sentiment is expressed by Eva Brann in her discussion of Late Geometric and Early Protoattic shapes and motifs that are strikingly Mycenaean in their appearance.She states:"The Athenians certainlyoften came acrossMycenaean antiquities, so there would have been no lack of models."65In discussing an early-7thcentury B.C.hydria from the Agora (Fig. 12), together with severalrelated vessels that echo Protogeometric tradition, Rodney Young wrote: The hydria gives a false impression that it is Protogeometric; but the pale clay, creamy slip, and general appearanceof the fabric are entirely different from true Protogeometric. The body ... is seen to be deeper and more pointed, narrowerat the bottom, and with a higher, less flaring ring foot. The decoration too is not usual for Protogeometric; the multiplication of glaze bands on the shoulder, the solid glaze on the handles instead of the canonical glaze bands ending in "tails,"and the single wavy line in the handle zone instead of the double or triple wavy bands always used in true Protogeometric, are evidence that our hydria belongs to a different fabric. The hydria too is comparativelyrare in Protogeometric.66
I68
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Young noted that a related Subgeometric hydria from grave X in the Agora (Fig. 13) was also very Protogeometric in appearance,but that its fabric is unlike any true Athenian Protogeometric vase.67In her discussion of the same hydria, together with a number of related examples, Brann noted that it was not clear whether the banded hydriai were "survivalsor revivals of their Protogeometric prototype."68Elsewhere, Brann referred to old-fashioned hydriai, of light-ground ware, considered "aMycenaean type revived in the 8th century and found on many Greek sites, especially eastern ones."69As for Protogeometric "models,"Brann considered it likely that pottery of this period was still to be found as "heirlooms,"and concluded: "At any rate, there is a whole group of Late Geometric pots done in a style which seems like a return to Protogeometric clay-ground technique."70 The dark-groundtechnique began to take over from the light-ground during the latest stages of the Protogeometric period, and is well established by Early Geometric I, continuing through the end of Middle Geometric. Although this is the case for many shapes, it is important to stress that for other specific vessel forms there is continuity in both style and production technique. This is especially true for pottery traditionallyconsidered as "utility ware,"though many such pots are found in exactly the same contexts as more elaborately decorated ("non-utility")vessels and were often used for the same function. A good example is the plain banded neck-handled amphora from the fill of well K 1:5, found together with several vessels decorated in the standard dark-ground technique of the period (Fig. 14).71Indeed, such plain vessels stand out in comparison to their more elaboratelydecorated contemporaries from the same context. Whereas many of the larger and smallervessels of the Athenian Early Iron Age repertoire develop rapidly in terms of the fashion of the day,
Figure13. Subgeometrichydria, AthenianAgoraP 4980, two views. CourtesyAgora Excavations
67. Young 1939, pp. 42-43, fig. 27. 68. AgoraVIII, p. 34; Brannalso notes two relatedhydriaifrom the Kerameikosdating to the first half of the 8th centuryB.c.;Kerameikos V.1, pl. 50.
69. Brann 1961, p. 100. 70. Brann 1961, p. 100. 71. The three pots illustratedin Fig. 14 were all found in the period-ofuse fill of this well, recentlyexcavated in the areanorth of the Eridanos, which can be assignedto Early Geometric II/Middle GeometricI; see Camp 1999, pp. 266-267.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
Figure 14. Athenian Agora, selected Early Geometric II/Middle Geometric I vessels from the period-ofuse fill of well K 1:5, in the area north of the Eridanos. CourtesyAgoraExcavations
i69
AMPHORA
f
other shapes remain remarkablyconservativeand consistent. Some of these display little noticeable change over a period of several decades or even centuries, while others develop in terms of nuances of shape, but retain a more traditional decorative idiom. Examples include a number of Geometric banded hydriai that date to various stages of the Early Iron Age.72 Similarly,plain banded neck-handled amphoras,ranging in date from Protogeometric through Late Geometric, are commonly found discarded in Early Iron Age wells in the area of the Classical Athenian Agora.73In the case of such pots, there was never any real break in tradition: rather than survival or revival there was continuity. To illustrate this point, I have assembled a representativesample of plain banded neck-handled amphoras in Figure 15, togetherwith a Protogeometric neck-handled amphoradecorated with the canonical mechanically drawn concentric semicircles on the shoulder (Fig. 15:a). Such banded vessels, found throughout the Early Iron Age, not only were very common, but they also kept alive a traditional light-ground technique of decorationthroughout the Earlyand Middle Geometric periods, when a dark-groundtechnique predominated, and into the Late Geometric period, characterizedby elaborate patterning. The banded amphoras assembled in Figure 15 range in date from Developed Protogeometric through Late Geometric I. The two Protogeometric amphoras (Fig. 15:a-b) were found in the fill of well A 20:5, assigned to a developed phase of the Protogeometric period;74two vessels (Fig. 15:c-d) were assigned by Coldstream respectivelyto Early Geometric II and Middle Geometric I;75the two examples from well L 6:2 (Fig. 15:e-f) areMiddle Geometric;76and the two from well I 13:1 (Fig. 15:g-h) can be assigned to a transitional phase between Middle Geometric II and Late Geometric I.77The same type of simple banded amphorawas used 72.Thesearefullydiscussedin 1998. Papadopoulos 73.Thesewillbe publishedin full in forthcoming volumesof the Athenian Agora.
forthcoming. Papadopoulos, in 75.Theywerefound,respectively, well C 18:6(EG II) andwellB 18:9 of (MG I);forthe relativechronology thesetwodeposits,see Coldstream
21;Papadopoulos, forthcoming. 77.WellI 13:1is oneof several Geometricwellsin the areaunderthe laterMiddleStoa;forthe morerecently Geometricdepositswithin discovered
74. For well A 20:5 see Young 1951b, p. 144; with full details in
1968, pp. 13, 16. 76. See Coldstream1968, pp. 16,
this area,see Camp 1999, pp. 260-263 (well I 13:4 and grave1 13:5).
I70
K. PAPADOPOULOS
JOHN
AND
EVELYN
L. SMITHSON
b
a
d
g
c
f
e
h
Figure15. AthenianAgora,selected bandedneck-handledamphoras: a-b) P 17455,P 17456 (wellA 20:5, PG); c) P 19012 (wellC 18:6,EG II); d) P 19037 (wellB 18:9,MG I); e-f) P 6411, P 6410 (wellL 6:2, MG); g-h) P 27938, P 27939 (wellI 13:1,MG II-LG I). Scale ca. 1:6. CourtesyAgora Excavations
CULTURAL
Figure16. AthenianAgora,banded neck-handledamphorasfrom EarlyIronAge tombs:a) P 6997 (tombB 10:1,LPG);b) P 24791 (tombN 16:4,EG I). Scale1:6.
OF A GEOMETRIC
BIOGRAPHY
AMPHORA
I7I
/
a
DrawingA. Hooton
b
not only for drawingwater from wells, but also could be used in tombs, as the Late Protogeometric and Early Geometric I amphoras illustrated in Figure 16 attest.78Related amphoras,often smallerthan their predecessors, continue into the late 8th century and throughout the 7th century B.C. (Fig. 17).79It is worth adding that in the Archaic period, as in the Geometric, pots decorated with bands were produced almost everywhere in the Greek world.80 Precisely the same range of shapes-hydriai, neck- and belly-handled amphoras-with decoration harking back to a Mycenaean and Protogeometric aesthetic, is common in the Cyclades during the Late Geometric period, as numerous examples from the Purification Trench on Rheneia and elsewhere attest.8 The pots assembled in Figures 12-17 highlight the continuity of a conservative tradition, which was widespread throughout the Greek world. Such conservatism was perhaps most acute in the Cyclades, not only during Late Geometric, but for later periods as well. Indeed, a series of cups in Geometric and even Protogeometric style, with metope panels and concentric circles, are common in the Cyclades, especially on Paros, Naxos, and Delos, some as late as ca. 500 B.C.82 Island isolation, to use a cliche, may be one of the contributing causes. This is well put by Coldstream,who penned, with specific regardto Thera, a statement that may be applied to other islands: "The potters of this remote island were slow to learn a Late Geometric style, and slow to forget it."83 Be that as it may,a similar phenomenon, albeit at different scales, is found on the mainland, including centers, like Athens, that have long been 78.The LateProtogeometric amphoraP 6997 comesfromthe B 10:1,andthe disturbed"tomb" Early Geometric I amphoraP 24791
wasfoundin tombN 16:4.See also othersimilarplainbandedamphoras fromthe tombsin Kerameikos I, pl. 42, T 12-13,inv.602-603;Kerameikos IV,
pl. 6, grave28, inv.910. 79.The vesselsshownin Fig. 17 are fullydiscussedinAgoraVIII,p. 34, pl. 3,
Coldstream 1968,pp.164generally, 195.SeefurtherYoung1939,pp.2728;Sheedy1985,pp.153-159,esp.
nos. 29,31-36.
p. 156, n. 18.
80. See CookandDupont1998, p. 132. 81.DelosXV,pls.II-XI,XV,no.27; cf. TheraII, p. 229, fig.427;and,more
82. Morris1997,pp.68-69, with fig.4. 83. Coldstream 1977,p. 216.
JOHN
I72
K. PAPADOPOULOS
AND EVELYN
L. SMITHSON
11. U II. * II
a
c
b
Ir
g
consideredas "trend-setters."84 In addition to the Cyclades,Protogeometric survivals in the Geometric period are well known in Thessaly, Boiotia, Skyros, Euboia, and elsewhere, as Coldstream has established,85and the same is true for much of the Peloponnese and western Greece generally,as well as Macedonia and other northern regions.86
THE ORIGIN
e
Figure17. AthenianAgora,banded amphorasof the late 8th through 7th centuriesB.c.:a) P 21578 (secondhalfof the 8th century); b-e) P 12444, P 12445, P 20731, P 26242 (late 8th century); f) P 23465 (thirdquarterof the 7th century);g) P 3469 (lastquarter of the 7th century).Courtesy Agora
p
f
d
OF THE AMPHORA
Although Agora P 14819 was never previously published in detail, those who have examined it have come to accept it as an import to Athens of Cycladic or East Greek origin.87Indeed, the uncertainty about the precise origin of the vessel is to a large measure the result of the close affinity between the two regions (Fig. 18), especially between individual centers such as Naxos and Rhodes in the Early Iron Age, a relationship effectively explored by Zapheiropoulou.88In the case of both regions, assigning individual pieces, or even stylistic groups, to a particularisland workshop is far from straightforward.89 Although much East Greek pottery (including the finer wares) in the Geometric and Early Archaic period is relatively coarse, I have not come across an East Greek vessel with a fabric sufficiently similar to that of
Excavations
84. For a long-lived Subgeometric style of pottery in Athens see, most recently,Papadopoulos,forthcoming, and in the Cyclades,Morris 1997, p. 68. 85. Coldstream1968, pp. 148-157, 196-211. 86. For the Protogeometricand Subprotogeometricpotteryof Troy, especiallythe neck-handledamphoras, with discussionof comparativematerial from other north Aegean sites, see Catling 1998b; also Lenz, Ruppenstein, Baumann,and Catling 1998. 87. See above,n. 32. 88. Zapheiropoulou1994. 89. See esp.Jones 1986, pp. 643673. Note also the fabricdescriptionsof some of the regionalvarietiesof chevronskyphoigiven in Descoeudres and Kearsley1983.
CULTURAL
AMPHORA
OF A GEOMETRIC
BIOGRAPHY
I73
MACEDONIA
,I A 0 Sardis.
,I
KARIA
0
4-I V4. CRETAN
SEA
RHODES
Knossos
CRETE
0
o
100 StatuteMiles 100 Kilometers
Figure18. Map of the Aegean showingthe Cycladesin relation to the Greekmainland,the Dodecanese,and East Greece generally. R. Finnerty
Agora P 14819 to provide a compelling visual match.90The distinctive fabrics of Chios and Klazomenai can be ruled out,91as can "Rhodian,"92 along with the distinctive sandy texture and light brown color of"Ionian" cups, many of which are local to Samos.93The generally coarse and gritty fabric of the "Wild Goat" and "Fikellura"styles, much of which can now be assigned to Miletos thanks to the work of Pierre Dupont,94 is a little
90. For a useful overviewof East Greek pottery,see Cook and Dupont 1998; for the Geometric period, Coldstream1968, pp. 262-301, remains fundamental.For the fabricof various East Greek wares,particularlyof the Archaicperiod, see TocraI, pp. 41-66; Jones 1986, pp. 660-673. I hasten to stressthat the following fabricdeterminations are made on the basis of visual criteria;the problemsinherent with the scientific determinationsof the clays of the Cycladesand East Greece, including Ionia and the Dodecanese, arewell laid out in Jones 1986. 91. For Chian Late Geometric and
Archaicpottery,Boardman1967 is fundamental;see also Boardman1998, pp. 144-146; and furtherTocraI, pp. 57-63; TocraII, pp. 24-28. For Klazomeniansee Cook and Dupont 1998, pp. 95-107, 121-128 (with full references);Boardman1998, pp. 148149. 92. By "Rhodian"I essentiallymean the "Bird-KotyleWorkshop"as defined by Coldstream1968, pp. 277-279; see also Boardman1998, pp. 51-52; Tocra I, pp. 41-57; TocraII, pp. 16-20. This distinctivefabricis characteristically brown-varying from darkorangeto coffee brown accordingto Coldstream 1968, p. 279-and containspredomi-
nantlywhite impurities,but only a relativelysmall quantityof silverymica. 93. Cf. Jones 1986, p. 665; Boardman1998, pp. 146-147. For potteryfrom Samos, see further Technau 1929, esp. pp. 6-37; Eilmann 1933. 94. Dupont 1986. See furtherCook and Dupont 1998, pp. 32-70, 77-91; Boardman1998, pp. 147-148. Although the home of Fikellurais Miletos, the Wild Goat Style is produced at variouscenters,including Miletos. The earlierstudies of the Wild Goat Style by Kardara(1963) and Schiering(1957) arestill important, especiallyfor shape and style.
i74
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
closer to the general appearanceof the fabric of P 14819, particularlyin the fact that the fired color of the clay can vary considerablyfrom a sandy brown through pink and red. Nevertheless, the color of the clay and range of impurities, especially the mica content, of the fabric of P 14819 are different from those of any pottery that can be assigned with confidence to Ionia. There are a number of fairly large belly-handled amphoras, especially three examples from Kameiros and Ialysos on Rhodes that are assigned by Coldstream on stylistic grounds to Early Geometric,95of which one is illustrated here (Fig. 19). None of these Rhodian amphoras, however, has the same shape and decoration as P 14819; they are all smaller than P 14819 and their decoration closer to true Protogeometric, despite their Subprotogeometricdate. By East Greek Middle Geometric, the bellyhandled amphora is exceedingly rare,and indeed one of the few examples is Thera 825 (Fig. 11), which Coldstream assigned to East Greece (see above). We have seen, however, that the fabric of Agora P 14819 is very different from that of Thera 825, as is that of the Rhodian amphoras assembled by Coldstream. As for the Cyclades (Fig. 20), a number of islands can also be quickruled out. The fabric of P 14819 is not the same as that which can ly be confidently assigned to Paros, Melos, Thera, or Siphnos.96The excavations directed by Lila Marangou at Minoa on Amorgos have brought to light an impressive arrayof Geometric, Archaic, and later material,but there is nothing from that island that I have seen that is similarto P 14819.97 The pottery found, to date, on Tenos includes a number of vessels made of a clay not unlike that of P 14819, but there is nothing that I know from that island that provides a satisfactorymatch.98The Geometric settlement site at Zagora on the island of Andros has yielded a wealth of Geometric pottery,which awaitsproperpublication.99The materialfrom Zagora high95. Coldstream1968, pp. 265266, pl. 58:a (Ialysos);Jacopi 1933, pp. 119-120, figs. 133-134; pp. 204205, figs. 244-245; two of these amphorasare also illustratedand discussedin Zapheiropoulou1994, p. 248, fig. 17, p. 250, fig. 19. 96. For the fabricof Parianand Melian, see Jones 1986, pp. 643-660; see also Coldstream1968, esp. pp. 176185; Zapheiropoulou1985; Sheedy 1985, the latterimportantin helping to clarifythe confusionin the literature between potteryvariouslyassignedas "Melian"or "Parian,"much of which was found in the PurificationTrench on Rheneia.Note also the pottery from Paroikia(Rubensohn1917, esp. pp. 7388) and the Delion publishedin Rubensohn 1962, pp. 83-129. For the highly distinctiveclay ofThera, which contains no mica, see Coldstream1968, pp. 185-189; 1977, pp. 216-217, and
discussionabove;see also Kontoleon 1958, esp. pp. 127-137. For Siphnian Geometric and Archaicpottery,including importedfabrics,see Brock and MackworthYoung 1949, pp. 3353; TocraI, pp. 73-78; TocraII, pp. 3438;Jones 1986, p. 644. For useful notes on the ceramicfabricfrom Melos, Kimolos,Paros,Mykonos, Naxos, Siphnos, as well as that fromThasos, see Villard 1993; Gautier 1993. I have not come acrosssufficientGeometric and EarlyArchaicpotteryfrom the northwestCyclades(i.e., Kea, Kythnos, and Seriphos)to comment on the productsof these islands;for a recent overviewof Kea and Kythnos,see the variouspapersin Mendoni and MazarakisAinian 1998. For a generaloverview of Cycladicpottery,the seminal studyby Dugas (1925), though out of date, is still useful in a numberof importantrespects,as is Buschor 1929.
97. Annual preliminaryreports appearin Ergonand Prakt;most recently,see Marangou1996; 1997; 1998 (which providesa useful historical introductionto the site, with references to earlierreports);and Marangou,in prep.For the publicationof two Protogeometricvases from Amorgos, see Catling andJones 1989. 98. ForTenian Geometric,see the overviewin Coldstream1968, p. 166 (with references);also Levi 1925-1926; Desborough 1952, pp. 158-161. I am gratefulto Nota Kouroufor discussing Tenian Geometric potterywith me in the Tenos Museum, and for showing me her recentfinds from Exombourgo. For the excavationsat Exombourgo (also -cEjlxtooPyo),see Kourou1996; 1999b. 99. In the meantime,see Zagora1; Zagora2; also Cambitoglou,Peirce, Segal, and Papadopoulos1981.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I75
Figure 19. Early Geometric belly-handled amphora, Rhodes, inv. 15533, from Ialysos, grave 43 (H. 0.560 m). CourtesyKBEphoreia of Prehistoricand ClassicalAntiquities, R hodes Rhodes
?%.^ ..^ *;-<^ ;:;-:A,,",,a'ii^ *:*:,...*.,....:;: ...*;.:: i:;;ii7i.... fst.''
lights one of the fundamental problems facing the study of Cycladic Geometric pottery, namely the incidence and great range of imported fabrics. While the comparativelynumerous Athenian, Euboian, Naxian, and Parian products can be distinguished with relative ease, there remains an abundance of fabrics both of local manufacture and imported from other islands and the mainland. Among this wealth of Geometric material, a local Andrian fabric can be distinguished, especially for a series of smaller vessels such as cups, which share a brown, micaceous fabric.100This appears to be similar to much of the Late Geometric and Archaic pottery from the site of Ypsili (T?YXiq,also known as Aprovatou) on the west coast of Andros, north of Zagora, recently excavated by Christina Televantou.101The fabric that I refer to as Andrian is, however, not the same as that of Agora P 14819. Naxos, the largest of the Cycladic islands, is known to have produced a significant and well-defined Geometric style, and several Middle Geometric Naxian amphoras have alreadybeen discussed for the similarity of their shape to that of P 14819 (Figs. 8-9).102The attempt to define with clarity the fabric of Naxos has led to some controversyin the past, particularlywith regardto the Cesnola kraterin NewYork and a numberof closely 100. I had occasion to study much of
thepotteryfromZagoraduringthe of the guidebookforthe site preparation Peirce,Segal,andPapa(Cambitoglou, dopoulos1981).I amespeciallygrateful to Dick Green,who is responsiblefor
the study of the Zagorapottery,as well
I am gratefulto ChristinaTelevantou
see Televantou1993; 1996; 1999.
195.
withme as KenSheedyfordiscussing of the localand the complexities importedpotteryfromZagora. atYpsili, 101.Forthe excavations
forshowingme the findsfromYpsili andfordiscussingwithme a varietyof Cycladicproblems. 102.See Coldstream 1968,pp.164-
I76
K.
JOHN
PAPADOPOULOS
AND
EVELYN
L.
SMITHSON
,YIAROS "" Kardiani
Ktikad
''ros
Galessas. V
(Dellagrazia
i
''
..
RHENEI
Poseidonia'i.
KYTHNOS
.MYKONOS
SYROS
"
"
#
'.7A! e7
NE '
DELOS
SERIPHOPOULO
SERIPHOS DONOUSA
~:NF?
ERIMOMELOSKI Helenika'
POLIVOS SIKINOS
MELOS
s/ANHYDROS
PHOI LEGANDROS
i THERA ;
THERASIA :'"'p
:;Thera
... -
0 0
Akrotir'"
25 StatuteMiles
I Kilo
I
25 Kilometers
C HRISTIANA
,..tANAPHE .
"
'
???
%. I:.-'i...
Figure20. Map of the Cyclades showingsome of the sites mentioned in the text. R. Finnerty
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I77
related pieces.?03Following Nikolaos Kontoleon, Coldstream originally assigned the Cesnola krater to Naxos, but later changed his opinion in favor of Euboia.104This reassignment did not meet with universal favor, and Walter-Karydi reaffirmedthe claim of Naxos.l05Other scholars, primarilyJohn Boardman, maintained a Euboian origin, but noted that "one has to deal almost with a 'koine' of decoration-Boiotia, Euboia, the The Cyclades-which has links which are yet to be properly explored."'06 of the Cesnola krater must be and this is Euboia, origin strongly supported by recent scientific analysis.107 Moreover, our understanding of Naxian Geometric pottery is now on a much more secure basis, thanks to the recent contributions of Zapheiropoulou, Lambrinoudakis,and Kourou.'10 It was, indeed, the characteristic"brick-red,"micaceous fabricof Naxos that led Smithson to suspect that Agora P 14819 was a Naxian import to Athens, and underlay, in part, her designation of the pot as "Anaxilas's amphora."Despite the close affinities, particularlyin shape and, to a certain extent, fabric, between Naxian pottery and P 14819, the latter is unlikely to be Naxian. Its fabric is coarser than normal Naxian fabrics, and none of the pottery published from the island matches the style of P 14819 in all respects. On the nearby small island of Donousa, an important Early Iron Age settlement has been excavated on the west coast by Zapheiropoulou. The definitive report on the excavationshas not yet appeared,but the excavator has provided a number of important summaries.109 Located east of Naxos, Donousa lies about halfway between Euboia and the Dodecanese. The Geometric settlement, dating to the third quarterof the 9th century B.C., was built on a secure anchorage and has yielded material displaying influences from other Cycladic islands, especially Naxos, but also Kos and Rhodes.110Among the pottery recoveredfrom the excavationson Donousa, a number of large Geometric amphoras have been illustrated in preliminary publications and several are on display in the Naxos (Chora) Museum. Two belly-handled amphoras in particularare close to the Rhodian example illustrated above (Fig. 19), and both have double handles similar to those of P 14819. They are decorated with two registers:the respective shoulder registers have mechanically drawn concentric semicircles, while their belly zones have fill concentric circles. One of these, inv. 4804 (Fig. 21),111has crosshatched panels in the center of the belly zone, with much 103. Kontoleon 1949, p. 12, fig. 4; Coldstream1968, pp. 172-174, where a numberof other pieces, including DelosXV, pl. 44, Bc 8, aregrouped aroundthe same "painter." See also discussionin LefkandiI, pp. 74-76. 104. Coldstream1971. 105. Walter-Karydi1972, esp. pp. 402-409. 106. Boardman1969, p. 112; a number of scholarsstill subscribeto this view
(Kourou, personalcommunication). 107. Jones 1986, p. 659. For Euboian EarlyIron Age pottery see Boardman1957; 1969; and, generally, LefkandiI, pp. 27-79 (for Euboian PG-LG potteryfrom the settlement) and pp. 281-354 (for the pottery from the cemetery);Coldstream1977, pp. 192-195. It sufficesto state that Agora P 14819 cannot be Euboian. 108. Zapheiropoulou1983; 1994;
Lambrinoudakis1983a; 1983b; cf. Lambrinoudakis1988; Kourou1984; 1992; 1994; 1997; 1999a. 109. See Zapheiropoulou1970; 1971; 1973a; 1973b; 1975. 110. Zapheiropoulou1994, p. 231. See furtherthe preliminaryreportsin ArchDelt,beginning in 1967, and n. 109 above. 111. Zapheiropoulou1975, pl. 473:y; 1994, p. 249, fig. 18.
I78
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Figure21 (left).EarlyGeometric belly-handledamphora,Donousa, of the lower body painted, whereas the other, inv. 4538 (Fig. 22), has short zigzags, or squiggles, alternatingwith the circles on the belly zone.l12The squiggles are similar but not identical to those on P 14819. The more rounded form of Donousa 4804 and 4538, together with their more elaborate decoration, indicates that the vessels are earlier than Agora P 14819, probably closer to the date of Rhodes 15533. Although the evidence for the date of these vessels has not yet been fully presented, the information provided in preliminaryreports indicates a 9th-century B.C. date for these amphoras, and more specifically the third quarter of that century.l13The fabric of Donousa 4804, as far as I could judge in the light of the Naxos Museum, is paler than that of Agora P 14819, and contains less mica, whereas the clay color of Donousa 4538 varies from a pale brown through red. Parts of the latter more closely resemble the fabric of Agora P 14819, but the match is not perfect; the same is true of other pots from the island.114
112. Zapheiropoulou1970, pl. 401:a; 1994, p. 247, fig. lb. 113. See Zapheiropoulou1994, p. 231. 114. A numberof relatedlarger vessels from Donousa areon displayin the Naxos Museum.These include two largeneck-handledamphoras(inv. 7905 and 4796) decoratedwith bands not unlike some of the Athenian amphorasdepictedin Fig. 15; one of
these (inv. 7905) is of a red fabric,the other with a more pale fired clay.Of similarfabricto the latteris a hydria (inv.4788, see Zapheiropoulou1973a, pl. 367:oc-p)and a shoulder-handled amphora(inv.4543, see Zapheiropoulou 1970, pl. 402:y), the latter essentiallydark-ground,except for the concentricsemicircleson the shoulder.
inv. 4804. CourtesyPh. Zapheiropoulou
Figure22 (right).EarlyGeometric belly-handledamphora,Donousa, inv. 4538. CourtesyPh. Zapheiropoulou
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
4 :.~~:~;rth!. ?C
ii
AMPHORA
o
I79
m [
Figure23. Geometricbelly-handled amphora,probablyfromSyros; Athens,NationalMuseum,inv.53. Scale ca. 1:6. CourtesyMuseum
bric to that of Agora P 14819 is a solitary amphora, now in the National Museum in Athens (inv. 53), which is said to be "probablyfrom Syros" (Fig. 23). Originally published by Sam Wide,"5 the vessel was fully described by Desborough, whose description of the vase is worth citing in full:
115. Wide 1897, p. 245, fig. 16;
CollignonandCouve1902,p. 26, sectionVIII,no. 137,inv.2201 (53), statedasbeingfrom"Syra?" Thiswas one of severalvases,saidto be perhaps fromSyros,includingsomeMycenaean (seepp.26-27, nos.136-138,142143).
116.Desborough1952,pp.32-33; see alsop. 212,whereDesborough reiterates thatit "iscertainlylaterthan Attic It wasthe any Protogeometric." onlyvesselfromSyrosknownto Desborough. Althoughthe fabricof the amphorais described byDesboroughas its fired coloris not brown," "light consistent(seebelow).
The clay is light brown and the paint chocolate-brown to redbrown. The body is slender-ovoid in shape, with the point of greatest diameter set high; the foot is low conical, coming away sharply from the body; the neck is high and slim, rises abruptly and almost vertically from the neck, and widens out at the top into an everted lip of the usual Class I type with a flat rim. Over this rim there are stripes; the neck is painted over; the shoulder has sets of full circles (eight circles to each set), and dividing them, except in one case, groups of three vertical wiggly lines. Below the shoulder, one thin band, one very thick, and one thin; the same system below the belly; on the belly, three sets of full circles between the single-loop handles; these are connected by similar groups of wiggly lines, now set horizontally; the usual paint splashes go over the handles and down the lower body, which is left free of paint. The height is 0.572 m. The context of the vase is unknown; it is obviously a later development of a vase such as [Kerameikos] 1089 from Tomb 38, but it is not so very far from it in shape; the whole scheme of decoration is Protogeometric in spirit."l6
I8o
JOHN
K. PAPADOPOULOS
AND
EVELYN
L.
SMITHSON
DIDYMI
..
STRONGYLO
Figure24. Map of Syrosshowing some of the principalarchaeological sites. R. Finnerty
Although Athens NM 53 is clearly Subprotogeometric, assigning a firm date is far from straightforward.It is similar in both shape and style to the two belly-handled amphoras from Donousa, especially inv. 4538, and its resemblance to these, as well as Rhodes 15533 and a number of other amphoras from Naxos, suggests a 9th-century B.C.date, more likely in the second half of the century.Whatever its precise date, its fabric is virtually identical to that of P 14819, particularlyin the range and quantity of visible impurities and mica content. Its reserved surfaceshave fired colors ranging from red through a dull gray-brown,and the color and range of the dull paint is also similar.Smaller than P 14819, Athens NM 53 will be fullypublishedby Kourouin a forthcomingfascicleof the CorpusVasorum Antiquorum.l17A solitary amphora supposedly from Syros would by itself be inadequate to establish the island as the provenience of Agora P 14819; recently,however, additional material has been discovered at Galessas (see below), which strengthens the attribution. Of all the larger Cycladic islands, Syros (Fig. 24) is perhaps the least known in terms of its Geometric, Archaic, and Classical history, despite the fact that the early prehistory of the island is well known, thanks to the pioneering work of Christos Tsountas.1l8The island appears in Homer briefly.In Odyssey15.403-414, for example,Homer has Eumaios, the swineherd of Ithake, describe his home as lying above Ortygia (Delos); an island good for cattle and sheep, full of vineyards and wheat raising. Eumaios
117. I am gratefulto Nota Kourou for discussingthis amphorawith me. 118. Tsountas 1898; 1899. See also Bosanquet1895-1896; Bossert 1967; Hekman 1994; and esp. Marthari1998 (with furtherreferences).The most recent overviewof the earlyCycladesis by Broodbank(2000).
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I8I
mentions two cities, but these are not named, and this has led to some speculation as to their location.119 Whatever the names and locations of the two cities ruled by Eumaios's father, Ktesios, the site that has to date yielded the most copious evidence for habitation during the Geometric period is Galessas (raXroaa&q) on the west coast of the island.120Antiquities, primarily surface pottery ranging in date from the Geometric through Roman periods, were reported by Antonis Manthos from the hill overlooking the Bay of Galessas, near the small chapel of Aghia Pakou.121Among the most notable of the surface sherds was one inscribed -)ov 0rp(X7o(v)--o(g--, in what is clearly early epichoric.122Systematic excavationsat Galessas, especiallyin the oLx6riEoo KLxiXta,since 1995 have been directed by the currentEphor of Prehistoric and Classical Antiquities of the Cyclades, Mariza Marthari.123These excavations have uncovered a thriving settlement dating at least from the Geometric period through the Classical era and later. A large quantity of pottery has been recovered from the site, much of which is stratified. A local fabric used for the wheelmade and painted Geometric pottery is virtually identical to the fabric of Agora P 14819 and Athens NM 53.124 There are even a number of fragments of large horizontal handles, undoubtedly from belly-handled amphoras or hydriai, of a size, shape, and decoration very similar to those of P 14819.125 In addition to the wheelmade and painted pottery, the site yielded large quantities of locally made Geometric coarsewarepottery, including pithoi, some of which were decorated. The quantity and quality of the imported pottery is not unlike that of Zagora on Andros. Among the 119. See discussionin RE IV, A, 2, 1932, cols. 1789-1794, s.v. Syros (W. Zschietzschmann).The two cities are most often equatedwith the modern Syros (Hermoupolis)on the east coast and Poseidonia-also known by its popularname Dellagrazia (immortalizedby MarkosVamvakaris's song Frankosyriani)-on the southwest coast.There is also the modern town of Phoinikas (oi'vlxas;) about a mile (less than 2 km) from Poseidonia,and sharingthe same harbor;Phoinikas and Poseidoniahave yielded tracesof ancient settlement.For the coinage of Syros,dating from the 3rd to the 1st centuriesB.C. and later,with the ethnic
further,ATL 1, 1939, p. 553, s.v. SoptoL. 120. The name Galessasas an ethnic is known epigraphically;see RE IV, A, 2, s.v. Syros (W. Zschietzschmann) and RE VII, 1, 1910, col. 601, s.v. Galessioi (ot rFaXAooaoL) (L. Burchner).In IG II2, 814a, lines 1819, there is named a certainfIIpaveib
(ET, EYP, EYPI, or STPION, as well as
122. The inscribedsherdis illustratedin Manthos 1979, p. 46, fig. 11. A photographof the inscription was shown to Henry Immerwahr,then Director of the American School of ClassicalStudies at Athens, who read it as:-)on, son of Therapontos. According to my notes taken in the SyrosMuseum, the letterswere inscribedon a fragmentof what
OEQNKABEIPQNSEPIQN),see Head 1911, pp. 491-492; for discussion of the location of the "Kabeirion," see Manthos 1979, pp. 42-45. For the inscriptionsfrom the island, see IG XII, 5, nos. 652-713, with addendaon p. 335, testimonia and notes on pp. xixxii, and map on p. xxxi;IG XII, Supplementum,pp. 117-118. See
note also Iplriv?ubS Z6opto raXlOooCto;; Z6Upto in IG II2,815. See further Fraserand Matthews 1987, p. 387, s.v. nIIlpVEOS (377-359 B.C.).
121. Manthos 1979. The small chapel of'AyLa Ilaxo--from
Ilavcayla
' a&xoUe--isillustratedin Manthos 1979, p. 40, fig. 1.
appearsto be the neck and rim of a Geometric oinochoe. 123. I am gratefulto Mariza Marthari,who was kind enough to allow me to view the materialfrom the excavationsby the Ephoreiaof Antiquities at Galessasin the Syros (Hermoupolis)Museum. 124. It is characterizedby a fired color of the clay that variesfrom red through a pale light brown.It contains numerousinclusionsof variouscolors, with white predominating,and a great deal of mica.The paint can fire differentshadesfrom red through black,with a reddishbrownbeing standard.On some of the smaller vessels, especiallycups, the visible impuritiesare fewer,but the mica content remainsthe same. 125. One of these comes from the Kikiliaplot, trenchVI, pass 10, level 4. Another fragmentfrom the body of an amphora,also from the Geometric level in trenchVI, must have been from a large closed vessel very similarto P 14819.
I82
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
easily identified imports were Athenian, Corinthian, Euboian, and East Greek (including Rhodian) vessels; a variety of island fabrics, including several pieces not easily assigned to any particularworkshop, were also found. The site also yielded smaller quantities of bronze, iron, and bone artifacts,as well as obsidian tools.126Of the Early Iron Age material that I inspected,the majorityis Late Geometric, though a numberof pieces should be earlier,including Middle Geometric. Further details will be provided by the excavator,but what is crucial to stress here is that Galessas is an important, hitherto unknown, Geometric settlement. Moreover, it is a Geometric settlement on an island little known for its post-Early Cycladic antiquities.Located more or less in the geographicalcenter of the Cyclades, and well situated with regardto communication with the Greek mainland, especially Attika, as well as the islands of the archipelago,Syros must have played a more important role in the early history of the Cyclades than is currentlyconceded.According to Homer (Od. 15.415-416), the island once ruled by Eumaios's father was also frequented by Phoenicians: "famous The importance of this passage seafarers,gnawers at other men'sgoods."127 lies in the fact that it suggests the possibility that commodities-including people-from the Cyclades, and Syros in particular,may have been carried by middlemen, such as the Phoenicians. It should therefore come as no surprise that a large amphora found slightly to the northwest of the historic center of Early Iron Age Athens derives from the island of Syros. This said, it is important to stress, as a word of warning, that the identification of Agora P 14819 as being of "Syrian"manufacturemust remain tentative. Future finds in one or other of the less well known islands of the Aegean, or reanalysis of material long known, may well add new evidence to the story of this remarkable amphora.128
ISLANDERS IN EARLY ATHENS: OF METICS PREHISTORY
TOWARD
A
In the foregoingdiscussionI suggestedthatAgoraP 14819camefromthe Cyclades, probablySyros,that its date is Late Geometric, and that it served at one time as a funeraryvase. Against this backdrop,it is tempting to ask, first of all, about the nature of the tomb. The context of the vase, or rather of its fragments as found, suggests that it was disturbed by the builders of the Hephaisteion from its not-so-ultimate resting place in a tomb. This impression is supported by the very shape of the vessel, since the bellyhandled amphora in Athens appearsto be exclusively funeraryor, at least, this is the case with those examples that are complete.129The very large version of the belly-handled amphora, with a height of ca. 0.50 m, while occasionally used as a cineraryurn, especially in Early Protogeometric and again in Early Geometric II and Middle Geometric I, seems more commonly to have stood over the grave as a marker,or else was used in other aspects of funerary ritual. By Late Geometric the shape had disappeared from the Attic repertory,apartfrom the giant "Dipylon amphoras,"which would themselves be gone by Late Geometric II.130The possibility that
126.Forthe incidenceanduseof obsidiantools in EarlyIron Age contexts at Zagora,see Runnels 1988; for the more generaluse of flaked-stone artifactsin Greece in historical contexts,see Runnels 1982. 127. For a criticallook at Homer's Phoenicians,see Winter 1995. 128. Had I not come acrossthe amphorafrom Syrosnow in the National Museum, or had I finished this articleseveralyearsago, I probably would have been content to assign Agora P 14819 to either Naxosfollowing Smithson-or, more likely, Donousa. 129. Domestic and industrial depositspriorto the 8th centuryB.C. in Athens areknown largelyfrom well or pit deposits.The few fragmentsof belly-handledamphorasfrom such contexts come from dumped filling that included,in additionto industrial debrissuch as potters'waste, cemetery refuse,which is plentiful in most parts of the areaof the laterAthenian Agora; see Papadopoulos1996; Papadopoulos, forthcoming. 130. See Coldstream1968, pp. 2990.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I83
P 14819 once served as a funerary markercannot be dismissed. The fact, however, that so much of the amphora was found indicates that its use as a marker is less likely.131 The use of P 14819 as a cinerary urn also seems since the rite of cremation dropped from favor after the close problematic, of Middle Geometric I and was relatively rare in Late Geometric, when metal urns seem to have been preferredby those who could afford them.132 It is unlikely, therefore, that P 14819, though of a once-popular urn type, would have been so used by any Athenian still practicing cremation. After reviewing Cycladic island burial ritual, Smithson concluded a decade ago that P 14819 was similarly unlikely to have been used as a cremation urn by an Aegean islander residing in Athens.133She noted that although the practice of cremation still continued, the normal custom in many of the Cycladic islands, and specifically Naxos, was not to parcel up human remains and personal valuables in an urn in quite the same way as in Athens. She thereforeleaned towardthe belief that it would seem equally improbable that a Cycladic visitor or immigrant to Athens would have turned to a rite unfamiliar in his or her homeland, and to one no longer practiced in Athens. It has to be stressed, however, that Cycladic island funerary ritual is This is often the case even on individual islands, such extremely diverse.134 as Naxos, where the standardinhumation tombs of coastalNaxia (Chora)135 cannot easily be reconciled with the idiosyncratic burial customs at Tsikalario in the interior of the island, which accordwith Macedonian tradition more than central or southern Greek traditions.136Elsewhere in the Cyclades, where cremation was practiced, the rite was rarelyperformed in the same way as in Athens. For example, Theran cremations were usually in family tombs, where severalurns would be placed in a chamber tomb; a similar practicewas followed in other Dorian islands, including Crete and Kimolos.137In the Ionian Cyclades, particularlyin the northern Cycladic islands such as Tenos, Andros, Rheneia, and parts of Naxos, inhumation appearsto have been generally preferred.l38On the islet of Donousa there are two large pyre deposits, one over 7.0 m long, presumablywith multiple cremations, if human bodies were indeed cremated there.139At Paroikia on Paros Zapheiropoulou has recently excavated two remarkablelarge mass tombs, referred to as polyandreia,in which numerous amphorasprimarily,if not exclusively,neck-handled amphoras-each contained the 131. The possibilitythat the amphorastood as a gravemarkermore or less undisturbedon the Kolonos Agoraios until the time of the construction of the Hephaisteion-a period of over 250 years-seems remote. 132. For a useful overviewof Athenian EarlyIron Age burial customs, see Wiesner 1938; Snodgrass 1971, pp. 140-212; Kurtzand Boardman1971, pp. 21-67; Desborough 1972, pp. 268-277; Coldstream1977, passim;Krause1975; Morris 1987.
133. Unpublishednotes. 134. See, for example,Kurtzand Boardman1971, pp. 177-179. 135. See, most recently,Kourou's meticulousreviewof burialcustoms in the South Cemetery of Naxos: Kourou 1999a, pp. 141-182. 136. For the cemeteryat Tsikalario, see Doumas 1965; PapadopoulouZapheiropoulou1968. See also the overviewsin Kurtzand Boardman 1971, p. 179; Snodgrass1971, pp. 156157,195; Coldstream1977, pp. 92, 375.
137. Coldstream1968, p. 186; Kurtz and Boardman1971, pp. 177-178. For Kimolos see Mustakas 1954-1955; Courbin 1954, p. 146, figs. 41-44 for the cemeteryat Limni (Hellenika);see also Coldstream1977, p. 91. 138. See Kurtzand Boardman1971, p. 179; Snodgrass1971, p. 159. For good illustrationsof inhumationcist tombs on Tenos, see Levi 1925-1926. 139. See especiallyColdstream 1977, pp. 91-92; see further Zapheiropoulou1970; 1973a; 1975.
I84
JOHN
K. PAPADOPOULOS
AND EVELYN
L. SMITHSON
cremated remains of an individual.'40Elsewhere at Paroikiaboth inhumation and cremation tombs are found,141and at Melos the plundered cemetery at Trypiti appearsto have contained cremation tombs.142The recent excavations at the site of Minoa on Amorgos by Marangou have uncovered cremation pyres, in addition to a solitary pot inhumation, ironically in a large banded neck-handled amphora.143In the light of the more recent evidence for Cycladic Early Iron Age graves, and in view of our lack of knowledge of burialcustoms in many individualislands,including Syros, it would be hazardous to venture generalizations or to speculate on the basis of negative evidence. Be that as it may,P 14819 does seem an unusual vessel for a cremation urn in the context of Late Geometric Athens and alternativepossibilities areworth exploring, especially the common occurrence of Late Geometric pot-inhumation. 'Ey-ctxtpa 6;g,pot-burial of unburned bodies, first appears in postBronze Age times in Attika in Late Protogeometric. It is limited to infants, as opposed to children,and appearsto be the primaryarchaeologically visible rite attested for them at the time.44 The common cooking amphora, with an average height of about 0.40 m, often fire-stained from presumably domestic use, is the standard container.The few skeletal remains that have been analyzed from such burials are from premature or newborn babies. Although no infant burials have been reported from the earlierphases of Geometric,145Late Geometric examples are numerous.146 A total of 39 pot burials from the Agora, the Kerameikos, and the cemetery on the Acropolis South Slope near the Odeion inspire a number of interesting observations. 47There is diversity in the shape and size of the container,and while newborns or prematurebabies predominate, there are a few older infants and perhaps even a young child or two.148The distinction between "infant"and "young child" is, of course, a gray area, I am 140.Theseawaitpublication.
1974, p. 373. Child gravesthroughthe
gratefulto YannosKourayosfor bring-
EarlyIronAge arediscussedmorefully
attention. 141. See Snodgrass1971, p. 157. 142. Smith 1895-1896, esp. pp. 7071; Coldstream1977, pp. 91,210. 143. Marangou1993, esp. p. 207, fig. 7:a-3, pl. 125:p,for the banded amphora;the burialcustoms of Minoa are discussedin detail in Marangou, in prep.I am gratefulto Professor Marangoufor giving me a copy of her manuscriptpriorto publication. 144. For the distinctionbetween infants and children,which is an importantone, see Papadopoulos2000, p. 111. For an ongoing study of child graves,especiallyin EarlyIron Age Athens, see Haentjens 1999, though some of the assumptionsconcerning the sexing of child graveson the basis of kterismataarequestionable. 145. See the discussionin Smithson
in Haentjens'dissertation(in prep.). 146. See Haentjens 1999. Various interpretationshave been suggestedfor this increase.For drought,famine, disease,and infant mortality,see Camp 1979, pp. 399-401; for alternative interpretations,see Morris 1987; also Papadopoulos1993. 147. This figureis based on a minimumnumberof fairlycertain child graves.The Agora graveswill be discussedmore fully in a forthcoming volume in the AthenianAgoraseries; see also Camp 1979, p. 400, n. 10; the Kerameikostombs include Kerameikos V.1, graves46, 51-53, 64-68, 85, 97, 100. For the cemeteryon the Acropolis South Slope nearthe Odeion, henceforth the "OdeionCemetery,"see Charitonides1975. 148. Of the relevantskeletalremains that have been analyzedfrom the
tombsto my ing theseremarkable
Agora, there is an individualaged sixteen months at death, anotheraged eighteen months, and a third aged three yearsand nine months.The relevant gravesareAgora B 21:10, Areiopagos ValleyCemeterygrave1 (Young 1951a, pp. 82-83: Late Geometric IIb); and G 12:14,Tholos Cemetery graveIX (Young 1939, pp. 36-41: Late Geometric lib), which containeda newborn and the infant or young child.The burialcontainerof the latteris a large pithos, with a height of 0.80 m, but still smallerthan the largestsuch vessel used in an Attic tomb,which is probably that from Odeion CemeterygraveVIII (H. 1.035 m). It should be addedthat a numberof EarlyIron Age pithoi from domestic contexts elsewherein the Greekworld are considerably larger:one fromToronewas, in its originalstate, probablyover 2.0 m high; see Cambitoglouand Papadopoulos 1994, pl. 21, no. 5.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I85
though in the case of Classical Athens it is a distinction that may be reasonablyset at the age of three years.'49Be that as it may, smaller burial containers were broken to admit the body, and the same appearsto be the case for many, if not most, of the larger containers.150A clear preference existed for the latter:22 of the 39 containers are over 0.50 m tall, and 13 of them are over 0.65 m. The most common shape was the neck-handled amphora, accounting for 23 of the 39 containers. Its popularity,in all likelihood, was due to its being one of the common Athenian shapes (apart from the bulky coarse pithos) that tended during the second half of the 8th century to be very large, providing ample room for the body of the deceased and any accompanyingkterismata.Particularlyfavoredin the Odeion Cemetery were large transportjars, the antecedents and close relatives of the SOS amphoras that became common in Subgeometric times.151 Against such a backdrop,Agora P 14819 fits well in the cluster of very large burial containers. Even though its shape is exotic in Athens at this time, its size recommends it. Whatever the shortcomings of its decoration, it was more elaboratelydecorated than the contemporarypithoi used as burial pots, and its deformed mouth (see above) would not have greatly mattered.152Its fragmentary state preserves no clear evidence of intentional breakage for the purpose of burial, but even without such breakage the mouth of the amphora would have been wide enough (0.203 m) to admit the skull and body of an infant and any of the more common burial gifts.153 The evidence for Agora P 14819 presented above is largely circumstantial. We have a large fragmentaryamphora, best assigned to the Late Geometric period, which seems to have been used in a tomb, perhaps, though not conclusively, for the inhumation of an infant. What remains beyond doubt is that the vessel was imported to Athens, with the island of Syros providing, at present, the strongest case for its origin. Against such a backdrop, it is important to stress that Cycladic imports to Athens in the Early Iron Age are exceedingly rare.As Brann noted: "Roughly speaking, in the 8th century the Athenians exported pottery, in the 7th century they imported."l54But even in the 7th century B.C.,Cycladic imports to Athens are most unusual.155For the Protogeometric and Geometric periods, Emil Kunze briefly mentioned a number of exports of Cycladic Geometric pottery,including examples exported to Athens, and the quantity of these was
149. Papadopoulos2000, p. 111. 150. In some cases the breakwas neatly executedso that the broken segment could be easilyreplaced(e.g., Tholos CemeterygraveX, Young 1939, pp. 42-44), but in other cases such neatnesswas either avoidedor not achieved.The wider mouth of coarse pithoi made fractureunnecessary(e.g., Agora gravesB 21:10 and G 12:14). Priordamageto a vessel maywell have favoredits selection as a burial container;on this aspect,see further Papadopoulos1998, pp. 115-116.
151. For these, see Johnston and Jones 1978; Docter 1991; Papadopoulos and Paspalas1999. 152. Smithson liked to referto the decorationof Agora P 14819 as "homely." 153. In orderto test that the diameterof the neck of the amphora was sufficientto accommodatethe skull of an infant, Smithson took a seriesof measurementsof the craniumof Carl Mauzy at variousstages throughouthis first yearof life in 1988. All of these measurementswere well below the
diameterof the neck of P 14819. 154. AgoraVIII, p. 27; for the exportof Athenian potteryin the 8th V.1, century,see also Kerameikos pp. 283-298. 155. Apart from P 14819, there is only one fragmentof a wheelmade and paintedvessel from the Agora that is perhapsCycladic(AgoraVIII, p. 106, pl. 41, no. 658, dating to the late 7th centuryB.C.;the fragmentmay be East Greek) and two possible,but uncertain, decoratedpithos fragments(AgoraVIII, p. 101, pl. 40, nos. 605-606).
I86
JOHN
K. PAPADOPOULOS
AND EVELYN L. SMITHSON
slightly swelled by Humfry Payne on the basis of unpublished fragments from the Athenian Acropolis.156Apart from these examples, few imports from the Cyclades to Athens are known. It is, therefore, all the more surprising to find such a massive amphora as P 14819 in Athens. How did this amphora make its way to Athens in the Late Geometric period? Any number of possibilities can be entertained. Of the many,the two that most often appear in the literature are trade and gift-exchange. In discussingthe presenceof largeAthenian amphorasat Knossos,Coldstream noted that these were not likely "to have been hawked overseas by casual traders."157 Indeed, Agora P 14819 does not appearto be part of any particular consignment. It is, for example, not part of any coordinated set of drinking or banquet crockery,such as drinking cups and other vessels, easThe Athenian imported pottery in one of ily stacked and transported.158 the Knossian tombs, Tekke tomb J, consisted of a "dinnerset"of26 vases.l59 In a similar vein, it is not a standard commodity container, such as the north Aegean neck-handled amphora found in quantity at Troy and exported as far afield as Bassit-Posidaion in North Syria.160Had it served as a commodity container for Cycladic oil or wine, one would expect to find more such vessels on the Greek mainland. Moreover, as a "one-off"-a hapax-in Early Iron Age Athens, P 14819 is difficult to place within the framework of the broader economic processes at work in the Mediterranean in the early first millennium B.C., which have been effectively explored by Andrew and Susan Sherratt, as well as Ian Morris.161 Perhaps the explanation of exchange mechanisms in the Early Iron Age Mediterranean that has loomed largest in recent studies is gift-exchange. This topic, particularlygift-relations between elites, has been well explored by a number of scholars.162Alternative scenarios stress ceremonial exchange involving sanctuaries,whether commodities were dedicated 156. Kunze 1931, p. 262, n. 58, mentions:"TheraischeScherbevon der Akropolis"(= Graef and Langlotz 1925, p. 31, no. 312; Gauss and Ruppenstein1998, p. 34, pl. 7, no. 2). In his reviewof Kunze 1931, Payne (1933, p. 123) noted:"Exportof Cycladicgeometricpottery:there are otherTheran, and some Siphnian, geometricor probablygeometric sherds from the Acropolis (unpublished),and Theran and other Cycladicfrom Perachora."See furtherGauss and Ruppenstein1998, p. 34, pl. 7, nos. 12, who discussand illustrateboth the Theran fragmentsfrom the Athenian Acropolis and the fragmentof a Cretan "pithos."Here it is also worth mentioning two pendant semicircleplates I, pl. 52, T 29, publishedin Kerameikos inv. 590; Kerameikos IV, pl. 34, inv. 1265. Although the shape,which is normallyassociatedwith the pendant
semicircleskyphos,is moreat home elsewhere,particularlyin Euboia and the Cyclades,the two examplesfrom the Kerameikosdo not appearto be imported.Desborough (1952, p. 118, pl. 12) noted nothing unusualabout them, and assumedthem to be Attic. Of the two, I had occasion to inspect one in the storeroomsof the KerameikosMuseum, and there is nothing in my notes to indicate a provenienceother than Athens. For the pendant semicircleplate, see also Gjerstad,Calvet,Yon, Karageorghis, andThalman 1977, p. 25, nos. 18-27, pl. 11:2-12;Popham 1994, p. 27, fig. 2.12. 157. Coldstream1995, p. 401; also Coldstream1990. 158. Cf. the existenceof such Athenian pottery sets at Knossos; Coldstream1995, p. 400. 159. Coldstream1995, p. 400.
160. For these amphoras,see Catling 1998b;Lenz, Ruppenstein, Baumann,and Catling 1998; Courbin 1993, for fragmentsof such amphoras at Bassit;for the lattersite, see further Courbin 1990. 161. Sherrattand Sherratt1993; also Sherrattand Sherratt1991; Morris 1986. For the Bronze Age, Knappand Cherry 1994 is particularlyuseful. 162. See, e.g., Coldstream1983; cf. Coldstream1986; 1989; 1994; 1995; Morris 1986; Rupp 1988; 1989; Colonna 1995. More recently,Crielaard (1993, p. 145) has arguedthat virtually all of the Greek EarlyIron Age pottery found on Cyprusand in the coastal Levant can be explainedthroughgiftrelationsbetween Euboian and local elites, an argumentthat has been taken to extremes;see Papadopoulos1997a, p.199.
CULTURAL
163.The bibliography on this subjectis growing;see de Polignac 1984; 1992; 1994; Langdon 1987;
Zaccagnini1987;KyrieleisandRollig 1988; Muscarella1989; Strom 1992; Catling 1995, pp. 127-128;
1997a,esp.p. 199. Papadopoulos
164. Coldstream1990, p. 30; 1993; 1995, p. 401. 165. E.g., Od. 17.441-444; see furtherPapadopoulos1997a, pp. 199200. 166. See Smithson 1968, esp. pls. 30-33; Morris and Papadopoulos,
forthcoming. 167.GaussandRuppenstein 1998, p. 34.
168.See above,n. 5. 169. Smithson 1968, p. 83.
170.Cf. Coldstream 1993. 171.SeefurtherBurkert1984; Purcell1990;Morris1992a;Morris 1992b,pp.xvii-xviii;Sherrattand Sherratt1993;Papadopoulos 1997a, p. 207. 172. For"involuntarycolonization" see Fernandez-Armesto1995, p. 269; for trade diasporas,see Cohen 1969; 1971. See furtherLyons and Papadopoulos2002.
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I87
as diplomatic gifts or prizes accumulated by wandering heroes, as well as "gift-exchange"between rulersand deities or even the importation of ritual equipment specifically for cult purposes.163In all of these explanations involving tradeor gift-exchange, it is usuallythe commodity itself that moves. Human agency is relegated to the background,or else seen as indirect. For example, in dealing with the large amphoras imported to Knossos from Athens and the Cyclades, alreadynoted, Coldstream concluded that they make more sense as gifts between guest friends in Athens, Knossos, and the Cyclades.'64Once the gifts were exchanged or the votive dedicated, the social actors involved withdrew. It would be wrong, however, to identify P 14819 too quickly as a gift exchanged between an Athenian and Cycladic family. Its context is such that it is difficult to picture the vessel-or, for that matter, any pot-in the context of gift-exchanges between elites or guest friends. If the evidence of Homer can be trusted, gift-exchange in the Late Bronze and Early Iron Age Mediterranean involved items of real value: gold, silver and bronze vessels, jewelry, objects of ivory,precious stone and faience, as well as livestock, such as horses or cattle, and human slaves.165The "rich Athenian lady"of the Areiopagos was rich not because of the many pots that were buried with her, but because of her gold and bronze jewelry, her necklace made of faience and glass beads, as well as the ivory stamp seals and disk, Moreover, if P 14819 was used to conamong other exotic kterismata.'66 tain an infant inhumation, it is important to remember that such burials were, more often than not, made in used cooking pots, in transport or storage vessels, not in containers of intrinsic value. Furthermore,the context of P 14819 would argue against ceremonial exchange involving sanctuaries, though such an explanation might contribute toward an understanding of the Cycladic imports found on the Athenian Acropolis.167 In her attempt to make sense of P 14819, Smithson leaned toward an interpretation that took into account all aspects of the life history of the This was, after all, a massive damaged amphora, its cultural biography.168 and repaired pot that was finally used in a burial, far from its place of origin. It appearsto have served various functions and to have been, to a certain extent, or at certain times, prized. In the same way that Smithson personalized the wealthy tomb on the Areiopagos as that of a rich Athenian lady-the daughter,conceivably,of apentakosiomedimnos, perhapseven an archon'swife, the yuvh 'Appoppovo; of the Athenian "king list"169-so too did she attempt to personalizeP 14819. She saw a person-a womanaccompanying the amphora, or, rather the amphora accompanying the woman. Smithson favored the idea of a marriage,not at the "frontiers"of the early Greek world,170but in the very heart of Greece. The amphora was seen as part of a dowry of a young Cycladic woman, or part of the domestic chattels, the objetsmobiliersof a Cycladic metic family-a large pot perhaps ultimately used to hold the remains of a deceased offspring. In so doing, Smithson visualized a reality with fluid boundaries, one which allowed not only for the movement of goods and ideas, but of people as well.171She envisioned people moving with goods, not as part of any predetermined colonizing enterprise, whether voluntary or involuntary,nor, necessarily,as part of any greater trade or other diaspora.172
I88
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Immigrant communities were a reality of the Classical polls, and one of the factors that contributed to the success of various Greek cities, including Corinth and Athens.173Several -oikoswords are attested for such immigrants, with metoikosthe most common. As David Whitehead has argued,with the reforms of Solon having created only indirect incentives to immigration, Athenian metic status probablyowes its formal origins to Kleisthenes.174The canonical view of metics in Athens on the basis of literarysources is most succinctly stated by Helen Pope in her 1935 dissertation: ... the metics were a part of the glory that was Athens. Solon gave the franchise to Txoiscavoveot; 'A60Ovare jLrxotxtLolevot; ?eit TrXvyn Solon Kleisthenes rcoXXo6SCopuXetouaeeivoo; xat (Plutarch, 24), 8o6Xoos;iTorxoTxo(Aristotle, Politics1275B 36), and rumor has it that Themistokles granted immunity from taxation to metics (Diodoros XI.43.3). Xenophon advocated methods to attract them (Revenues2.1), Isokrates hoped that o66l0cx6a8? TY]vro6lv ... ie?
xaci EvGovxali ?toLXXv, xv vSv ut7r6pcov Peace (On xa0?ocnxrv 21), and the most famous of all metics, F?pyilYn of said them 6 o6 Tx ?axotO; ?poi6Y0eoav xoaxrctopooajxov Lysias, 8tJC (AgainstPhilon 20).175
One of the earliest, and certainly most vivid, monuments in Athens relating to a metoikos-specifically one from the Cyclades-is the inscription block from the grave monument of Anaxilas of Naxos (Fig. 25).176 Found in the south tower of the Dipylon Gate, and dating to ca. 510 B.C., the block is roughly contemporary with the reforms of Kleisthenes. Its Cycladic honorand, however, must have resided in Athens for some years before. The inscription, as originally published, reads: 'AvaXoXXa ?8' 6Xo91puv6v 8axpoo6v 7toXo7r?evO?; X&dov?aT?xa tv?{[taxaxa(p9L0t.|?1<,?>vo: NaXaLoov tcLoxov 'AO?vaOt zLe?T?cOXOV
?XlooXaaoo90oo6v?s;?V?Xv E?'ap??r;: Toi ,t' ?ri TL[6lo[LaXO; ypacp6v XT?-0pa0ola av6vxt 0?X?V'AlpoCTovog7catxl X.apLOt6[Jvog;77
Here I stand, fraught with grief, sorrow,and lamentation, the monument of the deceased Anaxilas, the Naxian immigrant whom Athenians esteemed outstandingly for his prudence and virtue. Timomachos erected this majestic funerarygift (y?pap6v xT?pac;)as a kindness to the son of Ariston, who had died.l78 In this inscription, the critical word, pLr?TzoXOV, has been variously and some Whitehead, interpreted requires explanation. following Willemsen's original reading, was dubious of the historical value of private tombstones as evidence for metic studies, and he concluded that the term was descriptiveand that it was used in a "non-technical"way.179Keiji Baba, however, established that the reading meteoikoswas wrong, and that the
173. Whitehead 1984, esp. p. 59.
Forforeigners in Rome,see nowNoy 2000. 174. Whitehead 1977; 1984; see
alsoWhitehead's lucidsummary in OCD3,p. 969, s.v. metics. See also Clerc 1893; Gauthier1972. 175. Pope 1935, p. 46. 176. Willemsen 1963; SEGXXII, 79; IG 3, 1357; Baba 1984; Bakewell 1997,p. 221. 177. Athenian Kerameikos,inv. I. 388: Willemsen 1963, pp. 141-145, Beil. 72,2 and 73,3, from the south tower of the Dipylon Gate. 178. The English translationis my own. 179. Whitehead 1977, p. 64, n. 44; Bakewell 1997, p. 221.
CULTURAL
Figure25. Funeraryinscriptionfrom the monumentof Anaxilasof Naxos, AthenianKerameikos,inv.I. 388. CourtesyDeutsches Archaologisches Institut,Athens
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I89
stonecutter engraved at first [eTrzoLxov, but later corrected it to read [LeTdcLxov.l80Baba went on to state that immigrants were called metaoikoi, but were only defacto inhabitants of Athenian society, and that "an immigrant enjoyed there no better legal position than xenoi in general."'l8 More recently, Geoffrey Bakewell, in dealing with the metoikia in the Supplices of
Aiskhylos, noted that the use of the term in the case of Anaxilas probably had more to do with his burial place than with his status in life. And he concluded: "In the Ceramicus he will lie among the Athenians forever; only in this sense has he shifted residence."'82In all of these analyses, the focus has been, narrowly,on establishing the first incontrovertibleevidence of metoikia to mean, technically, a "resident alien living with the citizen
body" in a formal sense, as opposed to an "immigrant."'83But this philological and epigraphical search for the ur-metic loses sight of the fact that both metaoikosand metoikosderive from ie-rd + olxco, and both share the basic meaning of"an immigrant."Words such as metaoikoiand later metoikoi developed in the course of the Late Archaic period and earlier and, as As Whitehead has cogently such, enjoyed their own pre-writing history.184 noted, "it would be naive, surely,to imagine the metoikiaspringing fullyformed from Clisthenes' head."'85 Anaxilas, however, was only one of many metoikoior metaoikoi-immigrants-in Archaic Athens. Whatever the complexities of Kleisthenic and Solonian economics, and whatever the realities of Solon's reforms, there is growing evidence in the thriving Early Iron Age settlements like Lefkandi, Knossos, Corinth, and Athens for metics-resident aliens-in an erabefore recordedhistory.l86For example, in reviewing Protogeometric 180. Baba 1984, pp. 1-2, fig. 1. 181. Baba 1984, p. 4. Baba also notes that the historyof Athenian metic statusbegan with the reformsof Kleisthenes,and that the "oldterm" metaoikoswas appliedto this new status. 182. Bakewell 1997, p. 221. 183. Although, morphologically,the formeris evidentlya prototypeof the latter,Baba (1984, pp. 4-5) arguesthat
metoikos,which is used exclusivelyin later sources,had acquiredthe added sense of formal"metic-status." 184. Although the funerary monument of Anaxilasis clearlyLate Archaic,there is no archaeological evidence to pinpoint its date more accurately,beyond the subjectiveand vague criteriaof letterforms.Most classicistsdate the inscriptionto shortly after the reformsof Kleisthenes
(traditionally508/7 B.c.), and Baba (1984, p. 3) goes to greatpains to suggest a date of ca. 506 B.C.Although he may well be right, it is useful to rememberthat this is not a date "setin stone,"and that Anaxilasthe metaoikos certainlylived, and perhapsdied, before the reformsof Kleisthenes. 185. Whitehead 1977, p. 145. 186. For Corinth see, most recently, Morris and Papadopoulos1998.
190
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
burial customs at Lefkandi, Coldstream has recently suggested that several of the graves in the Toumba cemetery "seem to be those of Athenian residents."'87Similarly,another burial at Lefkandi, cast by its excavatorsas that of a "Euboian warrior-trader,"may very well represent a Phoenician, Cypriot, or North Syriantraveleror residentwho met his demise in Euboia and was buried there.188 The possibility of immigrant eastern craftsmen at Knossos has been more strongly argued in recent literature,including reference to the work of a so-called-if unlikely-"Phoenician jeweler"in an Orientalizing tomb at Tekke near Knossos.l89Elsewhere in Crete, there is growing evidence for the presence of Phoenicians at Eleutherna and at Kommos.190The whole question of Near Eastern contact with Crete, including immigrant presence, is effectively explored in Gail Hoffman's exhaustive study.19The evidence for Levantine resident aliens is even stronger in Early Iron Age Corinth, where severalPhoenician-style stele shrines are found in the Potters' Quarter.'92Indeed, Phoenician presence in the Corinthia was well outlined alreadyby classicists working in the late 19th and early 20th centuries.'93Building on the earliercontributions of scholars such as Ernst Maass, Henry Theodore Wade-Gery, Lewis Farnell, and others, Thomas Dunbabin wrote: An older generation, accustomed to seeing Phoenicians everywhere, regarded them as the bearers of Oriental culture to Greece. For them, Melikertes, the hero whose tomb was shown at the Isthmus, was Melkarth; the cult of Aphrodite, with its sacred prostitutes, was of Phoenician origin; Medea also and the cult of Hera Akraia, with its human sacrifice implied in the slaughter of Medea's children, were Phoenician. Athena was worshipped at Corinth with the title for what Phoinike [Schol. Lyc. 658; Steph. Byz. s.v. DotvtxocLov], that is worth; and there was a month Phoinikaios at Corinth and in Corinthian colonies. These Oriental elements, more strongly concentrated at Corinth than elsewhere in the Greek world, must have been introduced now, in the second half of the eighth century, when Corinth's commerce expands and orientalizing influences appearin her art. Whether they imply that men from Phoenicia or Syria settled at Corinth may be debated.194 The evidence in our literary sources for resident aliens, including the very articulation of words such as metoikos,is most abundant and clear for Classical Athens. It is, therefore, no surpriseto find growing evidence for metics-as in immigrants-in the pre-Classical city. In his publication of a Corinthianizing bowl made in Athens, Dunbabin cogently argued that the vessel was made by a Corinthian potter who emigrated to Athens.l95 Dunbabin placed the bowl at the head of a Corinthianizing current that 187.Coldstream 1995,p. 401. 188.PophamandLemos1995. 189.Coldstream 1988,p. 30; Hoffman(1997,pp. 191-245)discussesthe identityof the occupantof this tombin somedetail.Cf. Burkert1983. 190.ForEleutherna, see especially 1990a;1990b.ForKomStampolides
mos,see especiallythe SteleShrine, publishedin Shaw1989. 191.Hoffman1997,esp.pp.153189,252-260. 192.MorrisandPapadopoulos 1998. 193.MorrisandPapadopoulos 1998,p. 257.
194.Dunbabin1948,p. 66. 195.Dunbabin1950.The bowl is madeof Athenianclayandwas decorated by a pottertrainedin the workshopof the CorinthianSphinx Painter;see Dunbabin1950,p. 194 wasoriginallymade (theattribution byPayne).
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I9I
reached its height in the second quarterof the 6th century B.C.Following Plutarch (Sol. 24), Dunbabin went on to place the bowl and its maker against the backdrop of Solon's offer of Athenian citizenship to immiHere was straightforwardevidence, grants who came to practice a trade.196 not for the movement of pots, but for the relocation of the people who made the pots. More recently, further evidence has come to light for Corinthian pottery production, by way of a kiln dating to the later 8th and 7th century B.C., in the heart of the Early Iron Age Athenian Kerameikos, in the areathat was to become the ClassicalAgora of the city.197 Corinthians, however, were not the only emigre potters in Athens. Brann commented on several Protoattic potters, and considered them either as foreigners or, at least, well traveled. She wrote: The Ram Jug Painter shows some signs of having learned abroad about such things as Cycladic outline protomes and the use of brown paint, but one likes to think of him as an Athenian, albeit a traveler.The Polyphemos Painter, on the other hand, was an Aeginetan, or at least he worked in Aegina.l98 In discussingthe strong Cycladic influence on Protoatticpottery,Brann went on to note the general absence of contemporary Cycladic pottery in Athens and concluded that "the explanation must be that the potters and not the pottery traveled."'99Indeed, there is a long list of potters who migrated both to and from Athens, particularlyduring the Archaic and Classical periods.200The mobility of potters, and other artisans, can be traced back at least as early as the Late Bronze Age on the evidence of the Linear B tablets.201If the Ram Jug Painter was a metic or xenosin Athens he was, perhaps, not the first to boast a Cycladic ancestry. Kenneth Sheedy has cogently argued that the ParianAd Painter worked in Athens at the end of More specifically,Sheedy suggeststhat the Ad Painter the 8th centuryB.C.202 in the Workshop of the Wurzburg Group, and vases learned to decorate that he was also aware of the work of the Analatos Painter.203 The latter, one of the great pioneers of the Protoattic style, may have spent the later part of his working careerproducing pottery at the settlement at Incoronata in the choraof the later city of Metapontion in South Italy, according to the penetrating analysis of Martine Denoyelle.204 The evidence of emigre potters to Early Iron Age Athens may represent the tip of a large iceberg, given the tenacity of pottery to survive in abundance in most archaeological contexts. Moreover, it is important to stress that not all resident aliens were craftspersons,let alone potters. In 196. Dunbabin 1950, pp. 196,200. 197. Papadopoulos1996; Papadopoulos,forthcoming. 198. Agora VIII, p. 24. Indeed, the RamJug, the name-vaseof the RamJug Painter,was assumedto be Aiginetan when it first appeared,and scholarssuch as Pfuhl, Payne,Beazley,and Dunbabin doubtedwhether it was Attic (see Morris 1984, p. ix, with pl. 10 for the Ram Jug [Aigina Museum 566]). The
Aiginetan origin of much of the "Protoattic"Black and White style has been most fully arguedby SarahMorris (1984), though severalscholarscontinue to believe that this potter lived and worked in Athens; see especiallyWalterKarydi1997; Kyrkou1997, esp. p. 432. 199. AgoraVIII, p. 28. Brann also cited Schweitzer's(1955, p. 105) view that the painterof the Aristonothos kraterwas a travelingAttic potter,once
trainedin the Athenian Kerameikos. 200. Full details in Papadopoulos, forthcoming. 201. Papadopoulos1997b. 202. Sheedy 1985, pp. 170, 173, 189190. 203. Sheedy 1985, pp. 189-190. 204. Denoyelle 1996. See also the case of the later Pisticci Painter,one of the pioneersof the Lucanianredfigure style (Denoyelle 1997).
192
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
Peace 296-298, Aristophanes calls upon a whole slew of foreigners and tradesmen: &a', c) ytopyol
xa7topoaL
xaoL &qlLtooupYo xoaL (JtzoXOL
xOCL T-XTov?; xao evoL
xaoi vqatcxoTat, 6?bp' IT', X) 7av<'cg XecO
You farmers and merchants and carpenters and craftsmen and immigrants and foreigners and islanders, come here, all you people, as quick as you can.205 Some of these foreigners,like Anaxilas,were highly esteemed for their prudence and virtue; in later times, such a deserving metic could become an isoteles.206The funerarymonument of Anaxilas further establishes that the word metoikos-or more accuratelymetaoikos-does not first appearin the second quarterof the 5th century B.C., as has been suggested in earlier studies,207but can be traced back at least to the late 6th century B.C.More to the point, the very quantity of foreigners living and dying in historic Athens and the Piraeus that can be traced archaeologically,through the evidence of funerary inscriptions and iconography, as well as our literary sources, is staggering.208Ktesikles' figures for the time during the ascendancy ofDemetrios of Phaleron in the late 4th century B.C. include 21,000 Athenians and 10,000 metoikoi.209For the construction of the Erechtheion a century earlier,metic workmen outnumbered the citizen workmen and civic officials by a ratio of almost 2:1, if the Erechtheion building accounts are accurate.210 Although the evidence for the size of the metic population in Athens in earlier times is meager and controversial,211the surviving literatureon resident aliens alone attests their importance in the social and economic fabric of the city. 205. J. Henderson,trans.,Loeb edition (1998). 206. For the isotelesand isoteleia,see Whitehead 1977, pp. 11-13. 207. See Whitehead 1977, p. 7. 208. For bibliographyon the literary sources,see Whitehead 1977; 1984. For foreignersburiedin Athens, see Salta 1991, pp. 161-239. For a recent overview of Thraciansand Skythiansburiedin Athens, see Babler 1998; note also Thraciansin Athenian iconography, assembledand discussedin Tsiafakis 1998 and, generally,Papadopoulos 2000. For the metic in Athenian courts ofjustice, see Patterson2000. 209. See Whitehead 1977, p. 97, with full discussion,including comments on the veracityof these figures. 210. On the basis of G 12373, 374, and IG I2 1654, Pope (1935, pp. 52-53) calculated,among the citizens working on
the Erechtheion,three sculptors,17 workmen,and three civic officials,as opposed to 38 metics, four of whom were sculptors.The metics were distributedacrossseveralof the demes, includingMelite, Alopeke, Skambonidai,Agryle, Kollytos,Koile, and Kydathenaion.One of the 38 metics was a woman, Satyra.For the distributionof metics in demes, see also Clerc 1893, pp. 450-456; Gomme 1933, pp. 39-40. For furthercomments on metics in Attika, see WilamowitzM6llendorff 1887. In Randall'slater study,among the total recordedworkmen and specialistsinscribedon the Erechtheionaccounts,the breakdownis as follows:24 citizens, 42 metics, 20 slaves,21 unknown (107 total);see Randall1953, p. 201, table 1. To the list of demes, mentioned above,where metics were domiciled,Randall(1953,
p. 204, table 3) adds Phyla and Semachidai (the largestconcentrationof metics, 12, lived in Melite). For the inscriptionsassociatedwith the construction of the Erechtheion,see Stevens and Paton 1927, pp. 277-422 (text and commentaryby L. D. Caskey);Randall 1953 (with references);also Scranton's(1960) generalaccount.In the recentwork along the South Slope of the Athenian Acropolisin 2001 as part of the "Unificationof Athens"project,a large numberof the employed workmenappearedto be resident aliens,citizens born outside Greece, or native speakersof languagesother than Greek. Of the languagesspoken, Albanian,Pontic Greek, and perhaps also Serbo-Croatianand Russianwere more commonly heardthan modern Greek. 211. Whitehead 1977, pp. 97-98.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
I93
Potters like the Parian Ad Painter and the Ram Jug Painter, among others, provide a potential glimpse of Cycladic immigrants working in Athens generations before the reforms of Solon. Whether we call them "metoikoz? or "foreigners"(EvoL) or even, simply, "islanders"(voar)TEs) does not really matter.212 They, together with Corinthians and other foreigners-craftspersons, traders,and others-who were to become numerdemonous in Athens in the course of the 7th and 6th centuries B.C.,213 strate that resident aliens enjoyed a venerable prehistory in the city.They further suggest, through direct human agency,the manner in which a large amphora like P 14819, with its idiosyncratic life history, may have made its way to Late Geometric Athens: a Cycladic amphora accompanying a Cycladic immigrant.
ACKNOWLEDGMENTS
212. For these terms, and other types of foreignerswho might be residentin Athens, includingproxenoi, see Whitehead 1977, pp. 6-20. 213. For the variouscategoriesof non-Athenians in Attic inscriptions, see Pope 1935. See also the evidence assembledin Papadopoulos,forthcoming, ch. 3.
Many friends and colleagues have assisted in the preparationof this article and to all I extend my warmest thanks. Within the Agora, I am especially grateful to John Camp for his unfailing encouragement and support, to Craig Mauzy for his help with many of the illustrations used in this study, to Annie Hooton for her fine drawings, and to Jan Jordan and Sylvie Dumont for all sorts of assistance. Numerous colleagues in the Cyclades have graciously shown me and discussed material in their care, including much that is unpublished. I am particularlygrateful to the Ephor of the Cyclades, Mariza Marthari, for her constant assistance, as well as to the following: ChristinaTelevantou, Dick Green, and Alexander Cambitoglou (Andros); Nota Kourou (Tenos); Yannos Kourayos and Robert Koehl (Paros); Vassileios Lambrinoudakis and Konstantinos Zachos (Naxos); Photeini Zapheiropoulou (Donousa); Mariza Marthari and Maria Katri (Syros); Christos Doumas (Thera); Lila Marangou (Amorgos); and Richard Catling, Irene Lemos, and Petros Themelis (Euboia). I have benefited from discussions both in the past and more recently with four colleagues in particular,who have taught me a great deal about the islands of the archipelago: Lila Marangou, Nota Kourou, Richard Catling, and Kenneth Sheedy. For discussing a range of ceramic problems in East Greece I am indebted to Lela Walter-Karydi and Despoina Tsiafakis.Thanks are due to Fred Cooper for sharing with me his thoughts on the Hephaisteion. I am also grateful to Sarah Morris and Stavros Paspalas for all they have done and continue to do on my behalf, to Hans Rupprecht Goette and Raimund Wuensche for assistance with illustrations, Robert Finnerty for preparing the maps, and the anonymous Hesperia referees, who have done much to improve this study. Throughout Evelyn's notes and correspondenceon Anaxilas'samphora,the following individuals'names stand out: Nicolas Coldstream, Nota Kourou, Ken Sheedy, and the late Vincent Desborough, all of whom offeredhelp and sage advice.I am grateful to all of the above and also to Eve Harrison and Susan Rotroff. Finally, this study would not have been possible in its present form without the excavation notebooks of the late Dorothy Burr and Homer A. Thompson: they were the excavatorsof Anaxilas's amphora.
194
JOHN
K. PAPADOPOULOS
AND
EVELYN
L. SMITHSON
REFERENCES AgoraVIII = E. T. H. Brann,Late Geometric and ProtoatticPottery, Mid 8th to Late 7th Century B.C.,
Princeton 1962. Appadurai,A., ed. 1986. TheSocialLfe of Things,Cambridge. Baba,K. 1984. "On KerameikosInv. I 388 (SEGXXII, 79): A Note on the Formationof the Athenian Metic-Status,"BSA 79, pp. 1-5. Babler,B. 1998. FleissigeThrakerinnen undwehrhafteSkythen:Nichtgriechen im klassischen Athenund ihre archdologische Hinterlassenschaft, Stuttgart. Bakewell,G. W. 1997. "MeTroxiain the Supplicesof Aeschylus,"ClAnt16, pp.209-228. Bastea,E. 2000. The CreationofModern Athens:PlanningtheMyth, Cambridge. Boardman,J. 1957. "EarlyEuboian Potteryand History,"BSA 52, pp. 1-29. .1967. Excavationsin Chios, 1952-1955: GreekEmporio (BSA Suppl. 6), London. . 1969. "EuboeanPotteryin West and East,"DialArch3, pp. 102-114. . 1998. Early GreekVase Painting,London. Bosanquet,R. C. 1895-1896. "Prehistoric Gravesin Syra,"BSA 2, pp. 141-144. Bossert,E.-M. 1967. "Kastriauf Syros: Vorberichtiibereine Untersuchung der prahistorischenSiedlung," ArchDelt22, A', pp. 53-76. Brann,E. T. H. 1961. "LateGeometric Well Groups from the Athenian Agora,"Hesperia30, pp. 93146. Brock,J. K. 1957. Fortetsa:Early Greek TombsnearKnossos(BSA Suppl.2), Cambridge. Brock,J. K., and G. MackworthYoung. 1949. "Excavationsin Siphnos," BSA 44, pp. 1-92. Broodbank,C. 2000. An Island Archaeology of theEarly Cyclades, Cambridge. Brouskare,M. S. 1980. "ADark Age Cemeteryin ErechtheionStreet, Athens,"BSA 75, pp. 13-31.
Burkert,W. 1983. "ItinerantDiviners and Magicians:A Neglected Element in CulturalContacts,"in The GreekRenaissanceof theEighth Century B.C.:Tradition and Innovation.
Proceedings of the SecondInternational Symposiumat the SwedishInstitute in Athens,1-5 June1981, R. Hagg, ed., Stockholm,pp. 115-119. . 1984. Die orientalisierende Religion Epochein dergriechischen undLiteratur,Heidelberg. Buschor,E. 1929. "Kykladisches," AM 54, pp. 142-163. Cambitoglou,A., S. Peirce,O. Segal, andJ. K. Papadopoulos.1981. MuseumofAndros. Archaeological Guideto theFindsfromtheExcavationsof the GeometricTownat Zagora,Athens. Cambitoglou,A., andJ. K. Papadopoulos. 1994. "Excavationsat Torone, 1990,"MeditArch7, pp. 141-163. Camp,J. McK., II. 1979. "ADrought in the Late Eighth Century B.C.," Hesperia48, pp. 397-411. . 1999. "Excavationsin the Athenian Agora, 1996 and 1997," Hesperia68, pp. 255-283. Catling, H. W. 1995. "HeroesReturned?SubminoanBurialsfrom Crete,"in TheAgesofHomer: A Tributeto Emily TownsendVermeule,J. B. Carterand S. P.Morris, eds., Austin, pp. 123-136. Catling, R. W. V. 1998a. "Exportsof Attic ProtogeometricPottery and Their Identificationby NonAnalyticalMeans,"BSA 93, pp. 365-378. . 1998b. "TheTypology of the Protogeometricand SubprotogeometricPotteryfromTroia and Its Aegean Context,"StudiaTroica 8,pp. 151-187. Catling, R. W. V., and R. E. Jones. 1989. "ProtogeometricVasesfrom Amorgos in the Museum of the British School,"BSA 84, pp. 177185. Charitonides,S. I. 1975. "E6op4iVaTa T 7TpO)TOYCLtp,>CLX
XOC YCOt[?ETp1LX
TCoX; TSY avaoxaXlcpT VOti); 1T5 'Axpo7o6Xo;q, 1955-1959,"ArchDelt
28, 1973, A' [1975], pp. 1-63.
Clerc,M. 1893. Lesmetequesathenians: Etudesurla conditionlegale,la situation moraleet le rolesocialet economiquedesetrangersdomiciliesa Athenes,Paris. Cohen, A. 1969. Customand Politicsin UrbanAfrica:A StudyofHausa Migrantsin YorubaTowns,Berkeley. . 1971. "CulturalStrategiesin the Organizationof TradingDiasporas,"in TheDevelopmentofIndigenousTradeandMarketsin West Africa,C. Meillassoux,ed., London, pp.266-281. Coldstream,J. N. 1968. GreekGeometric Pottery.ASurveyof TenLocalStyles London. and TheirChronology, .1971. "The Cesnola Painter: A Change of Address,"BICS 18, pp. 1-15. . 1977. GeometricGreece, London. . 1983. "Gift Exchangein the Eighth Century B.C.," in The Greek
Renaissanceof theEighth CenturyB.C.: Traditionand Innovation.Proceedingsof theSecondInternationalSymposiumat the SwedishInstitutein Athens,1- June 1981, R. Hagg, ed., Stockholm,pp. 210-207. . 1986. "Kitionand Amathus: Some Reflectionson Their Westward Links duringthe Early Iron Age," in Actsof theInternational Symposium"Cyprus Archaeological betweenthe Orientand the Occi1985, dent,"Nicosia,8-14 September V. Karageorghis,ed., Nicosia, pp.321-329. 1988. "SomeMinoan Reflectionsin CretanGeometric Art,"in Studiesin Honourof T B. L. Webster 2, J. H. Betts, and T. J. R. Green, eds., Hooker, J. Bristol,pp. 23-32. . 1989. "EarlyGreek Visitors to Cyprusand the EasternMediterranean,"in Cyprusand theEast Mediterraneanin theIronAge: of the SeventhBritish Proceedings MuseumColloquium, April 1988, V.Tatton-Brown,ed., London, pp. 90-95. . 1990. "Cycladicand Euboean Imports in the North Cemetery of
CULTURAL
Knossos,"in ErMOrIA: Ceramic and IconographicStudies in Honour ofAlexander Cambitoglou (MeditArch
Suppl. 1),J.-P. Descoeudres,ed., Sydney,pp. 25-30. . 1993. "MixedMarriagesat the Frontiersof the Early Greek World,"OJA12, pp. 89-107. .1994. "Pithekoussai,Cyprus, and the Cesnola Painter,"in APOIKIA. Ipizi antichi insediamenti Greci in Occidente:Funzioni e modi dell'organizzazione politica e sociale. Scritti in onore di Giorgio Buchner (AON. AnnArchStorAnt n.s. 1),
B. d'Agostino and D. Ridgway,eds., Naples, pp. 77-86. .1995. "The Rich Lady of the Areiopagosand Her Contemporaries:A Tributein Memory of Evelyn Lord Smithson,"Hesperia 64, pp. 391-403. Coldstream,J. N., and H. W. Catling, eds. 1996. KnossosNorth Cemetery: Early GreekTombs (BSA Suppl. 28),
London. Collignon, M., and L. Couve. 1902. Catalogue des vasespeints du Musee National dAthenes, Paris.
Colonna, G. 1995. "Etruschia Pitecusa nel'orientalizzanteantico,"in L'incidenza dell'antico: Studi in memoria di Ettore Lepore 1, A. Storchi
Marino, ed., Naples, pp. 325-342. Cook, R. M., and P. Dupont. 1998. East GreekPottery, London.
Cooper, F. A. 1985. "ThreeYearsin Athens: A New Look at Old Buildings,"unpublishedmanuscriptof a lecturedeliveredat the American School of ClassicalStudies at Athens, 27 March 1985. Courbin,P. 1954. "Chroniquedes fouilles et decouvertesarch6ologique en Grece en 1953: Premiere partie,"BCH 78, pp. 95-157. . 1990. "Bassit-Posidaionin the Early Iron Age," in Greek Colonists and Native Populations: Proceedings of the First Australian Congressof ClassicalArchaeologyHeld in Honour of Emeritus ProfessorA. D. Trendall, Sydney, 9-14 July 1985, J.-P. Des-
coeudres,ed., Oxford, pp. 503-509. . 1993. "Fragmentsd'amphores protog6om6triquesgrecquesa Bassit (Syrie),"Hesperia62, pp. 95-113.
BIOGRAPHY
OF A GEOMETRIC
Crielaard,J. P. 1993. "The Social Organizationof EuboeanTradewith the EasternMediterraneanduring the 10th to 8th CenturiesB.C.," Pharos1, pp. 139-146. Dalongeville, R., and G. Rougemont, eds. 1993. Recherches dansles Cyclades:Resultatsdestravauxde la RCP 583, Lyon. Davis, R. H. 1997. Lives of Indian Images,Princeton. DelosXV = C. Dugas and C. Rhomaios, Les vasesprehelleniques etge'ometriques,Paris 1934. DelosXVII = C. Dugas, Lesvasesorientalisantsdestylenon melien,Paris 1935. Denoyelle, M. 1996. "Le peintre d'Analatos:Essai de synthese et perspectivesnouvelles,"AntK39, pp. 71-87. .1997. "Atticor non-Attic? The Case of the Pisticci Painter," in Oakley,Coulson, and Palagia 1997, pp.395-405. de Polignac,F. 1984. La naissancede la cite'grecque: Cultes,espaceet societe VIIe sieclesavantJ.C., Paris. 1992. "Influenceext6rieure ou evolution interne?L'innovation culturelleen Grece g6om6trique et archaique,"in KopckeandTokumaru 1992, pp. 112-127. . 1994. "Mediation,Competition, and Sovereignty:The Evolution of RuralSanctuariesin Geometric Greece,"in Placingthe Gods:Sanctuariesand SacredSpacein AncientGreece,S. E. Alcock and R. Osborne,eds., Oxford, pp. 3-18. Desborough,V. R. d'A. 1952. ProtogeometricPottery,Oxford. .1972. The GreekDarkAges, London. Descoeudres,J.-P.,and R. Kearsley. 1983. "GreekPotteryat Veii: Another Look,"BSA 78, pp. 9-53. Dinsmoor,W. B. 1941. Observationson theHephaisteion(HesperiaSuppl. 5), Princeton. Docter, R. F. 1991. "Athenavs. Dionysos:Reconsideringthe Contents of SOS Amphorae,"BABesch 66, pp. 45-49. Z-' Doumas, C. 1965. "KoxXd&;gs: ArchDelt18, 1963, B'2 'AXwv6cxla," [1965], pp. 279-280.
AMPHORA
I95
des Dugas, C. 1925. La ceramique Cyclades,Paris. Dunbabin,T. J. 1948. "The Early History of Corinth,"JHS68, pp. 59-69. . 1950. "AnAttic Bowl,"BSA 45, pp. 193-202. Dupont, P. 1986. "NaturwissenschaftlicheBestimmungder archaischenKeramikMilets," in Milet 1899-1980: Ergebnisse, einer Probleme,undPerspektiven Ausgrabung.KolloquiumFrankfurt am Main 1980 (IstMitt-BH 31), W. Muller-Wiener,ed., Tubingen, pp. 57-71. Eilmann, R. 1933. "Fruhegriechische Keramikim samischenHeraion," AM 58, pp. 47-145. Eiteljorg,H., II. 1980. "The Fast Wheel, the Multiple-BrushCompass, and Athens as Home of the ProtogeometricStyle,"AJA84, pp. 449-452. Fernandez-Armesto,F. 1995. Millennium:AHistoryof theLast Thousand Years,New York. Fraser,P. M., and E. Matthews. 1987. A Lexiconof GreekPersonalNames1: TheAegeanIslands,Cyprus,Cyrenaica,Oxford. Gauss,W., and F. Ruppenstein.1998. "Die Athener Akropolisin der friihen Eisenzeit,"AM 113, pp. 160. Gauthier,P. 1972. Symbola:Les etrangers et lajusticedanslescitesgrecques, Nancy. Gautier,J. 1993. "Les cycladesantiques: Caracterisationde centresde productions c6ramiquespar microscopie optique,"in Dalongeville and Rougemont 1993, pp. 167-204. Gjerstad,E., Y. Calvet,M. Yon, V. Karageorghis,andJ. P.Thalman. 1977. GreekGeometric andArchaic Found in Pottery Cyprus,Stockholm. Gomme, A. W. 1933. ThePopulation ofAthensin theFifth and Fourth CenturiesB.C.,Oxford. Gosden, C., and Y. Marshall,eds. 1999. TheCulturalBiographyof 31.2), London. Objects(WorldArch Graef,B., and E. Langlotz. 1925. Die antikenVasenvon derAkropolis zu Athen I, Berlin.
196
JOHN
K. PAPADOPOULOS
Haentjens,A. M. E. 1999. "AtticGeometric Childgraves:More than Bones in Pots,"in ClassicalArchaeologytowardsthe ThirdMillennium: ProceedReflectionsand Perspectives. the XVth International ingsof Congresson ClassicalArchaeology, Amsterdam, July 12-17, 1998, R. F. Docter and E. M. Moormann, eds., Amsterdam,pp. 182-183. Head, B. V. 1911. HistoriaNumorum: A Manualof GreekNumismatics,2nd ed., Oxford. Hekman,J. J. 1994. "Chalandrianion Syros:An EarlyBronze Age Cemetery in the Cyclades,"ArchEph 1994, pp. 47-74. Hoffman, G. L. 1997. Importsand Immigrants:NearEasternContacts with IronAge Crete,Ann Arbor. Hommel, P. 1959-1960. "Die Ausgrabung beim Athena-Tempelin Milet, 1957: II. Der Abschnitt bstlich des Athenatempels,"IstMitt 9-10, pp. 31-62. Jacopi,G. 1933. Esplorazionearcheologicadi CamiroII (ClRhVI-VII), Bergamo. Johnston,A. W., and R. E. Jones. 1978. "The 'SOS'Amphora,"BSA 73, pp. 103-141. Jones, R. E. 1986. Greekand Cypriot Pottery:AReview of Scientific Studies,Athens. Kardara,C. P. 1963. 'Po&osaxdryyeoAthens. ypacopia, Karo,G. 1934. "ArchaologischeFunde vomJuli 1933 bisJuli 1934," AA 1934 (JdI49), cols. 123-254. Karouzou,S. 1968. NationalArchaeologicalMuseum:Collectionof Athens. Sculpture, -&o6 Kawadias,P. 1890-1892. rourrra 'EOvtxoiMoovaioo,Athens. I = W. Kraikerand Kerameikos K. Kiibler,Die Nekropolendes12. bis 10.Jahrhunderts, Berlin 1939. Kerameikos IV = K. Kiibler,Neufunde ausderNekropoledes11. und 10. Berlin 1943. Jahrhunderts, Kerameikos V.1 = K. Kiibler,Die Nekropoledes10. bis 8.Jahrhunderts, Berlin 1954. Knapp,A. B., andJ. F. Cherry.1994. ProvenienceStudiesand Bronze Age Cyprus:Production,Exchange, andPolitico-Economic Change, Madison.
AND
EVELYN
L. SMITHSON
Koch, H. 1955. Studienzum in Athen,Berlin. Theseustempel -oO Kokkou,A. 1974. "ToxLov6xpavov vaoo TSoSZouvLa&oS 'A60lv&qE. M. 4478 xatlI SuZoyioltoi KavTaxoolqvob,"ArchEph 1974, pp. 102-112. Kontoleon,N. M. 1949. "FeorrI-rpLxo6 x Nadou,"ArchEph a&ICpopebg; 1945-1947 [1949], pp. 1-21. 1958. "Theraisches,"AM73, pp.117-139. Kopcke,G., and I. Tokumaru,eds. 1992. GreecebetweenEast and West: 10th-8th CenturiesB.c.:Papersof the Meetingat theInstituteof FineArts, New YorkUniversity,March1516th, 1990, Mainz. Kopytoff,I. 1986. "The CulturalBiographyof Things: Commodization as Process,"in Appadurai1986, pp. 64-91. Kourou,N. 1984. "LocalNaxian Workshopsand the Import-Export PotteryTradeof the Island in the Geometric Period,"in AncientGreek and RelatedPottery:Proceedings of the InternationalVaseSymposiumin Amsterdam,12-15April 1984, H. A. G. Brijder,ed., Amsterdam, pp. 107-112. . 1992. "Apropos d'un atelier geometriquenaxien,"in Lesateliers depotiersdansle mondegrecaux epoques geometrique, archaique,et classique(BCH Suppl.23), F. Blonde andJ. Y. Perreault,eds., Paris, pp. 131-143. 1994. "'H vatlaxi 7tapouoLca aTOaOLyaLo xaoca T xai T-1[L?Ca6y?eo in npcaxxnxa yecolieptxpLx izo0,"
TroA' 7raveMrjvtoaovesp[oo m' OCEMa "JNdroio rca peiaoo rTv acovci)v,"(tAdort,3-6 2Ere!4-p[oOU
1992, I. K. Promponasand S. E. Psarras,eds., Athens, pp. 263311. 1996. "Avaox%aqp?s or EL7c.oupyo Tjvou (1995-1996)," Prakt 1996, pp. 261-270. . 1997. "ANew Geometric Amphorain the BenakiMuseum: The InternalDynamics of an Attic Style," in Greek Offerings:Essays on
GreekArt in HonourofJohnBoardman, 0. Palagia,ed., Oxford, pp. 43-53.
. 999a. Avaaxaoc Nadoo: T6 VOTiovexpo-racElo r'5; Ndaoo xara
T yC?CW[rpitx7rzpi(o. "EpEove C3ev&r6v 1931-1939, Athens.
1999b. "Ta &poXata Ire(rli roo Ectproupyou," 'Eraipec'a TveIaxxtv
MeAcz3v2, pp. 93-114. zu Krause,G. 1975. Untersuchungen den dltesten Nekropolen am Eridanos
inAthen,2 vols. (HBA Suppl.3), Hamburg. Kunze,E. 1931. Kretische Bronzereliefs, Stuttgart. Kurtz,D. C., andJ. Boardman. 1971. GreekBurial Customs,Ithaca, N.Y. Kyrieleis,H., and W. Rollig. 1988. "Ein altorientalischerPferdschmuck aus dem Heraion von Samos,"AM 103, pp. 27-75. Kyrkou, M. 1997. "H ctoPx-oax-rTtx
N?eSxzpalix?g aocpTu7rp6%x.oq7M: pLr;,"in Oakley,Coulson, and Palagia1997, pp. 423-434. Lambrinoudakis,B. K. 1983a. "N?oc oatotXrLa yta r] yvwarqT-J NoaLxal 7TpCt.ig[l yc(jX-rpLtx;S axScq &apxdaix;xepaortlxig," ASAtene 61 (n.s. 45), pp. 109-119.
. 1983b. "Les ateliersde ceramiquegeometriqueet orientalisante de Naxos:Perspectivespour in Les l'analysearcheometrique," Materiauxpour uneetude Cyclades: degeographiehistorique,G. Rougemont, ed., Paris,pp. 165-175. . 1988. "Venerationof Ancestorsin Geometric Naxos,"in Early GreekCultPractice.Proceedingsof theFfth InternationalSymposiumat theSwedishInstituteat Athens,26-29June 1986, R. Hagg, N. Marinatos,and G. C. Nordquist, eds., Gbteborg,pp. 235246. Langdon, S. 1987. "Gift Exchange in Geometric Sanctuaries,"in Gifts to the Gods:Proceedings of the Uppsala T. Linders and 1985, Symposium, G. Nordquist,eds., Uppsala, pp. 107-113. Lefkandi I = M. R. Popham,
L. H. Sackett,and P. G. Themelis, eds., LefkandiI: TheIronAge (BSA Suppl. 11), Oxford 1979-1980. Lenz, D., F. Ruppenstein,M. Baumann, and R. W. V. Catling. 1998. "ProtogeometricPotteryat Troia,"StudiaTroica8, pp. 189222.
CULTURAL
Levi, D. 1925-1926. "Lanecropoli geometricadi Kardiania Tinos," ASAtene8-9, pp. 203-234. Lyons, C. L., andJ. K. Papadopoulos. 2002. "Archaeologyand Colonialism,"in TheArchaeology of Colonialism, C. L. Lyons andJ. K. Papadopoulos, eds., Los Angeles, pp. 1-23. Manthos, A. D. 1979. "E6piocrxa ?mL(pavEiag aor6 Tov raXFcooaa-i E6pou,"AAA12, pp. 39-46. Marangou,L. 1993. "'Avaoxaocp Mlvcas 'Ajiopyo5,"Prakt1993, pp.188-208. 1996. "'EpyaoieSq rcepeala, oouv p0loq1;xal LXAvTqa&vaaxaqrcp
Mlv(xa 'Ajiopyo5,"Prakt1996, pp.277-301. 1997. "'EpyaeoiEc orepecQGn, oV7ipvrtlrcqxal, ric? vs ; avacxaccp MLvc(ag'Ajiopyoo,"Prakt1997,
pp.183-189. . 1998. "The Acropolis Sanctuaryof Minoa on Amorgos:Cult Practicefrom the 8th Century B.C. to the 3rd Century
A.D.,"
in Ancient
GreekCultPracticefromtheArchaeologicalEvidence:Proceedings of the FourthInternationalSeminaron AncientGreekCult,Organizedby the SwedishInstituteatAthens,22-24 October1993, R. Hagg, ed., Stockholm, pp. 9-26. . In prep. "'O-aqxpxoS;7rrpipoXog V pOLwV LJtooLx&v iTOplXCoV Xp6vov xO6p'4zv X(pOVGJ a-ov Mtvca 'AiLopyo6."
Marthari,M. 1998. Syros,Chalandriani, Kastri:FromtheInvestigationsand Protectionto thePresentationofan Site,Athens. Archaeological Mendoni, L. G., and A. Mazarakis Ainian, eds. 1998. Kea-Kythnos: HistoryandArchaeology. Proceedings Keaofan InternationalSymposium, Kythnos,22-25 June 1994 (Meletemata 27), Athens. Morris,I. 1986. "Gift and Commodity in ArchaicGreece,"Man 21, pp. 117. .1987. BurialandAncient Society:TheRise of the GreekCityState,Cambridge. Morris, S. P. 1984. TheBlackand WhiteStyle:AthensandAigina in the OrientalizingPeriod,New Haven. . 1992a. Daidalosand the Origins of GreekArt, Princeton.
BIOGRAPHY
OF A GEOMETRIC
. 1992b. "Introduction.Greece beyond East and West: Perspectives and Prospects,"in Kopckeand Tokumaru1992, pp. xiii-xviii. . 1997. "Greekand Near Eastern Art in the Age of Homer,"in New Light on a DarkAge:Exploring the Cultureof GeometricGreece, S. Langdon, ed., Columbia,pp. 5671. Morris, S. P., andJ. K. Papadopoulos. 1998. "Phoeniciansand the Corinthian PotteryIndustry,"in ArchiologischeStudienin Kontaktzonender antikenWelt(Ver6ffentlichungder JoachimJungius-Gesellschaftder WissenschaftenHamburg87), R. Rolle, K. Schmidt, and R. F. Docter, eds., Hamburg,pp. 251263. .Forthcoming. "Of Granaries and Games:An Egyptian Stowaway in an Athenian Chest,"in X4iptl: Essaysin Honorof SaraA. Immerwahr (HesperiaSuppl.),A. P. Chapin, ed., Princeton. Muscarella,O. W. 1989. "KingMidas of Phrygia and the Greeks,"in Anatoliaand theAncientNearEast: Studiesin Honorof TahsinOzguc, K. Emre et al., eds., Ankara, pp.333-344. Mustakas,C. 1954-1955. "Kimolos," AM 69-70, pp. 153-158. Noy, D. 2000. Foreignersat Rome: Citizensand Strangers,London. Oakeshott,N. R. 1966. "Horned-Head Vase Handles,"JHS86, pp. 114132. Oakley,J.H., W. D. E. Coulson, and O. Palagia.1997. AthenianPotters and Painters:The Conference Oxford. Proceedings, Papadopoulos,J.K. 1993. "To Kill a Cemetery:The Athenian Kerameikos and the EarlyIron Age in the Aegean,"JMA6, pp. 175206. 1994a. "EvelynLord Smithson, 1923-1992,"AJA 98, pp. 563564. 1994b. "EarlyIron Age Potters'Marksin the Aegean,"Hesperia 63,pp. 437-507. . 1996. "The Original Kerameikos of Athens and the Siting of the ClassicalAgora,"GRBS37, pp. 107-128.
AMPHORA
I97
1997a. "PhantomEuboians," JMA 10, pp. 191-219. .1997b. "Innovations,Imitations, and Ceramic Style:Modes of Productionand Modes of Dissemination,"in TEXNH: and Craftsmen,Craftswomen, in Craftsmanship theAegeanBronze Age. The6th InternationalConference Organizedby the Universityof Liegeand TempleUniversity, 18-20April 1996, R. Laffineur and P. P. Betancourt,eds., Liege, pp. 449-462. .1998. "ABucket,by Any Other Name, and an Athenian Strangerin EarlyIron Age Crete," Hesperia67, pp. 109-123. .2000. "Skeletonsin Wells: Towardsan Archaeologyof Social Exclusionin the Ancient Greek World,"in Madness,Disability,and SocialExclusion:TheArchaeology and Anthropology ofDifference,J. Hubert, ed., London, pp. 96-118. . Forthcoming.Ceramicus Redivivus:TheEarlyIronAge Potters' Field in theArea of the Classical AthenianAgora(HesperiaSuppl.31), Princeton. Papadopoulos,J.K., and S. A. Paspalas. 1999. "Mendaianas Chalkidian Wine," Hesperia68, pp. 161-188. Papadopoulos,J.K.,J. F. Vedder, andT. Schreiber.1998. "Drawing Circles:ExperimentalArchaeology and the Pivoted Multiple Brush," AJA 102, pp. 507-529. Ph. Papadopoulou-Zapheiropoulou, 1968. "KuxXa8?;:N6aoc;,"ArchDelt 20, 1965, B' 3 [1968], pp. 515522. A. 1994. Athens: Papageorgiou-Venetas, TheAncientHeritageand theHistoric Cityscapein a ModernMetropolis, Athens. Patterson,C. 2000. "The Hospitality of AthenianJustice:The Metic in Court,"in Law and SocialStatus in ClassicalAthens, V. Hunter and J. Edmondson,eds., Oxford, pp. 93112. Payne,H. G. G. 1933. Review of Kunze 1931, inJHS 53, pp. 121123. Pelekides,S. 1917. "'AvaoxaqoOaX2,1916 [1917], yipoo,"ArchDelt pp. 13-64.
198
JOHN
K. PAPADOPOULOS
Pfuhl, E. 1903. "Der archaischeFriedhof am Stadtbergevon Thera," AM 28, pp. 1-288. Pope, H. 1935. Non-Atheniansin Attic Inscriptions,New York. Popham,M. R. 1994. "Precolonization: Early Greek Contactwith the East," in TheArchaeology of GreekColonization:EssaysDedicatedto SirJohn Boardman,G. R. Tsetskhladzeand F. De Angelis, eds., Oxford,pp. 1134. Popham,M. R., and I. S. Lemos. 1995. "AEuboeanWarriorTrader," OJA14, pp. 151-157. Purcell,N. 1990. "Mobilityand the Polis,"in TheGreekCityfromHomer toAlexander,O. Murrayand S. Price,eds., Oxford,pp. 29-58. Randall,R. H.,Jr. 1953. "The Erechtheum Workmen,"AJA57, pp. 199210. Rubensohn,0. 1917. "Die prahistorischen und friihgriechischenFunde aufdem Burghugelvon Paros," AM 42, pp. 1-96. . 1962. Das Delion von Paros, Wiesbaden. Runnels,C. N. 1982. "Flaked-Stone Artifactsin Greece duringthe HistoricalPeriod,"JFA9, pp. 363373. 1988. "The Flaked Obsidian Artifacts,"in Zagora2, pp. 245249. Rupp,D. W. 1988. "The 'RoyalTombs' of Salamis(Cyprus):Ideological Messages of Powerand Authority," JMA 1, pp. 111-139. .1989. "Puttin'on the Ritz: Manifestationsof High Status in Iron Age Cyprus,"in Early Society in Cyprus,E. Peltenburg,ed., Edinburgh,pp. 336-362. Salta,M. 1991. "AttischeGrabstelen mit Inschriften:Beitragezur Topographieund Prosopographie der Nekropolenvon Athen, Attika, und Salamisvom Peloponnesischen Kriegbis zur Mitte des 4. Jhs. v. Chr."(diss. Eberhard-KarlsUniversitatTiibingen). orienSchiering,W. 1957. Werkstitten talisierender KeramikausRhodos, Berlin. Schlorb-Vierneisel,B. 1966. "EridanosNekropole.Berichteuber die Grabungen1964 und 1965 sfidlich
AND
EVELYN
L. SMITHSON
der heiligen Strasse:I. Graberund OpferstellenhS 1-204,"AM 81, pp. 1-111. Schweitzer,B. 1955. "Zum Kraterdes Aristonothos,"RM 62, pp. 78-106. Scranton,R. L. 1960. "GreekArchitecturalInscriptionsas Documents," HarvardLibraryBulletin14, pp. 159-182. Shaw,J. W. 1989. "Phoeniciansin SouthernCrete,"AJA93, pp. 165183. Shear,T. L. 1936. "The American Excavationsin the Athenian Agora, Ninth Report:The Campaignof 1935,"Hesperia5, pp. 1-42. 1937. "The American Excavationsin the Athenian Agora, Twelfth Report:The Campaignof 1936,"Hesperia6, pp. 333-381. Sheedy,K. A. 1985. "ThreeVaseGroupsfrom the Purification Trenchon Rheneia and the Evidence for a ParianPotteryTradition,"BSA 80, pp. 151-190. .1990. "Atticand Atticizing Potteryin the Cycladesduring the Eighth Century B.C.," in
ErMOYrIA:CeramicandIconographicStudiesin HonourofAlexanderCambitoglou (MeditArch Suppl. 1),J.-P. Descoeudres,ed., Sydney,pp. 31-40. 1992. "The Late Geometric Hydria and the Advent of the Protoattic Style,"AM 107, pp. 11-28. Sherratt,A., and S. Sherratt.1991. "FromLuxuriesto Commodities: The Nature of Mediterranean Bronze Age TradingSystems,"in BronzeAge Tradein theMediterranean,N. H. Gale, ed.,Jonsered, pp. 351-386. 1993. "The Growth of the MediterraneanEconomy in the EarlyFirst Millennium B.C.," WorldArch 24, pp. 361-378. Smith, C. 1895-1896. "Excavationson Melos,"BSA 2, pp. 63-76. Smithson, E. L. 1961. "The ProtogeometricCemetery at Nea Ionia, 1949,"Hesperia30, pp. 147-178. 1968. "TheTomb of a Rich Athenian Lady, ca. 850 B.C.,"
Hesperia37, pp. 77-116. . 1974. "AGeometric Cemetery on the Areopagus:1897,1932, 1947,"Hesperia43, pp. 325-390.
Snodgrass,A. M. 1971. TheDarkAge An Archaeological of Greece: Survey to theEighth Centhe Eleventh of turies B.C.,Edinburgh.
Stampolides,N. 1990a. "Eleutherna on Crete:An Interim Reporton the Geometric-ArchaicCemetery," BSA 85, pp. 375-403. . 1990b. "AFuneraryCippus at Eleutherna:Evidence of Phoenician Presence?"BICS 37, pp. 99-106. Stevens, G. P., andJ. M. Paton. 1927. TheErechtheum (with contributions by L. D. Caskeyand H. N. Fowler), Cambridge,Mass. Strom, I. 1992. "Evidencefrom the Sanctuaries,"in KopckeandTokumaru 1992, pp. 46-60. Technau,W. 1929. "Griechische Keramikim samischenHeraion," AM 54, pp. 6-64. Televantou,C. A. 1993. "'AvpoS;: '0 yO?()[tg?Tex6; olixLo.[6l;
-i
'TXrl ," 'Av8ptiaxrXpovwxa21, pp. 187-208. . 1996. "Andros:Antico insediamento di Ipsili,"in Le cicladied il mondoegeo,E. Lanzillottaand D. Schilardi,eds., Rome, pp. 79100. .1999. ""Av8pos:T6o[ep06'trrlx;," in )QZKrKAAAIKON: l Mvrj
Nixo6AaooZacqptpeioTo oo,
Athens, pp. 132-139. TheraII = H. Dragendorff,ed., Thera. und Vermessungen, Untersuchungen, in den 1895Jahren Ausgrabungen 1902 II: Theraeische Graeber,Berlin 1903. Thompson, D. B. 1937. "The Garden of Hephaistos,"Hesperia6, pp. 396425. TocraI = J. BoardmanandJ. Hayes, Excavationsat Tocra,1963-1965. TheArchaicDepositsI (BSA Suppl. 4), Oxford 1966. TocraII = J. BoardmanandJ. Hayes, Excavationsat Tocra,1963-1965: TheArchaicDepositsII and Later Deposits(BSA Suppl. 10), Oxford 1973. Tsiafakis,D. 1998. H Opox7raoqv a-rrx>jcxovoypacpiavcoo5OS atcova 7r.X.: HpooarCyyGarc'r
AO ag(7X?ct Ovac; xact@QOpax, Komotini. Tsountas,C. 1898. "KvxXa8Lxd," ArchEph1898, pp. 137-212.
CULTURAL
BIOGRAPHY
OF A GEOMETRIC
AMPHORA
. 1951b. "AnIndustrialDistrict of Ancient Athens,"Hesperia20, pp. 135-288. Zaccagnini,C. 1987. "Aspectsof Ceremonial Exchangein the Near East duringthe Second Millennium
.1899. "KoxXaLxa, II,"
ArchEph1899, pp. 73-134. Villard,F. 1993. "Lalocalisationdes atelierscycladiquesde ceramique geometriqueet orientalisante,"in Dalongeville and Rougemont 1993, pp.143-165. Walter-Karydi,E. 1972. "Geometrische Keramikaus Naxos,"AA1972 (JdI87), pp. 386-421. . 1997. "AiginaversusAthens? The Case of the ProtoatticPottery on Aigina,"in Oakley,Coulson, and Palagia1997, pp. 385-394. Whitehead, D. 1977. TheIdeologyof theAthenianMetic,Cambridge. . 1984. "ImmigrantCommunities in the ClassicalPolis: Some Principlesfor a SynopticTreatment,"AntC11984, pp.47-59. Wide, S. 1897. "NachlebenmykenischerOrnamente,"JdI22, pp.233-258. Wiesner,J. 1938. GrabundJenseits: im dgdischen Raum Untersuchungen zur BronzezeitundfriihenEisenzeit, Berlin. Wilamowitz-M6llendorff,U. von. 1887. "Demotikader attischenMetoeken," Hermes22, pp. 107-128,211-259. Willemsen, F. 1963. "ArchaischeGrabmalbasenaus der athenerStadtmauer,"AM78, pp. 104-153. Winter, I. J. 1995. "Homer'sPhoenicians:History, Ethnography,or LiteraryTrope?[A Perspectiveon Early Orientalism],"in TheAgesof Homer:ATributeto Emily TownsendVermeule, J. B. Carterand S. P. Morris, eds., Austin, pp. 247271. Young,R. S. 1939. Late Geometric Gravesand a SeventhCenturyWell in theAgora(HesperiaSuppl.2), Princeton 1939. . 1951a. "Sepulturaeintra urbem,"Hesperia20, pp. 67-134.
B.C.," in Centre and Periphery in
theAncientWorld,M. Rowlands, M. T. Larsen,and K. Kristiansen, eds., Cambridge,pp. 57-65. Zagora1 = A. Cambitoglou,J.J. Coulton, J. Birmingham,andJ. R. Green, Zagora1: ExcavationSeason1967: StudySeason1968-9, Sydney 1971. Zagora2 = A. Cambitoglou,A. Birchall,J. J. Coulton, andJ. R. Green, Zagora2: Excavationof a Geometric Townon theIslandofAndros.Excavation Season1969: StudySeason 1969-1970, Athens 1988. Zapheiropoulou,Ph. 1970. "KuxXa8es: Aovo6o," ArchDelt24, 1969, B' 2 [1970], pp. 390-393. 1971. "rFo,zTpLx o6X6ptooG)a
it;KuxXa8eS,"AAA 4, pp. 210-216. .1973a. "KuxXarO;: Aovo6oa," ArchDelt25, 1970, B' 2 [1973], pp.426-428. 1973b. "'Anr6 T6vy?to)zx?ptx6v ouvotxLo[a6v-S; Aovo6ovsi,"A4A6, pp.256-259. .1975. "KuxXadrs:Aovo6oac," ArchDelt26, 1971, B' 2 [1975], pp.465-467. . 1983. "rFcoziTpLxa OCr6 a&yyrxoc Ti1Neo,"ASAtene 61 (n.s. 45), pp.121-136. 1985. IlpoSpAlTara T17 , Athens. r4Alaxrjayy?Ctoypcco(ac .1994. "noXTTlOtLX(x; arxColS
tro Nadou [Le P68o,"in 1paxrwxaC A' ravevX7vtIoo t, aovv8pcoo O!a ",
Nd6o (; (a6?0 ao0 T(rv aitvowv," OtAot,3-6 Ze'rrrcypp(oo 1992,
I. K. Promponasand S. E. Psarras, eds., Athens, pp. 229-261.
and Evelyn LordSmithsont JohnK. Papadopoulos UNIVERSITY OF CALIFORNIA, DEPARTMENT THE A2IO LOS
COTSEN
OF CLASSICS INSTITUTE
Los ANGELES
AND OF ARCHAEOLOGY
FOWLER ANGELES,
CALIFORNIA
[email protected]
I99
90095-151O
HESPERIA
71 (2002)
Pages 201-207
A
PUBLIC DRUM
COLUMN FROM
A
CORINTHIANQUARRY
ABSTRACT Thisstudypresentsatwo-letterinscription,AA,anabbreviation ofthe Greek carvedon the liftingbossof a drumabandonedin a wordmeaning"public," evidence The inscriptionconstitutesthefirstepigraphical Corinthianquarry. for publicownershipof Corinthianlimestonequarriesandaddsto ourunderstandingof the legalstatusof quarriesin pre-RomanGreece.
1. I would like to thank the three anonymousHesperiareviewersof this articlefor their helpful comments and suggestions.I am also gratefulto Leda Costaki andJeannieMarchand. 2. In July 2001, I showed the quarry to Chris Hayward,who will include it in his geological study of the oolithic limestone quarriesof the Corinthia. 3. Neither Sakellariouand Pharaklas (1971) nor Wiseman (1978) make any referenceto it. 4. More indicativefor dating purposes is the form of the alpha,with a horizontal,ratherthan slanted,crossbar.In this form it commonly occursin Corinthianinscriptionsfrom the mid5th centurydown to the Roman period, when it is increasinglyreplacedby the alphawith a brokencross-bar.
The marly terracesabove the west bank of the Longopotamos River,south of the modern village of Assos in the northern Corinthia, preserve at an altitude of 112 masl a pocket of limestone extensively quarriedin antiquity (Fig. 1).1The quarrysite comprises an area of ca. 25,000 m2 and includes contiguous pits of a maximum visible depth of 3 m (Fig. 2). Some quarrying also took place along the edge of the cliffs toward the Longopotamos gulley to the east and the lower terrace to the north. The area is now cultivated with olive trees, while the surroundingarea,where there are no visible signs of quarryingactivity,is planted with vineyards.It was manifestly an important quarryof oolithic limestone.2 Despite the size of the quarry, it has escaped the attention of surveyorsof the Corinthian countryside.3 On the northern side of the quarrystands an unfluted column drum with three lifting bosses preserved (Fig. 3). The diameter of the drum is 1.29 m and its height 0.98 m. A fourth lifting boss is not preserved;the corresponding part of the drum has been removed by an irregular cut, triangular in section, ca. 0.34 x 0.30 m, extending throughout the height of the stone. Part of the remaining surfaceof the drum shows considerable weathering in the form of cavities and holes-a sign of the friability of the stone. Perhaps because of this deficiency, the quarryoperators decided to abandon the drum; at a later date someone apparently came and cut a small section of it from its least weathered side. One of the lifting bosses (H. 0.27, W. 0.26 m) preserves two carved letters: AA (Fig. 4). The inscribed surface is 0.12 (H.) x 0.21 m (W.). The delta is 0.12 x 0.12 m, the alpha 0.12 (H.) x 0.08 m (max. W.). The strokes of the letters are ca. 5 mm wide and 3-4 mm deep. Based on the form of these letters, the inscription can be dated to between the 5th and 2nd centuries B.C.4
202
YANNIS
A.
LOLOS
quarry
*
rivers streams
t ^ 7R
'n
l6lmo
/' v
/
A /
W[
A
r"~' ., 1~ \^\ 'fl100 "~
contours ancient boundaries
modern roads
contourinterval ~m
Figure 1. Map of the area based on the 1:50,000 sheets of the Hellenic Army Geographical Service. Y. A. Lolos
Figure 2. Partial view of limestone quarries south of the village ofAssos. Looking north, toward the Corinthian Gulf and Perachora. Photo author
A PUBLIC
Figure3 (above).Columndrumwith inscribedliftingboss.The Corinthian Gulf is in the background. Photo author
Figure4 (right).Detail showingthe inscribedliftingboss. Photoauthor
5. For examplesin the Doric dialect, see SEGXLVI 568: &a(67otov)carved on a vase from Hyampolis;SEG XL 298bis: 8a(i6otov) stampedon storage amphorasfound in the temenos of the Temple of Aphaia;Pagano 1995: stampedtiles from Pallantion(Arcadia), all belonging to the roof of Temple C (cf. SEGXLV 364-383); Themelis 1969, p. 352: AAinscribed on roof tiles from the 6th-centuryB.C. Sanctuaryof Poseidon near Kalamata; Lolos, forthcoming:AAMOZIO, rupestralinscription. 6. Guarducci1974, pp. 378-379. Contractors'markscarvedon blocks of the Argive Heraion and the Temple of
COLUMN
DRUM
203
It is clear that we are dealing here with a typical abbreviation of the word for public (AAMOSION) in the Doric dialect. Many examples of "publicinscriptions"are known, often found in abbreviatedforms, on potsherds and roof tiles; one example was recently discovered on a vertical rock-cut surface in the Sikyonian countryside.5No such inscription, however, has been previously reported from a quarry site, or from a quarry block elsewhere. Although hundreds of lifting bosses on ancient monuments were left untrimmed, perhaps the most well known being those on the Athenian Propylaia, again no inscription such as that found in the Corinthia has been reported. The rare pre-Roman inscriptions (as opposed to graffiti) that we do know from quarrysites are names or initials of individuals, such as a certain Pytharchos who claims ownership of a quarry not far from Persepolis: IIYOAPXOEIMI. Guarducci dated this inscription to the 5th century B.C., and maintained that Pytharchos was a contractor directing work in the area of the quarry.6 Apollo at Delphi have been dated to the 5th and 4th century,respectively (Pfaff 2001). Pfaff (2001, p. 152) raises the possibilitythat the names of the contractorswho suppliedthe material for the Argive constructionsite "survive from the time the blocks were first extractedfrom the quarry."
Fromthe Romanperiodin Greece
we have at least two cases of a personal name carvedon the side of a quarry. Ober (1981) presentstwo rock-cut inscriptionslocated in an areaof ancient marblequarrieson Mt. Hymettos. They both readKEOHrOT,the genitive form of Ker0lyo;,which Ober rightly takes as the Greek form of the Latin
cognomen Cethegus, who was perhaps a memberof the Roman patrician branchof the Cethegi. Ober (1981, pp. 68-73) makes the attractivesuggestion that Kethegoswas either the owner or the lessee of this plot of land, possiblyduringthe Augustanperiod, and that he intended to exploit it as a quarry.The second name associated with a quarryand dated to the Imperial period is AIOKAHC,inscribedon the wall of a quarryoutside Karystos. Lambraki(1980, pp. 46-47), who publishedthe inscription,does not discuss the statusof Diokles, who was presumablythe contractorof the specific quarry.
YANNIS
204
A.
LOLOS
The uniqueness of the inscription from the Corinthian quarrymakes its interpretation difficult. The main question that we need to address is whether the drum only or the quarryas a whole was public. In the former case the state of Corinth would have ordered a certain number of blocks, each of which was labeled "public,"from the private quarry.The second possibility is that the entire quarrybelonged to the state and that the blocks coming out of it bore the "public"signature.The lack of published comparandafrom the Greek world does not allow for a definite answer.During the Roman Imperial period, official inscriptions (dittacia) in quarries belonging to the emperor were carved or painted on extracted blocks.7 One is tempted to believe that a similar practice was common in earlier periods. Accordingly, our column drum would have been the product of a public quarry. This suggestion should be checked against what we know about the legal status of quarriesbefore the arrivalof the Romans, a longstanding issue in the scholarly discussion of ownership of mineral lands in Greek antiquity. We have very few literary sources and inscriptions containing explicit or implicit reference to ownership and administration of quarries for the Classical and Hellenistic periods. The extensive stone quarriesof Syracuse served as a prison for the defeated Athenians and their allies, which indicates that they were owned by the state (Thuc. 7.86.2). A similar fate was later decreed for the Syracusanprisonerswho "wereimmured in stone quarriesin Peiraieus"(Xen. Hell. 1.2.14). Once again the quarries in question must have been state property.Quarries of Peiraieus appearin epigraphical documents of the 4th century. In a decree concerning the administration of the Sanctuary of Asklepios, the demos resolved that the administratorsof the sanctuaryshould use the money generated from working the quarries in order to pay for the preliminary sacrifices and the construction of the sanctuary.8A second public document, which contains records of confiscated properties, refers to "anothersurety for a stone-quarry in Peiraieus."9Here again, the quarriesin question were not in private hands. Two decrees, also of the 4th century,indicate that quarriesat Eleusis These decrees deal with the lease of quarries belonged to the sanctuary.10 to Herakles-in-Akris for a period of one or sacred (,ia0ootL Xaxoro[tv) two years.The monies of the lease were collected by the demarch and subsequently used for the festival of Herakles-in-Akris. This procedureindicates further, as pointed out by Koumanoudis and Gofas, that quarries belonging to sanctuarieslocated in the territory of a deme were administered by that deme. Quarries are mentioned in one more case in the context of leasing sacred property, this time from the town of Herakleia in 7. See Dworakowska 1983, pp. 99-
a&7 Toi &opyupio To ex Tx XlOro-
104; Fant 1989, passim. 8. IG II247 (cf. Syll.3144), lines 2832: ?)( pioeaL-rx)t 1lt -roU; oVe mLxTpo06ev -& arxaro; xoO 'A7xlIX7ri?Lio
iE[i]o[.. .]o.l[.]o EatlooUiLvo,xO8 &Xko&pyoplov[xxa]x-a[pa]XXl[e]v S; qrivo[xoouidavTxo lp06. 9. Meritt 1936, p. 401, no. 10, lines
& elyTxc 06uoaxa-r
138-140: xa. ?x-?p[a]vivy6rlv Xlt0oo-
[E6]068YIillo;
v x[ai] 4[lav ?L IEllI]aei T?TzXPTn ic?Vnqv 68o0-Ta6[tax ?xalco]iv T]V
xarapoX/iv. 10. Koumanoudisand Gofas 1978; the decreesdate to 332/331 B.c.
A PUBLIC
COLUMN
DRUM
205
southern Italy, in a 4th-century inscription concerning lands sacred to Dionysos (IG XIV 645). According to one clause of the contract (lines 137-138), the lessee is not to open any tufa quarries in the sacred land (oi68
-To90CvcaS?gV T&LLap&tcy&at7otrloE
o68e
rXXov iao?CE);other-
wise, he will be held responsible for having damaged the land. Similar prohibitions are found in a record of the lease of public land from Ephesos (I.Ephesos3). According to that contract, which dates to ca. 290 B.C., the state rents out the land but keeps the quarries(hard stone and poros) that exist within the area in order to exploit the stone for building roads: ?Yoxalpa& 7C LCpva 0ooa ?V?7Ttv IeivTYL YTLt TcoaCvtL, xao6[eOa 8? XaToz6talo, (lines 11-12). Such prohibitions and reservations should not surprise us, given the constant need of Greek states for a supply of stone. When stone was not availablein public lands, the state tapped private estates. Thus, an inscription from Troizen, dating to the mid-4th century B.C., records that stone used for the city walls was extracted from private land.1' The rarity of leases of publicly owned quarriesin official documents of Greek cities indicates that renting out quarrieswas not a normal practice.12As Burford observed, "there could after all be little point in leasing out quarriesto private contractors if the stone they were to extractwas to be used for public works in the city."13Indeed, large quarriesyielding sizable blocks could hardly serve any other purpose in Greek antiquity than to be used in state-commissioned projects for the erection of religious and secular monuments.14 The drum inscribed "public" from the Corinthian quarrymust have been destined for a large building, such as a stoa, propylon, or temple.15It does not fit any known building at Corinth,16although most ofpre-Roman Corinth was destroyed by Lucius Mummius, and we therefore do not have many Greek buildings with which to attempt to associate the drum. It is possible that the drum was meant for export, for example to a Panhellenic sanctuary-Delphi, Epidauros, or Olympia.17The 4th-century accounts of the naopoioi of the 11. IG IV 823, line 33: - - - -][y 6v oLo1100Cpoo00, t) oLX0Oo &op[c]C i]
12. As arguedby Osborne (1985, who makesthe observation 104), p. that even in the case of the Eleusinian quarrysacredto Herakles-in-Akris, "the quarrycertainlyhad not been rented out before,since the deme honored the man who suggested the leasing for making the suggestion." 13. Burford1969, p. 174. 14. Cf. Burford1969, p. 174: "thereis nothing to show what charges or controlswere imposed on quarries at Corinth.The quarrieswere very likely state-owned,having been exploited from the first chiefly for public works."
15. The fact that the drum is labeled "public"does not necessarilymean that it was made for a secularmonument; temples were also consideredpart of the public domain.Thus, fragments of roof tiles inscribedAAMOZIOS AOANAEwere found at the Menelaion of Sparta:Catling 1975, p. 267; cf. the "public"tiles of Temple C in Pallantionand from the Sanctuaryof Poseidon near Kalamatamentioned above,n. 5. 16. The diameterof the drum is significantlylargerthan that of any column found at Corinth, with the exception of the monolithic columns of the Temple of Apollo and the column fragmentof what Dinsmoor (1949) called "thelargesttemple of
the Peloponnese,"found rebuilt into the Venetianfortifications ca. 500 m northwestof the Temple of Apollo. 17. We should excludethe neighboring cities of Kleonai and Sikyon as potential importersof the stone. Kleonaihas in its territoryan enormous quarry,situatedon modern Mt. Drymoni and recentlymapped and studiedby Marchandas part of her dissertationresearch(Marchand, in prep). Sikyon also possessed ample sourcesof limestone and conglomerate quarries;I am currentlystudyingthese quarriesand plan to include a discussion of them in a monographon the topographyof that city.
206
YANNIS
A.
LOLOS
Temple of Apollo in Delphi and the Tholos accounts of the Asklepieion in Epidauros come readily to mind: payments made to Corinthian stonecutters (Xaro6toL)and stone carriers(XL0aywyoi)of poros stones are mentioned in four Delphic inscriptions, while transportation of stone from Corinth to Epidauros is mentioned in two Epidaurian inscriptions, proof that Corinthian limestone did travel extensively abroad.l8 Burford suggested that Greek states "mayhave retained quasi-regalian rights over all quarries";the examples of Herakleia, Ephesos, and Troizen point in this direction, and also bring to mind the status of mines claimed by the state.19Burford'ssuggestion may also apply to the status of quarries under the Roman Empire. From the reign of Augustus onward the most important quarriescame increasinglyinto the emperor'spossession.20The inscription presented here from Corinth constitutes the first epigraphical evidence for public ownership of Corinthian limestone quarries.21At the same time, the fact that the column drum was marked as public suggests that there was private activity in quarriesas well, at least in the Corinthia. Such activity presumablytook place in smaller quarries,and certainly not in a large quarrysuch as that in which this drum was found. 18. Bousquet 1989, no. 31, lines 98, 101-102; no. 56 III, lines 15-19; no. 59; no. 62 IIA, lines 1-2. For Epidaurossee IG IV2 103, lines 10-11, 40-41, and an inscriptionfound in the mid-1980s and brieflyreportedby Kritzas(1987, p. 14). Moreover,Burford (1969, passim) recordsseveral occasions,not only in the Sanctuaryof Asklepios but also in the Sanctuaryof Apollo Maleatas,in which Corinthian limestone was used.The formulasused in the Delphic inscriptionfor quarrying and transportingthe Corinthianporos (rotlYixai xoIlt8l) have been understood by Francotte(1900) as indicating public ownershipof the Corinthian for the quarriessince no price (xtCCJ) materialis mentioned. Referencesto price and transportation(-rliqxai xoLLai/)of quarried stones appear in the
accountsof the epistataiof the sanctuaryat Eleusis (IG II2 1672, lines 5354; 1673, lines 2 and 5). According to Francotte(1900, p. 180), the specific Eleusinianquarrieswere private whereasCorinth owned the quarries and chargedthe sanctuaryat Delphi only for the cutting and transportation of the stone. Francotte'sargumenthas been shattered by Ampolo (1982), who showed that the price of the stone was included in the price of its quarrying:whenever
the price of the stone (rtlIl) is mentioned, the price of the quarrying is omitted, and vice versa. (TolaYl) Even more persuasiveis the evidence of IGII2 1672, lines 49-50: the cost of purchasingeach of the 304 stones (xItiaxro XOoo)is given, and then the total cost of their quarrying(roil), which correspondsto 304 times the price of the stone. In other words,the cost of quarryingincludes the price of the quarriedstone:Ampolo 1982, p. 255, n. 31. We still, however,do not know why a few inscriptionsmention and xo[iL8i while the majority refer to Tolyl and xojitL&.
-tji
19. Burford1969, p. 174. The whole region of Laurionwas owned by the Athenian state, and mining sites were let to privateentrepreneursfor a stated numberof years:see Francotte1900, p. 177; Michell 1957, pp. 104-107. 20. See Dubois 1908, pp. 99-133; Larsen 1938, pp. 462-465; Ober 1981, p. 71; Dworakowska1983, pp. 27-28; Marc 1995, p. 34. 21. Glotz's thesis (1926, p. 152), that "hard-stonequarries[such as marble]were workedby the state,while pits of soft stone [such as conglomerateand limestone] belonged to the owner of the ground,"should now be put to rest.
A PUBLIC
COLUMN
DRUM
207
REFERENCES Ampolo, C. 1982. "Le cave di pietra dell'Attica:Problemigiuridici ed economici,"Opus1, pp. 251-260. Bousquet,J. 1989. Les comptesdu quatriemeet du troisiemesiecle (Corpus des inscriptionsde Delphes II), Paris. Burford,A. 1969. TheGreekTemple Buildersat Epidauros,Liverpool. Catling, H. W. 1975. "Excavations of the British School at Athens at the Menelaion, Sparta,1973-75," LakonikaiSpoudai2, pp. 258269. Dinsmoor,W. B. 1949. "The Largest Temple in the Peloponnesos,"in Commemorative Studiesin Honorof LeslieShear(Hesperia Theodore Suppl. 8), Princeton,pp. 104-115. Dubois, C. 1908. Etudesurl'administrationet exploitationdescarrieres, marbres, porphyre,granit, etc.,dansle monderomain,Paris. Dworakowska,A. 1983. Quarriesin RomanProvinces,Wroclaw. Fant,J. Clayton. 1989. Cavumantrum Phrygiae:The Organizationand Operationsof theRomanImperial MarbleQuarriesin Phrygia(BAR-IS 482), Oxford. Francotte,H. 1900. L'industriedansla GreceancienneII, Brussels. Glotz, G. 1926. AncientGreeceat Work:
An EconomicHistoryof Greecefrom theHomericPeriodto theRoman Conquest,London. Guarducci,M. 1974. EpigrafiagrecaIII, Rome. Koumanoudis,S., and D. Gofas. 1978. "Deux decretsin6dits d'Eleusis," REG 91, pp. 289-306. Kritzas,C. 1987. "Z6ro[yo XpoVlx6 -Tovepeuvcv oTo 6apao Tn; Elmin H Srod Too Apdaroo aa6opoo," o-ro AaxArcmdo 'qq Emrtaxpou (Oia6&a Epyaoiaqa yla Tnv Suovrix-ov Mvrqicixv Txo AoxXrcTlpYqoYq
eLooEmTCaupou), Athens, pp. 1114. Lambraki,A. 1980. "Le cipolin de la Karystie:Contributiona l'etude des marbresde la Grece exploites aux epoques romaineet paleochretienne,"RA 1980, pp. 3162. Larsen,J.A. 0. 1938. "RomanGreece," in An EconomicSurveyofAncient RomeIV,T. Frank,ed., Baltimore, pp.259-498. Lolos, Y. Forthcoming."FTr ar8oatoa: Mioaimtypaoplxi uapxupia,"Horos.
Marc,J.-Y. 1995. "Who Owned the Marble QuarriesofThasos during the ImperialPeriod?"in TheStudy ofMarbleand OtherStonesUsedin Antiquity(Asmosia3), Y. Maniatis,
YannisA.Lolos AMERICAN SCHOOLOF CLASSICAL STUDIES AT ATHENS 54
SOUIDIAS
ATHENS
io6
GREECE
[email protected]
STREET
76
N. Herz, and Y. Basiakos,eds., London, pp. 33-37. Marchand,J. In prep."Well-Built Kleonai:A History of the Peloponnesian City Based on a Surveyof the Visible Remains and a Study of the Literaryand Epigraphic Sources"(diss. Universityof California,Berkeley). Meritt, B. D. 1936. "GreekInscriptions,"Hesperia5, pp. 355-441. Michell, H. 1957. TheEconomicsof AncientGreece,2nd ed., Cambridge. Ober,J. 1981. "Rock-cutInscriptions from Mt. Hymettos,"Hesperia50, pp. 68-77. Osborne, R. 1985. Demos:TheDiscovery of ClassicalAttika,Cambridge. Pagano,M. 1995. "Catalogodei reperti dello scavo 1940,"ASAtene 68-69, 1990-1991 [1995], pp. 132143. Pfaff, C. 2001. "Fifth-CenturyContractors'Marks at the Argive Heraion,"JHS121, pp. 150-153. Sakellariou,M., and N. Pharaklas. 1971. KopLev(a-KAsctval (Ancient Greek Cities 3), Athens. Themelis, P. 1969. "'IepovHoostL&Ovoq AAA 2, tl; 'AxopiLxxaKaXataxxag," pp.352-357. Wiseman,J. 1978. TheLand of the AncientCorinthians,Goteborg.
HESPERIA
71 (2002)
Pages 209-229
AND INSCRIPTIONS IN ICONOGRAPHY THE
MONUMENTSOF
THE
TH
RACIAN
RI
DER
ABSTRACT The Thracian rider monuments are either funeraryor dedicated to various deities.The inscriptionsprovidethe only certainway to identifythe deities or the monument'stype. After examiningthe relationshipbetween inscriptions and iconography,I suggest in the presentstudythat the horsemanis an iconographicalconvention for a god/hero, and that his iconographyis borrowed from Greek art. Interpretingthe horsemanas a conventionalimage obviates the currentview that he representsa multifunctionalgod conflatedwith nearly every Greek, Roman, Thracian, or Eastern divinity, and produces a better understandingof both the monument type and cult. The monuments of the so-called Thracian rider present an extreme case of the relationship between epigraphyand art:the inscriptions are the only certain way to clarify the iconography, identity, and cult of the Thracian horseman.1Moreover, the inscriptions frequently provide the only reliable evidence to determine the type of monument (votive or funerary),since in many instances the findspot is of little help-most reliefs are found in a secondary context, and their function is unclear.2 The term "Thracian rider"relief is used to describe monuments of varying size, most typically stone slabs 30-40 cm wide and 20-30 cm high, which depict a rider, turned to the right (seldom to the left), his horse 1. The present study originatedas a seminarpaperfor Carol Lawton's courseon votives, given at the American School of ClassicalStudies at Athens in the spring of 2000. An abridgedversionwas deliveredin San Diego on 6 January2001 at the Annual Meeting of the ArchaeologicalInstitute of America, as part of the joint APA and AIA panel "Epigraphyand the Arts." I owe immense gratitudeto both
Carol Lawton of LawrenceUniversity and Kevin Clinton of Cornell University for encouragingme to pursuemy researchon this fascinatingsubject and for providingextremelyhelpful and detailed comments on numerous occasions.I would also like to thank Dimitar Boyadzhievand Nikola Theodossiev of Sofia Universityand JeffreyRusten of Cornell Universityfor their valuablesuggestions.I greatly appreciatethe expertcriticismof Zlato-
zara Goceva and Dimitar Popov, despite our disagreementregardingthe so-called interpretatio graeca.I am also to the grateful anonymousHesperia reviewersfor their corrections,all of which substantiallyenhancedthe clarityof my argument. All translationsare my own. 2. See the discussionof specific monuments and their findspotsbelow, pp. 211-220.
NORA
210
DIMITROVA
walking, galloping, or standing still. The rider is called "Thracian"because he appearson more than 2,000 reliefs3from at least 350 locations in Thrace, its neighboring territories,and other places characterizedbyThracian presence.4The earliest surviving monuments are Hellenistic, but the majority date to Roman times. The most popular scenes have been classified as A) the horseman facing a woman, an altar,and a snake-entwined tree, his horse walking or standing still; B) the horseman galloping and attacking a boar; and C) the horseman returning from hunting, carrying a deer.5In addition, there are many variants,and type C can be viewed as a subtype of type B. There is no strict geographical or temporal pattern in the distribution of these types. The identity of the depicted horseman is unknown. Fortunately,approximately a third of the monuments have inscriptions. They are typically below the relief, in Greek or Latin, and are generally straightforward in meaning and form: the name of the deity in the dative, usually followed by the name of the dedicant(s) in the nominative; or the name of the deceased, sometimes preceded by Dis Manibus or OeoE;KtocX0oviot;, and followed by typical epitaphial information. The inscriptions are either votive (about two-thirds of the sample) or funerary.The votive examples exhibit an unparalleled variety of names and epithets, all referring to the rider. He is called 0e6o, pox;, x6potos 0e6, x6po0;otpco, Apollo, Hades, Asklepios, Hephaistos, Sabazius, Iuppiter Optimus Maximus, Silvanus, and the Dioskouroi, in addition to numerous local names and epithets,6 such as Karabasmos, Keilade(i)nos, Manimazos, Vetespios (Outaspios), Aularchenos, Aulosadenos, and Pyrmeroulas.The words 0O6eand yipog; are often used interchangeably or both together.7 Scholars have remarkedon the extraordinaryvariety of attributes apparently possessed by the Thracian horseman. Ivan Venedikov's description is representative:the Thracian horseman appearswith the attributes of nearly every Graeco-Roman deity, ranging from Aphrodite's shell to Ares' helmet, from the Sun's radiate crown to Asklepios's serpent staff, from Apollo's lyre to Silvanus'ssaw, to name only a few.8 3. This numberis based on published monuments;since many others are as yet unpublishedand others are being excavated,a precisetotal is impossible to calculate.The main collections (corporaand catalogues),which complementeach other,include CCET;LIMC Kacarov"Denkmaler"; VI.1, 1992, pp. 1018-1081, pls. 673719, s.v. Heros Equitans (H. KoukouliChrysanthaki,V. Machaira,P. Pantos, et al.);IGBulg;Gerov,Inscriptiones. Subsequentreferencesto KoukouliChrysanthakiet al. in LIMC VI.1 are abbreviatedas LIMC VI.1. 4. Thraciansservedin the Roman armyand are attestedin Pannonia,
Britannia,southernRussia,Egypt, Asia Minor,Tunisia,and Rome; see
CCETI,p. 1. 5. These types are defined and applied by Kacarov,"Denkmaler,"passim. They areby no means strict categories, but used for convenienceby Kacarov and the editorsof CCET.This classification is not used in the presentpaper. 6. On the epithets of the Thracian horseman,see especiallyGoceva 1992. 7. E.g., CCETII.1 335 is a dedica331 Oe0CKotion Ozx pclt'Aro6Xcovt; 11.2655 365 'pol Ka7xprvcp; rpOlvcp; and 656 Orx6 pxolBacoxl8L0a.
8. Venedikov1979, p. 2: "Obwohler nur aufwenigen Reliefs drei Kopfe hat
und von dem dreikopfigenHund der Unterwelt begleitet wird wie die trikephale Gottin des Todes, Hekate, so tragt er auf anderen-allerdings auferst selten vertretenenDenkmilern-eine Strahlenkroneauf seinem Haupte und ist flankiertvon den Busten des Sol und der Luna. In anderenMonumenten wird derThrakischeReiter aufgefait als Gott der Fruchtbarkeitund halt wie Aphrodite eine Muschel oder hat sie als Nimbus hinter seinem Kopfe. Er kann fernerangetansein mit dem Helm des Ares, kann in seiner Hand den fur Asklepios typischen Schlangenstabhalten, erscheintmit der Kitharades Apollon, dem Waldmesser
MONUMENTS
THRACIAN
OF THE
RIDER
2II
The prevailing hypothesis is that the Thracian rider is an advanced case of religious syncretism, conflated with nearly every Greek, Roman, Thracian, or Eastern divinity.9In most of the relevant scholarship he is viewed as an all-purpose god, a significant step on the way to monotheism, if not monotheism per se.'1 The main problem with the concept of religious syncretism in this case is that, although one can easily understand the phenomenon whereby two or three deities from different cultures,but with more or less similar functions, are conflated as a result of historical interaction, it is much more difficult to imagine one deity being merged with numerous other gods or heroes, whose cults and traditions are hardly compatible. How did this religious concept function? How did worshippers who set up a thank-offering or made a vow in hope of well-being, or individuals who erected a grave monument, perceive this deity?
INSCRIPTIONS
AND ICONOGRAPHY
My approach to the problem outlined above is based on a monument-bymonument examination of the relationship between inscription and iconography.The difficulties encountered can be illustratedby two groups of examples:Group I consists of different types of monuments with identical iconographical features; and Group II exhibits the same divinity's name on monuments with different iconography. GROUP
des Silvanusund dem Szepter des Zeus. So erscheintdieser eigenartige und recht ungew6hnlichethrakische Gott in Funktionenund ausgestattet mit Attributen fast allergracoromischen G6ttheiten." 9. The extensiveliteraturein which the Thracianrideris considereda result of syncretismis not quoted here for purposesof economy.One of the earlier and brieferaccountscan be found in Venedikov1963. For a more recent discussion,see Werner 1999, pp. 59121. This is also the view expressedin LIMCVI.1, p. 1066, n. 28. 10. E.g., Fol and Marazov 1977, "Onewonders if this hero was 17: p. not the representationof a universal god, reveredby all the Thraciantribes, each of which gave him a local name."
I
The first group includes the two most popularscenes in the Thracian rider's iconography: the horseman facing a snake-entwined tree and the horseman as a hunter.The first of these scenes is illustrated by CCETI 34, 40, 162, IV 29, and IGBulg IV 2134 (Figs. 1-2). These monuments share identical iconographical elements: the horseman, wearing a chlamys and holding the reins in his right hand, is turned to the right, facing a snakeentwined tree and an altar;the horse is walking (not standing still or galloping, as on other monuments). Normally such similarity would suggest that the figures depicted are one and the same. The inscriptions, however, reveal a different picture. CCETI 34 from Odessus, northeast Bulgaria, is a dedication to the hero Karabasmos: 'AyaOLtxT6X[rlt] oi 7tEQp 'HpoL Kocapao3ao[LeXoaptrgrptLOvXOLVoVOL
MCvav8pov'ATcoXtovio xcal0e6&o0ov 0Eo&0pou [xal 'At7o]jWvtov 0elTocra xaoltOItO6voV ItXOSeVOo 3' [xal -- -]pa ?Evovog xa 'Apr?fitopov
&av0rlxav. The monument was found in a Late Antique necropolis in the vicinity of Pazaren ploshtad ("MarketSquare"),apparentlyin a secondary context. If it had lacked an inscription, scholars would reasonablyhave considered it a grave monument.
212
NORA
DIMITROVA
Figure1.Thracianriderrelief. CCET 140, Odessus.National ArchaeologicalMuseum,Sofia, inv. 180. CourtesyMuseum
CCET I 40 (Fig. 1), also from Odessus, is a dedication to the hero Manimazos: above relief. "Hpto Macvtlajico below relief. 'Ec-raooq;N?Lxiov O6rct -COvoivv NLtxiov xatX'Ayaov. 06yvopog; Xaptoi7tor The monument was found at 8 Noemvri Street, near the ruins of the Roman thermae, and dates to the late 2nd century A.c.11 CCETI 162 (Fig. 2) from the Burgas region, southeast Bulgaria, is a Latin grave inscription dating to ca. A.D. 150-200: aboverelief. D]is Manibus below relief. L. Titovio L. lib(erto) Diadumeno Flavia Vera coniugi bene merenti et sibi et suis viva fecit. The inscription was found in a secondary context, in the vineyards north of the village of Balgarovo. CCETIV 29 (Tomis, southeast Rumania) is a eucharisterionfor good fortune: 'Ayce0i TUXo'q v "Hpxcoa eoXoactoxrplto av0qxav ... of (list 27 names) Its findspot 1-...-r is not specified.
11. SeeIGBulgI 78 (= CCETI40). Mihailovconsidersthe suggestionthat the ruinsin questionbelongedto a of ApolloKarabasmos far sanctuary fromcertain.
MONUMENTS
OF THE
THRACIAN
RIDER
2I3
Figure2. CCET I 162, Burgas. NationalArchaeologicalMuseum, Sofia, inv. 618. CourtesyMuseum
IGBulg IV 2134 (Pautalia region, southwest Bulgaria) is a dedication to Asklepios: BeLOusAtrOC?v[oS;] 'AoxXqItmC e?Di&l?vo; xaCl?ertLTX[oWv &v?-]
on 12. SeeMihailov'scommentary IGBulgIV 2134. 13. CCETI 30, 31, 34,40, 57, 58, 59,60,61,83,84,85,86,87,111,158, 162; II 200, 534, 586; IV 3, 5, 26, 30, 48, 63, 64, 108; V 23,25, 65, 71, 72.
The monument was found near the village of Dolna Dikanja, but the exact location of the findspot is controversial.'2In this example, as in the others listed above, if there were no inscription, one would not know to whom the monument was dedicated or if it was votive or funerary. The snake-entwined tree is one of the major iconographical features of the Thracian horseman reliefs. In CCETit appearson at least 33 of the roughly 340 monuments with inscriptions published in this corpus (a precise figure is impossible to determine, since some reliefs are broken).13The inscriptions list a great variety of names to whom these reliefs are dedicated:Karabasmos,Manimazos, Propylaios,Dosaenos, Katoikadios,Asklepios, Apollo, Paladeinenos, Tato. In addition, many of these 33 monuments are funerary. A natural approach that scholars have taken in order to explain the tree and the serpent on these reliefs has been to examine each particular case and try to understand the symbolism of the image in its various contexts. Thus the snake has been interpreted in one way in a dedication to
NORA
214
DIMITROVA
Asklepios-as one of his attributes-and in a differentway in a gravemonument-as a "chthonic"symbol.14The tree, on the other hand, has been interpreted as "atree of life" symbolizing the rebirth of nature,15a sacred tree,16or a landscape background.17There seem to be two problems with this approach.The first is the lack of good evidence for Thracian religion and especially the cult of its numerous local deities. This gives free room for speculation, and little can be proved (or, of course, refuted) with certainty.The other difficulty is the presence of the snake-entwined tree on similar-looking rider monuments from the same region and the same period. Why should one then assume that it meant different things in each case? Another way to look at this problem is to consider the snake-entwined tree not as an organic element of the ideology behind each scene, but as an imported iconographicaldetail.The image of a horsemanfacing a snakeentwined tree is well known from earlier monuments found throughout the Greek world. It occurs on a 3rd-century B.C. relief from Athens, a 2nd-century B.C. grave relief from Pergamon, a 2nd-century B.C. relief from Ephesos, and many others.18It seems logical to assume that the image was borrowed by Thracian artists under the strong and well-documented influence of Greek culture and did not evolve by itself from a native tradition. If the image of the horseman itself was borrowed from Greek art,19then it is not surprising that it appears in Thrace together with one of its most common iconographical attributes. The other extremely popular scene in the first group of examples represents the Thracian rider as a hunter.20Typically he carries a spear and is accompanied by a dog and other animals associated with hunting: boar, hare, deer, lion. As in the case of the snake-entwined tree, hunting attributes are seen on various monuments, with no apparentconnection to a specific cult. For instance, CCETI 54 (Fig. 3), dating to the late 2nd century A.C., is a grave monument found in a secondary context, on the corner of Pop Chariton and Knjaz Cerkazki Streets in Odessus: ALov6otog 'AvTcpXou 6 xoatZxptoL;.
It is noteworthy that the back of the monument is decoratedwith another relief, exhibiting a funeral banquet. This is one of the few cases in which the type of monument can be surmised from information other than the inscription. CCETI 85, from the Odessus region in northeast Bulgaria,is a dedication to the hero Karabasmos: "Hp&oL Kapaoaoapc [sic] 'Ea[xltaog] 'Aya0-qvopos ipT~C
ooTpLa(xS[T[q i]8iag xa~iTxv
Iicov &av?OixcV vacat eoxapotorptov.
The dedicant may have been the grandson of EEortato NeLxtoomentioned in CCETI 40 (Fig. 1), whose son was named 'Ayao06vop.The monument was found in the Early Christian basilica in Galata.This basilicawas built upon the ruins of a pagan sanctuary,where 16 rider reliefs had been found
14. See, e.g., Georgieva1965, pp. 119-120. She interpretsthe serpentin the Thracianhorsemanreliefs as a symbol of the forcesof nature,fertility, the underworld,and healing deities. 15. Fol and Marazov 1977, pp. 1819. 16. Georgieva1965, p. 121. 17. Georgieva1965, p. 121.
18.Forthe reliefsfromAthens, Pergamon,and Ephesos, see LIMC VI.1, p. 1025, nos. 6, 8, 9, pl. 674. See also LIMC VI.1, pp. 1031-1032, nos. 113-129, pls. 680-681, which include examplesof the snake-entwinedtree from Rhodes, Kos,Thessaly,and TroadicAlexandreia,in additionto those fromThrace and its immediate vicinity. 19. See the discussionbelow, pp. 220-224.
20. Cf. Decev1945.
MONUMENTS
OF THE
THRACIAN
RIDER
215
,
Figure3. CCET I 54, Odessus., ArchaeologicalMuseum,Varna, inv. 1545. CourtesyMuseum
(CCETI 80-95). Numbers 83-89 are inscribed and dedicated to the hero Karabasmos;80-82 and 90-92 areuninscribed;93 is corrupt;94 is a thankwhile 95 is a dedication "HpcoL Tocaocay.The IlpoortoXacp, offering "HptoL remains of the sanctuary are so meager that no meaningful information can be derived from them; the only information about the names or epithets of the worshipped deities derives from the inscriptions. The authors of CCETconclude that the hero Karabasmoswas worshipped in this sanctuary,since most of the inscriptions revealhis name. This conclusion seems logical, and one would be inclined to think that the uninscribed monuments were dedicated to Karabasmosas well. Yet, surprisingly,reliefs 94 and 95 bear other deities' names. Thus, one cannot rely completely on the findspot to indicate the deity worshipped, even in cases where the monuments come from a specific sanctuary. CCETI 123 is a dedication to the god Eisenos, 147 to Apollo, II 246 06c 0pcot,415 to the god Eitiosaros, 457 to Apollo Aulosadenos, 458 Oea AuXoaocL,483 to Apollo, 612 to Asklepios, 655 Oe itpcoLBaoaxLLOLa,IV 1 Dis Manibus, 48 Ero et D[omno], 52 and 105 are epitaphs, V 53 a dedication to Artemis, and 65 a grave monument. All of these monuments portray a hunting rider,but his identity is known only through the inscriptions.
2I6
NORA
DIMITROVA
GROUP II Let us consider now the second group of examples, in which the same divine figure-as identified by the inscriptions-appears on different reliefs. One of the most frequently named deities in the monuments of the Thracian horseman is Apollo. Among the better-preserveddedications to him are CCETI 33, II 200, and 445. CCETI 33, from Odessus, has the following text: [0Oec'Aorr6X]ovt [sic] [Kao]af3ca[aotjp] IlpioociaOv A[ -- ---]
'AI6vrcVtoMEN[- -----] [----
-]TS] 'Apto?j[t8a&poo-
-]
[a&v?0ox]av. The relief depicts a horseman dressed in a chiton and chlamys, facing an altar and a tree and holding a patera, his horse walking. CCETII 200, from Marcianopolis, northeast Bulgaria, reads: Z?VL;AoXoreveoo eoaOeC0'Ar6oXXovLt pLOiptLOV aV?0rlX?v.
This relief represents a hunting riderwearing a chiton and chlamys, holding a spear,his horse galloping; a dog, lyre, tree, snake, and altar are also portrayed. CCETII 445, from the Targovishte region, northeast Bulgaria, reads simply: Atoyevv
.....
'Aco6XXco[vI]. It portrays a rider wearing a chiton, his right hand in the gesture of "benedictio latina,"facing a woman, his horse walking. These monuments do not exhibit an identical iconography.More significantly,they do not depict an image of Apollo as distinct from any other Thracian rider. These observations also apply to monuments dedicated to other frequently named deities or heroes: Asklepios (e.g., CCETI 8, 111; II 542, 612; V 23, 24); Karabasmos(e.g., CCETI 28, 30, 33, 34, 83, 84, 85, 86, 87); Aulosadas/Aulosadenos (e.g., CCET II 457, 458, 460, 483, 484); Pyrmeroulas(e.g., Kacarov,"Denkmaler"375,602,603,608); Manimazos (e.g., CCETI 31, 40; IV 36). All these deities are represented as riders, with no discernible iconographical patterns that might reveal their identity in contrast to other Thracian horsemen. Only the inscriptions permit recognition of the addressee of the vow or the type of the monument, whether votive or funerary.21 Reliefs dedicated to Asklepios are especially illuminating. IGBulg II 569 (Fig. 4) is a thank-offering to Asklepios and Aphrodite: ['AaxXYl]7cLIxaC 'A(ppoSeiTyl.XoaptcrTPtOV.
21. The pictureis furthercomplicated by the fact that sometimes a foreign god receivesa local name or epithet:thus one can see dedicationsto Apollo Karabasmos,Asklepios Keilade(i)nos,Apollo Aulosadas/ Aulosadenos,Apollo Tetradeenos,and so on.
MONUMENTS
Figure4 (left).IGBulgII 569, Glava Panega.NationalArchaeological Museum, Sofia, inv. 3906. Courtesy Museum
Figure5 (right).IGBulgII 512, GlavaPanega.NationalArchaeologicalMuseum,Sofia,inv.3739. CourtesyMuseum
22. See, e.g., Fol and Mihailov 1979, pp. 260-261. Other examplesof the same phenomenon arefound in the Sanctuaryof Apollo at the village of Trud,Plovdivregion:IGBulgIII 1458 shows Apollo in his traditional, horselessiconography,while IGBulgIII 1457, 1460-1466, 1468-1470 show a rider.
OF THE
THRACIAN
RIDER
2I7
The relief shows a rider,turned to the right, wearing a chlamys and holding a patera, and a standing female figure of the same height, clad in a chiton and mantle.The monument dates to the 3rd centuryA.C.and comes from the Asklepieion at Glava Panega in the Vraca region of northwest Bulgaria. It was found together with numerous dedications to Asklepios, some of which represent him as the Thracian rider, others in his traditional iconography,such as IGBulg II 512 (Fig. 5): aboverelief: 'AoxXrlrCtc SZocaXrlvC below relief: 'Io6XtLoKevoou eoxaptoorlpLov. The standardinterpretationof these reliefs is to suppose that Asklepios has been conflated with the Thracian rider,who was worshipped locally as a healing deity.22There seems to be no difference, however, in function, meaning, and date between the two kinds of depictions of Asklepios. If his iconography were ideologically meaningful in each particular case, then one would have to assume that only for some of the worshippers at this Asklepieion was he a product of syncretismwith the Thracian rider,while for others he was still the traditionalAsklepios in his traditionaliconography. Such a situation does not seem plausible. If, however, one supposes that the image of the horseman was simply a convention, then only those reliefs in which Asklepios is represented in his traditional iconography, without a horse, should be considered as especially significant-perhaps an expression of greater care on the part of the worshippers, who did not set up a standard image used for other gods as well. The above examples can thus be explained by the hypothesis that the horseman is an iconographical convention for a god or hero, and that the inscriptions serve to personalize (and identify) this otherwise nameless conventional image.
218
NORA
DIMITROVA
Figure6. CCETV 23, Krupac. EthnographicalMuseum,Pirot. This hypothesis also accommodates CCETV 23 (Fig. 6), a late-2ndcentury A.C. dedication to Apollo and Asklepios found in Krupac,in eastern Yugoslavia.The relief depicts two identical horsemen facing each other, while the inscription beneath the relief reads: 'AT6oXXcovt xol 'AoaxXIqm rFato;Hp6xXoo BepocxeXrvoLg Ei0a?eVO;
a&v?0iX.23
Thus one horseman is presumablyApollo, and the other Asklepios. The relief is most easily understood if we explain the rider as a convention for a divinity of some kind, personalized only by the inscription. IGBulg III 1467 is a very interesting case:it shows Apollo standing (in his traditional iconography), and next to him a horse-as though the dedicant felt that a relief portrayinga deity would be somehow incomplete without it. The monument comes from the Sanctuary of Apollo at Trud, where Apollo is depicted in other reliefs on horseback and-only rarelyin his traditional imagery.4 The rider occasionally assumes features and attributes of the deity from whose sanctuarythe relief derives:thus, in a sanctuaryof Apollo one might find riders with a lyre.25Sometimes the rider is represented as a 23. This examplerules out the possibilitythat the Thracianrider might have been the dedicanthimself. If the horsemanwere the dedicant, then we would have to assumetwo dedicantsfor this particularrelief.The inscription,however,clearlytells us that the dedicantis one, and the gods two: "Gaius,son of Proclus,dedicated[this monument] to Apollo and Asklepios Berakelenoi,having made a vow." Another eloquentpiece of evidenceis
CCETIV 29. The text reads:"So-andso dedicateda pOS;to (or for) good fortune."The numberof the dedicators is 27, so it is impossibleto identify them with the single riderdepicted. Moreover,the text says that "theydedicated a hero,"and not "themselves." Another counterargumentis the frequent presenceof adorantson the reliefs.Their figuresare significantly smallerin scale than the horseman,and this speaksin favorof his higher status.
24. See above,n. 22. 25. See, e.g., the discussionof the Sanctuaryof Apollo Aulosadasin Konstantinovet al. 1980, pp. 142-172. On sanctuariesof the Thracian horsemansee, among others, Concev 1941; Boteva 1985; KoukouliChrysanthakiand Malamidou 1989; Ovcarov1972. A very helpful summary of sanctuariesin the Roman provinceof Thrace is given in Goceva 2000.
MONUMENTS
26. E.g., Silvanus:LIMCVI.1, p. 1046, no. 371, pl. 697; Asklepios: IGBulgII 529, 535.
OF THE
THRACIAN
RIDER
219
bearded man, usually in the context of a sanctuary of a named bearded deity.26It is impossible, however, to see a clear pattern in the adoption of attributes,or to understandwhy in some cases figures with attributeswere preferred over unspecified stylized images of horsemen. The occasional presence of riderswith specific attributesis perhaps a sign of a rathercarefully done and expensive relief. Sometimes the inscription appears incompatible with the image depicted. IGBulg III 1319 shows only one Thracian rider, represented as a hunter,while the text reads 0eoScAtoaxopot;. What seems to be even more surprisingis that on a fewThracian rider reliefs only female deities such as Artemis (CCETV 21, 53) or the nymphs are inscribed as receiving dedications. In IGBulg III 1368, for example, the relief represents a horseman, but is dedicated to the nymphs, as the inscription tells us. The relief comes from a sanctuaryof the nymphs at Bourdapa (Plovdiv region), where numerous votive reliefs have been found, the most common type of which represents three female figures. In such cases it appearsthat iconography was of even less importance-perhaps because of ignorance or economic reasons related to mass production. Mass productionmay have played an importantrole. In many instances it seems that dedicants chose a relief that had been made earlier,adjusting it for the specific purpose they wanted it to serve. Unfortunately, it is impossible to establish a chart of workshops that specialized in producing certain kinds of reliefs, since most monuments are found in a secondary context, and we possess no evidence about specific ateliers. In general, images of riders facing a snake-entwined tree are more common along the Black Sea coast than are those of hunting riders,but no strict patterns can be established. It is also difficult to discern particularchronological patterns in the distribution of iconographical types, since most reliefs are by no means securely dated. Bearing in mind the role of mass production and related factors, we must nevertheless remark that the few monuments dedicated only to female deities are difficult to interpret because the reliefs are broken, and one cannot determine whether a female figure was depicted apart from the horseman, or whether the inscription included a male deity's name as well. Other, better-preserved reliefs show both a male and a female deity, and in those cases only the male deity is represented as a rider: see, for example, Kacarov,"Denkmaler"301 (Silvanus and Epona); 309, 331, 345 (Asklepios and Hygieia); 310 (Asklepios and Aphrodite); 318 (Silvanus and Diana). These reliefs confirm the natural assumption that the rider was typically a convention for a male deity. The Thracian practice of using a standardimage for different divinities occurs in other contexts as well. The above-mentioned Asklepieion at Glava Panega, for example, contains an abundanceof votive reliefs depicting a god and a goddess. These deities have been identified as Asklepios and Hygieia (see, e.g., IGBulg II 514-525), and indeed many of the reliefs bear dedications to them and exhibit their typical iconography:Asklepios as a bearded man, holding a serpent-staff, with Hygieia frequently holding a snake.They are usually accompanied by the subsidiaryhealing deity Telesphoros, represented as a minuscule figure between them or in one of
NORA
220
DIMITROVA
the composition's corners. Some monuments in the same sanctuary,however,depict Asklepios and Hygieia, but arededicatedto Silvanusand Diana, who were worshipped by the Romanized population in the region (Gerov, Inscriptiones197, 208). This suggests that the original iconography was not overly important to the dedicants: they used an already-made relief showing Asklepios and Hygieia, but invested it with new meaning, thereby satisfying their need for a dedication to Silvanus and Diana. What mattered was only the basic formal resemblance,consisting in the depiction of a male and a female deity.A similar phenomenon occurredcenturies later, when Thracian rider reliefs were used by Christians in the cults of St. Demetrius and especially St. George.27
THE THRACIAN HERO RELIEFS
RIDER AND GREEK
Understanding the rider as an iconographic convention for a divinity is generally consistent with the suggestion that the image is borrowed from Greek art and, more precisely,from funeraryreliefs of the heroized dead. Moreover, the word qpwo;,which is so characteristicof Greek epitaphs, is frequently associated with the Thracian rider. One of the most detailed and important discussions on the subjectwas provided in 1955 by Ernest Will, who demonstrated that the image of the Thracian rider with his main attributes is a later variant of the Greek hero reliefs, which had already become a widespread iconographic convention.28 The ubiquity of rider images can be illustratedby numerous examples. Pannonia,Dacia, and Moesia arerich in monuments of the so-called Danubian riders,dating from the 1st (or early2nd) to the 4th century A.c.29The extreme scarcity of inscriptions impedes precise analysis of their nature, but three iconographical elements clearly distinguish the Danubian riders from the Thracian riders and other deities on horseback: the prostrate enemy trampled by the horse's hooves, the fish, and the ram.30Dumitru Tudor viewed the Danubian riders as mystery cult deities of Dacian origin.31I would not venture an opinion on the complicated issue of their origin and nature, but only mention that the notion of victory played a significant role in the cult and that the monuments' iconography and distribution share noteworthy similarities with Mithraic religion. For our purposes, it is important to emphasize the fact that images of deities on 27. See Hoddinott1981,pp.174175:the reliefsof theThracianrider wereconsidered iconsof St. Georgeor one tabletwasusedas an Demetrius; iconin a privatehousein Plovdiv,and aslateas 1907pilgrimssoughta cure at GlavaPanegaon St. George'sDay. 28. Will 1955, pp. 78-79: "Avant
de devenirle motifle plusbanalde thracede l'Empire, l'iconographie
etaitdeja l'imagedu Heroscavalier banaledansle mondegrecsurles deux rivesde 1'Egee.Auxfaussespr6cisions fourniesen generaldansce debat,on uneformulesimpleet peutsubstituer nette:en definitive, le motifthracene variante tardived'une repr6sente qu'une serie plus vaste dont on ne sauraitle
celleduHerosgrec."Will d6tacher, furtherconcluded(pp.105,116)that
in ThraceandMoesiatherewere no rider-gods, butonlygodsdepicted on horseback andgodsof riderpeo"Denkmaler," ples.Cf.alsoKacarov, p. 1;FolandMihailov1979,pp.261262. 29.Tudor1976,pp.78-84. 30. Tudor 1976, pp. 58-59.
31.Tudor1976,pp.232-276.
MONUMENTS
32. Petsas 1978. 33. Horsley 1999, p. 43. 34. See, e.g., Metzger 1952; Robert 1946. 35. Horsley 1999. 36. See Pfuhl and Mobius 1977. 37. See, e.g., LIMC VI.1, pp. 1032, 1037, nos. 134-136,229-230, pls. 681, 686. Coins depicting riderswere also common in Macedonia andThessaly. 38. See Thompson 1963, pp. 5657, pls. XXVII-XXVIII; Barr1996, pp.133-157. 39. Barr1996, p. 133. 40. Barr1996, p. 138. 41. Calaghan 1978, p. 21. 42. Calaghan 1978, pp. 21-22. 43. LIMC VI.1, p. 1043, no. 345, pl. 695. 44. LIMC VI.1, p. 1061, no. 597, pl. 715.
OF THE
THRACIAN
RIDER
22I
horsebackwere widespreadon monuments fromThrace'sneighboringprovinces. Images of horsemen also occur throughout the Aegean, across a wide geographical area. Photios Petsas, for example, interpreted certain rider reliefs in Macedonia as "prototypes"of the Thracian rider.32Indicative of the monuments' geographical range are examples from Asia Minor, on the one hand, and Crete, on the other, which add weight to the assumption that the Thracian rider'siconography is borrowed from Greek art. The examples from Asia Minor include reliefs of Herakles and Kakasbos represented as rider gods. Their cult probably existed in the 1st century A.C., and reached its acme in the 2nd and 3rd centuries.33The Herakles and Kakasbos reliefs were popular in Lycia, Pisidia, and Pamphylia.34A recent study on the subject has been provided by G. H. R. Horsley, who discusses in detail the "RiderGod" stelai at the BurdurMuseum in southernTurkey.35 These stelai, numbering over 100, arepredomiand the indigenous god Kakasbos, as the inof Herakles nantly images on roughly half of them, indicate. In addition to scriptions, present appearingon votive reliefs, riderswere commonly depicted on gravestones, especially in the western part of Asia Minor, as is evident, for example, from Ernst Pfuhl and Hans M6bius's published collection of reliefs.36 The horseman iconography, so popular in Asia Minor, was not restricted to lapidary monuments, but appearedalso on coins and terracotta plaques. Coins with rider images were minted in Dardanus, Magnesiaon-the-Meander, Colophon, Skepsis, and elsewhere.37AtTroy deposits of several hundred terracotta plaques have been found, depicting a hero on horseback.38They represent a beardless rider mounted on a rearing horse, with a snake beneath the horse,39and are dated on stylistic grounds from the 3rd to the 1st century B.C. It is unclear whether there was a specific hero associated with the plaques from Troy.Amy Barr concluded that the plaques themselves were a local phenomenon, while their iconographical tradition belonged to a much larger context, and cannot be explained satisfactorily, "since we find images of horse, rider, and snake in so many different topographic areaswithout an explicit pattern."40 Similar terracottaplaques have been found at Knossos, in the so-called shrine of the hero Glaukos.41They are Hellenistic in date and portray a rider (armed or unarmed), typically mounted on a rearing horse, with a snake beneath. P.J. Callaghan viewed the plaques as elements of an initiation cult for youths in honor of the foundation hero. He also noted that such images were found in "widely scattered parts of the Greek world and have long been recognized as representationsof heroes."42 The broad territory and time span over which one can find reliefs of heroes on horseback can be further illustrated by a few other examples: a 5th-century B.C. relief from Cumae (Fig. 7) with a riding hero, dressed in a short chiton and chlamys, turned to the left, his horse galloping, accompanied by a heroine and facing six adorants;43a 4th-century B.C. relief from Peiraeus with a hero, dressed in a short chiton and chlamys, turned to the left, his horse galloping, facing a bearded adorant and an altar,accompanied by two dogs;44a 3rd-century B.C.relief from Athens showing a
222
NORA
DIMITROVA
hunting horseman,wearing a chlamys and armor,turned to the right, holding a spear and attacking a boar, his horse galloping;45a 2nd-century B.C. relief from Ephesos (Fig. 8) showing a horseman, turned to the right, holding a spear,clad in a chiton and a chlamys, facing a round snake-entwined altarand a tree, his horse walking;46and a 1st-century B.C. relief from Pergamon depicting a hero turnedto the right, dressedin a chiton and chlamys, with his horse walking, facing a round table and a snake-entwined tree.47 H. Koukouli-Chrysanthaki et al. have adduced votive offerings to heroes on horsebackboth in cult places dedicated to heroes alone (Sparta,Knossos, Corinth, Pylos) and in sanctuaries of many gods (e.g., Artemis Orthia in Sparta,Athena in Lindos and Ilion, Demeter in Pergamon).48 The above examples reveal the widespread iconographical connection between horse and hero. It must be noted, however, that the image of the horse was not part of the specific cult associated with a particularhero, but rather a general attribute indicating superior status. In many societies the horse was a symbol of nobility. In discussing aristocraticconstitutions of military origin in Eretria, Chalcis, Magnesia, and "manyothers throughout Asia,"49Aristotle wrote that only the wealthy could rear horses (Pol. 1289b, 1297b, 1321a). It was no accident that hippeiswas a term applied to the Spartan elite or an Athenian property class. The horse was a signifier of a higher status, whether a member of the social elite or a hero. The significance of the horse as a heroic attribute is lucidly summarized in LIMC: "One of the characteristiciconographical types of the hero is the riding hero: its iconography is found from the Iberian Peninsula ... to
Figure 7. LIMC 345, Cumae.
Antikensammlung,Staatliche Museenzu Berlin-Preussischer Kulturbesitz inv. SK 805. Courtesy Museum
45. LIMC VI.1, p. 1052, no. 457, 704. pl. 46. LIMC VI., p. 1038, no. 254, pl. 688. 47. LIMC VI., p. 1034, no. 167, 683. pl. 48. LIMC VI.1, p. 1065. The authorscontinuewith the following observation:"Alssich die Heroisierung derToten seit der hellenistischenZeit verallgemeinerte,wurde das Grab des Toten zum Heroon, zu seiner Kultstatte." 49. See Arnheim 1977, p. 54. The authorcites furthera fragmentof HeraclidesPonticus,accordingto which each memberof the ruling class in Cumae was obliged to keep a horse.
MONUMENTS
OF THE
THRACIAN
RIDER
223
Figure8. LIMC254, Ephesos. Antikensammlung,Staatliche Museen zu Berlin-Preussischer Kulturbesitzinv.SK 810. Courtesy Museum
Asia and from Central Europe to North Africa, and comprises two basic iconographical types: I. the hero leading a horse, II. the hero on horseback."50 These two types appearin the Greek world in the late 6th century and B.C., crystallize in the 4th century B.C.Together they form an ideal image for depicting the heroized dead all over the Greek world-even on islands with poor conditions for horse-breeding.51 There can hardly be any doubt, therefore, that the iconography of the Thracian horseman belongs to Greek artistic tradition. It might be assumed that since most of the reliefs date to the Roman period, we should look for their origin in Roman sculpture. Hellenistic examples exist as well,52however, and given the exact iconographical parallels with Greek hero reliefs,it is impossible to justify a Roman originof the Thracian horseman reliefs (without ruling out iconographical similarities and artistic influence). This is consistent with the special position that Greek culture held in Thrace. The preservedGreek and Roman inscriptions from Thrace show that Roman civilization was never as wholly embraced by the local 50. LIMC VI.1, pp. 1065-1066. 51. LIMC VI.1, pp. 1065-1066. As far as the Thracianrideris concerned, however,the authorsbelieve that he was a deity with a specific cult, and not an anonymoushero (p. 1066). Additional supportfor perceivingthe horse as a typical element in Greek hero reliefshas recentlybeen providedby Larson (1995, pp. 43-53), who cites the following examples:a relief from Rhodes, ca. 400
B.C., showinga femalefiguremeetinga a contemporary reheroon horseback;
lief fromTanagradepicting a heroized woman pouringwine to a hero leading
a horse;a 5th-centuryB.C.relieffrom Cumae(seeabove,Fig.7);anda relief
from Pharsalus,with youth, horse, and
a heroine.Larsontoo regardsthe image of the horseas a symbolof nobilityand drawsparallelswith the "herostones" from India.The Totenmahlreliefs,on
whichhorsesarefrequently depicted, werededicatedfirstto heroes.Afterthe 3rd centuryB.C. they were typicallyset up for dead people as well, and so the horse, havingoriginatedas an element of aristocraticiconography,eventually became a common funerarysymbol (Larson 1995, pp. 43-53; cf. Dentzer 1982, pp. 429-452). 52. E.g., CCETI 28,29, IV 61, 77. See also CCETIV, pp. 7-8, with n. 2.
NORA DIMITROVA
224
population as was Greek: it was considered much more an alien element than the traditional Hellenic cultural presence, which dates back to the times of Greek colonization on the Black Sea and Aegean coasts. The Sapaean royal line, for example, was especially Hellenized: Rhoemetalces III was eponymous archon in Athens in 36/7 and financed bull-wrestling contests there.53
IMAGE OF THE HEROIZED
DEAD
A natural question that arises is why the image of the heroized dead was especially attractiveto the Thracians. An answer can be found in the evidence for the Thracian attitude toward death. The most famous literary account is found in Herodotos 4.94: 8' 6v6s -r6vtp63tov.ou-re 0&rtoOv,GxeLv &&outob; O0avoc-TtiouoaL TOy L E ' LivrXL &4tOVOx vo~LLomo -ce'ToX?.L4V OtapOX 6kl(XO~LV 0!. 6 cXUT&)VTOV OCL)OVToOTOV VO[ILLroumY Fr&OO'Lv. BLOC 1roual &yyeXov 7tcXp&o6'vZoXk.totv,E &&)V,T0L.rr4.ucoucn OF6J&3eoi [iv SXoucJL,0CxXol 8, &otXC4vtZ
zakxLoSLvt&;S xsCipocq xoci't obq
[LerCEXpov PiTt-ooaCF F-; t&; gi7C OU hS
ov .X6,[LsvoLT)V ccv exacoaTot
-CptO cci'rrcvTCxX0Cv-rs; &x6wvTLc
-robi ,Ciort0 j1roL:v0o u nTOCPOx -Ov 7rc6&x;, &xvaCxLv70CV-ts;OCcT6v
X6y),x;C.i"V[A' i'j&r7CO06Xcvvx S Oi' 0X OISL ?i'VO LZL. "V 5' 0'T nOO'V-q 'i 6 zoa6i2~Ao; 0E6; ~ox~rc SLtXoL.?iv as ~i 6cnto0ivn, oXLL&wcXLGj?o xocx6veivOC. &V5VAp ct6zzbvt6ov CyyrXoV,qx .trevot IIu1v
TOZOL
'LIF
They "immortalize"in the following way: they think that they do not die, but the dead person goes to the god Zalmoxis; some of them think that he is identical with Gebeleizis. Every five years they send an envoy, chosen by lot among them, to Zalmoxis, instructing him each time about their needs. They send him in the following way: some of the appointed people hold three spears. Others, having seized the hands and the legs of the person who is being sent to Zalmoxis, throw him on top of the spears, after swinging him to and fro in the air.If he dies pierced by the spears, they say that the god is favorablydisposed to them; if he does not die, they blame this messenger, saying that he is a bad man.54 53. See Robert 1982. 54. An exceptionallydetailedrecent discussionof this passageis provided by Boshnakov(2000, pp. 11-93). He favorsthe readingye BeXeL"LV, and interpretsthe name *Beleizisas meaning"god-king,"cognatewith Greek fxarlhe6;.The readingrtepXSLrLv,however, is preferable, since the
position and usage of the particleyr would be ratherawkward:yr expresses
concentration,limitation,and intensification, and is commonlyrenderedas "atleast";see Denniston 1996, pp. 114162. Its position is typicallyafter the word it emphasizes,while in the above passage vo[iLrouOL is the least emphatic
word. Furthermore,it seems too bold to deem *Beleiziscognatewith Greek r3wJLkro6, given the lack of sufficient evidencefor Thracianetymology. Boshnakovunderstandsthe messenger's
death as a "gradualinitiation of the young dynastinto priestlyranksand his renderingin serviceto the Goddess and the God duringthe fifth, sixth, and seventh year,at the end of which he became one of the principalpriestsof the male deities, i.e. Zalmoxis,... worshippedas a'god-king"'(p. 81). His interpretationis based on the famous passageaboutmysteryinitiation in Apuleius'sGoldenAss (11.23), which he
MONUMENTS
OF THE THRACIAN
RIDER
225
Later authors, such as Pomponius Mela, supply similar information, with minor variations: una gens Thraces habitant, aliis aliisque praeditis nominibus et moribus. Quidam feri sunt et paratissimi ad mortem, Getae utique. Id varia opinio perficit; alii rediturasputant animas obeuntium, alii etsi not redeant non extingui tamen, sed ad beatiora transire,alii emori quidem, sed id melius esse quam vivere. Itaque lugentur apud quosdam puerperianatique deflentur, funera contra festa sunt, et veluti sacra cantu lusuque celebrantur. The Thracians have different names and rites. Some of them are savage and meet death with delight, especially the Getae. This is because of their different beliefs: some believe that the souls of the dead come back, others that they [sc. the souls] do not perish, even if they do not come back, others believe that souls die and that this is better than if they continued to live. For that reason some lament birth and newborn babies; on the contrary-burials have a festive characterand are celebrated like sacred rites with songs and games.55 Solinus writes: concordant omnes ad interitum voluntarium, dum nonnulli eorum putant obeuntium animas reverti, alii non extingui, sed beatas magis fieri. All Thracians unanimously value voluntary death, as some of them believe that the souls of the dead come back to the upper world, while others believe that souls do not perish, but become happier after death.56
thinks elucidatesthe imageryof the Letnitsa plaques (a seriesof images associatedwith hunting, on some of which a rideris depicted;see below). This interpretationseems unconvincing, given the lack of evidence to supportthe idea that either Herodotos's passageor the Letnitsa plaqueshad anythingto do with mysteryinitiation, and the temporaland geographical divide between the texts of Herodotos and Apuleius. 2.18. 55. De chorographia rerummemorabili56. Collectanea orum10.2-3. 57. IGBulgII 796. For a detailed discussionsee Mihailov 1951. 58. See IGBulgII, p. 179.
Taken by itself, the information given by Herodotos and later sources may be judged as dubious in terms of its historical value. What it says about the Thracian attitude toward death is supported, however, not only by archaeological data, namely numerous and rich grave inventories, but also by epigraphical evidence revealing an unusually strong belief in immortality among the Thracians. A monument from the land of the Getae (Belogradec, northeast Bulgaria) carriesthe following text:57 6veOo PloxooTropsog "Hpcot "EvOoc AltvLt; 6[tzXVO xLO N6ai(pyqS[sic] T? xo6poaL rcep o 7? XX6t?XVO 0eaocz? iadqjvog' eyeveTo yap 7CoXuXoXoCTog ipCo5g &a6vaxog.
Here Dinis, son of Reskouporis, who outlived his child, dedicated [a monument] to the hero and the goddesses nymphs, after praying [to them]. He [the deceased] became a famous immortal hero. The editor, G. Mihailov, aptly commented: "The deceased has become an immortal qpco. The Thracian believed that he became after death a hero or god, i.e., a hero-horseman, and lived the life of that god. Therefore the inscription is at the same time grave and votive."58This explains why there is no iconographical distinction between funeraryand votive reliefs in the
226
NORA
DIMITROVA
caseof the Thracianrider.The gravemonumentsfunctionas dedicatory; the deadpersonhas been immortalizedand is thereforethe recipientof the dedication,even though this is not explicitlyreflectedin the text of mostinscriptions. In Greekgraveinscriptions,on the otherhand,the word &Odvoacog is almostneverused for a deadperson.The only (unlikethe word pwO;) comparableexampleknown to me is AnthologiaGraeca16.294, where Homer is called &Oavdrxot;'ao,;Oo;.
This usage, however,is quite differ-
ent fromimmortalizinganydeadpersonandthe expressionusedis weaker: not "immortalhero,"but "heroequalto the immortal[gods]."59 On the human is contrasted with txvvalx/xX6os contrary, mortality typically &O6vatov.60 Expressionssuch as "immortalsoul"and "mortalbody"are alsousedinfrequently.61 An interestingepigramfromthe Romanperiod,indicativeofThracian belief,is preservedon a gravestelefromMesambria(Burgasregion,southeast Bulgaria);one reliefon the stele representsthe deceasedas Hecate:62 'EvO6deEycbx?[i? 'Ex&Tm eos6(q ?oopas;. -tlYv xO TrcXal 3poto;, vVv ?? &OdvaTo; xal apYlpo;' 'IooXia Nc?xiou. ..
I, the goddess Hecate, as you see, am buried here. I was a mortal in the past, but now am immortal and ageless; Iulia, daughter ofNikias... The strength of the Thracian belief in immortality can be surmised also from reliefs in which the horseman shows individual features of the deceased,63for example in CCETI 15,60, and V 32, where the faces of the horsemen bear distinctive characteristics.This phenomenon can be interpreted as an attempt to personalize the impersonal, conventional image of the rider.
THE HERO ON HORSEBACK If Thracian beliefs in immortality account for the popularity of the iconography of the heroized dead, they do not yet explain why the Thracians chose from the options availableto them in traditional Greek art precisely the image of the hero on horseback (and not, for instance, the cenafunebris or any other stylized representationof the deceased). A reason for this can be found in the important role that horse-breeding had in Thracian culture, richly documented in the ancient sources. Suffice it to recall Hesiod's o (Op. 507) or Sophocles' (pLXhCotg;OpYL (Tereus,fr. O,pxlg; irc7roTp@6po 523). As noted earlier,horses generally symbolized a superiorstatus.Apart from the literary sources, images of horsemen are found on coins,64frescoes (e.g., the newly found Alexandrovo tomb),65the Lovec belt,66and the Loukovit and Letnitsa plaques, interpreted as depicting the figure of"the tribal ancestor and hero, possessor of many horses, a hunter and mighty warrior."67 This figure of the (presumablydeified) ruler-horseman can be
59. On the difference between Greekbeliefsin immortality of the soul andThracianbeliefs in deification of the deceased,see also Popov 1995, pp. 52-53. 60. E.g., CEG I 6,103,177; II 486, 645. 61. E.g.,AnthologiaGraeca7.61, 108. 62. IGBulgI 345. 63. See LIMC VI., p. 1067; Fol and Mihailov 1979, p. 263. 64. See, e.g., Fol 1979, pp. 214-215, nos. 5, 8. 65. The discoveryof the Alexandrovotomb was officiallyannouncedat the InternationalSymposium"Odryssian Kingshipand Nobility,"held in Karlovo,Bulgaria,15-18 January2001. 66. E.g., Marazov 1979, pp. 244245. 67. Hoddinott 1981, p. 171. For a new interpretationof the Letnitsa plaques,see n. 54.
MONUMENTS
OF THE THRACIAN
RIDER
227
tracedback to the 5th-4th centuryB.C.on the basisof the above-menandwaslaterrepresented tionednumismaticandartisticevidence,68 through the borrowedpictoriallanguageof the Hellenisticheroreliefs. The Thracianfondnessfor horsesand horse-breedinghas also been seen as supportfor a localoriginof the rider-reliefs.69 EarlyThracianreligion, however,was apparentlyaniconic.Ralph Hoddinott convincingly concludedin the caseof the Letnitsaplaquesthatthey"mustbe the result In addition art."70 of masteringthe importedfashionof anthropomorphic of the to the absence a nativeiconictradition, strikingsimilaritiesbetween the GreekheroreliefsandtheThracianhorsemanmonumentsin termsof featuressuchastree,serpent,andhuntingattributesstrongly iconographical a argueagainst locallyevolvediconography. CONCLUSION The reliefsof theThracianriderexemplifyimportantaspectsof the reception of Greekartby a neighboringnon-Greekpopulation.Suchreception is a complexand difficultissue,which has founddiverseexpressionsin a territoryextendingfromSpainto India,andfromthe BlackSeato Africa, dependingon the form of contactbetweenthe Greeksand non-Greeks, the stabilityand natureof localartistictraditions,andthe needsandtaste of the nativeinhabitants.The Thracianridermonumentsrevealthe process by which a certainiconographyis adopted,investedwith new meanAs John ing, andimbuedwith indigenousbeliefsandculturalpreferences. Boardmanobservedregardingthewayin whichnon-Greekculturestreated Greekart,"theirreactionsweredeterminedby theirneeds,theiropportuHe nities, and for many,not least, by the idioms of their nativearts."71 furtherremarkedthatThracianartwas "tingeddeeplyby the Greek,not leastin the ubiquitousreliefsof the Thracianriderherowho appearsas a At the same time Boardmanpoints out that ClassicalGreekcavalier."72 one shouldnot hastilylabelThracianartasprovincialGreek,butbe aware of its nativeand Orientalinspiration.73 Thracianriderreliefsare dedicatedto numerousdeities and heroes. They also appearon gravemonuments,but functionas dedicatoryin that contextas well, in view of the clearlydocumentedThracianbelief in the of the dead.The proposalpresentedhereis thatthe horseimmortalization man is best understoodas a conventionalimage for a (male)divinityof somesort.The originof this imageis to be foundin the Greekheroreliefs. Interpretingthe imageof theThracianrideras an artisticconventionfor a god or hero that does not by itself imply anythingabouta specificcult challengesthe currentview that the image representsa multifunctional for 68.I thankNikolaTheodossiev conflatedwith nearlyeveryGreek,Roman,Thracian,or Easterndigod attention. this to my bringing 69.See,e.g.,Marazov of this all-embracing 1979,p.353; syncretismcanproducea betvinity.Reconsideration Konstantinov et al.1980,passim; Popov ter understandingof both monumentand cult.I do not intendto imply konnik. 1993,s.v.Trakijski that therewas no earlierThraciancult connectedwith riders.The exist70. Hoddinott 1981, p. 171. ence andnatureof sucha cultremainanunsolvedproblem.Clarifyingthe 1994,p.10. 71. Boardman 72. Boardman iconographyof the Thracianriderthroughinscriptions,however,should 1994,p. 192. 73. Boardman1994, p. 192.
facilitate a solution to this problem.
228
NORA
DIMITROVA
REFERENCES in Arnheim,M. T. W. 1977. Aristocracy GreekSociety,London. Barr,A. 1996. "Horseand Rider Plaquesat Ilion: A Preliminary Study of the Hellenistic Hero Cult in Asia Minor,"StudiaTroica6, pp. 133-157. Boardman,J. 1994. TheDiffusionof ClassicalArtin Antiquity,Princeton. drevnost: Boshnakov,K. 2000. Trakijska Istoriceski ocerci[Thracianantiquity: Historical essays],Sofia. Boteva, D. 1985. "Obrocnipametniciot svetilishtetona Asklepij Limenos krajSlivnica,Sofijskiokrag"[Votive monumentsfrom the Sanctuaryof Asclepius Limenos nearSlivnica, Sofia region],Arheologija27:4, pp. 31-38. Callaghan,P.J. 1978. "KRS1976: Excavationsat a Shrine of Glaukos, Knossos,"BSA 73, pp. 1-30. CCETI = Z. Goceva and M. Oppermann, CorpuscultusequitisThraciiI: Etudespreliminairesaux religions orientales,Leiden 1979. CCETII.1 = Z. Goceva and M. Oppermann,Corpuscultusequitis ThraciiII.1: Etudespreliminaires aux religionsorientales,Leiden 1981. CCETII.2 = Z. Goceva and M. Oppermann,Corpuscultusequitis Thracii11.2:Etudespreliminaires aux religionsorientales,Leiden 1984. CCETIV = N. Hampartumian,Corpus cultusequitisThraciiIV: Etudespreliminairesaux religionsorientales, Leiden 1979. CCETV = A. Cermanovic-Kuzmanovic, CorpuscultusequitisThraciiV: Etudespreliminairesaux religions orientales,Leiden 1982. CEG = P. Hansen, Carminaepigraphica graecaI-II, Berlin 1983, 1989. Concev,D. 1941. Le sanctuairethrace presdu villagedeBatkoun,Sofia. Heros Decev, D. 1945. "Trakijskijat kato bog-lovec"[The Thracianhero as a hunting god], BdlgarskaAkademijana Naukite[Bulgarian Academy of Sciences] 70 (Klon 33 [Historical istoriko-filologicen and philologicalsection]), pp. 185199.
Denniston,J. D. 1996. TheGreek Particles,2nd ed., Indianapolis. Dentzer,J.-M. 1982. Le motf du banquet couchedans le Proche-Orient et le mondegrec du VIIeau IV siecle avantJ.-C., Rome.
Fol, A. 1979. "Krizatav elinisticeskija svjati obshtijatikonomiceskii politiceskiupadakna Trakijaprez III-I v. pr.n.e."[The crisisin the Hellenistic world and the general economic and political decline of Thrace in the 3rd-lst centuries B.C.], in Velkov et al. 1979,
pp.211-217. Fol, A., and I. Marazov.1977. Thrace and the Thracians, Sofia.
Fol, A., and G. Mihailov.1979. "Pogrebalniobicai:Religija" [Burialrites:Religion], in Velkov et al. 1979, pp. 247-267. Georgieva,I. 1965. "Derwildtragende thrakischeReitergott,"Eirene4, pp. 115-121. Gerov, Inscriptiones = B. Gerov, Inscriptiones latinae inter Oescumet latrum repertae,Sofia 1989.
Goceva, Z. 1992. "Les epithetes du cavalierthrace,"Linguistique balkanique35:3-4, pp. 155-180. .2000. "Lestemples dans les villes at les sanctuairesthracessur leur territoiredans la provincede Thrace," in 'AyaO6c; a'cycov. Mythes et cultes.Etudes d'iconographie en l'honneur de Lilly Kahil
(BCH Suppl. 38), P. L. de Bellefonds et al., eds., Athens, pp. 185191.
Hoddinott, R. F. 1981. TheThracians (Ancient Peoples and Places 98), London. Horsley,G. H. R. 1999. TheRider God Steles at Burdur Museum in
Turkey,New England,Australia. IGBulg = G. Mihailov, Inscriptiones graecae in Bulgaria repertae,Sofia. Kacarov, "Denkmiler" = G. Kacarov,
"Die Denkmalerdes thrakischen Reitergottesin Bulgarien,"Dissertationes Pannonicae, ser. II, fasc. 14,
Budapest1938. Konstantinov,K., et al. 1980. Monumenta Thraciae antiquae 2: Trakiiski svetilishta [Thracian
sanctuaries],Sofia.
MONUMENTS
OF THE
THRACIAN
Koukouli-Chrysanthaki,H., and D. Malamidou. 1989. "Ipo6too To ApXoto'HpcocAuXowvreiT," Aoytx6dpyo on7rMaxesov(a xat Opcxry3, pp. 553-561. Larson,J. 1995. GreekHeroineCults, Madison. Marazov,I. 1979. "Hudozhestveni zanajati.IzkustvoI duhovenzhivot" [Crafts.Art and cultural life], in Velkovet al. 1979, pp. 226-246, 349-363. Metzger, H. 1952. Cataloguedes monumentsvotifsdu MuseedSAdalia (Etudes orientales11), Paris. Mihailov,G. 1951. "Epigramme funeraired'un thrace,"REG 64, pp.104-118. Ovcarov,D. 1972. "Trako-rimsko selishte i svetilishtena Apolon pri selo Gorno Ablanovo,Targovishtko"[The Thraco-Roman settlement and sanctuaryof Apollo at the village of Gorno Ablanovo, Targovishteregion],Arheologija 14:2, pp. 46-55. Petsas,P. 1978. "SomePicturesof MacedonianRidersas Prototypes of the ThracianRider,"Pulpudeva2, pp.192-204. Pfuhl, E., and H. M6bius. 1977. Die ostgriechischen 1, Grabreliefs Mainz.
Nora Dimitrova CORNELL UNIVERSITY DEPARTMENT 120
GOLDWIN
ITHACA,
NEW
OF CLASSICS SMITH YORK
[email protected]
HALL 14853
RIDER
229
Popov,D. 1993. "Trakijskikonnik" [Thracianhorseman],in Kratka na traki/skatadrevnost Enciklopedija [Brief encyclopediaof Ancient Thrace], Sofia. .1995. Bogdts mnogoimena [The god with many names], Sofia. Robert,L. 1946. "Un dieu anatolien: Kakasbos,"Hellenica3, pp. 38-73. . 1982. "Deuxepigrammesde Philippe de Thessalonique,"JSav, pp.139-162. Thompson, D. B. 1963. Troy:The Terracotta Figurinesof theHellenistic Period(Troy,Suppl.3), Princeton. Tudor,D. 1976. Corpusmonumentorum religionisequitumdanuviorum2: TheAnalysesandInterpretationof theMonuments(EPRO 13), Leiden. Velkov,V., et al. 1979. Istorijana Bdlgarija [History of Bulgaria]1, Sofia. Venedikov,I. 1963. "Le syncretisme religieuxen Thrace a l'epoque romaine,"ActaantiquaPhilippopolitana,pp. 153-166. . 1979. "DerThrakischeReiter," in CCETI, pp. 1-6. Werner,R. 1999. "Aspekteder thrakischenKultur,"Chiron29, pp. 51-121. Will, E. 1955. Le reliefcultuelgrecoromain:Contributiona l'histoirede l'artde l'empireromain,Paris.
';,
'
>
*
e
9%
-H
*
S
JOSEPH W. SHAW, ALEYDIS VAN DE MOORTEL, PETERM. DAY, VASSI LIS Kl LlKOGLOU
RecentlyPublished
LM
A
IA
Ceramic
South-Central and
175pages,66 figures,16 tables Hesperia Supplement 30. ISBN 0-87661-530-2. July2001.Paper$35.00
*
Kiln
in
Crete:
Pottery
Function
Production
An in-depthstudyof theLateMinoanIA cross-draft kilnfoundin excavationat Kommos,Crete.The kilnis of a typepopularduringtheNeopalatialperiod,anditsgoodstateof preservation hasallowedtheauthors to speculate aboutits originalinternallayoutanduseaswellasaboutthe roofthatcoveredit. Muchof the largequantityof obviously locallyproducedpotteryfoundassociated withthekilnis analyzed in detail,allowing forthefirsttimethestudyof theshapes,decoration, andtechnical characteristicsof vasesknownto havebeenfiredin a specificLM IA kiln.The bookpresentsan integrated program of analytical techniques usedto illustratethe rangeof firingtemperatures, the compositional similarities anddifferences in theclayused,andaspectsof thefiringprocessandthe upperkilnstructure. Offeredhereis avaluable contribution to ourknowledgeof thetechnology andorganization of ceramicproduction atthebeginningof theLateMinoanperiod,whichwillforma basisforstudiesof potteryprovenience andexchange.
.#.
-.,
,-
.
'''4'.
*
#
0
w
%, q 4 ,,
te .
_
*
*
s
}
t z
>
6
*
-
*
> d
/
-
a ,
.
,
.
0
*.
.. a
.
'.
*
*
. .
a 44
.
.
o
GO'
@
C
CHARLES K. WILLIAMS 11 AND N A N CY B00 KID I5, ED ITO RS Corinth,
the
Centenary:
I896
Forthcoming
I996
attheDecember Ca.500 pages,400 illustrations of thepaperspresented twenty-five Thisbookpublishes anniver- CorinthXX theone-hundredth heldin Athensto celebrate 1996symposium at Ancient ISBN0-87661-020-8 saryof the AmericanSchoolof ClassicalStudiesexcavations July2002.Clothca. $75/ £57.50 therangein subjectmatter Corinth.Thepapersareintendedto illustrate by scholarsof AncientCorinth, beingundertaken currently of research workfor andtheirinclusionin onevolumewillserveasa usefulreference Eachof the topics whichvarywidelyfromCorinthian nonspecialists. by the geologyto religiouspracticesto Byzantinepottery is presented expertin thatarea. acknowledged of articlesandvolumes Thebookincludesa fullgeneralbibliography of onehundred at Corinth.As a summary excavated material concerning to come. of scholars to generations useful it willbe yearsof research,
HESPERIA SUPPLEMENTS 13 MarcusAurelius:Aspects of Civic and CulturalPolicyin theEast, byJamesH. Oliver(1970). $15.00 14 ThePoliticalOrganizationof A2ica, byJohnS.Traill(1975). $15.00 16 A Sanctuaryof Zeuson MountHymettos,byMerleK.Langdon(1976). $15.00 17 Kalliasof Sphettosand theRevoltof Athensin 286 B.C., byT. LeslieShearJr.(1978). $15.00 19 Studiesin A2ic Epigraphy,History,and Topography Presentedto Eugene Vanderpool (1982). $15.00 20 Studiesin AthenianArchitecture, Sculpture, and Topography Presentedto HomerA.Thompson(1982). $15.00 21 ExcavationsatPylos in Elis, byJohnE. Coleman(1986). $25.00 22 A2ic GraveReliefsThatRepresentWomenin theDressof Isis, by ElizabethJ.WMters(1988). $40.00 23 HellenisticReliefMoldspromtheAthenianAgora,by ClaireveGrandjouan (1989). $25.00 24 ThePrepalatialCemeteries at Mochlosand Gourniaand theHouseTombsof BronzeAgeCrete,byJeffreyS. Soles(1992). $35.00 25 Debrisproma PublicDining Placein theAthenianAgora, by SusanI. RotroffandJohnH. Oakley(1992). $35.00 26 TheSanctuaryofAthenaNike in Athens:Architectural Stagesand Chronology,by IraS. Mark(1993). $50.00 27 Proceedings of theInternationalConference on GreekArchitectural Terracottas of the Classicaland HellenisticPeriods,December12-15, 1991, editedby NancyA. Winter(1994). $120.00 28 Studiesin ArchaicCorinthianVasePainting,by D. A. AmyxandPatricia Lawrence(1996). $65.00 29 TheAthenianGrain-TaxLatv of 374/3 $35.00
B. C.,
by RonaldS. Stroud(1998).
30 A LMIA CeramicKiln in South-CentralCrete:Functionand Pottery Production,byJosephW. Shaw,AleydisVande Moortel,PeterM. Day, andVassilisKilikoglou(2001). $35.00
I
VOLUME
r
7I:
NUMBER
I
| 11
I
I
I
I
I
I
I
THE
JOURNAL OF
O F
C L A S S I C A L
O
I,
1
11.
1
1
,
O
L
I, I I
I
3
THE
I
I
I
.
11
11
l
AMERICAN SCHOOL
S T U D I E S
A T
A T H E N S
W_-
and TheAmericanSchoolof ClassicalStudiesatAthensis a research studyof thearchaeology, to advanced dedicated teachinginstitution of Greeceandthe andliterature language, art,history,philosophy, of nineAmerican in 1881by a consortium Greekworld.Established studentsandscholars the Schoolnowservesgraduate universities, actingasa collegesanduniversities, frommorethan150affiliated andstudyin Greece.The mainbuildingsof the baseforresearch and Schoolandits libraryarelocatedin Athens,withadministrative NewJersey.As partof its mission, off1cesin Princeton, publications in theAthenianAgoraandat the Schooldirectsongoingexcavations andsurveys excavations CorinthandsponsorsallotherAmerican-led linkbetweenAmericanarchaeologists on Greeksoil. It is theoff1cial Serviceof the GreekMinistry andtheArchaeological andclassicists of to thewisemanagement of Cultureand,assuch,is dedicated of the of knowledge andto thedissemination resources cultural in the Schoolorparticiaboutmembership world.Inquiries classical pationin the SummerSessionsshouldbe sentto theAmerican Schoolof ClassicalStudiesatAthens,6-8 CharltonStreet, NewJersey08540-5232. Princeton, bytheAmericanSchoolof Classical quarterly Hesperiais published to the StudiesatAthens.Foundedin 1932anddevotedprimarily andSchoolof reportson School-sponsored timelypublication fromallscholars Hesperiawelcomessubmissions directedprojects, history, art,epigraphy, workingin thefieldsof Greekarchaeology, Hesperiais a timesonward. fromearliestprehistoric andliterature, journal. refereed
VOLUME
7I: NUMBER
3
JULY-S EPTEM B ER 2002
I PUBLICATIONS STAFF
UX-a
I
THEJOURNAL OFTHEAMERICAN SCHOOL OFCLASSICAL STUDIES ATATHENS
A. J. GRAHAM The Colonization of Samothrace
231
EDITOR-IN-CHIEF Kerri EDITOR, Tracey EDITOR,
Cullen MONOGRAPHS
Michael
Sarah MANUSCRIPT
Architraval Arrogance? Dedicatory Inscriptions in Greek Architecture of the ClassicalPeriod
Fitzgerald
PRODUCTION
Carol
GRETCHENUMHOLTZ
Cox
Hesperia
MANAGER George
Figueira
EDITOR
A.Stein
PUBLICATIONS COMMITTEE ThomasG. Palaima(Chairman) Universityof Texasat Austin DariceBirge LoyolaUniversityof Chicago Margaret Miller Universityof Toronto JeniferNeils CaseWesternReserveUniversity JamesP.Sickinger FloridaStateUniversity KathleenW. Slane Universityof Missouri-Columbia RhysF.Townsend(exofScio) ClarkUniversity
261
MARKL. LAWALL EarlyExcavations at Pergamon andthe Chronology of RhodianAmphoraStamps
295
Submissions:Manuscriptsand communicationsshould be addressedto Dr.TraceyCullen,Editor,Hesperia,AmericanSchoolof ClassicalStudiesat NewJersey08540;tel.609-683-0800; Athens,6-8CharltonStreet,Princeton, Manuscripts andphotocopiesof ilfax609-924-0578;[email protected]. originalartworkandphotographs lustrations mustbe submittedin triplicate; aremadewith the editor.A shouldnot be sent unlesspriorarrangements themajorconclusions ofthe articleshouldalsobe shortabstractsummarizing reviewprocessandauthors included.Articlesaresubmittedto a double-blind accordingly, withouttheirnameor arerequestedto preparetheirmanuscripts notes,bibliograpreparation, affiliationappearing. The stylefor manuscript on submissions canbe foundin the Guidelinesfor phy,andotherinformation Authorspublishedin Hesperia62, 1993,pp. i-xvi;on the School'sWeb site at the aboveaddress. (www.ascsa.org); orbywritingto ASCSAPublications The AmericanSchoolof ClassicalStudiesatAthenswillnotknowinglyprint orinitialscholthe announcement in Hesperiaoranyof its otherpublications byanymeans acquired after December 30,1970, arlypresentation of anyobject otherthanthroughan officiallysanctionedexcavationor survey,unlessthe existingcollectionorwaslegallyexportedfrom objectwaspartof a previously the countryof origin.
Copyright(C)2002 The AmericanSchoolof ClassicalStudiesatAthens Producedat EdwardsBrothers, Inc.,AnnArbor,Michigan. Designby EllenMcKie. Findspotof the Coverillustration: PergamonDeposit,Pergamon. andT.Wiegand, AfterG. Kawerau Die PalastederHochburg(AvP V.1),
Berlin1930,textpl.20. Photo courtesyW. Radt. postagepaidat Periodicals Princeton,NewJersey,andat additionalmailingoffices. Sendaddresschanges Postmaster: to Hesperia,P.O.Box529, Canton, MA 02021-0529,U.S.A.
price The annualsubscription Subscriptions: Hesperiais publishedquarterly. $70 forinstitutions($80,internais $60 forindividuals($70,international), Paymentmust ISSN 0018498X tional),andX33forstudents(proofof studentstatusrequired). beinU.S.dollars,drawnon aU.S.bankorbymoneyorder,andsentto Hesperia, SubscriberServices,P.O. Box 529, Canton,MA 02021-0529;tel. (U.S.) 800-821-7823; (outsideU.S.) 781-828-8450; fax 781-828-8915; e-mail of [email protected]. Singleissues(currentandbacknumberswhenavailable) Hesperiaareavailable for $15 eachpluspostagefromthe DavidBrownBook Company,P.O.Box 511, Oakville,CT 06779;tel. 800-791-9354,860-9459329;fax 860-945-9468;or (outsideNorthAmerica)OxbowBooks,Park End Place,OxfordOX1 1HN, U.K.;tel. +44 (0) 1865-241249;fax +44 (0) 1865-794449.IndexII andSupplements13-17 and19-30 arealsoavailable fromDavidBrownBooks.Reprintsof IndexI, Supplements1-12 and 18, b.v., andearlyissuesof Hesperiashouldbe orderedfromSwetsandZeitlinger, [email protected]. P.O.Box810,2160 SZ Lisse,Netherlands;
HESPERIA
71
(2002)
Pages 231-260
THE
COLONIZATION OF
SAMOTH
RACE
ABSTRACT Threeaspectspertainingto the Greekcolonizationof Samothraceareaddressed:the originof the colonists,the foundationdate,and relationsbetweenGreeksandnon-Greeks.Relevantliterarysourcesandotherindications makeit clearthatthe GreekcolonistscamefromSamos;the current theorythattheywereAeoliansshouldbe abandoned.No foundationdateis remains,especiallyfrom preservedin the ancientsources,butarchaeological cemeterieson theisland,pointstronglyto thefirsthalfofthe6thcenturyB.C. Evidencefor a Greektakeoverof a non-Greekcult,and,especially,for the useofa non-Greeklanguage aswellasGreek,makesthecoexistence of Greeks andnon-Greeksa plausiblehypothesis. AlthoughtherearemanyGreekcoloniesof theArchaicperiodaboutwhich we know very little, we usuallyknow the city or cities from which the colonistscame,and in manycasesthe date of foundation.In the case of Samothrace,'the prestigeof the Sanctuaryof the GreatGods hasensured that therearemanyreferencesto Samothracein ancientliterature.Lewis found241 items for his collectionof the literarysourcesin Samothrace 1, and Burkertfound two others that had been missedby Lewis.2Yet, in spite of this abundanceof literaryreferences,the origin of the colonists and date of the foundationare mattersof disputeor, as Lazaridisput it, obscure.3Cole considered these issues an open question.4 It is my aim in
this articleto discuss,first,thesetwo basicquestions,andthen to consider the relationsbetweenthe colonistsandthe non-Greekinhabitants,which areat Samothraceboth interestingandpotentiallysignificant. 1. Special thanks aredue to Dimitris Matsas for his kind reception when I visited the island in 1995. Apart from showing me archaeological sites, he also gave me much invaluable information.I am gratefulto the ResearchFoundationof the Univer-
sity of Pennsylvaniafor a grant to finance my travel.I would also like to thank Kevin Clinton, who kindly read a draft of this paperand made helpful suggestionsfor its improvement; SaraOwen, for adviceaboutThracian matters;and the anonymousHesperia
reviewersfor their useful comments. In this article,all referencesto Lewis are to Samothrace 1. 2. Samothrace 1; Burkert1993, pp. 179, 181-182. 3. Lazaridis1971, p. 18. 4. Cole 1984, pp. 10-11.
A. J. GRAHAM
232
THE ORIGIN
OF THE COLONISTS
THE LITERARY EVIDENCE Most of our information about Samothrace from the literarysources concerns myth, so there are few passages that are valuable for the history of the Greek colonization.5 Of these the earliest is a fragment of Antiphon's speech On the Tributeof the Samothracians,which states that the colonists came from Samos, whence they were expelled by tyrants.The full passage (MinorAttic OratorsI, Loeb, fr. A.2; Lewis 40 [as throughout, numbers refer to Lewis's passages in Samothrace1]) is as follows: xoct yap o. Tiv apto)Yv oixiLoav'e
oauXvS[zLtoL, E ()V
I?TlEl
8i avayxn, o6x irOuVi,a T(S v?qaoo. i?yv6O.ItOx.xaOrCxio06OtaCv TaoTx6T' ?Sr?oov yaxp 7ro TropavvCov ex Ca[Loo x(XLTOrU)q iEX^ycaxVro O-TE]Opcaxqr a(pLxvoov-rTac XLiocvXaoc6vprcT 7r6 eq Tiqv vroov.6
For those who originally settled the island were Samians, and from them we are descended. They were settled by necessity, not by desire of the island; for they were expelled from Samos by tyrants and experienced this misfortune; having taken booty from Thrace they arrivedat the island. (Lewis, trans., adapted) This is a clear statement of origin. While Athenian oratorsfreely perverted historical truth if it suited their advocacy,they could not make false statements about well-known facts in speeches designed to be heard by a large audience,7if they were not to lose all credibility.In this case, it is hard to doubt Antiphon's information about the origin of the colonists. Antiphon's statements are, moreover, supported by the Aristotelian politeia of the Samothracians,which we know from Herakleides' epitome (FGrHist548 F5b; Lewis 41) and,less fully,from the scholion to Apollonius Rhodius, 916-918 (FGrHist 546 Flb; Lewis 37). The passage of Herakleides states: 6 1 oOExo0p -axl T[ov
' a xcs exaXeizo Aeoxocavax, e?&Op 6L&T
X?oxi
ElvOct' oT?'cpov 8, Optx6v XaTC'roOVTCo, Opgxi.x. TOO'TcV8? S aC.LoL xxoCT.cxtoav axThqv exXLtOVTtcOV,OT?'rpOV?T?LoLV eiTaCrcxooLO io
v,l olxiLac, xai Ea[co0o6paxrlvexdOX?7av. ix7cao6vrcT?s Samothrace was originally called Leukania, because of being white, but later,when Thracians occupied it, Thrakia. When these had left it, seven hundred years later Samians settled it, when they had been expelled from their own country, and called it Samothrace. While Ps.Scymnus does not explicitly state that the Samians had been expelled from their homeland, his account of the historical colonization agreeswith Antiphon and the Aristotelianpoliteia on the Samian origin of the colonists (679-680, 690-695; Lewis 58):
5. The relevantpassagesaregiven in FGrHist548 (Anhang),with characteristicallyacute and learned commentsbyJacobyin IIIb (Text), pp. 470-475 (Noten), pp. 279-281. There is an extendeddiscussionof these passagesin Prinz 1978, pp. 187205. Although mainly interestedin mythology,Prinz also treatsthe origin of the colonists at length. 6. I providehere the text ofJacoby, FGrHist548 F5a, who regardsthe transmittedtext as sound, but breaking off too soon; see IIIb (Text),p. 475, Kommentarto F5a. It is understandable, however,that editorshave suspecteda or added xci before lacunaafterTrocn6-, Xsiav,becauseof the asyndeton.While as Jacobyis right to take T-r6iT-CTx6 xrX.,he does referringback to &avaryxf not considerthe problemof the asyndeton. A certainconclusion on the matterseems unattainable. 7. See Meiggs 1972, pp. 240-241; Rhodes 1972, pp. 90-91; Harris2000, p. 496.
THE
COLONIZATION
OF SAMOTHRACE
233
repaocv Saio0paCxq ' ecoTnvjaoo Tp6Ctxl, eXouoa tYv olxVqav avaori[tiLy?LY lv' TOo;S ? Saoa[ti6Opxa;, Tpooa; ovTracTz) y?V?t aCrt ToOT607ou 8? OpXaOCX ?S7TCXXoU[L?voOU, TC T6tCOO. 61' ?I7E?PLO?tV?yXaTa[XS?LVaL
T&CV 6' atOOLT; ?V oTTO&LiX SCa[LLcV 7xOTE ?7rapXeoGaVTOv, TYvtLxaoOTE'?X TY;q &C[Lou aoVOLXOLoeoX)OaV. ?rC?Ea[IEVOL TLvaqO
Opposite is Samothrace, a Trojan island, which has mixed inhabitants.... But the Samothracians,being Trojan by race, but called Thracians from the geographical position, stayed in the place because of piety. But when in a famine once, the Samians supplied them, at that time they received some from Samos and had them as fellow-inhabitants. The Samian origin of the Greek colonists is also found in Iliadscholia, most notably Eustathius's commentary on 24.78 (Lewis 51), where he writes: .... . A?oxvta, (paC, r7rpT?pov ixaX?iTO, ?ITOa 6Ot a[Lo0poCxqv Wv Ta a SaOcLIxv oLXCLo0eLaC, oaxdc(YarT6Oti &t)iatXCTa OpalCTOca xaocTxavooav, Sa[to0p.axyqCv6oc7Taom. Samothrace ... which, they say,was formerly called Leukonia, and then, having been settled by Samians, whose boats captive Thracian women burnt there, it was named Samothrace.8 There is general support for the Samian origin of the colonists of Samothrace in Herodotos's account of an incident in the Battle of Salamis (8.90.1-3; Lewis 97, but breaking off too soon): 8 xoCa T-c6O?v TcOOopO6pc TcOv TcV -rV?C; TOroOT. 'Ey'?V?To 'DOLVxcov, at v?? ?X06vT?; Le0(p6dpazo, 7rapCc PaoTXC?a ?p[cXXauovTob; "ICovag,(t) &8'?XELVOoaCTCOXoiocro oaiVi??, 0; 7tpod6VTccv. (ov)V?IX? oV oVT'co caT?'Ic3vCvT? TOib;oTpaq7yoxb [.Y &xroXooLat Ea TrobtapaXXovT Oolvixcov T? Xcap?Lv TcOL6v8? [LRO66v.'?TCTO6T'Co
8 TaoTraX?y6v-rcovi?V3aX?vrli'ATTL-rx Za[zo0opqLxYvYvO.q -cT? ?O xaC ?Trct(pEpOL?VYl 'ArTLXi xa?Cr cro AlyLtailVwvoS XacT??0oy? TCOV r'v v?oa. 'aT-r? ?OVT?caxovTrcacA o' Ea[,ioop0Ylx?; SaojyoOpqixcov To r xoTarcaood&cr ; v; TooS; ?S7LpaTrac; arTc PV oXXaovT?;d &n7paCcav xac ?iTC?pY3adv T?xa ?CoXovaoTqv. Tao'T y?VOL?VOaobV "Icovag 8. The same informationis found in the ScholiaTownleianaon the same passagefrom Homer; see Lewis 45.
?pp6ocaoo cd ya&p?L? oap?as Eipxlqr ?pyov ?liya ?pyaoaccaivoo;, T?cxaL TrcvTac IoivLxaqcola 67TpXo7?rj?tv6O; E?pdp&?Tor7p6 TCobO; x?ctaX&( TO; ?X?X?OE? aCLT6l[l?voS, xaiL op?CSoV a7oTraC[i?v rvaV[L aOtCOLxaxol
y?vo[i?voI
TOVS;aX[LLvovac; Laca3dXXcoOL.
A. J. GRAHAM
234
It happened also amid this disorder that certain Phoenicians whose ships had been destroyed came to the king and accused the Ionians of treason, saying that it was by their doing that the ships had been lost; the end of which matter was, that the Ionian captains were not put to death, and those Phoenicians who accused them were rewarded as I will show. While they yet spoke as aforesaid, a Samothracian ship charged an Attic; and while the Attic ship was sinking, a ship of Aegina bore down and sank the Samothracian;but the Samothracians,being javelin throwers, swept the fighting men with a shower of javelins off from the ship that had sunk theirs, and boarded and seized her themselves. Thereby the Ionians were saved; for when Xerxes saw this great feat of their arms, he turned on the Phoenicians (being moved to blame all in the bitterness of his heart) and commanded that their heads be cut off, that so they might not accuse better men, being themselves cowards. (A. D. Godley, trans., Loeb) From this it is clearthat for Herodotos the Samothracianswere Ionians. And Herodotos had himself been to Samothrace, as the passage 2.51.2-4 (Lewis 140) shows: oCtcl &
T-a Saopo0oitx T'xKopt3eitovopyLa[i?l6J-roTaL,
E7CL-rsTeooLt7aocaXa36vT?-; tocpa HIeXaOy&v,o6oTO;dV)qpol[e TOXeyco'-'v yap Zaqio0opqxilvoL'xeovTpo-cetpovHeXaoyoTo6Tzo avotxot eyevovTo, xai rapa To6ztov oliep 'AYOrvoLtoal EaCoopLxe?;
TC o6pyLo 7a TopaXarpdvouoCt.6poOaov eXeLv T&
ro 'EpcAOjvaotot trocT 'EpTO c3ocaxTacyaXTa cpCrot 'EXXlvwv o TrapoxHleXaoyviroLloGaxvTo'olo8e HeXoayoi ip6v ntva jia0O6vT?S; X6yov TCpl OCi'rob ?Xexav, Ta& V'cTooa Ev ZaOo0opQYxn [UoTYpLiolaL
Anyone who has been initiated into the secret rites of the Cabiri, which the Samothracians derived from the Pelasgians and now practice, knows what I mean. Samothrace was formerly inhabited by those Pelasgians who used to dwell among the Athenians, and it is from them that the Samothracians derive their secret rites. Thus the Athenians, who learned it from the Pelasgians, were the first of the Greeks to make ithyphallic statues of Hermes. The Pelasgians told a kind of sacred story about this, and it is disclosed in the mysteries at Samothrace. (Lewis, trans.) Since Herodotos here shows knowledge of secret information known only to initiates, information which, significantly, he is careful not to divulge, it is correctto conclude that he was himself initiated into the SamothracianMysteries,9and, therefore,that he had visited Samothrace.Jacoby poses an interesting question, when he wonders if Herodotos's initiation into the Samothracian Mysteries had anything to do with Herodotos's stay on Samos and the relations between Samos and Samothrace,10 but it
is, unfortunately,a question to which no certain answer can be given.
9. See RE Supp. II, col. 259, s.v. 9. SeeRE Suppl.II, col.259, s.v. Herodotos (E Jacoby).It is strangethat Lateiner(1989, p. 65) doubted this.
10.FGrHistIIIb(Text),p. 467.
THE
COLONIZATION
OF SAMOTHRACE
235
The only serious doubt about a Samian origin in our literary sources seems to be a passage in Strabo (10.2.17 [C 457]; Lewis 42): xal 7Tv Op,axav, 9 vvv oSapo0paxrqv xaocXi ' 6 tcoLqrTl SC&pJov ol6 Tv 8' 'IcvLxqv xaoXobitv. [pLv, cd ELX6o'-xat yap t1v 'IovtxY Y atotLxiotv EiLSvat (paivcvTaL ox av avxtCTL8octEXe? 6v ]"v ovu.iLav,
. -qEao08oparxY T7Epl
Tco?mcL0?T-c TOT?[L?V XEycov,
64oo ?7r'axpoTTYrl; xopuopy Zd[ouj 6Xrl?aoq;,
TOT? o?Y or,uytoa
TCv 7Xqhoiovvolotov'
e?; Zaplov ?; T'"I[3ppov xaci A9IJvov &a[cxOXaX6aooav'
xaoLraXtv [Ls?oaqy6;TE?Sa[Loto xal "Iip[pou 7rcaLraXo?aooqs. 6?&[pI?V a oi068' ?xaX?TO T(CaOxco o6v, oVx v6[IXax? 6'Oj'/v' Ei 6v6ollat rcpOT?pov,aXXa& M?Xa[ipuXXo, ?x' 'AvO?Eti;, HIapO?via &o76TOOroTa[oL ToOIlapO?vioo, 6; "Ii`o3poaoo Lx?TovouraO07Y. E??t ov xaTa Tz Tptox lx dos Jl?VxaoLq r ?xaX?io (oO xacL K?spaX)Xrvia q EaioOpaxqr yap av 'Exadpy OTL to6 XSyooCOa, 7raia;S avTTg ?iYqily?TO : 7epvXaaX', ov x? Xdpotl, ?5;Ea[ov ?; T'"I,[ppov), 'IovLxh 3' o6x aXTrCOCxT6 7TC,3qXov 3' oTr h7cr6-Trv tcpOT?ptovTcLV;-v 6Cxovuxiav ?oX?V ?E Jv XaX?LVO 89]Xov, oTL7rapt T7v aPXaLaovloxoTplav O X?youoiv ot (poqoav?Xg, [?Tx tV 'I1)VLXYV dX7roLxtav xalt v T T?OipptioVoCapoouoiav dtoixoou ?X0?[v ExSd6ou xal ovo[ptaoL Za&ov tv xvSaioo0opaxjv, (g oi dJILOL otobT' ?rXdoaavTOo 86q;s Xdpotv. 7rcLav(OT?poLo' ?lolv ol
?(paivz?To WaocaJ ?V"I0Y, xal vi?S; 'Axatcv. paLiv?To HlptQI.otom7rC6.Xt cTLV?e5? SZdxiovxaX?coOaxa(taotv anob Sa'ov, tCO oLxoovwov ?OxoVT-v r7pOo?X(y xTX. Op.axcv 7Tp6T?pov,o0 xaL Tiv 7rLctpOV
The poet also uses the name "Samos"for the Thracian island, which we now call Samothrace. And it is reasonable to suppose that he knows the Ionian Samos, for he also appears to know of the Ionian migration; otherwise he would not have differentiated between the places of the same name when referring to Samothrace, which he
designates at one time by the epithet "high on the topmost summit of woody Samos, the Thracian,"and at another time by connecting it with the islands near it, "unto Samos and Imbros and inhospitable Lemnos." And again, "between Samos and rugged Imbros." He therefore knew the Ionian island, although he did not name it; in fact it was not called by the same name in earlier times, but Melamphylus, then Anthemis, then Parthenia, from the River Parthenius, the name of which was changed to Imbrasus. Since, then, both Cephallenia and Samothrace were called Samos at the time of the TrojanWar (for otherwise Hecabe would not have been introduced as saying that he was for selling her children whom he might take captive "unto Samos and unto Imbros"),and since the
A. J. GRAHAM
236
Ionian Samos had not yet been colonised, it clearly got its name from one of the islands which earlier bore the same name. Whence that other fact is also clear,that those writers contradict ancient history who say that colonists came from Samos after the Ionian migration and the arrivalof Tembrion, and named Samothrace Samos, since this story was fabricatedby the Samians to enhance the glory of their island. Those writers are more plausible who say that the island came upon this name from the fact that lofty places are called "samoi,""forthence all Ida was plain to see, and plain to see were the city of Priam and the ships of the Achaeans."But some say that the island was called Samos after the Saii, the Thracians who inhabited it in earlier times, who also held the adjacent mainland, etc. (H. L. Jones, trans., Loeb, corrected) The above passage constitutes the end of a long discussion of the name Samos, which is included in Strabo'streatment of the Ionian islands. The topic begins with Kephalleniaat 10.2.10. Strabowas using two authorities here, Demetrios of Skepsis and Apollodoros, and Schwartz thought that we have Demetrios's account combined with excerpts from Apollodoros.l1 The two authorities are named in 10.2.16, where they disagreed. Schwartz thus attributed large sections to Demetrios,12but not 10.2.17, which he gave to Apollodoros.13 Jacoby also attributed that chapter to Apollodoros, and printed it, in his smallest type, as FGrHist 244 F178b. Arguments in favor of the attribution arethat Apollodoros is the last authoritynamed in Strabo (10.2.16), and Apollodoros's discussion about Homer's ways of distinguishing homonymous places cited the example of Samothrace (Strab. 8.3.6; Lewis 84; FGrHist244 F181), as in 10.2.17. But there is a serious obstacle to attributing 10.2.17 to Apollodoros. We know from the scholiast to Iliad 13.12 (Lewis 46; FGrHist 244 F178a) that Apollodoros stated that Samothrace was colonized from Samos 209 years after the TrojanWar,whereas in 10.2.17 colonization from Samos is expressly stated to be unhistorical and invented by the Samians to gain glory.Jacoby'sexplanation of this contradiction, that the scholiast was mistaken and did not realize that Apollodoros rejected the Samian colonization, is complicated and unconvincing.14
A detailed comparison strengthens the case against Jacoby's attribution. There were apparentlytwo different accounts of the mythical colonization of Samothraceby Samians. In one, attributedto Apollodoros, as we have seen, the colonization is dated 209 years after the Trojan War, i.e., 976/5. The motive given is an oracle from Delphi. In the other, the colonization is the result of a successful Ephesian attack on the Samians, some of whom fled to Samothrace (Paus. 7.4.2-3; Lewis 43). Not only is the motivation different, but also the date. The leader of the Ephesian attack on Samos was Androkles, the founder of Ephesos, which implies a close connection to the Ionian migration, the accepted date of which was 1044/3.15 Strabo'swords in 10.2.17, Ie?Tc' vv 'ICOVLXv xoaLrv T?ippeiovos O&CoLXLOCV aocpouoiav,may be taken to show that it is the second story that is referred to here, because Tembrion was a founder of Ionian Samos (Strab.
11. RE IV, 1901, col. 2809, s.v. Demetrios(78) (E. Schwartz)= Schwartz1957, p. 108. 12. 10.2.8-10 to iv 7rp6TcpovZS(IOov atwhichpointStraboturns xaXeSoEv, to Apollodoros,who is named;10.2.13 to IlHpcvrloog xoc Kp&vtoL, where Strabo turnsto his own times; 10.2.14; 10.2.16 to
qji(pi&uoot,where Strabo turns from
Demetrios to Apollodoros. 13. RE I, 1894, col. 2869, s.v. Apollodoros (61) (E. Schwartz) =
Schwartz1957, p. 272. Prinz (1978, p. 203), oddly,cites this passageof Schwartzto supporthis opinion that the statementthat Samianorigin was inventedby the Samiansto gain glory came from either Apollodoros or Demetrios. 14. See FGrHistIID, p. 787, Kommentar to 244 F178.
15. See Sakellariou1958, pp. 30724.
THE
COLONIZATION
OF SAMOTHRACE
237
14.1.3 [C 633]). In that case,it was not Apollodoros'saccount of the Samian colonization of Samothrace to which Strabo refers in 10.2.17. The problem could be resolved by assuming that Strabo was still using both authors in 10.2.17. The early part of the chapter could be Apollodoran, while at Esrce ouv xaxCtO Tptotxa Strabo may have reverted to Demetrios, in orderto returnto the comparisonwith Kephallenia,where he did draw on him. The style of argument in the latter part of the chapter is the same as that of the earlier passages attributed to Demetrios. That Demetrios was in general interested in Samothrace is shown by his discussion of the rites of the Cabiri (Strab. 10.3.20 [C 472]; Lewis 163). The search for lost sources rarely leads to certain conclusions, but if Demetrios of Skepsis is indeed the source for the denial of Samian colonization of Samothrace, the authority of the statement is not enhanced. Schwartz may have judged that Demetrios was not a Falscher,16but he was, for all his learning, a pedant, preparedto use very bad sources and to reach very perverse conclusions. He could, for example, deny that Xerxes dug his Athos canal (Strab. 7 fr. 35 [C 331]),17 and his treatment of the geography of the Troad notoriously contains forced arguments based on a combination of pedantry and local patriotism.18 Apart from one dissentient voice of dubious authority,therefore, our few relevant literary sources agree that Samos was the origin of the Greek colony of Samothrace. In recent times, however, that consensus has been rejected. Prinz, for example, argued that the tradition of Samian origin derived from a local historian, who invented the story from nothing more than the name of the island.19As we have seen, this is not an adequate assessmentof the informationin the relevantancient literarysources.Prinz's discussion is also vitiated by his failure to take into account either the dates or the relative historical worth of those sources.20 OTHER
16. RE IV, 1901, col. 2812, s.v. Demetrios(78) (E. Schwartz)= Schwartz1957, p. 112. 17. Herodotos'sinformation(7.2224) has been shown to be correctby recent geophysicalinvestigations;see Isserlin et al. 1996; Blackman19992000,pp.86-87. 18. Leaf 1916-1918. 19. Prinz 1978, pp. 193-205. 20. Prinz'streatmentdoes not take into accountthe modern excavations and the publicationsin the Samothrace series.His only mention of archaeological evidence is in his discussionof the city wall, where he cites very old authoritiesfor an erroneousdating;see Prinz 1978, p. 193 with n. 10. 21. Still in Guide6,p. 19. 22. Samothrace 2.1, no. 1. Cf. also the remarkson p. 3.
EVIDENCE
A much more serious argument for rejecting the tradition of Samian origin has been proposed by Karl Lehmann and those who have followed him in the Samothracepublications.21Lehmann's first argument was based on a fragmentaryinscription, found on the city site, the text of which was published by Fraser22as follows:
------------------------------
__
... .5.. . . j 5 z p-
_-____-_____-____________--------____------- _-- ?e S -------------
[o---
----------
[jio0potxov---
[xSag;---
---
a]; TOXtO;TC; EC-
To------o
'ETLxXeoo;gTao-
H(pa]taoxTvaxai 'EcrL-
-----------------------scVTcy(L
-----------------------oc_oog
Tcpo[e]-
T----------
The inscription is thought to record a Samothracian decree, and contains two Aeolisms, viz., tr6XLog Ta-&and o6vTEoaL. This inscription has there-
238
A. J. GRAHAM
fore been interpreted as evidence that the colonists of Samothrace were Aeolians. Lehmann's second argument came from the distinctive painted pottery,called G 2-3 ware from the squareof the excavationsat Troywhere it was found, which occurs in quantity at Samothrace and has generally been dated to the first half of the 7th century.23On the basis of the places where it has been found, this pottery has been thought to emanate from the Aeolian area.The two arguments led to the received opinion of the Samothracepublications that the island was colonized by Aeolian Greeks earlyin the 7th centuryB.C. See, for example,the words of Phyllis Lehmann: "The pottery found in this burned, black layer ranged from fine subgeometric ware of the first half of the seventh century B.C. [i.e., the G 2-3 ware] to coarse, handmade vessels. The contrast between the elegant wheelmade kantharoiwith their thin fabric and restraineddecoration and the crude, undecorated, handmade cups found in this stratum suggests that it was at this period that the Greek-speaking settlers came to Samothrace and mingled and worshiped with the native, Thracian-speaking inhabitants."24
What is the strength of these arguments?It is not at first sight obvious that the fragmentary inscription with Aeolic dialect forms records a Samothracian decree, since it could well be a decree of some other city, which honored a Samothracian citizen and was therefore inscribed at Samothrace. Fraserrecognized that possibility, but argued that the restoration Za[jio0pdtxcov]is required,because there is no other Aeolian city that begins with Sa. The fragility of the further assumption that the decree must be Samothracian was well exposed by Bernard.25The city of Samothracecould easily have been named in a decree of another city,which was inscribed at Samothrace.Thus the argument from the Aeolisms in the fragmentaryinscription carries no conviction. That leaves the argument based on the G 2-3 pottery found at Samothrace.This fine painted pottery with attractive Subgeometric decoration has been found at Troy, Lesbos, Lemnos, Thasos, Samothrace, Neapolis, Eion, and Skyros.26By no means have all these finds been published. In particular,the very important and extensive Lemnian material,discovered by the Italian excavationsat Hephaistia and Chloi, has not been fully published, though there are useful summary descriptions by Beschi.27As a result, the best account of G 2-3 ware as a whole still seems to be that of Bernardin his publicationof the finds from the deep soundings atThasos.28 Bernard divided the G 2-3 pottery at Thasos into two categories, G 2-3 ware proper,which he called "Aeolian G 2-3 ware,"and a variety known to have been made on Lemnos, which he called "Lemnian G 2-3 ware."That this variety was made on Lemnos is shown not only by the great quantities found there, but also by technical aspects and by the occurrence of some badly formed or overfired examples.29There is general agreement that the place of manufacture of Aeolian G 2-3 ware cannot, with our evidence, yet be determined. But that does not matter for our present argument, because all the G 2-3 ware at Samothrace has been identified by Beschi as made on Lemnos.30At that time Lemnos was inhabited by pre-Greek people who are usually called, on the basis of Thucydides 4.109.4, Tyrrhenians. Far from being, as Karl Lehmann thought, the first Hellenic pottery found on Samothrace, the G 2-3 ware found there was not in fact made by Greeks.
23. Beschi (1996, p. 40) thinks that K. Lehmann'sdate for the G 2-3 pottery at Samothraceis too high. 24. Samothrace 5, p. 269. 25. Bernard1964, p. 92, n. 1. 26. Koukouli-Chrysanthaki1992, pp. 572-574. She was not, however, awareof the find from Skyros,for which see Beschi 1985, pp. 58-59. 27. Beschi 1996, pp. 30-34, 38-41; 1998, pp. 70-75; 2000, pp. 77-79. 28. Bernard1964, pp. 88-109, followed by Graham1978, pp. 67-69 (= Graham2001, pp. 176-177). Consult also M. B. Moore in Samothrace 5, pp. 318-371; Koukouli-Chrysanthaki 1992, pp. 572-574; Beschi 1985; 1996; Fisher 1996. Fisher did not know of or Beschi's Koukouli-Chrysanthaki's mention so she does not publications, the finds from Neapolis, Eion, or Skyros,but she has useful information on new finds and contexts at Troy, which come close to stratigraphy, without, however,close dating. 29. See Beschi 1996, p. 39. 30. Beschi 1996, pp. 40-41, acceptedby I. Love in the discussion (p. 65).
THE
COLONIZATION
OF SAMOTHRACE
239
Both Lehmann's arguments in favor of colonization of Samothrace by Aeolian Greeks are thus shown to have no substance.There are also other possible indications that the Greek colonists were Ionians. A possible indication of Samian origin was acutely seen byJacoby in the Samothracian eponymous basileus attested on the inscribed lists of theoroi and initiates, because Herodotos dates an event of 6th-century Samian history errL 'A,iupLxpal-eoS P3aloXsrovTro (3.59.4).31Jacoby only posed the question, but the idea is worth taking into account. There is also an important piece of sculpturalevidence that supports the tradition of Ionian colonization. This is the well-known Archaic relief, now on permanent display at the Louvre, that bears the figures of Agamemnon, Talthybios,and Epeios, who arenamed.32Historians of Greek sculpture date the relief to about 550 B.C.,33 and the script of the inscriptions was called by L. H. Jeffery "a fine example of eastern Ionic lettering.34 This relief was found on Samothrace and sold to Choiseul-Gouffier about 1790. Its Samothracianprovenanceis quite certain,35and C. Fredrich long ago suggested that it supported the tradition of Samian origin.36 We may conclude that the consensus of the ancient literary sources is strengthened by other indications, and not contradicted by any good evidence or arguments, and we should therefore accept that the Greek colonists of Samothrace came from Samos.37
THE DATE OF FOUNDATION As we saw above, two mythical dates for the Samian colonization of Samothraceare recordedin our literarysources:about 1044/3 (Paus. 7.4.23; Lewis 43) and 976/5 (Schol. to Iliad 13.12; Lewis 46). These dates have nothing to do with the historical record and may be set aside. The only other apparentdate in the literary sources is the one implied by the interval of 700 years given by Herakleides (Politeiai21; Lewis 41; see above). It is uncertain whether this interval separatedthe Samian colonization from the departureof the Thracians, as Lewis translated, or from their arrival, as in the translationgiven above, but, in any case, without additional chronological information, the interval does not give us a date for the Samian
31. FGrHistIIIb (Noten), p. 281, n. 43. For the lists, see Samothrace 2.1, nos. 22 and following. 32. Louvre697; IG XII.8 226. 33. See Richter 1949, p. 96, who dated it ca. 560-550; Boardman (1978, caption to fig. 264) writes "about550."The serpentlikefigure on the right, called a griffin by Boardman, of which we now possess only the spiralemanatingfrom the head, can be clearlyseen in Millingen 1826, pl.l. 34. Jeffery1990, p. 299. 35. See the unequivocaland circumstantialstatementby Dubois
(1818, p. 40, no. 108); cf. Conze, Hauser,and Benndorf 1880, p. 108; Bousquet 1948, pp. 112-113, n. 1. Cole (1984, p. 108, n. 64) was wrong to say that the Samothracianprovenance is uncertain.She was also misled by a mistakeby Bousquet,who stated that, accordingto S. Reinach, the reliefwas seen by Cyriacusof Ancona. The relief to which Reinach (1891, p. 90) refers("Cyriaqued'Ancone ... signalaun bas-reliefqui est aujourd'huiau Musee de Louvre")is manifestlythe Frieze of the Dancing Maidens (now renamedChoral Dancers),part of which was then in
the Louvre (Guide6,pp. 123-124 and fig. 33), which was mentioned and illustratedby Cyriacus;see Bodnarand Mitchell 1976, pp. 37-39. The absence of the relief of Agamemnon,Talthybios, and Epeios from Cyriacusis thus completelyirrelevant. 36. Fredrich1909, p. 25. 37. This has been the opinion of good judges;see, e.g., Fredrich1909, p. 25, and Bernard1964, p. 92, n. 1, though neither arguedthe matterin detail.I expressedthe same opinion brieflyin CAH II1.3,2nd ed., pp. 117118.
240
A. J. GRAHAM
colonization.38Thus the ancient literary sources do not provide us with a historical date of foundation, and we are thrown back on indirect indications and archaeologicalevidence. zuovvcov were thought byJacoby to be an inAntiphon's words Or6O direct way of referring to the most famous Samian tyrant, Polykrates.39 Unduly influenced byJacoby, I once suggested that the foundation probably occurred during the second half of the 6th century.40Since we now have clear evidence for the existence of a tyranny on Samos from early in the 6th century,and we know the names of three tyrantsbefore Polykrates,41 the chronological implications behind Antiphon's words are much broader than Jacoby thought, and are probably too vague to give a precise date of foundation. Successivelyearlierterminiantequemfor the establishmentof the Greek colony may be assembled as follows. Samothrace is certainly Greek during the Persian Wars and at the time when Herodotos wrote. By the end of the 6th century,the community was striking coins, some of which bear an abbreviation of their name in Greek.42It would require very far-fetched and improbable hypotheses to account for the presence of the Archaic relief mentioned above before a Greek community was established on Samothrace, so we may raise the terminusante quemto ca. 550. When we turn to consider the archaeologicalevidence for the date of Greek settlement, we can distinguish four areas that have been, at least partly,investigated: the sanctuary,city, countryside, and cemeteries. The sanctuary.Lehmann gave dates in the Archaic period to several structures in the sanctuary (Fig. 1), but they have all been comprehensively repudiated except that of the so-called Hall of Votive Gifts (Fig. 1:16),43which has retained the date of ca. 550 B.C., and is now in isolation.44That date was based on architectural considerations and stratigraphy. The dating by architecturalconsiderations depended on the analogies suggested by Lehmann for his reconstructionof the building.45He thought that these pointed to a 6th-century date, and he opted for 560-540 on the basis of a fragmentary Doric capital.46Lehmann's reconstruction of the original building contained many unusual or unique features.47J.M. Cook described his reconstructed building as "really astonishing,"48and J. J. Coulton commented that "the evidence for the form of the Hall is so fragmentary that Lehmann's fascinating stylistic analysis . . . is largely hypo-
thetical."49Most important, however,was Kuhn'sreexamination of all the evidence, and his fundamental reconsiderationof the arguments on which Lehmann'sreconstructionwas based.50This work showed conclusivelythat the building cannot be reconstructed as Lehmann proposed, and that, far from being a stoa with a colonnade along its west side, it was in fact enclosed on all four sides, and clearly designed for banqueting. Lehmann's chronological conclusions from architecture thus fall to the ground. The only architecturalindication of date seen by Kuhn is the profile of the echinus of a fragment of a Doric capital, which he regarded as Late Archaic.51
38. Jacobyfound the figure 700 inexplicable,and thought it might be corrupt;see FGrHistIIIb (Text),p. 475, Kommentarto 548, F5b. Prinz (1978, pp. 104-105) did not think therewas any reasonto see the figureas corrupt, since any numberwould do for a theory of Thracianorigin based on the name Thrake;but we have alreadyseen the weaknessesof his treatment. 39. FGrHistIIIb, p. 474, Kommentar to 548, F5. 40. CAH 111.3,2nd ed., p. 117. 41. See SEG XXXVII 729; TolleKastenbein1976, pp. 34-35; Boardman 1984, no. 320 with comments. 42. See Schwabacher1938; cf. Schonert-Geiss 1996, pp. 271-272. 43. CompareGuide',pp. 40-41, with Guide6,pp. 52-87. 44. Apart from some tracesof earlierstructures,which, though not dated, are attributed,not always confidently,to Archaictimes; see, e.g., Guide6,p. 85. 45. See Samothrace 4.1, pp. 56-73. 46. See Samothrace 4.1, pp. 76-79; pl. IV. For the date, see p. 96. 47. See Samothrace 4.1, pp. 11-54. 48. Cook 1964, p. 118. 49. Coulton 1976, p. 33, n. 2. 50. Kuhn 1985, pp. 187-199. 51. Kuhn 1985, p. 198; see Samothrace 4.1, p. 29, fig. 28.
THE
COLONIZATION
OF SAMOTHRACE
Figure1. Samothrace.Plan of the Sanctuaryof the GreatGods: 1-3) LateHellenisticbuildings; 4) unfinishedEarlyHellenistic building;6) Milesiandedication; 7) diningrooms;8) unidentified niche;9) archaisticniche; 10) unidentifiedniche;11) Stoa; 12) Nike Monument;13)Theater; 14) AltarCourt;15) Hieron; 16) Hall of VotiveGifts;
24I
17) Hall of ChoralDancers; 20) Rotundaof ArsinoeII; 22) Sacristy;23) Anaktoron; 24) dedicationof Philip III and AlexanderIV;25) theatralarea; 26) Propylonof PtolemyII; 27) South Cemetery;29) Neorion.
After Guide6,plan IV. DrawingJ. Kurtich. CourtesyInstitute of Fine Arts, New York University.
A. J. GRAHAM
242
10
r~~~~~~~~~~~~~~~~~~~~~~~Oo
lood~
vi
5.. vi
8~~~~~~~~~~~~~~ ,A ',
I'~~~~~~~~~~~~~~~~~~~~'
/ , ..J'
:
-r
~ ~
~
~
-
~
y.
vi
YE
?~~~,. t
1 Xl
X2X3
,
/3 4
X
5
X
6
X
X10g X
7
X
X8
9
meters~~~~~~~~~~~~~~~ 1 0~~~~~ II
.
a
0
0 .
0
O
0
6
0
7
0
8
0
9
0
10
I!II
Y2
Y2
meters 0
100
200
300
400
500
600
700
B00
900
1000
Figure 2. Samothrace. Plan of city and sanctuary: 1) ancient harbor; 2) the city; 3) Sanctuary of the Great Gods; 4) North Cemetery; 5) Early Christian church; 6) 15th-century A.C. fortress of Gattilusi; A-D) wall gates; E) small gate. AfterLazaridis 1971,fig.34. CourtesyAthensCenterof Ekistics.
THE
52. Samothrace 4.1, p. 110. 53. Samothrace 4.1, pp. 117-122. 54. Samothrace 4.1, p. 110, but some of the potterywas later thought by Love to be of early-4th-centurydate; see p. 123. 55. Samothrace 4.1, pp. 123-154, cat. nos. 13-104. 56. Kuhn 1985, pp. 197-198. 57. Guide6,pp. 46-47. 58. Letter of 27 April 1977. 59. See Lazaridis1971, p. 19. 60. Bouzek and Ondrejova1985, 142 and fig. 113. p. 61. See Lehmann-Hartleben1939, pp.141-145. 62. See Samothrace 5, p. 320, n. 15; p. 344, ad no. 23. 63. Guide6,pp. 173-176; Karadima 1995. 64. E.g., Lazaridis1971, pp. 19, 83; Scranton1941, pp. 31-33 and 161, A.4.11; Guide6,p. 19. 65. See Sehrig 1927. Lawrence (1979, p. 353) is hardlyvery persuasive with his "conjecturaldating of shortly after 350," in view of his extraordinary reason:"noone before Philip (II) would have built a fortificationwall over a kilometrein length aroundthe town of the non-Greek islanders." 66. Kern 1893, pp. 381-384; see also Guide6,pp. 178-179; Matsas, Karadima, and Koutsoumanis1993.
COLONIZATION
OF SAMOTHRACE
243
Stratigraphicalevidence was provided first by a small amount of pottery recoveredfrom gaps between the large and widely spaced boulders of the foundation.52Twelve sherds of this pottery are described in the catalogue (by Iris C. Love), but they included some of 5th-century character, which were regarded as later intrusions.53The second stratigraphicalevidence came from under the floor in the northern half of the building, which was dated to the end of the 5th century by the latest pottery found beneath it.54All the pottery found under the floor was assumed to be from earliervotive gifts stored in the building, and some of it went back to the 6th century.55Kuhn, on the other hand, who regarded this pottery as the sole stratigraphicalevidence, decided, on the grounds that no evidence of an earlierfloor was discovered,that the stratigraphypointed to a construction date for the building near the end of the 5th century.The fragmentary Doric capital mentioned above, however, led him to conclude that the building was erected about the end of the 6th century or the beginning of the 5th, and provided with a new floor about 100 years later.56 The early date given to the Hall of Votive Gifts is, therefore,extremely insecure, and there is no building in the sanctuary that has a confirmed Archaic date. In any case, it is probably vain to hope for good evidence from the sanctuary for the date of the Greek colonization. The area has been excavated officially several times,57not to mention its being the target of illicit digging, and J. R. McCredie once informed me that he had "foundno stratum anywhere,in more than a dozen years of digging, that is earlier than the middle of the fifth century B.C."58 Plenty of 6th-century Greek pottery exists among the finds from the sanctuary,but that is at best only indirect testimony to the date of the Greek colonization. The city.To turn to the city, it was situated on a strong, well-watered site in the northwest of the island, and had a small harbor (Fig. 2).59There has been a little archaeologicalwork in the harbor,but the structuresdiscovered were assumed to be of Roman date.60There was also a little work done on the city site as a whole in 1938, but nothing earlier than the Hellenistic period was found.61More recently,a few sherds of G 2-3 ware were found by chance on the Acropolis.62Since 1995 excavations have been conducted on the city site, but the only structuresso far reported date to the 4th and 3rd centuries B.C. Some sherds of G 2-3 pottery were also found.63The magnificent standing city wall has been dated, at least in part, to the Archaic period, on the basis of its "Cyclopean"style,64but there are strong reasons to assign the whole construction to Hellenistic times.65So the city site has apparentlyyet to yield the information on the Archaic period that it surely possesses. The countryside.The only site in the countryside (Fig. 3) that offers any information likely to bear on the question of the colonial foundation date is the rural shrine of Mandal' Panagia. This remote mountain site, where Dimitri Matsas has recently conducted excavations,was first investigated in the 19th century by a local inhabitant, Nikolaos Fardys. His finds were seen and in part described by Otto Kern.66Later, some of the finds came to the Museum of Art in Bonn, and a selection of these were
244
A. J. GRAHAM
Figure3. Map of Samothrace.
described by Carl Fredrich.67To these we can now add the much more numerous finds of Matsas, recovered by surface exploration and regular excavation.A large proportion of the finds are easily recognizable as Greek dedications, and the very numerous terracottafemale heads are characteristic offerings to a Greek goddess. The letters IAIon a statue base found by Fardysnearby68have invited the restoration'ApT-ci']8t,and form the basis for the tentative identification of the shrine's goddess as Artemis.69The earliest of these dedications date to the 6th century, and that is taken as the time when Greek worship at the shrine began. The cemeteries.Two cemeteries at Samothrace have been partly excavated, the North Cemetery (Fig. 2:4) in the area outside the northwest corner of the ancient city, and the South Cemetery (Fig. 1:27), which lies close to the sanctuary,to the east of the southeast corner of the Hieron and to the south of the Propylon of Ptolemy.70 The term North Cemetery can now be taken to include the various cemeteries in the area separately designated by Dusenbery as Hotel (H), Area K (K), Road (R), and Waterpipe (W).71The earliest graves from the North Cemetery known to Dusenbery dated from the middle of the 5th century,72but we now have a much earlier burial,73which will be considered below. The pattern of use in the North Cemetery seems to have been entirely normal for an ancient Greek cemetery.74 In the South Cemetery,by contrast,the pattern of use was most abnormal. "A roughly rectangulararea less than three hundred meters square"
After Guide6,plan II. DrawingJ. Kurtich.
CourtesyInstituteof FineArts,NewYork University.
67. Fredrich1909. 68. IG XII.8 234. 69. The letters MIon a plain Roman sherd discoveredby Matsas could also belong to the name Artemis;see Matsas, Karadima,and Koutsoumanis 1993, p. 650. 70. Guide6,pp. 91-94. 71. See Samothrace 11, p. 5. Thus two cemeteriesonly,North and South, are distinguishedby Karadimaand Koutsoumanis(1992), who include Dusenbery'sH cemeteryin the North Cemetery (p. 677, n. 2). 72. Samothrace 11, pp. 9, 438-443. 73. Karadimaand Koutsoumanis 1992. 74. See Samothrace 11, p. 6.
THE
75. Samothrace11 p. 6. 76. Samothrace11, p. 8. 77. Samothrace11, pp. 8, 11691173. 78. Samothrace11, p. 8. 79. Samothrace11, p. xxiii, and vol. 2, passim. 80. Samothrace 11, pp. xxiv,no. 7. Thus the fragmentsof large"karchesia" (see below) mentioned by Love (1964, pp. 213, n. 54,221, n. 92) do not appear in Samothrace 11, pp. 743-744. 81. See Samothrace 11, pp. 3, 6-8. 82. Samothrace11, p. 7. 83. Samothrace 11, pp. 6, 702. 84. See Dusenbery 1967, pp. 117, 122. 85. See Samothrace 11, pp. 701-702; Beschi 1985, p. 52. 86. See Samothrace11; Dusenbery's prefaceis dated 1994. 87. See Karadimaand Koutsoumanis 1992, which did not appearuntil 1995. 88. See Johnston andJones 1978; for the dating, see pp. 103, 133. 89. See Johnston andJones 1978, p. 133, and, for the date of the Francois Vase,Cook 1997, p. 73. 90. In a letter dated 22 January 1996. 91. See Rizzo 1990, pp. 68-70, figs. 92 and 363.
COLONIZATION
OF SAMOTHRACE
245
was extraordinarilycrowded with graves, and continually used, reused, and overused.75Old graves were frequently disturbed and their contents dispersed by more recent burials, as the contents of the fill of the cemetery revealed.76In some cases the ash containers of earliergraves were carefully placed in groups, called by Dusenbery "accumulations,"by the diggers of later graves.77When the prevailing burial rite changed from cremation to inhumation during the 4th century B.C.,new earth was brought in to raise the level of the cemetery area.78These circumstances make it impossible to associate all the material found with specific graves, and Dusenbery uses the letter X to designate pieces that cannot be attributed to a grave group, but which she includes in her catalogue of material by categories.79In addition, large quantities of sherds from the South Cemetery,which are preserved in the Samothrace Museum, are not mentioned in the final publication.80 The exceptional pattern of use of the South Cemetery shows that burialtherewas seen as especiallyimportant and desirable,and Dusenbery's assumption that that was connected with its proximity to the sanctuary seems very likely to be correct. On the other hand, as she rightly emphasizes, nothing about the occupants or character of the individual graves distinguishes them from those in graves elsewhere at Samothrace, so only guesses can be made about the possibility that burial in the South Cemetery was reserved for a distinct group.81 Dusenbery notes that "cremationwas, with few exceptions, the method of disposal in the S Nekropolis during the sixth, fifth, and parts of the fourth century."82After the body was burned, the ashes were placed in a large ceramic vessel and one or more smaller pots were frequently also deposited in the grave. Dusenbery dates twenty of these cremation graves, nos. S1-S20, to the second quarter of the 6th century, and thinks that to be the time when the South Cemetery began to be used.83She has thus abandoned her earlier date, in the 7th century, which she once thought was indicated by the Lemnian stamnoi used as ash containers in some of the early graves, S2-S5 (though the last is called "?Lemnian").84 Although these stamnoi may not be closely datable, it does seem possible that they were made in the first half of the 6th century,85the date indicated by the best evidence in these early graves, the Attic pots in S1 (ca. 570), S7 (ca. 560), S9 (ca. 550), and S10 ("?Atticlekanis,"ca. 550). Although the nonAttic pots in these graves cannot be as securely dated, they can reasonably be attributed to approximatelythe same period. Since Dusenbery completed her text,86another early grave has been reported,this time in the North Cemetery. It is an infant's grave for which an SOS amphora was used as the container.87These distinctive transport amphoras were produced from the late 8th century into the 6th.88It is generally believed that production ceased sometime before 566, as Panathenaic amphoras were modeled on the a la brossetype, which replaced the SOS. The depiction of an SOS on the Francois Vase, however,which is normally dated ca. 570, and some finds in the West, suggest that the SOS was still around in the second quarter of the 6th century.89Alan Johnston informs me90that, while the Samothracianamphorais not closely datable, it is similar to one associated in an Etruscan grave at Cerveteri with painted pottery of the second quarter of the 6th century.91We may
246
A. J. GRAHAM
Figure4. GraveS252, as excavated.
After Samothrace 11, p. 410. Courtesy Institute of Fine Arts, New YorkUniversity.
Figure5. Restoredkarchesion, S252.1. H. 0.087 m. AfterSamothrace11, p. 410. CourtesyInstitute of Fine Arts, New YorkUniversity.
conclude that this grave, though certainly one of the earliest found at Samothrace, is not necessarily earlier than the earliest graves discovered previously.It does, however, show that these early burials were not confined to the South Cemetery. An exception to the rule that all of the early graves in the South Cemetery were cremations seems to be presented by grave S252, an inhumation of a young adult male in a pithos found in situ (Fig. 4). The grave contained one pot, the special sort of kantharos that we may follow Love in calling a karchesion (Fig. 5).92 The fabric of this vase identifies it as G 2-3 ware.93Dusenbery originally thought that the sherds belonging to this vessel had been swept into the pithos during a later period,94but she evidently changed her mind and decided that the karchesion was part of the original assemblage of the grave. Dusenbery included this grave in her category "Burialsof Uncertain Date," but she calls it "probablyClassical." Such a date is not consistent
92. Love 1964. Love expressedher pleasurethat Beschi (1996, discussion, p. 65) has adoptedher term,but it is not universallyaccepted.While Gebauer(1992, pp. 79-80) uses it in referenceto the finds at Assos, Stupperich (1990, p. 22) proposedinstead the use of Homer's words, depas amphikypellon,
and he has been followed by Utili (1999, nos. 685-690). 93. Samothrace 11, p. 744. Love (1964, p. 207, with n. 22) providesa fullerdescriptionof the vessel than Dusenbery'sat pp. 410, 744. 94. As reportedin Love 1964, n. 75. 217, p.
THE
95. As Samothrace11, p. 744. 96. Love 1964, pp. 212-215. 97. Beschi 1996, pp. 33-34; Di Vita 1979-1980, pp. 444-445. The destructionof the sanctuaryat Hephaistia had alreadybeen dated to 512/11 by Mustilli (1940, pp. 157158). 98. Beschi 1996, pp. 34,43; Beschi 2000, p. 77. 99. See Samothrace11, pp. 409-410. 100. See Samothrace 11, pp. 35-39, where S252 is not even mentioned. 101. See Archibald1998, pp. 61-63. The best exampleseems to be grave 15 at Ravna. 102. See Archibald1998, pp. 174176.
COLONIZATION
OF SAMOTHRACE
247
with the chronological indicators for the karchesion. It may well be true that in a general sense the shape had a very long life,95but the particular examples found at Samothrace can be placed in a reasonably restricted chronological range. Love used comparable examples from other sites to date the Samothracian karchesiafrom the last decades of the 7th century to the middle of the 6th,96but Beschi found numerous examples on Lemnos in the destructionlevels of the sanctuaryat Hephaistia and the Cabirion at Chloi. At the former site the destruction level was dated, chiefly by Attic pottery, to the last decades of the 6th century,and Beschi concluded that the destruction of both sanctuarieswas the work of the Persians,when they attacked the island in about 512/11.97 His discoveries have shown, therefore,that Love'slower terminuswas too high, and that these karchesia were in use throughout the 6th century.98Dusenbery's designation of "probablyClassical"is clearly impossible. The karchesion dates the grave, and that date must be the late 7th century to the end of the 6th. It is thus a unique early adult inhumation among the very numerous Samothracian cremation graves so far excavated. The karchesion is not the only exceptional feature associated with grave S252. Dusenbery notes that the top of the skeleton's cranium had been "split away,"and the top of another human cranium was also contained in the grave.These features are clearly visible in Figure 4.99Other examples of post-mortem mutilation of bodies are found in the South Cemetery, but they occur in graves so much later in date that they cannot serve as parallels for S252. In any event, no other case of this treatment of the skull was found, nor of the presence of another cranium.100 In her comments on the grave, Dusenbery aptly stated that the treatment of the skulls "recallsthe famous Scythian custom related by Herodotos 4.65," and there are Thracian parallels for the presence of other bodies or parts of bodies in graves.101It is often thought that such Thracian burialsshow that suttee was practiced,which is attested for some Thracians by Herodotos (5.5).102 We can only agree with Dusenbery's comment that "there is no certain explanation of the mysterious circumstances" in grave S252, but of one thing we can be confident: this was surely not a Greek grave. This is a very interesting fact, which inevitably stimulates speculation, but a single grave is an insufficient basis for historical hypothesis. Although grave S252 must be among the early graves discovered at Samothrace, the dating indication of the karchesion is too wide for us to conclude with confidence that it is among the earliest.Those are the burials placed by Dusenbery in the second quarterof the 6th century.This is the most reliable evidence that we at present possess for the date of the Greek colonization of Samothrace. It is also in agreement with the other chronological indications we have found: Antiphon's Samian tyranny;the relief of Agamemnon, Talthybios, and Epeios; the offerings at Mandal' Panagia; and the earliest evidence for Greek use of the sanctuary.For the moment, therefore,we can think of a foundation date for the Greek colony in the first half of the 6th century,but it is obviously possible that that date might be superseded by new archaeological discoveries.
A.
248
J.
GRAHAM
GREEKS AND NON-GREEKS IDENTITY
OF PRECOLONIAL
INHABITANTS
The first questionin the discussionof the relationsbetweenthe Greek colonistsand the non-Greekpopulationmust be, who were those nonGreeks?The literarysourcesfor the pre-Greekinhabitantsof Samothrace are very divergent.Diodoros (5.47.2) says that the originalinhabitants were autochthonous,and so therewas no traditionaboutthem or their leaders.As we sawin the firstsectionabove,accordingto Ps.Scymnusthey wereTrojansby descentbut calledThraciansfromtheirlocation;according to the Aristotelianpoliteiathey were Thracians;and accordingto Herodotos,Pelasgians.Herodotos,who had himselfbeen to the island, would normallyhavethe greatestauthority,but the historicalidentityof the Pelasgiansis very hardto pin down, and a strongargumentcan be madethat they arelargelya constructof Greekhistoriography.'03 The literarysourcesalso differon the questionof whetherthesepreGreekinhabitantswere presentwhen the Greeksarrived.Accordingto Ps.Scymnustheywere;accordingto the Aristotelianpoliteiathey hadleft beforethe Greeksarrived,with a time intervalthat,as we haveseen,canwhileHerodotoscannotbe pressed.His statementthat not be understood; tookoverthe cultof the GreatGodsfromthe Pelasgians, the Samothracians who lived there 7rpo6-pov,implies that the Pelasgians were present when the Greeks arrived, but, apart from the general uncertainty about the Pelasgians, Herodotos is clearly more interested in the takeover of cult than any historical coexistence. There is also some archaeologicalevidence for the precolonial inhabitants of Samothrace.The best-investigated site is Mikro Vouni (Fig. 3). Here a section 9 m deep through a small tell by the sea on the southwest coast of the island uncovered continuous occupation from the Neolithic, in the 5th millennium B.C., to the Middle Bronze Age in the second.l04 Very important though the discoveries are, they are clearly too early to contribute to the present discussion. Closer in time to the Greek colonization, but still a long way before it, is the Iron Age settlement on the hill of Vrychos, near the modern town of Chora (Fig. 3). The evidence is not rich, but the traces of the fortification on the summit of the hill, and of habitations lower down, appear to be similar to some on the Thracian mainland, and some of the pottery has similar connections. The site is taken to date from the 11th century B.C.105 The clearest evidence that the inhabitants were Thracians are the megalithic, so-called dolmen, tombs at Gialomandra, on the northern slopes of Vrychos.106 Similar tombs have been recorded in large numbers in parts of Thrace; they belong to the Early Iron Age.l07 At Mandal' Panagia (Fig. 3), in addition to the plentiful evidence of Greek worship mentioned above, surface finds included handmade pottery with incised decoration of"Thracian"type and bronze fibulas dating to the 8th and 7th centuries.These finds aretaken to show that the sanctuary was used by a Thracian population before the Greek colonization.'08If that is right, Mandal' Panagiaprovidesa little information about the inhabitants of the island in the period immediately preceding that colonization. The chronology at Mandal' Panagia is roughly the same as the chronology of the Sanctuary of the Great Gods,109where coarse handmade pottery,
103. See Lloyd 1976, pp. 232-234, 240-241, adHdt. 2.50.2, 51.1. Cf. Hall 1997, 72: "... Pelasgians-a name which occursthroughoutGreece and which would appearto be used without any particularlypreciseapplicationto indicate a populationthat was believed to be aboriginal." 104. See Guide6,pp. 165-169; Matsas 1984; 1991. 105. See Guide6,pp. 169-173. 106. Guide6,pp. 169-170; Moutsopoulos 1989, pp. 246-249; Andrioti 1929 (= Andrioti 1976); Moutsopoulou and Dimitrokalli1988. 107. See Archibald1998, pp. 64-66. 108. See Guide6,pp. 178-179; Matsas, Karadima,and Koutsoumanis 1993. 109. See Matsas, Karadima,and Koutsoumanis1993.
THE COLONIZATION
OF SAMOTHRACE
249
some of which had incised decoration, was also found.110It appears,therefore, as is only to be expected, that the same population was worshiping at the same time at both sanctuaries. Finally, as we have seen, grave S252 may have held a Thracian burial. The archaeologicalevidence for the precolonial inhabitants found to date is much less than we would ideally want, but what we have suggests that they were Thracians. If both the literary and the archaeological evidence are less than satisfactory,we need to turn to religion and language, which are partly intertwined. RELIGION
110. Samothrace 5, pp. 378-382. 111. Strab.10.3.20 (C 472). 112. FGrHist 107 F20. 113. See Hemberg 1950, esp. pp. 73-81. 114. As by Lloyd 1976, pp. 241242; Cole 1984, pp. 1-2. 115. In his later discussion,Burkert (1993) recognizedthat, and accepted the cult of the Cabiri;earlier,he had expresseddoubt:see Burkert1985, pp.282-285. 116. See Schwabacher1952. 117. See Hemberg 1950, pp. 160170 for Lemnos; pp. 37-43 for Imbros. 118. Beschi 1996, pp. 38-41; 2000, pp. 77-79. 119. Beschi 2000, pp. 77-79. 120. Beschi 1996, pp. 43-48; 2000, pp. 77-78. 121. Beschi (1996, pp. 40-41) gives reasonswhy the Lemnian cult may have predatedthat at Samothrace. Nevertheless,he sees the cult at Samothrace as originallyThracianwith Lemnian borrowings.Apart from the pottery and graffiti,he believes that the figure of Kadmilosat Samothracewas borrowedfrom Lemnos, where he is part of the genealogy of the Cabiri. That genealogy is given by Strabo (10.3.21 [C 472]), following Demetrios of Skepsis (named at 10.3.20 [C 472] and 21 [C 473]). Preferable,however,is the interpretationofJacoby,who regards Kadmilosas a Samothracian element inserted into a genealogy of the Cabiriinventedby Greeks for a cult to which such genealogizingwas foreign (FGrHistI, p. 378, Kommentarto 2 F20).
AND
LANGUAGE
As we saw in the first section above, the mystery cult of the Great Gods at Samothrace is stated by Herodotos to be rites of the Cabiri, which the Greeks took over from the Pelasgians. Demetrios of Skepsis, following his usual tendency, denied the connection of the Cabiri with Samothrace, using the argument from silence that there was no mythical story about the Cabiri at Samothrace.111He did, however, report that Stesimbrotos of Thasos, who was a contemporaryof Herodotos, stated that the Samothracian rites were for the Cabiri.112In modern times, Hemberg denied that the cult was of the Cabiri, primarily on the basis of another argument from silence, that their name has not been found on any extant inscriptions from Samothrace. He suggested that Herodotos was following his normal method of interpretatio.ll Hemberg has been followed,114but his argument does not seem good, since it requires the rejection of our two earliest authorities, both of whom had good reason to know the truth.115 In addition, there are representations of the Cabiri with pointed caps on Archaic coins of Samothrace.116While Hemberg could not know of this numismatic evidence, those who followed him could. It is important too that the cult of the Cabiri was practiced on the neighboring islands of Lemnos and Imbros.1l7Although the Italian excavations of the sanctuaryof the Cabiri at Chloi on Lemnos have not been fully published, we know that the cult was alreadyflourishing by the first half of the 7th century,and that the most common ceramic finds throughout the 7th and 6th centuries were drinking vessels (including karchesia), some of which were G 2-3 ware.ll8Beschi concludes that libations played a very important part in the cult of the Cabiri on Lemnos.l19That these vessels were also dedicatory offerings is indicated by the presence on many of them of graffiti and dipinti, which are in the language used by the Tyrrhenians,the pre-Greek inhabitants of Lemnos, and called, therefore, Tyrrhenian.l20By comparison with the brief Greek graffiti found on vases used by the later Greek inhabitants of Lemnos, which are dedicatory or indications of divine ownership, Beschi concludes that the earlier,Tyrrhenian, inscriptions were of the same type. In the cult at Samothrace, the same vessels were used during the same period, and the same types of graffiti arefound (see below). It is hard to doubt that the cult at Samothrace was the same as that of the Cabiri practiced on the neighboring islands, which are very close geographically and well in sight of Samothrace.l21 There is thus strong support for the literary evidence that the cult at Samothrace was of the Cabiri. That may be the truth that lies behind Herodotos's statement that the Samothracian cult of the Cabiri was taken
A. J. GRAHAM
250
Figure6. Inscriptionin non-Greek 2.1, pl.XXV, language. AfterSamothrace no.64. CourtesyInstituteof FineArts,New YorkUniversity.
over from the Pelasgians(see above),since Herodotos called the Tyrrhenian population of Lemnos Pelasgians.l22In any case, it is clear that the Greek colonists of Samothrace took over a pre-Greek cult. The same conclusion follows from the existence of a non-Greek language at Samothrace. EVIDENCE
FROM
INSCRIPTIONS
A relativelylarge number of graffiti have been found on pottery from various areas of the excavations. Some of these are straightforwardlyGreek, but others, although written in the Greek script, do not make sense as Greek.123 They cannot be examplesof the nonsense inscriptionsoften found on Greek vases; they were all incised after the pots were fired;they have no decorative quality;'24and the constant repetition of the same groups of letters shows that they made sense. There is also one inscription on stone, where the Greek letters do not make sense as Greek (Fig. 6).125 It is necessary,first, to consider attempts that have been made to read some of these inscriptions as Greek. Oikonomides126began from his interpretation of Diodoros 5.47.3: eoxarol o 8T arcXcarv 8iocv 8todex-cov oL TOOvov TlpoEiat. He stated oic ocroTX0ovwe, ;5roXX&?v O0T-Ois t(otLoat
122. Hdt. 6.136.2-140; cf. Thuc. 4.109.4; see Lloyd 1976, pp. 232-234, 240-241. 123. The pot graffitiknown up to that time were publishedby Lehmann in Samothrace 2.2. 124. As Lehmann, in Samothrace 2.2, pp. 22,61. 125. Samothrace 2.1, no. 64. 126. Oikonomides 1978, pp. 159161.
THE
COLONIZATION
OF SAMOTHRACE
25I
Figure7. Graffitoin non-Greek 2.2, pl.I, language. AfterSamothrace no. 1. CourtesyInstituteof FineArts,New YorkUniversity.
127. Samothrace2.2, p. 45, no. 1.
The forms of beta in Greek alphabets arevery varied;cf. Jeffery1990, p. 23. 128. Samothrace 2.2, pp. 47-52. 129. It is surprisingthat Cole (1984, pp. 107-108, n. 79) suggests that Oikonomides offers seriouscriticismof Lehmann. 130. See Georgiev 1977, pp. 143151, which largelyrepeatswhat he wrote in his earlierbook, Georgiev 1957,pp.31-39. 131. Line 6 in Fraser'stext: Samothrace2.1, no. 64.
132. Samothrace 2.2, pp. 8-13.
that 8dLc?x-covcannot mean language, so must imply a dialect of Greek. But that is simply false, as the article in LSJ, s.v. 8t&?XTxos;,shows. TFvb Oikonomides read Lehmann's no. 1 (Fig. 7) as [T?VS]?Tro-r'T? 'AvToxX?[ous;ov?Ox??v].For this reading two names and two arbitrary supplements were required.It was also necessary to take the seventh preserved letter, which Lehmann took as beta,127as epsilon, which it certainly is not. There are perfectly clear epsilons in the text. Oikonomides also proposed to read the AENTOAE/AINTOAEinscriptionsl28as 8v t66?. The word 8?v is assumed to be the neuter of 8si 8?v65, meaning no one or no thing (see LSJ, s.v.), which would give no intelligible sense.129 Although Georgiev accepted that many of Lehmann's graffiti on pottery are in a non-Greek language, probably Thracian,"30he also read Lehmann's no. 1 as Greek. He proposed the reading [L']?rx(oioa? or oi?t) His suggested abbreviationis unparalleled,and 6 cT?Xc?V?s 'AvTOx[X]?[Yl;]. T?X?V?o;is an invented word supposed to be equivalent to T-c?vitlr;.Like Oikonomides, he also readthe seventh preservedletter as epsilon. Georgiev also read the stone inscription as Greek and interpreted it as a thanks-forcure inscription, the main basis for which was his understanding of r3?xa in line 7131as the accusativeof Attic 3i1 meaning cough! Not surprisingly, he silently withdrew this suggestion from his later work, while still regarding the inscription as Greek. In order to read the inscription as Greek, however, he had to suggest Greek words that are very far-fetched and improbable, even with the help of of his supplements. These attempts should, therefore,be seen as very unconvincing.There is inevitably some residual uncertainty about the language of the most abbreviatedinscriptions, though even these can in most cases be identified by the longer versions of what are obviously the same designations. We may thus confidently accept that most of the non-Greek inscriptions represent a different language, even if we cannot certainly determine what that language was. Lehmann himself argued that the non-Greek language is Thracian.l32 The Samothracian inscriptions are, however, restricted in scope and the comparative Thracian material is very meager. In a careful discussion, Bonfante compared the frequencies of letters, diphthongs, and consonant
252
A. J. GRAHAM
clusters, and decided that, while the language could not be Greek, Tyrrhenian, or Etruscan, it could well be Indo-European and, more specifically,Thracian.l33 Apart from a few items in well-dated graves,l34none of the inscribed pottery came from closed deposits, and many of the sherds are not sufficiently distinctive to be datable.A substantial number of sherds, however, do belong to well-known categories of vases that can be dated. Imported vessels, in particular,are useful chronologically.'35While dating by the letter forms of such informal inscriptions is necessarilyeven more uncertain than usual, the pottery dates are sufficient to establish the chronology of the material in general and, in particular,the relative dating of the Greek and non-Greek graffiti. Lehmann himself did not generally give dates of the individual items in his corpus, but he followed a chronological order within his various groups,136and he made a general statement about the chronology of the material:"apartfrom some few E and OE inscriptions on vases, our excavations have yielded hardly any archaic or even fifthcentury Greek ceramic inscriptions, while the majority of the non-Greek inscriptions listed above are from these periods, in which the use of the The idiom for such purposes seems to have prevailed in the Sanctuary."'37 in the catahe for their of individual items left the republication dating It is a ratherlaborious task logues of the various areasof the excavation.138 to find the items in Lehmann's corpus that have been republished in later pottery catalogues, so I present a list in the appendix. By my count, 82 of his items have been so republished, and their dates provide as a whole strong support for Lehmann's chronological generalization. Some differences and other points of interest in the pottery catalogues should, however,be noted. The cataloguesof Love, Kopcke,and Dusenbery are here treated in turn,139 and account is taken, not only of the republications of items from Lehmann's corpus, but also of other pottery graffiti, which were not known to Lehmann. Love makes an important downdating of Lehmann's no. 1 (L1), a non-Greek inscription, which he put before 550 on the basis of the letter forms; see her A-A33.140The vase is an Attic glazed kylix of a type which had a "greatvogue in the second quarterof the fifth century B.C."She has two inscriptions securely dated in the 6th century, nos. V-A9 and A-A36 (L40 and L15). The vases are Attic kylikes of later 6th-century date. Two Ionian or local mortars, Love's V-A6 and V-A7 (L12 and L13), are assigned to the 6th or 5th centuries, but solely on the basis of the letter forms of the inscriptions. Of the presumed Greek inscriptions, her H-150 (not in Lehmann's corpus) is a fragment of a local unglazed beaker(?),and is dated to the 6th or 5th centuries on the basis of the form of the epsilon, which is the only letter inscribed. Love has an important note on L103, although it is not republished in her catalogues; Lehmann thought this might be of the 6th century,which would make it very early for a Greek inscription, but Love thought it was "probablyof later date."'14 Lehmann dated his no. 18, a non-Greek inscription, to the 5th-4th centuries B.C.,but Kopcke placed it in the second half of the 4th century.l42 This is a fairly late date for a non-Greek inscription. Kopcke also lowered the date of one of the Greek inscriptions, L98; see R-303. This is on part of an Attic lamp, and was dated to the 5th centuryby Lehmann, but Kopcke dated such "disklamps"to the 4th century.143
133. Bonfante 1955. Lehmann 2.2, pp. 9-11) presented (Samothrace
slightlyrevisedstatisticsof Bonfante's frequencies. 134. Samothrace 11, S103.1, S119.1, S128.1, S130.13, S178.1, Hl.l, H9.1, H9.2, H14.1, H17.1, H17.H. 135. Cole (1984, p. 11) was incorrectwhen she wrote that all the inscribedceramicdedicationswere of local manufacture. 136. See Samothrace 2.2, p. 5. 137. Samothrace 2.2, p. 62. The e and ?E inscriptionsareabbreviations of Oesv and by far the most common of the Greek inscriptions;see Samothrace 3, pp. 2112.2, pp. 21-23; Samothrace 214. 138. That is what is meant by his statement(Samothrace 2.2, p. 6) that his dates must be taken for grantedfor the time being, which Cole (1984, p. 108, n. 88) appearsto have misunderstood when she cites it in orderto throw doubt on Lehmann'schronology. 139. For these catalogues,see the appendixbelow. 140. L =Lehmann;for other abbreviations used in this section, see the appendix below. 141. Samothrace 2.2, p. 21; Samothrace3, p. 211, n. 5. 142. See Lehmann 1955, p. 98, no. 12; and Samothrace 7, pp. 298-299, where Kopckestates that none of the Attic or Atticizing low bowls, on one of which this graffito (R-110) was incised, need be earlierthan the middle of the 4th century. 143. Samothrace 7, p. 323.
THE
Figure8. Graffitoon a fragment karchesionfromthe South Cem11, etery, XS165. AfterSamothrace p. 1151. CourtesyInstitute of Fine Arts New YorkUniversity.
Figure9. Graffitofromthe Norti 2.2, p. 13( Cemetery. After Samothrace no. 319. CourtesyInstitute of Fine Arts, New YorkUniversity.
COLONIZATION
OF SAMOTHRACE
253
The most interesting graffito on pottery from the cemeteries is XS165 (not in Lehmann'scorpus).This is a fragment from a karchesion,on which are incised two letters, readby Dusenbery as IA (Fig. 8).144Theform of the letter that she takes to be an alpha, however, to judge by her illustration, could be a delta or possibly a lambda. Although Dusenbery thought that the inscription could be the end of a name, and thus Greek, it seems to me that it could be IA, i.e., Al retrograde,as in L17, or Al, as in L32, in which cases the inscription would be non-Greek. It would also be dedicatory. Dusenbery dates XS165 to the 6th century B.C., which suits the dating of karchesia discussed above, so, if these ideas about the reading are correct, we have another graffito, inscribed in the non-Greek language, securely dated to the 6th century. From the cemeteries there are two possible non-Greek inscriptions, which are late for that category. Dusenbery's H17.1 (L319) is a pithos with a number of letters incised on the shoulder,which Lehmann took to be a numeral. On the lip, however, are three signs that are well separated (Fig. 9). Two, a lambda and an iota, are clear.The third is taken to be a tau by Dusenbery,'14but the horizontal line is well below the top of the vertical, giving the shape of a cross. Lehmann did not take this mark for a letter, which left A and I, a combination found in several non-Greek inscriptions (cf. L28, L29, L32, L33, and possibly L34). The grave is well dated to the late 4th or early 3rd century, so, if this inscription is to be understood as AI, we have a non-Greek inscription of Hellenistic date. H14.1 (not in Lehmann's corpus) is a kantharos that bears on two of its four nonjoining fragments a delta and an iota (Fig. 10).146 The relationship between the two letters is not certain, but if we read AI, we have a combination well attested in the non-Greek inscriptions (cf. L3 to L21). The vase is dated to the late 4th or early 3rd century B.C., so this too would be a Hellenistic non-Greek inscription.
Figure 10. Graffito from the Non th I11, Cemetery, H14.1. AfterSamothrace p. 1157. CourtesyInstitute of Fine Arts, New YorkUniversity.
144. 1151. 145. 146. 147.
See Samothrace 11, pp. 115C Samothrace 11, p. 1156. Samothrace 11, p. 1157. See Samothrace 3, pp. 211-2 )14.
To summarizethe chronology of the pottery inscriptions,we have three (or four) examples that are securely dated to the 6th century.Where identifiable, they are non-Greek. We also have two possible non-Greek inscriptions of Hellenistic date, which slightly extends the period of use of the non-Greek inscriptions envisaged by Lehmann. There is no securely dated Greek inscription before the middle of the 5th century.The earliest of the G or OE inscriptions is a salt-cellar of the mid-5th century,147 and it seems doubtful if any Greek pottery inscriptions at Samothrace of earlier date have been discovered.To these chronological indications we may add the non-Greek inscription on stone (Fig. 6). The shapes of its letters were
254
A. J. GRAHAM
judged by its editor as not easy to date, but suggested to him the first half or the middle of the 4th century B.C.148 The simplest conclusion to be drawn from the above discussion is that a non-Greek language was in use at Samothrace for a very long period after the Greek colony had been established. There are two possible users of this language: a non-Greek people, who used Greek letters to write their own language, or the Greeks using their own alphabet but a foreign language.The latter possibility must be admitted, since Diodoros (5.47.3, quoted above, p. 250) states that many words of the earlier autochthonous language were used as a lingua sacrain the Samothracian cult down to his own day.It is economical to assume that our non-Greek inscriptions are in that language.149In view of Diodoros's methods, however, it is not safe to assume that his statement shows that the autochthonous language was still in use when when he was writing, in the 1st century B.C., as the words iEtP1T-cob v5v could have been taken from his source. Jacoby decided that the source of Diodoros's account of Samothrace cannot be known, but was probably a local historian who lived not long before Diodoros's own time. The information, however, probably came in part from high Hellenistic learning.?50If Jacoby is right, the earliest date to which -o vvv referswould be the 3rd century B.C., but it could be as late as the first. The content of the non-Greek inscriptions is consistent with the idea of a lingua sacra. Using the similar Greek pottery graffiti, Lehmann was able to show that most of the non-Greek inscriptions are likely to be the simple designations of divine ownership or dedications to be expected in a sanctuary.5'The Tyrrheniangraffiti and dipinti from the Cabirion on Lemnos provide, as we have seen, a very illuminating parallel.They are even very close in their detailed character,because most commonly they are single-letter abbreviations.'52In the Samothracian graffiti too, abbreviations, sometimes as short as a single letter, are commonly found, and they can also be identified from longer versions of what were obviously the same designations.l53As for the stone inscription, which shows similarities of terminations and uneven line endings (in spite of some attempt at vertical alignment of letters), it may well be a list of names, like the numerous later lists of initiates and theoroi on Greek inscriptions from Samothrace.154 It is theoreticallypossible, therefore,that the non-Greek language was used solely by Greeks as a lingua sacra.It is also possible that, as Lehmann suggested, the language was both spoken by its non-Greek speakers and used as a lingua sacra.When the population of Samothrace all became Greek speakers,the non-Greek language went out of everyday use. That would fit well the relative chronology of the non-Greek and Greek pottery inscriptions seen above.155In that case, the Greek colonists and their descendants lived side by side with the precolonial population and theirs. The Greek takeover of a pre-Greek cult shows that, when the Greeks arrived,the non-Greek people and the cult were there, and a short period of coexistence might seem to be the minimum requirement for an adoption of cult and language.The cult of the Cabiri on nearby Lemnos offers a suggestive comparison.There the cult also existed before the Athenians came to the island, and the Cabiri were worshiped at the same site for
148. See Samothrace 2.1, comments on no. 64. 149. Lehmannwas right to dismiss the idea that the languageof Diodoros might be differentfrom that of the inscriptions;see Samothrace 2.2, pp. 8-10. 150. See FGrHistIIIb (Text), pp. 470-471, Kommentarto 548 Fl. 151. Samothrace 2.2, pp. 13-18. 152. Of those that appearmost frequently,two, zeta and nu, seem likely to be abbreviationsof the words ZAPI[and NOFAIENA,which are also found;see Beschi 1996, pp. 43-48; 2000, pp. 77-78. 153. See Samothrace 2.2, pp. 13-18. Beschi (1996, pp. 46-47) gives useful tables,which make the comparison clear,but, unfortunately,he did not take accountof the republicationsof many items in K. Lehmann'scorpus,with the resultthat many of his dates arewrong. 154. This suggestionwas made both 2.1, commentary by Fraser(Samothrace to no. 64) andJeffery(1961, p. 299, n. 3), without referenceto each other. Their books appearedin consecutive years. 155. See Samothrace 2.2, pp. 11-12.
THE
COLONIZATION
OF SAMOTHRACE
255
many centuries by the Greek inhabitants. In that case, Herodotos (6.140) expressly states that the Athenians expelled the Tyrrhenian population. Beschi found a lacuna at the Lemnian Cabirion between the last archaeological material testifying to the Tyrrhenian practice of the cult and the first evidence for Greek worship.156He also sees significant differences between the conduct of the cult by the Tyrrheniansand that by the Greeks. The clearest of these is the importance of nocturnal rites in the Greek period, attested not only by literaryevidence (Cic. Nat. D. 1.42 [119]), but also by the many lamps found.157 Beschi envisages the lacuna lasting from the destruction of the Tyrrhenian Cabirion until the arrivalof Athenian cleruchs immediately after the middle of the 5th century.We know, however, that Athenian settlers arrived in Lemnos at about the beginning of the 5th century.l58Mustilli found the graves of the earliest Athenian settlers at Hephaistia, distinguished fromTyrrhenian,not only by their position, but also by the change of rite from cremation to inhumation, and by very significantly different grave goods. These graves are well dated to the beginning of the 5th century by their Attic lekythoi.159If Beschi's lacuna at the Cabirion is not merely the absence of evidence, which is so often deceptive, some time elapsed between the arrivalof the Athenians and the revivalof the cult of the Cabiri. In any case, the situation on Lemnos, where we know that the previous population was expelled, appearsto have been different from that at Samothrace,and that differencemaybe reflectedin the use at Samothrace of a non-Greek language.
CONCLUSION
156. Beschi 2000, p. 79. 157. Beschi 2000, p. 79. 158. Graham 1983, p. 175, n. 3. 159. Mustilli 1940, pp. 155-157. 160. I am gratefulto Alexandra Koukouzelifor drawingmy attention to this possibility.
We may conclude that, before the Greek colonists came to Samothrace, it was occupied by a population who were probably Thracians, worshiped the Cabiri, and maintained close contacts with the neighboring islands of Lemnos and Imbros. An analogy, which also concerns Samos, may be suggestive.'60When the Samian enemies of Polykrateshad finally to leave Samos, they exacted booty from two places and then settled at Kydonia in Crete, where, among other temples, they built one to the local goddess Diktynna (Hdt. 3.57-59). It is possible that an earlier group of Samian enemies of the tyrants exacted booty from Thracians, settled on Samothrace by invitation of the local inhabitants (to combine the information of Antiphon and Ps.Scymnus), and adopted the local cult of the Great Gods. We can imagine that they lived side by side with the precolonial inhabitants,who graduallybecame Greek-speaking and were absorbedinto the Greek community. But that is, admittedly, an imaginative, maximalist, reconstruction. The following conclusions can be drawn from the above discussion: first, that the Greek colonists came from Samos; second, that, on the best evidence we have, colonization took place in the first half of the 6th century B.c.; and, finally, that the relations between Greeks and nonGreeks at Samothrace are highly interesting and very important, but many uncertainties remain. It is possible that some of these uncertainties may one day be resolved by new archaeological discoveries.
A. J. GRAHAM
256
APPENDIX REPUBLISHED INSCRIPTIONS ON POTTERY Below arelisted the items from KarlLehmann'scorpus of ceramic inscriptions (Samothrace2.2) that have been republishedin the pottery catalogues of later Samothracevolumes. In order of publication, these are those of the Hall of Votive Gifts (V), the Altar Court (A), and the Hieron (H) by Love;6"'the Rotunda of Arsinoe (R) by Kopcke;162and the Nekropoleis (N) by Dusenbery.163No inscriptions were included in the catalogue of pottery from the Temenos published by Moore, even though a few of Lehmann's items originated there (nos. 32, 39, 88, 303), but she did not include finds from"disturbedcontexts"(apartfrom G 2-3 wareand karchesia), Also indicated are the and the "undisturbedfill"was restricted in area.164 or U(ncertain); the the of G(reek), N(on-Greek), inscription, language can be determined; or where this of the L(ocal), vase, I(mported) origin and the date (if given in the republication), either the number of the century B.C. or H(ellenistic). A question mark has been added if the suggested origin of the vessel is uncertain or a date is based entirely on letter forms.165 Lehmanns Republication Number Number Language
Origin
Date
I I I I
5 5 5 5
1 3 4 8
A-A33 A-A34 V-A1 A-A35
N N N N
12
V-A6
N
-
6or5?
13
V-A7
N
-
6or5?
15 16 18 19 25
N N N N N N N N
I I I? _
28 30
A-A36 V-A8 R-110 R-234 H-144 V-AlO A-A7 V-A2
I I
-
33
A-A14
N
L
4
36 38 40 46 47 49 64
V-A13 A-A22 V-A9 A-A23 A-A15 A-A16 R-235
N N N N N N N
L L I L L L
70 72 73 76 81
N V-All~~12~ 5^~ ~ N V-All 70~~66 N V-A N V-A3 G H-153 G H-127
~_27
6 5 4 _
161. I. C. Love, "Appendixto the Cataloguesof Ceramicsin Volume4.1 4.2, pp. 217-228; and 2," in Samothrace and "InscribedCeramics,"in Samothrace3, pp. 211-236. 162. G. Kopcke,"Ceramics,"in 7, pp. 275-326. Samothrace
I I 5
5 6 5 4 ~plate"
1~~I I I I L L
163. Samothrace 11.
5 5 4 4
164. M. B. Moore, "Catalogueof Finds: Ceramics,"in Samothrace 5, pp. 317-394; see esp. pp. 317-318. 165. One more item in Lehmann's corpushas been republished,but it is unfortunatelyconcealedby a misnumbering in Kopcke'scatalogueof the Rotunda.Kopcke'sno. 186, a "large inscribedwith a theta, is stated to be no. 200 in Lehmann's him by corpus.That cannot be, because200 is a glazed kantharos,and is actuallylisted by Kopcke (p. 288) with other such kantharoi.
THE
89 89b 91 92 93 95 96 98 102a 102b 107 111 114 122 123 124 125 131 135 140 150 154 161 162 167 177 178 179 182 194 201 211 213 214 216a 216b 228 229 245 249 252 253 258 259 260 262 297 317 318 319 320 321 321a 323
COLONIZATION
A-A24 H-129 A-A31 H-135 A-A25 A-A19 A-A26 R-303 H-148 H-152 R-102 R-122 R-183 A-A27 A-A28 A-A29 A-A30 H-130 H-133 H-128 H-131 H-132 H-126 H-120 H-121 A-A18 H-124 R-181 R-73 R-60 R-306 H-151 H-136 H-134 H-139 H-140 A-A17 H-147 R-9 H-145 A-A38 V-A5 H-149 R-255 A-A20 A-A32 R-233 N-H1.1 N-S128.1 N-H17.1 N-H9.2 N-H17.H N-S130.13 N-H9.1
OF SAMOTHRACE
G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G G U N? U U U U U U U U G N?
G? N? G G G
G?
257 L L L L L L L I L L I? I? I? L L L L L L L L L L L L L L I? L L I L L L? L L L L I? I I I L
4 H
4 4 4? 4 4 4 4 4 H
4 4 H 4 4
H 4 4 H H H 4 4 H 4 4 4 4 4 4 6 5 4
I I L? L? L?
L
5 H H 4 H H 4
258
A. J. GRAHAM
REFERENCES Andrioti, N. P 1929. "HIpo6iToptxoo Tccot0L?v ZSaciopaxix," Hpaxnxdr
r5CiaM7vlxiS avOp&(toAOY?X; ?ratp?cia 1929, pp. 54-64. 1976. "HpoLcroptxoo trapoLev
arov Soqio0opxn,"in 'AvriXapitata xaOrry1riNcxAco 1. 'Av8ptor7, Salonica,pp. 384-390. Archibald,Z. 1998. TheOdrysianKingdomof Thrace:OrpheusUnmasked, Oxford. Bernard,P. 1964. "Ceramiquesde la premieremoiti6 du VIIesiecle a Thasos,"BCH 88, pp. 77-146. Beschi, L. 1985. "Materialisubgeometricie arcaicidel Nord-Egeo: Esportazionidi Lemnos,"Quaderni del Centronazionaleper le ricerce112, pp. 51-64. 1996. "ITirreni di Lemno alla luce dei recenti dati di scavo,"in Magna GreciaEtruschiFenici.'Atti del trentatreesimo convegnodi studi sullaMagna Grecia,Naples, pp. 2350 (discussion,pp. 61-69). 1998. "Artee culturadi Lemno arcaica,"PP 53, pp. 48-76. .2000. "Gli scavidel Cabiriodi Chloi,"in Unpontefral'Italiae la Grecia:Attidelsimposioin onoredi AntoninoDi Vita,Padua,pp. 75-84. Blackman,D. 1999-2000. "Archaeology in Greece, 1999-2000,"AR 46, pp. 3-151. Boardman,J.1978. GreekSculpture: TheArchaicPeriod,London. , ed. 1984. CambridgeAncient III The History:Platesto Volume MiddleEast, the GreekWorld,and the Balkansto the Sixth CenturyB.C.,new ed., Cambridge. Bodnar,E. W., and C. Mitchell. 1976. in the CyriacusofAncona'sJourneys PropontisandNorthernAegean, 1441-1445 (Memoirs of the American PhilosophicalSociety 112), Philadelphia. Bonfante, G. 1955. "ANote on the SamothracianLanguage,"Hesperia 24, pp.101-109. Bousquet,J. 1948. "Callimaque, Herodote, et le tr6ne de l'Hermes de Samothrace,"RA29-30, pp. 105-131.
Bouzek,J., and I. Ondrejova.1985. Samothrace 1923, 1927, 1978, Prague. Burkert,W. 1985. GreekReligion: Archaicand Classical, J. Raffan, trans.,Oxford. .1993. "ConcordiaDiscors:The Literaryand ArchaeologicalEvidence on the Sanctuaryof SamoNew thrace,"in GreekSanctuaries. N. Marinatosand R. Approaches, Hagg, eds., London, pp. 178-191. Cole, S. G. 1984. TheoiMegaloi:The Cultof the GreatGodsat Samothrace, Leiden. Conze, A., A. Hauser,and 0. Benndorf. 1880. Neue archaeologische II, Untersuchungen auf Samothrake Vienna. Cook,J. M. 1964. Rev.of K. Lehmann, TheHall of VotiveGifts(Samothrace 4.1) in CR n.s. 14, pp. 117-118. Cook, R. M. 1997. GreekPainted Pottery,3rd ed., London. Coulton,J. J. 1976. TheArchitectural Formof the GreekStoa,Oxford. Di Vita, A. 1979-1980. "Attidella Scuola,1979,"ASAtene57-58, pp. 441-507. Dubois, L. J. J. 1818. Cataloguedesantiromaines, quitesegyptiennes, grecques, et celtiques,etc.,formantla collection deFeuM. le Cte.Choiseul-Gouffier, Paris. Dusenbery,E. B. 1967. "Samothrace: The South Nekropolis,"Archaeology 20.1, pp. 116-122. Fisher,S. M. 1996. "Trojan'G 2-3 Ware'Revisited,"StudiaTroica6, pp.119-132. Fredrich,C. 1909. "AusSamothrake," AM 34, pp. 23-28. Gebauer,J. 1992. "Die archaische geglattete graue Keramik,"in in Assos1990 (Asia Ausgrabungen Minor Studien 5), U. Serdaroglu and R. Stupperich,eds., Bonn, pp. 65-101. Ezik, Georgiev,V. I. 1957. Trakiiskayat Sofia. . 1977. Trakitei TechniyatEzik, Sofia. Graham,A. J. 1978. "The Foundation ofThasos," BSA 73, pp. 62-98.
THE COLONIZATION
.1983. Colonyand Mother City in AncientGreece,2nd ed., Chicago. .2001. CollectedPaperson Greek Colonization,Leiden. Guide1= K. Lehmann, Samothrace: A Guideto theExcavationsand Museum,1st ed., New York1955. Guide6= K. Lehmann, Samothrace: A Guideto theExcavationsand Museum,6th ed., Salonica 1998. Hall, J. M. 1997. EthnicIdentityin GreekAntiquity,Cambridge. Harris,E. M. 2000. "The Authenticity of Andokides'De Pace,"in Polisand Politics:Studiesin AncientGreekHistory,P. Flensted-Jensen,T. H. Nielsen, L. Rubinstein,eds., Copenhagen, pp. 479-505. Hemberg, B. 1950. Die Kabiren, Uppsala. Isserlin,B. S.J., et al. 1996. "The Canal of Xerxes:Investigationsin 19931994,"BSA 91, pp. 329-340. Jeffery,L. H. 1961. TheLocalScriptsof ArchaicGreece,Oxford. 1990. TheLocalScriptsofArchaicGreece,rev.ed., with a supplement by A. W. Johnston, Oxford. Johnston,A. W., and R. E.Jones. 1978. "The 'SOS' Amphora,"BSA 73, pp. 103-141. Karadima,H. 1995. "ApxaLoXoytxeg ?pycaeoi? otl Maprcveta xCatTn
To 1995," To ApXcaoSaooOp&axrl 2oylxo 'Epyoor Maxe&ovi'axac Opaxr79, pp. 487-496. Karadima,H., and M. Koutsoumanis. 1992. "ApXltOXoyLtxg; pyaoLiE SaEqio0paxrl1992," To Apcxaoxat Aoytxd'Epyoc7 Macxe5ovc'a epax7 6, pp. 677-683. Kern,0. 1893. "AusSamothrake,"AM 18, pp. 337-384. Koukouli-Chrysanthaki,H. 1992. Odiaos:Ta vexpoIpcoWoiTcroptxY zoO ou oxayo6 Trcxapa
Kacrorp,
Athens. Kuhn, G. 1985. "Untersuchungenzur Funktionder Siulenhalle in archaischerund klassischerZeit," Jd 100, pp. 169-317. Lateiner,D. L. 1989. TheHistorical MethodofHerodotus,Toronto. Lawrence,A. W. 1979. GreekAims in Fortification,Oxford.
OF SAMOTHRACE
Lazaridis, D. 1971. ZacoOp'axr7xaC 7 7rrpaia nq,, Athens.
Leaf, W. 1916-1918. "Straboand Demetrios of Skepsis,"BSA22, pp.23-47. Lehmann, K. 1955. "Documentsof the SamothracianLanguage,"Hesperia 24,pp.93-100. Lehmann-Hartleben,K. 1939. "Excavations in Samothrace,"AJA 43, pp. 133-145. Lloyd,A. B. 1976. Herodotus,BookII, 1-98, Leiden. Commentary Love, I. C. 1964. "Kantharosor Karchesion?A SamothracianContribution,"in Essaysin MemoryofKarl Lehmann,L. F. Sandler,ed., New York,pp. 204-222. Matsas, D. 1984. "Mxpo6BouvLSaoZLoOpa&xlg:MLa7TpodcropLtxxoLv6tYqxa o' va vqoLCt0xxo o ucr6o a Too BA
6, AtLyoou," AvOpo7ro2oycxod pp. 73-94. . 1991. "Samothraceand the NortheasternAegean:The Minoan Connection,"StudiaTroica1, pp.159-179. Matsas, D., H. Karadima,and M. Koutsoumanis.1993. "HavaT'vS oxacpi oanv Havayta t' MvrcaX' ZSao0opaxlSTo 1993," To ApXoaoAoytx6'Epyoaon Maxcovica xat Opdoxr7, pp. 647-655. Meiggs, R. 1972. TheAthenianEmpire, Oxford. Millingen, J. 1826. AncientUnedited Monuments:Statues,Busts,Basreliefs,and OtherRemainsof Grecian Art II, London. Moutsopoulos,N. C. 1989. "Tournee au Rhodope du Sud et a Samothrace,"in ThraciansandMycenaeans,J. G. P. Best and N.M. W. de Vries,eds., Leiden, pp. 246279. Moutsopoulou,N. K., and G. Dimitrokalli. 1988. "Tavxro6Xiev oq 'rhtaX6ojavtpoc't;
Saolxo0paxxg;,"in
H ao-ropx-r, xat capxacoAoy?cx AaoypacxpLx, EpEuvayea r7 Opaxr7.Irpaxexai, Salonica, pp. 17-37. Mustilli, D. 1940. "L'occupazione ateniese di Lemnos e gli scavi di Hephaistia,"in Studidi antichitaclassicaoffertida colleghie
259
discepolia EmanueleCiaceri,Rome, pp.149-158. Oikonomides,A. 1978. "Misread Greek Inscriptions,"AncW1, pp.159-166. und Prinz, F. 1978. Grindungsmythen (Zetemata72), Sagenchronologie Munich. Reinach, S. 1891. "LaVictoire de Samothrace,"GazettedesBeauxArts 5, pp. 89-102. Rhodes, P.J. 1972. TheAthenianBoule, Oxford. Richter,G. M. A. 1949. ArchaicGreek Art, New York. Rizzo, M. A. 1990. Le anforedi trasporto e il commercio Etruscoarcaico, Rome. Sakellariou,M. B. 1958. La migration grecqueen Ionie,Athens. Samothrace: ExcavationsConducted by theInstituteof Fine Arts ofNew York University 1 = N. Lewis, TheAncientLiter1), New York ary Sources(Samothrace 1958. 2.1 = P. Fraser,TheInscriptionson Stone(Samothrace 2.1), London 1960. 2.2 = K. Lehmann, TheInscriptionson CeramicsandMinor Objects (Samothrace 2.2), London 1960. 3 = P.W. Lehmann, TheHieron (Samothrace 3), Princeton 1969. 4.1 = K. Lehmann, TheHall of VotiveGifts(Samothrace 4.1), New York1962. 4.2 = K. Lehmann and D. Spittle, TheAltar Court(Samothrace 4.2), New York1964. 5 = P.W. Lehmann and D. Spittle, TheTemenos(Samothrace 5), Princeton 1982. 7 = J. McCredie et. al., The RotundaofArsinoe(Samothrace 7), Princeton 1992. 11 = E. B. Dusenbery,The 11), Necropoleis1-2 (Samothrace Princeton 1998. Sch6nert-Geiss,E. 1996. "Zur Minzpragung von Samothrake: Ein Uberblick,"in XAPAKTHP: AqVipwcao McdvrwOixovoTooApXOaoiuiov0(ArlO_loo aJLErTaca Xoylxo6AeXkiou57), Athens, pp. 271-276.
260
A. J. GRAHAM
Schwabacher,W. 1938. "Ein Fund archaischerMunzen von Samothrake,"in Transactions of theInternationalNumismaticCongress Organizedand Held in London,1936, J. Allan, H. Mattingly,and E. S. G. Robinson,eds., London, pp. 109120. 1952. "Cabirion Archaic Coins of Samothrace," ANSMN5, pp. 49-51. Schwartz,E. 1957. Griechische Geschichtschreiber, Leipzig. Scranton,R. L. 1941. GreekWalls, Cambridge,Mass.
Sehrig, H. 1927. "Surl'antiquitedes rempartsde Samothrace,"BCH51, pp.353-368. Stupperich,R. 1990. "Vorberichtiiber die Grabungin der WesttorNekropolevon Assos im Sommer in Assos 1989,"in Ausgrabungen (Asia Minor Studien 2), U. Serdaroglu, R. Stupperich,and E. Schwertheim,eds., Bonn, pp. 7-22. Tolle-Kastenbein,R. 1976. Herodotund Samos,Bochum. Utili, F. 1999. Die archaische Nekropole vonAssos(Asia Minor Studien 31), Bonn.
A.J. Graham UNIVERSITY DEPARTMENT PHILADELPHIA,
OF PENNSYLVANIA OF CLASSICAL PENNSYLVANIA
STUDIES 91O04
HESPERIA
71
(2002)
Pages 261-293
ITRAVAL
ARCH
ARROGANCE? DEDICATORY GREEK
INSCRIPTIONS
ARCHITECTURE
CLASSICAL
IN
OF THE
PERIOD
ABSTRACT Currentorthodoxyconsidersthe proliferationofarchitravalinscriptionsnaming the donors of architecturaldedicationsin the middle of the 4th centurya striking departurefrom Greek practiceof the High Classical period, when modest self-effacement is supposed to have been the rule. I argue,however, that a comprehensiveview of the evidence suggests substantial continuity ratherthan drasticchange:that inscribingpersonalnames on the architraves of Greek buildings is not the productof foreign influence or royalarrogance, nor an appropriationby individualsof rights previouslyexercisedonly by the state,but rathera naturaland predictablemanifestationofwidespreadGreek votive and epigraphicalhabits of long standing. Dedicatory inscriptions are attested on Greek architectural monuments from the 6th century B.C. onward.1They record dedications made both by individuals and corporate groups (usually cities), and appear on a variety of structures, including temples, treasuries, altars, stoas, gateways, fountain houses, and commemorative monuments. Extant examples for the 6th and 5th centuries, however, are far fewer in number, and sometimes less impressive in appearanceas well, than those from the 4th century and later.In fact, the second half of the 4th centurywitnesses a strikingproliferation of architecturaldedicatory inscriptions, most of them recordinggifts by individuals.2This proliferation has often been considered a product of the great changes in Greek social and political life from the Classical 1. I am indebted to more people than I can name for their comments on and responsesto earlierversions of this paperpresentedas talks at the Universityof California,Berkeley, the Universityof Virginia,McMaster University,and in Chicago at the 99th Annual Meeting of the Archaeological Institute of America.For advice, information,references,and comments
on written draftsI am particularly gratefulto Carol Lawton, Ronald Stroud,Hans RupprechtGoette, Molly Richardson,and anonymous referees.I wish also to thank the Arts ResearchBoard of McMaster University for financialsupport,and the Deutsches ArchaiologischesInstitut and Ecole francaised'Athenes for photographsand assistance.
2. No extant examplesof inscribed architecturaldedicationsby individuals areknown from the period 460-360 B.C., but more than a dozen date between 360 and 300. This dramatic increaseis not accompaniedby an analogousupsurgein dedicatoryinscriptionsnaming cities or corporate groups.
262
GRETCHEN
UMHOLTZ
to the Hellenistic period-an index of the decline of older civic values of the polis and the rise of the ambitious individual, no longer restrainedby the religious or social scruples of the Classical Age.3 In the archaeologicalrecordof the Classical period, architravaldedicatory inscriptions by individuals seem to make a sudden and spectacular debut, not in the heart of the Greek world, but rather in Caria, with the imposing dedications of Maussollos and his brother Idrieus in the sanctuary of Zeus at Labraunda (350s-340s
B.c.).4 These include two imposing
androns,one dedicated by Maussollos and one by Idrieus, a stoa dedicated by Maussollos, and a peristyletemple, a gateway,a suite ofoikoi, and another Doric building, all dedicated by Idrieus. Many of these monuments carryprominent, elegantly carved dedicatory inscriptions in the center of their marble architraves.5In his publication of these inscriptions, Jonas Crampa writes: "It was contraryto the older Greek spirit to record a public or private dedication of an edifice to (a) god(s) by an inscription on the walls or on the architrave,though some few instances are known."6In one form or another this view is shared by many scholars.7It continues to be widely accepted in part because it fits so well with much that we know, or think we know, about Greek (and particularly Athenian) political and social values of the 5th century and the hostility of Athenian democracy towardanyone setting himself abovehis fellows.8Accordingly,the phenomenon of private architectural dedication, and the prominent inscribing 3. For example,Lehmann (Samothrace4.2, p. 118) writes:"In the decade of 340-30 B.C. ... the recordingof such a personaldedicationof a building in a monumentalinscriptionon its entablaturewas a novelty attainableonly by a memberof a royalfamily.... The precedentfor dedicationsrecordedon the great altarsof Greek sanctuaries was offeredby city-states ... and, in the fourth century,only royaltycould assumesuch prerogatives."Similarly, with particularreferenceto temples, Hornblower(1982, p. 281) remarks: "Greekpracticeregardingtemple dedicationsis sociallyilluminating.In few other areasis it possible to trace so clearlythe developmentfrom the classical corporatismof old Greece to the assertivenessof the Hellenistic period." Other examplesinclude Morgan 1993, p. 19; and LabraundaIII.2, p. 5, both quoted below,as well as Botermann 1994, p. 182; Bean 1966, pp. 58-59; Picard 1965, p. 95. 4. The earliestof these is the andron dedicatedby Maussollos (Andron B) on the terracebelow the Temple of Zeus (discussedbelow).The monuments at Labraundamay soon have to surrendertheir status as the earliest
4th-centuryexamplesof architraval inscriptions.Ongoing work by H. R. Goette suggeststhat Athenian choregic monumentscould have taken the form of buildingswith inscribedarchitraves as earlyas the 360s B.C. (Goette 1999). 5. For these buildingsand inscriptions, see a fullerdiscussionbelow. 6. Crampa(LabraundaIII.2, p. 5) brieflyadumbratesa slow and sporadic developmentof the practice,which he sees becoming more common in the Hellenistic period and normalby Roman times. In light of eastern precedents(Egyptian,Lydian,and Persian),however,he concludesthat it was "naturalthat Mausollusand other Hecatomnids proudlyrecorded their dedicationsof monumental edificesby means of inscriptions, mostly placed on the architraves, where they could be readin the most impressiveway." The most comprehensivepresentation of the precedentsfor and significance of the dedicatoryinscriptions from Labraundais that of Hornblower (1982, pp. 286-288). The possible contributionsand significanceof Near Eastern(especiallyPersian)and Anatolian models arediscussedin
detail by Gunter (1985, pp. 118-119). 7. See, for example,Hornblower 1982, p. 274: "Byinscribingtheir dedications in conspicuouspositions on the architravesand facadesof religious buildings,the Hekatomnidsbrokewith a firm archaicand classicalGreek tradition.The traditionwas that which enjoinedself-effacementby the dedicants,howevergenerous,of such buildings.... Hence, precedentsfor Hekatomnidpracticemay be sought from beyond the Greekworld."Other examplesarelisted in n. 3, above. 8. The institution of ostracism,the cessationof the use of funeraryrelief sculpture,and the tendencyin portraitureto emphasizecommunally valued civic virtuesratherthan individual featuresall seem to supportthis generalview; see Scholl 1996, pp. 2629, n. 178. Yet, recent scholarshipalso highlights the complexityof ongoing tensions between individualand communalprestigeand the remarkable rangeof responsesto these tensions in variousarenasof Athenian publiclife; an excellentexampleis PeterWilson's (Wilson 2000, studyof the choregeia esp. pp. 11-49, 109-197).
ARCHITRAVAL
263
ARROGANCE?
in such a highlyvisiblelocationas of the nameof the donor,particularly hasbeen frequentlyviewedas an exampleof the citizensof the architrave, Greekpoleisadoptingthe practicesof foreigners,tyrants,and kings.9 of Greekarchitectural In the followingreconsideration dedications,I arguethatthe significanceandnoveltyof placinga dedicatoryinscription that foreigninfluon a building'sarchitraveshouldnot be overestimated, enceis negligibleforthis Greekpractice,andthatfundamentalpoliticalor religiouschangesneed not be invokedto accountfor the proliferationof survivingexamplesin mainlandGreecein the secondhalfof the 4th century.The inscribedmonumentsat Labraundahave deservedlyplayeda prominentrolein scholarshipon this topic (andwill be discussedin more detailbelow),but they arenot the sole, or even the primary,focusof this paper.The goalhereis to assemblea morecompletepictureof the rangeof possibilities,precedents,andexpectationsthatshapedGreekattitudestoloss dedications.In light of the irremediable wardinscribingarchitectural of one of the most essentialcategoriesof evidencefor classicalpractice, namelypaintedinscriptionson wood or plaster,it is veryimportantnot to ignoreor marginalizethe potentiallyrelevanttypes of evidencethat do remain.Accordingly,the followingdiscussiontakesa broadview,consideringepigraphical,aesthetic,andreligiousaswell as sociopoliticalfactors, I shallarand theirmanifestationsin othermediaas well as architecture. inscriptionsshouldbe viewednot as a departurefrom gue that architraval the traditionsof major(uninscribed)publicarchitectural monuments,but ratheras a consistentmanifestationof long-standingepigraphicalhabits applicableto personalreligiousdedicationsof all types.
EXAMPLES FROM THE ARCHAIC AND EARLY CLASSICAL PERIODS The extantexamplesof inscribedGreek architecturaldedicationsfrom the Archaicand EarlyClassicalperiodsarerelativelyfew in number,but theyprovideimportantbackgroundfor assessinglatermonuments.These earlyexamplesencompassa varietyof donors,buildingtypes, and geographicallocations,as well as of letter sizes, styles,and positionsof the inscriptionson the monuments.The followingsurveyillustratesthe variety and flexibilitythat characterizedthis epigraphicalform in its early stages,but also highlightsunderlyingaestheticprinciplesand sharedfeaturesthat continuein latermanifestations. One of the mostimpressiveearlyexamplesof aninscriptionon a Greek buildingis an imposing(ca. 8 m long) but imperfectlypreservedinscription carvedin largeletterson the verticalfaceof the limestonestylobateof the Templeof Apollo at Syracuse,constructedca.580 B.c.10Althoughthe stoneis badlyweathered,approximateconsensushasbeen reachedon the 9. Hornblower1982, p. 274: "Once the traditionhad been breached,private citizens followed the exampleof the satraps";see also Morgan 1993, p. 19 4.2, (quotedbelow), and Samothrace
p. 118 (quoted in partin n. 3, above). 10. IGXIV 1; SEGXXXI 841; Guarducci1985; letter height: ca. 0.150.18 m; drawing:Guarducci1967, p. 343, fig. 173.
GRETCHEN
264
UMHOLTZ
identification of most of the surviving letters, though not on their interpretation.11One possible reading of the text is: KXAeo[[..] e: iroiteae T6orXovL: ho Kv6i68aoc: xe?7x[X]?S TX?sac
: xoXa
FEpyc.12
Kleom[. .]es, son of Knidieidas, made (the temple) for Apollo, and Epik[l]es (made) the columns, beautiful works. The inscription is generally considered to be dedicatory, on the assumpbut it is not tion that TOIIEAONIshould be understood as -T 'A!o6XXcoVL, certain what is being dedicated-the entire temple, part of the temple, or something else entirely.13It is generally agreed that the principal dedicator is Kleom .. es son of Knidieidas, but his role and status are not clear.Likeor wise, it is not clear whether EIIIKAHis an adjectivemodifying oUToXerL the name of a second dedicator,Epikles.The use of the verb 7roi6o-e(rather than ave0-xs)) has suggested to some that Kleom .. es should be an architect, not simply a donor, but this need not be the case.14In conjunction with a new reading of the inscription, M. Guarducci has recently revived T.J. Dunbabin's suggestion that Kleom .. es was neither donor nor architect, but instead a local notable entrusted with responsibility for oversight of the project.l5I believe that the dedicatory natureof the inscription must imply some kind of personal financial contribution, whether direct (a contractor covering part of the costs) or indirect (an architect working without pay). Whatever his status and the precise nature of his contribution, the large scale of the lettering with which Kleom .. es advertisedhis role is not surpassedin any extant monument until the Stoa of the Athenians at Delphi at least half a century later. The contribution of the Lydian king Kroisos to the adornment of the great Archaic Temple of Artemis at Ephesos (before 547 B.C.) is attested
is 11.The extensivebibliography mostreadilyaccessedthroughEngelmann 1981 and Guarducci1987. 12. Guarducci1967, p. 343; on the basis of 19th-centurydrawingsshe has more recentlyproposedKXoJii[?v]?S: ?7roi6?a? TrcOXov : ho KvL&8Eia:
x?7ri?X?rtuXio: xa[X]&F6pya (SEG XXXI 841), readingiTriXEEas "seton foot, instigated"("esuscit6 i colonnati") as in Hom. Od.22.49 (Guarducci1987, expandingon her earlierpublicationof this readingin RendLinc37 [1982], pp. 13-20). Engelmann 1981, p. 94, based in part on word groupings suggestedby the interpuncts,reads: : ho .... .]?S : ?roiaCT?OrEXovL KXco[ xa[Xa] KvL&?[i]8a: x?7TLX?Ta-oXeLoc: es, son of Knidiei-
F?pYOc ("Kleo ....
das, made for Apollo also famous columns, beautifulworks").He considers xa[X(X)a]FoPyao("andother works")a
Svenson-Evers1996, pp. 382-383. possible alternatereadingfor the last 15. See n. 12, above.Dunbabin two words. 1948, p. 59, n. 3: "He was perhapsan Jeffery(1990, p. 265) notes that the official of the State chargedwith the inscriptionappearsto date laterthan the building itself (i.e., to the third or building of the temple, as Agathokles was of an earlyAthenaion";see Guareven fourth quarterof the 6th century). ducci 1987, pp. 44-45. Literary If the inscriptiontrulyis this late I referencesto individualsin analogous think it is likely to be a reinscriptionof roles include Agathokles at Syracuse an earliertext, perhapsoriginallycarved or painted elsewhere.Reinscriptionis (Diod. Sic. 8.11), Phalarisat Akragas attestedfor the Marathonbase beside (Polyainos, Strat. 5.1.1),Theron at the AthenianTreasuryat Delphi and Akragas(Polyainos,Strat.6.51 a close doublet of the Phalarisstory), for the Altar of the Chians, also at and the Alkmaionidaiat Delphi (Hdt. Delphi (discussedbelow). 13. For the Doric spelling'A7cXXcov, 5.62.3). Holloway (1991, p. 73) endorsesGuarducci'sview; he sees see Burkert1975, pp. 6-7. Kleom .. es as an epistatestaking par14. Svenson-Evers(1996, pp. 462ticular that this concludes 463, 466-467) pride in the columns as his own achievement:"Theywere an architect's not an is special inscription additionto what was expected,perhaps signature;for other examplesof even to what had originallybeen as causative(referring ?7TCOYlaE to the donor,not the artist),see planned."
ARCHITRAVAL
ARROGANCE?
265
by fragments of the inscribed base molding of four or five of the sculpted marble column drums: (1) [patALXei;S] Kp[oiooq &ave0Yxrv]
(2) [paaEixbSKpolaoosa&v]06x[?v] (3) (3a[aL?Xb;KpoiLOo;]av[?0rx?v] (4) [P3aL?Xeb;Kpoioo adV06]X]?V.16
Ki[ng] Kr[oisos] d[e]dicated (this). The placement of the inscription on these column drums is analogous to that of dedicatory inscriptions in other media. Greek votive gifts of all kinds often carried their dedicatory inscription on the base on which the gift was displayed, rather than on the object itself. Kouroi and korai, both votive and funerary,were frequently set on inscribed bases, although the figures themselves could also be inscribed.17This was both a practical and aesthetic matter: the base offered a plain, smooth surface suitable for inscribing, and the object itself was not cluttered or damaged thereby.In the case of Kroisos'ssculpted column drums, the inscribed molding under the feet of the figures functions visually as a base for the relief figures progressing around the drums. Similarly,at the Temple of Apollo at Syracuse, the stylobate (the base of the building as a whole) provides a clearlyvisible and readily accessible location for the dedicatory inscription. On the Greek mainland, a nearly contemporary marble Ionic treasury building at Delphi (mid-6th century) shows a different approach. This monument carried its dedicatory inscription on the architrave, in relatively small letters in a single line from near the left-hand edge, with a short, boustrophedon return at the end of the line. The text is quite fragmentary and no consensus exists on its restoration.Georges Roux has proposed the following: UIIoOLLo c6v 0Ylao0p6vTo6v8xao ta&y6aX[a[-a'ATc6XXcovt] [av0E6qx?]8sx6cT[av 6 aq&ioS6 Kv8i]cov.18
[The demos of the Knidians dedicated] this treasuryand the statue[s] to Pythian [Apollo] as a tith[e].
The restoration of the Knidians as the dedicators in this fragmentary text is based upon architecturalconsiderations in conjunction with a brief 16. IEphesos1518; Syll.36; London, British Museum B 16. See Hicks 1890, no. 518, with referenceto Hdt. 1.92 for Kroisos'shaving donated"the majority of the columns."For bibliographyand commentarysee Viviers 1992, p. 88, ns. 117-118, fig. 12; Schaber1982, pp. 13-18. Viviers (1992, p. 88, n. 118) notes that Schaber'sdrawing(1982, p. 14) is more accuratein some details than Guarducci1967, pp. 264-265, fig. 118.
17. Raubitschek(1949) presentsa wide selection. 18. Pouillouxand Roux 1963, p. 68; letter height: 0.03-0.06 m; photographsand drawings:Salviat1977, pp. 26-27. If Salviat'sreconstructionof the spacingof the blocks is correct(see below), then the actualtext must have been longer than this or any previous version. Syll.38 reads[ Kvi8LoL] x6v [ra 0lotxuopv -[6]v8? xaolxT&oXy6lX Hno0&i[a&vOev] E&x-raT[ 'ATc6XXwovL]
a&rc6 but Salviat(1977, Tx)y rToX?t?i]cov, p. 24) has establishedthat the first survivingletters must have been the beginning of the originaltext. Salviat (1977) does not offer a text himself, but a rough approximationbased on his discussionand drawings(esp. p. 32, fig. 6) might be as follows:TovOlqaau---ca. 6--p6v 6TOV8 xOaiTayaXFoja[Tx no0uico[------ca. 25-----] 'A6XXoMv] ex6x[av ------ca. 12------ ]v.
266
GRETCHEN
UMHOLTZ
mention in Pausanias(10.11.5).19The inscribed blocks are unusually shallow for an architrave,but the underside of the largest and best-preserved block has a smoothly finished soffit, with more roughly dressed resting surfacesat its ends only-clear indication that it was designed to span the opening between two supports.20With the exception of the final return, the lettering is generally confined to the upper third of the face of the blocks; this position, as well as the slight irregularityof the line, is very similar to that of the inscription on the stylobate of the Temple of Apollo at Syracuse.This is the earliest extant architravaldedicatory inscription; it may also be the earliest extant Greek architecturaldedicatory inscription by a corporate group. In Athens, parts of several inscribed altars survive from the Archaic period. The earliest of these (ca. 600-575? B.C.) is a large poros slab, the crowning element of an altar,or perhaps of a table, whose inscription ran from right to left across the long, smoothly dressed, front face: [-----ca.
: TaorxVt6ov 13------ avO]?exev 'A0evatxioaXcca[L]pov
KAXs8([o hoL6o].21
Cha[i]rion, [son of] Kledik[os, dedic]ated [the ... ] to Athena when he was treasurer. The dedicator may have been the same Chairion whose tombstone (in Eretria) identifies him as one of the Eupatridai, and who may ultimately have been connected to the family of Kleinias and Alkibiades.22 A poros altar of apparently somewhat smaller dimensions was dedicated to Athena Nike on the western bastion of the Acropolis at or shortly before the middle of the 6th century.23One block of the dado was found reused and in damaged condition; its front face was inscribed in five lines, beginning quite close to the top and left side of the block: ST;'A60[vcita<] -T; Ntx?s
Poll ,0. IoCTpox<X>?8[ES] ?7TCOt??EV.
vacat 24
Altar of Athe[na] Nike. Patrok(l)eid[es] made/dedicated (it). 19. Bommelaer1991, pp. 141-142; Pausaniasmentions the Treasuryof the Knidianstogetherwith those of the Athenians andThebans. 20. Salviat1977, pp. 28-29. The other,smallerfragmentsconformto this patternas well; for a drawingof the undersidesof the blocks, see Salviat 1977, pp. 26-27, fig. 2. Earliertheories about the placementof this course variedwidely;resting surfacesof the blocks,formerlyincorporatedin a reconstructionin the museumat
Delphi, were for many decades concealedby plaster(Salviat1977, pp. 23-24). 21. IG I3590; letter height: 0.060.07 m (roundletters smaller);height of face:0.30 m, length:2.30 m; Raubitschek 1949, pp. 364-365 (no. 330); photograph:Kirchner1948, no. 4. The placementof the letters,immediately below the upperedge, resemblesthat on the stylobateof the Temple of Apollo at Syracuse. 22. Raubitschek1949, pp. 364-365;
Davies (1971, pp. 12-15 [600, IV]) acceptsthe associationwith the funerarymarker,but does not follow Raubitschekin connecting Chairion's familywith that of Alkibiades. 23. For descriptionand drawingsee Mark 1993, pp. 32-34; his drawingis based on an earlierreadingof the dedicator'sname (Patrokles). 24. IG I3596; letter height: 0.030.04 m, lines 4-5 stoichedon;Raubitschek 1949, pp. 359-364 (no. 329).
ARCHITRAVAL
ARROGANCE?
267
Because of damage to the right-hand side of the stone, the termination of the dedicator's name is not certain. Here, as at the Temple of Apollo at Syracuse, rioL?a?((v) indicates responsibilityfor the dedication ratherthan direct authorship of the handiwork.25 From the second half of the century in Athens we have portions of two marble altars with dedicatory inscriptions on the front face of their crowning elements. The earlier of these (ca. 530-520 B.C.) is represented by three inscribed fragments; it was apparently dedicated by three individuals, but only the third name survives in full: [- - -!:?XMa]pXEt8s ii:.eo07?L
: ]otoV i][.
'
[-- -- zv Po]I6Ov.
vacat26
[..., Chai]releides, (and) Thopeithes made/dedicated [the al]tar [for/to... ]. The text runs in a single line from left to right, with letters very neatly carved in the upper half of the fascia; although the letters lie in the upper part of the band, they are set near the middle, rather than along the very top as was the case on the Temple of Apollo at Syracuse, the Knidian Treasury,and the altar dedicated by Chairion.27 The crowning fascia of a white marble altar of Apollo Pythios, dedicated by Peisistratos son of Hippias (in or shortly after 522/1 B.C.), bore an elegantly cut metrical inscription: jivuxa cr668?hS; &pX eneLoaoz[pa-rTo h7t7o 'At6XXovoq IIo0[]o ?VT?vl?.t.28
h]OL6oOixEv
In the temenos of Apollo Pyth[i]os, Peisist[ratos,] son [of Hippias,] dedicated this memorial of his archonship. The lettering runs in a single line, slightly above the center of a flat fascia immediately above a cyma reversamolding with lesbian leaf decoration.29 Like two of the earlier Athenian altars, and like the Knidian Treasury at Delphi, this monument is inscribed on a flat surface in a high position,
25. Raubitschek1949, p. 359: "Patrokleswas not the mason,but the dedicator";see also Mark 1993, p. 33, with bibliographyin n. 10. 26. IG I3605; letter height: 0.030.035 m (roundletters 0.02 m); height of face:0.115 m, length:2.30 m. Kirchner 1948, no.4. 27. For photograph,see Raubitschek 1949, p. 366 (no. 331). 28. IG I3 948 (IG I2 761; EM 6787); SEGXXXI 31; cf.Thuc. 6.54. The date, based on the archonshipof Peisistratos(Develin 1989, p. 47), has recentlybeen much discussed.Viviers (1992, pp. 87-88) rejectsImmerwahr's view (1990, pp. 18, 76) that the in-
scriptionmay have been recut at the end of the 5th century.Vivierspoints to similaritiesbetween this inscription and contemporaryor earlierIonian inscriptionsand associatesit with inscriptionson works by Endoios or his atelier.Dillon (1995, esp. pp. 62-65) surveysand rejectsvariousrecent arguments for dating after522/1 B.c.; Aloni (2000, pp. 84-86) also arguesagainst downdating. 29. The letteringbegan at the lefthand edge of the monument;at the end of the line the letteringbecame somewhat less tidy and precise-less regular in both horizontaland vertical positioning.
268
GRETCHEN
UMHOLTZ
rather than on the base; this is the location that would yield the greatest visibility. The impressive marble altar dedicated by the Chians in front of the great Temple of Apollo at Delphi, ca. 500 B.C., was also inscribed on its crowning molding (Fig. 1): Tov f3coLl6v.30 Xtot 'Ao6Xow.vL The Chians (dedicated) the altar to Apollo. Pierre Amandry has argued that the surviving inscription belongs to a Hellenistic version of the monument, but that its text and position, though not all details of its letter forms, replicate those of the Archaic original.31 The brief inscription consists of a single line running from the left edge and stopping well short of center;for maximum legibility, the letters nearly fill the height of the band in which they are located. The prominence and visibility of the inscription derived not only from its lofty location, but also from the color contrast between the white marble of the crowning molding and the blue-black stone of which the altarwas primarily made. Perhaps the most prominent dedicatory inscription of all at Delphi was that of the Stoa of the Athenians, just below the Altar of the Chians, against the great polygonal wall supporting the temple terrace.The precise date of the stoa is controversial,but should fall somewhere between 30. Syll.319a;FdD III.3, no. 212; Homolle 1893, p. 614 (briefmention); 1896, pp. 617-620; Amandry 1986, p. 206. 31. Amandry 1986, pp. 216-217. The architecturepreservesArchaic molding profiles,but uses hook-clamps and horizontalpour channels.The letters do not replicateArchaicforms-
Amandrynotes (1986, p. 217) that "if it were not for the moulding and Herodotus, this inscriptionwould have been dated to the fourth century."The monument as seen on site today is the result of a modernreconstructioncarriedout partiallyin 1920 and more fully in 1959 (Amandry1986, pp. 206-209).
Figure1. Inscribedcrowning moldingof the Altarof Apollo, Delphi. CourtesyEcolefrangaise d'Athenes, neg. photoP.Amandry, L.4187.51
ARCHITRAVAL
ARROGANCE?
269
Figure2. Stoa of the Athenians, Delphi. CourtesyEcolefran9aise d'Athenes, photoP.Amandry, neg.10.836
480 and 450 B.C.32The inscription on its stylobate is impressively large and beautiful, with letters among the largest to have survivedfrom the 5th century (Figs. 2-3). It is carved in a single line and the letters occupy nearly the full height of the stylobate, running from left to right. The inscription reads: xal Toxh6noX[ax]at -TcxxpoTzpto 'AOvcodot&aviOeocv Tc'voToV XOC hsX6vT?c; TOv 7oXes[JIO]V.33
The Athenians dedicated the stoa and the equipme[nt a]nd the stern ornaments, having taken them from their ene[mie]s. The Stoa of the Athenians follows the well-established practice of placing a dedicatory inscription immediately beneath the dedication itself. 32. Pausanias(10.11.6) states that the Athenians built a stoa out of the spoils taken from the Peloponnesians and their allies;he mentions ships' ornamentsand bronze shields and recordsthe names of the cities from which the Athenians dedicatedfirst fruits.Pausaniasinfers from the
accompanyinginscriptions(no longer extant) that the victoriesinclude those of Phormio (429 B.C.). Becausefeatures of the architectureand letter style suggest a much earlierdate, however, many scholarsbelieve that Pausanias has mistakenlyconflatedthe dedication of the Peloponnesianspoils with the
originaldedicationof the building. Amandryand Fomine (1953, pp. 112114) believe that this is a dedicationby Kimon of booty from Mykale and Sestos (479-478 B.C.),but Walsh (1986) arguesfor a later date.The 470s seem most probableto me. 33. Syll.329.
270
GRETCHEN
UMHOLTZ
But in the relation of letter height to height of inscribed surface, it is effecting at a much larger scale the aesthetic preferences shown in the Artemision column drums and in the crowning moldings of the altars of the Chians and the younger Peisistratos.The prominence of this inscription was achieved not only through the size of its lettering,but also through its position. Because of the steeply rising course of the Sacred Way, the inscription on the stylobate of the stoa would appearat eye level or higher as one approached.Thus the location on the stylobate is in no sense a sign of modesty-it is in fact an impressive example of conspicuous display, successfully exploiting particularlocal conditions.34 Several other, less well preserved, early instances may also be mentioned here. Thucydides (6.54.6-7) reports that Peisistratos the Younger dedicated the Altar of the Twelve Gods in the Agora of Athens, but that its inscription was obliterated in a later enlargement of the altar.No traces of this (or any subsequent) inscription now survive, but the expansion in which it vanished seems to have occurred in the final third of the 5th century B.C.35At Delphi, Plutarch reports that the Corinthians sought and received permission to inscribe the treasurydedicated by Kypselos as that of the city.36Plutarch'swording does not offer a clear indication of whether this inscription constituted a replacement of one dedicatory inscription with another.37One fragmentaryporos block inscribed with the 34. Of course,practicalconsiderations also play a role:the entablature of the stoa seems to have been of wood and thus not as suitableas the stylobate for a carvedinscription. 35. For chronologyof the altar,see Gadbery1992, esp. pp. 466, 470-472, 487. 36. Plut. De Pyth.or.13 (Mor. 400D-E); in responseto Sarapion's question tL8aqT6voTxovo6 KoqiXou -o5 avao0?vro;&XXaKopLv0iovovoPlutarchrepeatsthe story of OtcCoootv, the guides:-r ; Topavv83oS xaoraXocrr po6Xovro Kopiv0tot xa. T-rvev O?ioq
IIlc r )XPCoouv Oav8ptXavTaxociTbv ?vTaOuaToouovI 0qloaop6ov rcLypaCat xTl ToCX?oewG. AEXYolo [ICv o)v
i05ooav (oS
'HXeiouo;8 8Lxatov xaci ouvEcXOpjloav, (pOovyCocavTa-g; ecpiqtAavTo [MII[L?T?XELV
'Io0[ezov.See also Hdt. 1.14. 37. Scholarsassumingreplacement of an originalKypselidinscriptioninclude Flaceliere(1937, p. 60) and Schr6der(1990, p. 286); those rejecting a Kypselidinscriptioninclude Bousquet (1970, p. 671) and Partida(2000, p. 181). Herodotos (1.14) does not mention inscriptions,but does distinguish the roles of Kypselosand the
Figure 3. Detail of inscription on stylobate of Stoa of the Athenians, Delphi. CourtesyEcolefrangaise d'Athenes,neg.12.688
Corinthians.Without more precise informationon timing we cannot be sure,but if it came as late as the 5th century(see n. 38 below), the Corinthians'requestto inscribethe building as their own may have arisen primarilyfrom a desirefor greater visibilityfor their city'sname in the intensely competitivelandscapeof monuments in the sanctuary,rather than a need to make a specifically antityrannicalgesture.For examples of inscriptionslater added to existing treasurybuildings,see n. 41, below.
ARCHITRAVAL
ARROGANCE?
27I
letters KOPINhas previouslybeen taken as a possible remnant of the (dedicatory?) inscription referredto by Plutarch. Jean Bousquet, however, has associated this block with another closely similar fragment on which the letters AE survive; he offers the plausible restoration A4[Xpoie6oxocv] 7rpoowtvrTLarv] KopLv[0LoLS ("The De[lphians granted promanteia to the] Corin[thians]").38Accordingly,we cannot be certain whether Plutarch refers to this inscription or to another, nor whether Kypselos'sname originally appeared on the building; if it did, it would be the earliest known example of an inscribed architecturaldedication by an individual. At Olympia, Pausanias reports (6.19.15) that the inscription of the treasurynearest the stadium says that the treasuryand the images within it are a dedication of the Geloans. He does not indicate where this inscription was located, and no traces of it now survive.The Sikyonian treasury, dating to the first half of the 5th century,is inscribed on the front face of the anta, perhaps indicating a preference for display closer to eye level: The inscription was carved over the joint beZexuo)vLo ("Sikyonians").39 tween two blocks, of which only the upper survives;this failure to avoid the joint may indicate that the lettering began immediately below some object mounted on the anta above it. Without the lower block we do not know whether there was more to the inscription and cannot tell whether it is a dedicatory inscription for the building or for some other offering.40 Architectural fragments from Delphi and Olympia are not abundant enough to indicate whether most treasurybuildings carriedinscriptions or where these would have been located, but the examples that survive suggest that a range of positions were possible.41
GENERAL OBSERVATIONS ON EARLY EXAMPLES Several conclusions-epigraphic, aesthetic, and sociopolitical-may be drawn from this survey of architecturaldedicatory inscriptions of the Archaic and Early Classical periods. Although the inscription from Syracuse 38. Bousquet 1970, pp. 672-673. The scriptis Delphian, of the first half of the 5th century B.C.; the blocks
probablybelong to the horizontal cornice of the building.Letter height ca. 0.13 m; the letters fill most of the height of the flat face on which they are inscribed,like those on the stylobateof the Stoa of the Athenians.I disagree with Partida'sview (2000, p. 180) that "oftenthe awardof promanteiaor proxeny... was preferableto a plain dedicatoryinscription."Although both may confer prestige,these are two very differentcategoriesof inscription-one pertainingto a direct transactionwith the god and the other to a relationship with the Delphians or the sanctuary administration.The Altar of the
Chians carriedboth a dedicatoryand a promanteiainscription. 39.IvO 649; SEGXLII 393; brown sandstone. 40. IvO 650 (roughlycontemporary),carvedon the side of one of the antae of the same treasury,seems to recordanotherdedication.It too has lettering that crossesthe joint between blocks. 41. Amandry(1988, pp. 600-601) notes that poor preservationof the architraveof the SiphnianTreasury does not allow certainty,but concludes: "Le nom des donateursetait probablement gravea l'architravedu tresor;il l'etait sirement au linteau de la porte." The (partiallypreserved)inscription on the lintel is FdD III.1 197, a pro-
manteiainscriptionof uncertaindate, perhapsof the 4th centuryB.C. On the fragmentaryarchitraveof the Massaliot Treasury(second half of the 6th century B.C.)arepreservedthree letters carvedin the late 5th or early4th century:Z A A (a second 2 was later added);see Daux 1958. At Olympia, too, single-wordinscriptionswere added to some buildingslong after the originalconstruction(IvO 653: Mey[ap]?ov,on the architraveof the MegarianTreasuryin letters of the Roman period;IvO 652: Zariwxv,on a shallowblock from an unknown building,letters of the Hellenistic period);these may be more properly consideredas labels than as dedicatory inscriptions.
272
GRETCHEN
UMHOLTZ
poses difficulties in detail, all of the architectural dedications discussed above employ the patterns or formulas that were alreadywell established for votive gifts in other media. They include various combinations of the basic elements of Greek dedicatory inscriptions: the name of the donor, name of the recipient, identification of the gift, and occasion of the gift. The presence or absence of any of these elements seems to be determined by circumstances specific to the monument, rather than the result of any special rules or habits applicable to architectural dedications in particular.In short, nothing in the wording or formulas of these dedications distinguishes them from inscriptions on dedications in other media.42 Great variety is evidenced both in letter size and in placement of the inscriptions, but some apparent preferences may be noted. Well-defined, narrowbands, such as stylobates and bases (or base moldings), or crowning fascias of altars,were often favored locations; this preference is seen as well in the habit of inscribing flutes of dedicatory columns (Fig. 4).43Such bands offered convenient limits and structure for the lettering, giving it prominence,but also maintaining harmonywith the monument as a whole. Substantial variety in placement on the building is possible. Flat stone faces in highly visible positions are ideal, but it is particularcircumstances such as material, scale, and the nature of the site (and probably of nearby monuments and inscriptions as well) that most directly affect the placement of the inscriptions.44Clarity and prominence are consistent goals, but there are no fixed or universal rules governing how these should or should not be achieved. 42. It is worth noting, however, that althoughin offeringsof other sorts (generallyquite small in scale) the formulais sometimes reducedto just the name of the divinity,no certain case survivesof an architectural dedicatoryinscriptionthat did not include the dedicator'sname as well.
43. Examplesof votive columns thus inscribed:Raubitschek1949, pp. 5-28. On all of the monuments discussedthus far,with the exception of the Treasuryof the Knidiansat Delphi and the SikyonianTreasury at Olympia, the inscriptionsare positionedwithin a distinct band. 44. Highly effective areasfor display
Figure4. Votivecolumndedicatedby Ithidiketo AthenaPoliouchos, Athens (EM inv.6241). Courtesy Deutsches ArchaologischesInstitut,Athens, neg. NM 2403
included the crowningmoldings of relativelylow monumentslike altars, the stylobatesof buildingswhose scale or location renderedthis position more highly visible (Temple of Apollo at Syracuse,Stoa of the Athenians at Delphi), and the architraveor anta of smallertreasurybuildings.
ARCHITRAVAL
ARROGANCE?
273
Finally, it should be noted that extant examples of Archaic and Early Classical architecturalinscriptions are not restricted to any particulargeographicalor political center.Moreover,both individuals and poleis arerepresented as dedicators.45In both cases the inscription serves to record a relationship with the divinity in question and to present the achievement and status of the donor to the appropriateaudience.
ARCHITRAVAL DEDICATORY AT LABRAUNDA
45. The fact that no inscriptionsby poleis are attestedwithin cities'own territorieshas led many to conclude that poleis inscribedbuildings that were dedicatedoutside of their own territory, but not those within. I suggest below that the primarydistinctionsunderlying this phenomenon are not necessarily ones of geographyand audience,but ratherof the natureand funding of each project,i.e., the more technical details of the building'sstatus as a dedication. 46. For building phases, see Labraunda1.3, pp. 40-46. For early dedicationsat the sanctuary,see Hornblower1982, p. 278. 47. Hellstr6m 1996a, p. 134. 48. The superstructureof the stoa is not well enough preservedto assessthe degree of conservatismor innovationof its architecturalorder. 49. A good, recent overviewof the features,functions, and significanceof this monument,with currentbibliography,may be found in Hellstrom 1996a, pp. 134-136; the architectureis discussedby Westholm (Labraunda1.2, pp. 45-51, 84, 93-95, 103-108), as well as by Hellstrom (1981; 1989; 1996a; 1996b).
INSCRIPTIONS
There are no extant examples of architecturaldedicatory inscriptions from buildings of the second half of the 5th or the early 4th century. When they do reappearin the archaeologicalrecord,later in the Classical period, these inscriptions show greateruniformity in style and placement than did the Archaic and Early Classical examples. The building material is now almost always marble, and the architraveis frequently the favored location. This pattern (i.e., architravalinscription in marble) is used with great consistency, authority,and grace by Maussollos and Idrieus in their dedications at Labraunda. Examination of the relationship of these monuments to Greek traditions can contribute, at least indirectly, to a better understanding of architecturaldedicatory behavior on the Greek mainland in this period for which direct evidence is so scarce.To that end, the following account reviews the architectural dedications at Labraunda in light of the aesthetic principles and epigraphicalhabits identified above as characteristic of Greek practice in the Archaic and Early Classical periods. The Hekatomnid buildings at Labraunda include traditional Greek forms, Anatolian building types, and some unusual innovations. The heart of the sanctuarywas a small, Archaic temple of Zeus, ultimately replaced by Idrieus with a larger peripteral temple (perhaps completing work begun by Maussollos).46Around this earlier temple, Maussollos added two buildings, probablybeginning in the early 350s.47One of these, the North Stoa, was a very traditional type of building for any Greek sanctuary,but the other, Andron B, was unusual, perhaps unique, both in its design and its decoration; it employs Greek architecturalorders and ornament in a nonconventional way and its plan seems to owe more to Anatolian than to Greek building types.48In broad terms, the andron, distyle in antis, resembles a large treasurybuilding or small temple. It was, however, intended to accommodate ritualdining and also included a broad,elevated niche across the back of the interior, perhaps an indication of dynastic or ruler cult functions; the exterior featured a boldly mixed order,with a Doric entablature supported by Ionic columns.49Maussollos's brother and successor Idrieus dedicated an andron, gateway, and set of oikoi, and replaced the earlier temple with a larger peristyle structurein the Ionic order. These buildings are unusual in many respects, but in their dedicatory formulas they are utterly conventional and fully in accordancewith Greek
GRETCHEN
274
UMHOLTZ
traditions and practice.The inscription ofAndron B (before 352 B.c.) may be taken as representative(Fig. 5): Mot6oocoXXo;'ExocTJIyvco[a&v0qxe TOVa]vO8pvo Ail Aotipoc6v&cot.vacat50
[xoc]l tO& vE6vtra
Maussollos, son of Hekatomnos, [dedicated the a]ndron [an]d its contents to Zeus Labraundos. The other architecturaldedicatory inscriptions from the sanctuaryall follow the same pattern, with some slight variation in nomenclature. Although some of the other inscriptions are fairly heavily restored, in most cases the architecturalevidence for number and spacing of missing letters makes these restorations quite secure. The dedicatory inscriptions of Idrieus differ from those of his brotherin including the adjectiveMoXoCaEE6 in addition to the patronymic'Exato,6tvo.51Both Maussollos'sand Idrieus's inscriptions present a fairly full version of the standard Greek dedicatory formula,with dedicator'sname, verb,direct object,and indirectobject (Zeus Labraundos) explicitly included in almost all instances. The following examples illustrate how consistently and conventionally this formula is used. 50. Labraunda111.2,pp. 9-10, no. 14, pl. 2; letter height: 0.10-0.12 m. The term for the structurededicated, i.e., andron,is somewhatsurprising--in Greek contexts it generallyrefersto the receptionroom of a house ratherthan a freestandingbuilding.For discussionof the use of androns,with ancientliterary references,see Labraunda111.2,p. 10; see also Hellstr6m 1989; 1996b. Maussollos and Idrieuseach dedicatedan andron;the other structureswere designated in more familiarterms:naos, stoa, oikoi. 51. The relevantportions of the inscriptionsof the temple and of Andron A are not preserved,but those of the Oikoi, South Propylaia,and Doric buildingall include MuXaeoS;.
Crampa(Labraunda111.2,p. 6) suggests that by adopting this formula,which emphasizedlocal ties, Idrieuswas pointedly distinguishinghimself from his brother.Hellstrom (1996a, p. 138) gives greaternuanceto the distinction: Idrieuswas concernedmore with internal domestic consolidationand therefore stresseshis native Carianidentity, whereasMaussollos,looking primarily to his role on the internationalstage, had no reasonto stresslocal roots. It is also possible that Maussollos'somission of MuXaoes6was not intended to downplayhis Carianidentity,but instead to emphasizethat Labraunda was a pan-Carian,ratherthan a merely local, sanctuary.
Figure5. Dedicatoryinscription fromAndronB, Labraunda.Photo author
ARCHITRAVAL
ARROGANCE?
275
The North Stoa on the temple terrace,dedicated by Maussollos, carried its dedicatory inscription in three lines on the eastern anta: [M]a6.o.oo; 'Ex.a[.6.vo] TV CTOIqVv v a&vE0YX? ALLAal3pao6vLot.
vacat 52
vacat
[M]aussollos, son of Heka[tomnos], dedicated the stoa to Zeus Labraundos. The Oikoi dedicated by Idrieus on the temple terracecarrieda single-line inscription on a Doric architrave: 'I8pLS 'qExaorio6vco MoXaaoei a&vEOYX?E TOb; o'ixooS AtL Aaq[ipa6v8oiL. vacat53 Idrieus, son of Hekatomnos, of Mylasa, dedicated the oikoi to Zeus Labraundos. The Temple of Zeus was inscribed in a single line on the upper of the two fascias of its Ionic architrave; the direct object and the dedicator's ethnic are missing, but have been plausibly restored: 'I8pLeb;gExca[xoivCo MoXaoEsig avFOlxe T6v voc6v AL\. Aa4iP3pc6]vGoL. vacat54 Idrieus, son of Heka[tomnos, [Zeus Labrau]ndos.
of Mylasa, dedicated the temple] to
Three other monuments dedicated by Idrieus in the sanctuary carried analogous formulas on their architraves (the Propylaia, Andron A, and the Doric building), as did a gateway at Amyzon.55 The sanctuary at Labraunda is exceptional in the narrowly focused chronological range of its architectural development and in the degree to which that development was the work of two individuals. Yet the novelty of the circumstances of construction of the monuments should not lead us 52. This inscription(Labraunda III.2, p. 8, no. 13, pl. 1) is unique in being the only quasi-stoichedon inscriptionat Labraunda;letter height: 0.024 m. It begins at the top left cornerand is not verticallycentered on the block;dimensions of block: H. 0.565 m, W. 0.69 m. Later,below this dedicatoryinscription,was carved the text of a decreeof 267 B.c. honoring Apollonios, oikonomos of Ptolemy. If Doric, the columns of this stoa were probablyca. 4.5 m high: Hellstrom 1996a, p. 134. 53. Labraunda111.2,pp. 14-15,
no. 17; letter height: 0.06 m. The letters are set 0.16 m abovethe lower edge. 54. Labraunda111.2,p. 13, no. 16; architraveheight: 0.56 m; upperfascia: 0.29 m; lower fascia:0.22 m; letter height: 0.08 m, with letters set 0.105 m abovethe lower edge of the fascia.For the architectureof the temple, see Labraunda1.3. 55. Propylaia:Labraunda111.2, pp. 15-16, no. 18; Andron A: Labraunda111.2,pp. 11-13, no. 15; Doric building:Labraunda111.2, pp. 17-18, no. 19; Propylaiaat Amyzon: Robert and Robert 1983, pp. 93-
94, no. 1; OGIS235. Labraundahad two closely contemporarygateways, designatedby the excavatorsas the South Propylaiaand the East Propylaia (LabraundaI.1), though Crampaconcludes that tuX6vwas the more likely term.Westholm (Labraunda1.2, pp. 96-97, 111-112), followed by Crampa(LabraundaIII.2, p. 16), suggeststhat the East Propylaiawas the earlierbuilding,destroyedand reusedpartlyin the constructionof the South Propylaia.Partsof the lettering were removedin conjunctionwith this reuse.
GRETCHEN
276
UMHOLTZ
to overestimate the nature and extent of aesthetic and conceptual innovation represented by their inscriptions. The Hekatomnids at Labraunda (and Maussollos at his tomb) repeatedlycombined Greek with non-Greek architecturalidiom and brilliantly clad non-Greek practices and institutions in Greek-style artistic splendor. A contemporary Greek would undoubtedly have found the sight of an entire sanctuaryfilled with architectural dedications by two men arrestingand remarkable.But would he have found the form or location of the dedicatory inscriptions themselves surprising? I believe not. The following discussion highlights the closeness with which the aesthetic features of the Hekatomnid architravalinscriptions reflect contemporary Greek practice for scale and positioning of votive and commemorative inscriptions. The separate, and more complex, issue of how inscribed architecturalvotives fit with Greek ideas concerning entitlement and propriety is then considered.
AESTHETIC
PRECEDENTS
AND CONTEXT
We have already seen that, as Greek dedicatory texts, the Hekatomnid architecturalinscriptions are quite unremarkable:their formulas are in no way unusual for or inconsonant with Greek traditions.56In aesthetic terms as well, the inscriptions at Labraunda follow closely the same principles that governed Archaic and Early Classical architectural inscriptions throughout the Greek world. Some variety exists, but most are carved on the smooth face of a narrow horizontal band at a prominent location on the building. The architravalposition places them at a height above the viewer comparable to that of the inscriptions of the Altar of the Chians and of the Treasuryof the Knidians at Delphi. It is possible that the use of two fascias, rather than three, on the Ionic architrave of the Temple of Zeus at Labraundawas intended specifically to better frame and accommodate the handsome dedicatory inscription.57 More importantly, however, the Labraundainscriptions fit comfortably not only within the broad aesthetic parametersobserved in architectural inscriptions of the Archaic and Early Classical periods, but also with related monuments in their more immediate High and Late Classical context. Although thus far discussion has focused on freestanding structures, these were by no means the only kind of inscribed architecturein the Classical Greek world. From the later 5th century onward, many funerary monuments develop increasinglyelaboratearchitecturalframes.These can provide an overhanging brow, offering some protection from weather, but for the most part they are intended to highlight and set off the images within. It is easy to appreciatethe architecturalcomponent in the examples in which frame and relief are carved from a single block (Fig. 6), but many funerary monuments at a larger scale (usually less well preserved) offer even more impressive parallelsfor the inscribed architravesat Labraunda. In these larger monuments the relief sculpture was set into a separately constructed architecturalframe. Many of the most impressive reliefs became separatedfrom their architecturalframes long ago, but fragments of such frames, at all scales, indicate that the architrave,or an architravelike
56. This point is emphasizedby Gunter (1985, p. 119). 57. Hellstr6m andThieme (Labraunda1.3, p. 54, n. 66) survey precedentsfor architraveswith two fascias;they note that three fascias would mean that letterswould need to be smaller,and that the use of a Doric architraveon Andron B (despiteIonic columns) allowedletters to be larger than would be possible on an Ionic architravewith fascias.The letters on Andron B are0.10-0.12 m high, whereasthose on the temple are only 0.08 m high.
ARCHITRAVAL
ARROGANCE?
277
Figure6. Funeraryreliefstele of Damasistrate,Athens (NM 423). CourtesyDeutsches Archaologisches Institut, Athens, neg. NM 423
band, was the most frequently used location for inscriptions (Fig. 7).58 Imposing as these monuments were, they were not necessarilyintended to suggest the heroization of the deceased.59 Votive reliefs too made use of architravalareas to carry inscriptions. Their development seems, in loose terms at least, to accord closely with that of funeraryreliefs:both appearin the final decades of the 5th century and become more numerous and varied throughout much of the 4th century.But differences in function and context of display lead to differences in appearance as well. Architectural frames are rarer for votive than for 58.Theselarge-scale architectural framesaregenerallythought to date to
the middleof the 4th centuryorlater, butdatingcanbe difficultwhenthe sculpturalcomponentshave been lost.
of the useof Oneironicconsequence
larger,independentarchitecturalframes is that many of the figuresrepresented on the most imposing funeraryreliefs
are now nameless,while their less grandlycommemoratedcontemporaries remainclearlylabeled on one-piece stelai. Funerarymonumentswith independentarchitecturalframesseem to become largerand to occur more frequentlyas the 4th centuryprogresses,but their differencefrom the late-5th-centurystelai is one of scale
ratherthanconception. 59. Ridgway(1997, pp. 160-161) and Clairmont(1993, p. 40) emphasize that the naiskosframedoes not, in and of itself, imply heroization.This is importantfor the presentargument:it was not only "heroes"who could have their names inscribedon architraves.
GRETCHEN
278
[.
A
A _
A
A
1
A3
UMHOLTZ
A
-
I
_
f
TPATokIEOYC :OY NIE r
t
E0 O S
Ot::
E N OY
: OySI MAi:P:0tAi;Oi::Nd;fliOifYt:00 Yt MAPAO
Y:
A A i:: i;:50
X
O 0 Y Y
F
funeraryreliefs.60When they do appear they can occasionally be naiskoslike in form, but architecturalframes without a gable (resembling the side of a building, or perhaps a stoa) are far more common, in part because they can accommodate larger groups of figures and more complex actions. Although some votive reliefs feature inscribed architravesor architravelike bands, many more were set up on tall bases, which would have carriedthe dedicatory inscriptions.6'These raised bases seem often to have provided a more spacious and highly visible location for inscriptions, with architraval inscriptions appearing only on a few of the larger monuments, though even some of the smaller architecturalframes may originally have carried painted inscriptions, now lost.62There was apparently no single, correct place for inscriptions nor way of inscribing them, nor was there a hierarchy of ostentation in inscription location. Individuals chose how and where to inscribe, with practical circumstances determining how best to attain the goal of maximum visibility. These habits of funeraryand votive inscribing do not offer conclusive evidence for contemporary practice in full-size, functional, freestanding architecture,but they do suggest that, in other media at least, architraval (or similarly prominent) labeling was a standard part of the visual landscape of late-Sth-century Athens.63From the late 5th century onward, the city of Athens was surroundedwith inscribed funerarynaiskoi or naiskosstyle stelai. The Peiraieus may have been even more crowded with these impressive monuments, virtuallyall carryingthe names of the deceased on the architraveor on flat fascias or bands near the top.64In sum, inscribed architraves,including personal names in both funerary and votive contexts, had been commonplace in Athens, and probablymuch of the rest of the Greek world, for half a century before the start of the Hekatomnid building program at Labraunda. 60. Architecturalframesfor votive offeringsare discussedbrieflyby van Straten(1992, pp. 265-266). For votive reliefs,van Straten(1992, p. 265) observesthat "antaeseem to occurfrom ca. 420 B.C., the complete architectural
framesomewhatlater."Earlyexamples of funeraryreliefswith architectural framesinclude the stele of Ktesilas and Theano, ca. 410-400 B.C. (see
Stewart1990, fig. 430); the names of the deceasedareinscribedon the architrave. 61. For the practiceof setting votives up on a pillaror tall base
(at a level allowingfor effective visual display),see van Straten1992, pp. 248-249. 62. I am gratefulto Carol Lawton for these observations,and for further informationand bibliography concerningvotive reliefs.Document reliefsfrequentlyfeatureprojecting moldings (taeniaplus other forms) between sculptureand text, often inscribedwith the opening line or heading of the document (Lawton 1995, p. 11); for developmentof more explicitlyarchitecturalframes in the 4th century,and votive con-
I
0
fr
N I E N E.' Y
Y .
-
Figure7. Inscribedarchitraveof funerarynaiskosof Alexos,Athens (NM 2584). CourtesyDeutsches ArchaologischesInstitut,Athens, neg. 72/3870
notations of these frames,see Lawton 1995, p. 12. 63. This kind of labelingoccursin a broadcontinuum,including true architraves,fasciasresemblingarchitraves,horizontalgeisa, and flat surfacesor bandswith no particular architecturalanalogywhatsoever; modern attemptsto isolate architraval inscriptionsfrom this continuumand view them as a strikingnew departure impose an artificialisolation on one aspectof a much more broadlybased phenomenon. 64. See Scholl 1994; 1996.
ARCHITRAVAL
279
ARROGANCE?
ENTITLEMENT
AND PROPRIETY
Given the widespread use of architravalinscription in other media, it is clear that if the architraves of votive structures were never inscribed in Classical Greece, this would have represented a specific and very narrow exclusion-a notable exception to an otherwise widely used set of practices and expectations for epigraphicaldisplay of individuals'names. Many scholars believe that a specific exclusion of this sort did indeed exist, that the rules for architecturaldedications were different than those for dedications of other types, and that the inscription of a donor's name on an architecturalvotive, or on a temple in particular,would have been unacceptable in a way that inscriptions on gifts of other sorts were not.65The likelihood or extent of this exclusion is difficult to assess, since much of the relevant evidence is either negative or indirect.66In the following discussion I first address the question of what kind of limitations, if any, applied to individuals'dedication of buildings in Classical Greece, and then consider specifically epigraphical restrictions. There is no question that limitations of various sorts did apply to Greek votive behavior. Many categories of offerings could only be given by those entitled to do so. Obvious examples include the erection of military trophies, choregic monuments, and athletic victor statues. The right to offer these types of dedications had to be earned by conspicuous achievement in the appropriatefield.67The finds at most Greek sanctuaries suggest, however, that, apart from commemorative dedications of those types, almost anything could be dedicated by almost anyone.68 Fulfillment of vows, and expressions of gratitude for economic prosperity or for other divine favors received or hoped for, offered scope for a tremendous range and variety in personal dedications. Special local traditions of the cult or the requirementsof specific ritual occasions may have exerted some influence, but generally,for most dedicators at most sanctuaries, the principal constraint seems to have been their own financial 65. For example,Henner von Hesberg (1994, p. 39) writes:"In der Polis klassischerZeit hat es starkeWiderstandedagegen,dass eine Einzelpersonihren Namen auf den von ihr gestifteten Bau setzt." Similarly,Botermann(1994, p. 181): "Tempelwurden den Gottern von Gemeinden oder Individuengebaut, ohne dass diese ihrerEitelkeit durch eine BauinschriftAusdruckgaben." Amandry (1988, p. 600), by contrast,emphasizescontinuityof dedicatoryexpectationsand practice acrossall media:"Les tresorsetaient des monumentsvotifs, au meme titre que les groupesde statueset autres ex-voto. Quelque forme que revetit l'offrande,le donateurne laissaitpas
ignorerson identite."Amandryis not here addressingthe question of temples specifically,or of individualvs. corporate donations,but I would maintain that the same principlesapplyin all cases. 66. I am not aware,for example,of any instancein which a building known to have been a personaldedicationcan be conclusivelydemonstratedto have been uninscribed. 67. Even in the case of military monuments,however,there was apparentlyscope for both publicly awardedand privatelyinitiated commemoration.For example,a fragmentof a 5th-centurydedication from Eleusis, a relief depicting a man leading troops,is inscribed
with the text [I1u066opo]q'ETltXo i7[7x]apX[oa; ----ca. 8--
(IG I3
999). Other kinds of monuments may have been subjectto restriction accordingto context, ratherthan type. For example,in recognitionof distinguishedservice,the state could awardthe right to set up a herm in the northwestpart of the agora(AgoraIII, pp. 301-313); this was a carefully controlledactivityat this location, but it is not clearwhat restrictions, if any,would have appliedto the setting up of herms elsewhere. 68. For discussionof variousaspects of Greekvotive behavior,see van Straten 1981 (esp. pp. 78, 81, 88, 102); 1992.
280
GRETCHEN
UMHOLTZ
capacity.69This general situation of freedom and abundance in individual votive behavior is reflected in the complaint in Plato's Laws about the proliferation of dedications and founding of shrines for every conceivable reason.70 Social, political, or family status could be a requirement or limit in determining the nature of participation in various cult activities (e.g., the Panathenaic procession, the Eleusinian mysteries), but these do not seem to have been decisive factors in determining the kinds of votive offerings individuals could make.This point emerges clearlyin Diane Harris'sstudy of the inventories of dedications to Athena stored in the Parthenon and Erechtheion in the 5th and 4th centuries B.C. She notes that the most impressiveofferingsarenot necessarilymade by the dedicatorsof the highest social status, and that few of the dedicators recorded in the surviving inventories are of particularlyhigh status, or otherwise attested at all.71 But if, in Classical Greece, differences in status among individuals do not impose significant limitations on what may be dedicated by whom, what of that larger difference, considered by many to be absolutely fundamental: the difference between individuals and states?Were there kinds of votive gifts or offerings that states could make that individuals could not? Again, it would seem not. Greek states could awardhonors and privileges to individuals, and could appoint officers to regulatevarious aspects of the organization and financial affairsof cults, but there does not seem to have been any specific type of gift that a state alone could dedicate to a divinity. It has often been assumed that temples differed significantly from other dedications in this respect. For example, Catherine Morgan writes: From Archaic times at least, the right of temple commission was a prerogativeof the state (although individuals, such as the Alkmaionidai at Delphi, often contributed money or materials), and this remained the rule outside the east Greek world until changing concepts of kingship and personal rule during the Hellenistic period encouraged the kind of personal investment evident in, for example, the Philippeion at Olympia (c. 335 B.C.).72 In fact, however, although extant physical remains are scarce, literary and epigraphical texts attest a number of examples of private sponsorship of temples and other religious monuments in mainland Greece in the Classical period. Themistokles may have built a temple to Artemis(?) Herkane before the battle of Salamis.73He restored the telesterion of the Lykomidai at Phlya and decorated it with paintings (probably in the 470s B.C.).74 He also built a temple to Artemis Aristoboule near his home 69. Indeed, the traditionof aparche dedicationssuggests that financial prosperityvirtuallyrequiredthat thanksbe renderedwith a suitably impressivededication. 70. Leg. 10.909e-910a. This passage is cited and the phenomenon discussed by Uta Kron (1996, p. 166). For the introductionof new cults by individuals,see Parker1996, pp. 215-217.
71. Harris 1995, p. 228 (with specific referenceto the Erechtheion inventories). 72. Morgan 1993, p. 19. 73. SEG XXVI 121 (perhapsearly 1st centuryA.C.), concerningthe repair of sacredbuildingsin the Peiraieus, mentions (line 45) a shrine of -kane, which Themistokles set up before the battle of Salamis.Garland(1992, p. 76)
considersa shrine to Artemis Herkane a plausiblecomponent of Themistokles' religiousactivities;Parker(1996, p. 155, n. 9) urges skepticism:"theascription to Themistocles in a text of this characterhas little authority." 74. Plut. Them.1.3. Near the end of his life he also set up a shrineto the Dindymian Meter in Magnesia (Plut. Them.30).
ARCHITRAVAL
ARROGANCE?
28i
in Melite.75 Shortly before 419/8 B.C., a certain Telemachos (otherwise unknown to us) founded the sanctuaryof Asklepios on the South Slope of the Acropolis in Athens and set up a wooden propylon at its entrance.76 This propylon, perhaps depicted in the relief decoration of the monument Telemachos set up (ca. 400 B.C.) to commemorate his foundation, was probably replaced in stone during the course of the 4th century.77Konon built a sanctuaryto Aphrodite in Peiraieus as a dedication after the defeat of the Lacedaemonian navy off Knidos in 394 B.C.78 Xenophon dedicated a sanctuary,with temple and altarand cult image, to Ephesian Artemis on his estate at Skillous.79 Nor were these temples and sanctuariesthe only examples of private architectural sponsorship or adornment in mainland Greece in the 5th (and early 4th) century. In Athens, Peisianax built or sponsored the Stoa Poikile (ca. 460 B.C.) and Kimon decorated the Theseion with paintings (ca. 470).80At Delphi, Brasidas may have helped sponsor the Treasuryof Brasidasand the Akanthians (ca. 420 B.c.).1 This broad range of examples strongly suggests that the mainland Greeks did not need a Macedonian king (or even their own easterncousins) to set an example for temple building by individuals in the late 4th century. Moreover, Peisianax and the Stoa Poikile seem to show that, even in 5th-century Athens, building by individuals was not necessarily restricted to sanctuarycontexts alone.82 It is noteworthy that so many of the examples listed above pertain to the foundation of new cults by individuals. New cults may have offered a somewhat more flexible venue for prominently expressing devotion to a divinity (and engaging in conspicuous display) than did existing cults, especially majorones. But differenceswould have been primarilypractical, rather than moral, in nature: established cults are more likely already to 75. Plut. Them.22.1-2; deHer. Mal. 37. This temple should almost certainlybe associatedwith the remains of a small anta-templeexcavated byJ. Threpsiadesin 1959 and publishedby Eugene Vanderpool (1966). Identificationis based on an inscriptioncommemoratingthe renewal(apparently)of the sanctuaryby Neoptolemos of Melite, ca. 330 B.c.; the extant remainsdate primarilyto the time of this rebuilding. PierreAmandry'sdoubts (19671968) about the identificationand/or date of these remainsdo not affect the value of Plutarch'stestimonia for the originaldedication. 76. This foundationis recorded,in both words and relief carving,in the Telemachosmonument (IG II249604963; SEGXXV 226); Aleshire 1989, pp. 7-11; Mitroupoulou1975. For revisedtext and discussion,see SEG XLVII 232. The role of the playwrightSophokles in bringingAsklepios to Athens is
much better known than that of Telemachos (unattestedbefore the discoveryof SEGXXV 226). Sophokles was a priest in the healing cult of Amynos (Vit. Soph.II); SaraAleshire (1989, p. 9) suggeststhat his hosting of Asklepios may have been as a private household cult, as opposed to the formal,public one, for which Telemachos claims credit.SEG XXV 226 recordssome oppositionby the Kerykes: yicp?a3[T-[ovt6 ] Xop0ioxaLevta lTo)loa (lines 21-23). [7T?xtcX]Uoav Aleshire (1989, p. 9) notes that this oppositionwas probablybased not only on the Kerykes'ownershipof land within the Pelargikon,the ultimate site of the sanctuary,"butalso [on] their supportfor the private healing cult at the Amyneion." 77. For the propylon:Aleshire 1989, p. 34. SEG XXV 226, line 34: cf. lines 35-36: [oXo7ou[Xo]; T-C [oXOxl6XLa]. [?7Tc?x]s?oaC6o 78. Paus. 1.1.3. For discussionof the location and political significance
of Konon'stemple in the Peiraieus,see Funke 1983, pp. 175-189. 79. Paus.5.6.5-6; Xen. An. 5.3.9, with furtherdescriptionat 5.3.11-12. 80. Stoa Poikile:AgoraXIV, pp. 9094; Paus. 1.15.4. The Theseion is sometimes attributedto Kimon,but it is not known whether it was a personal projector a public one undertakenwith his leadership.See Plut. Thes.36.1-4; Plut. Cim. 8.5-7; Paus. 1.17.2-3, 6; Boersma 1970, p. 242. Plutarch'saccount of the role of Periklesin the constructionof the Parthenonpresents specialproblemsand will be discussed separatelybelow. 81. Plut. Lys. 1; Mor.400F. 82. Boersma(1970, p. 9) notes (in referenceto a somewhat shorter list):"These examplesgive the impressionthat privatecitizens were alloweda large measureof freedom in such matters.Yet this kind of activity can neverhave been very common, becauseordinaryAthenians could not affordto finance a big project."
282
GRETCHEN
UMHOLTZ
have temple buildings, which would make architecturalgifts by individuals complicated or even redundant.In any case, it is clear that private construction of temples, as well as other forms of architecturaldisplay in religious contexts, was in fact possible in mainland Greece throughout the Classical period and that those who could afford to and felt so inclined could apparently engage in it with little restriction, at least on privately owned property.83
NATURE AND ROLE OF DEDICATORY INSCRIPTIONS The freedom to dedicate buildings does not necessarilyimply the freedom to inscribe them. The physical remains of the buildings mentioned above aretoo scant to indicate whether they carriedinscriptions,but the fact that larger and better-preserved temples apparently did not has led many to conclude, afortiori, that smaller,private dedications must also have been uninscribed. Consideration of the nature and role of dedicatory inscriptions suggests that this inference is by no means necessary, and perhaps not even valid. The contractual nature of many aspects of Greek religion has long been recognized. Dedicatory inscriptions, because of their public and permanent nature,present a particularlystrong manifestation of this contractual outlook. At a minimum they declare whose property the dedicated object has become, but often they mark more fully the transferof ownership from the dedicator to the divinity by naming both parties and other details of the transaction,such as gift and occasion. The act of dedication, the giving of the gift, establishes a relationship between donor and deitya relationshipthat the dedicatoryinscriptioncommemoratesand announces to all.84Accordingly,the circumstancesin which such inscriptions arelikely to appear will be those in which the relationship, the transfer of ownership, is clear and straightforward.By this criterion, however, major,"publicly sponsored"temples are precisely the kind of building least likely ever to have carried dedicatory inscriptions. In the cases for which we have information it is clear that the funding of such projects was generally diverse, involving a variety of sources, often over a long period of time, and A temple frequentlyrelying at least in part on the divinity'sown resources.85 built using a god's own money can hardlybe presented as a gift from someone else to that god. In short, the larger and more expensive the temple, and more complex its funding, the less likely it would be ever to bear a dedicatory inscription. These observations have significant implications for how we understand smaller,privately funded architecturalgifts. Modern scholarship on this subject has tended to organize its categories and expectations based on the nature of the object (buildings vs. other types of votives) rather than the nature of the relationship (personal gifts vs. projects of other sorts).This is surelyan error.Whatever the reasons that major,corporately sponsored temples lacked inscriptions, these reasons do not necessarily apply to temples dedicated by individuals. On the contrary,temples (or other structures) dedicated by individuals fall into a different and more
83. WalterBurkert(1988, p. 43) considersanathemata(includingtemple buildings)"aform of display... which, in contrastto other such forms,does not raiserivalryor envy becausethe objectsare no longer privateproperty." I would not go quite so far-there was probablyno areaof Greek life from which rivalryand envywere entirely absent-but, to some degree,dedications did enjoy a special status.Restrictionsand limitationscould clearly have been imposed by the administrators of a given sanctuaryor area,but a lawmakeror assemblywould have had to be bold indeed to deprivethe gods of temples or gifts offeredon privately owned land. 84. ChristianeSourvinou-Inwood (1992, p. 265) writes:"Dedicationsby individuals,like polis and group dedications,were often commemorated throughinscriptions,which perpetuated the memory of the donor'saction and his-her claim to a specialrelationship with the divine." 85. The Temple of Zeus at Olympia was built from spoils when the Eleians took control of the sanctuaryfrom Pisa (Paus.5.10.2)-it may be that this monument represented,at least in part, a redeploymentof resourcespreviously dedicatedin the sanctuary.The funding of the Parthenonis a vexed question, but divine revenuesdo seem to have playedat least a small role (Ameling 1985, p. 51, n. 18; see also Kallet-Marx 1989; Samons 1993; Giovannini 1997). Contributorsto the 4th-centuryrebuilding of the Temple of Apollo were named at Delphi in accountsdisplayed in the sanctuary,but no simple dedicatory formulacould possiblyhave encompassedthis diverseand international group.
ARCHITRAVAL
283
ARROGANCE?
straightforwardcategory of religious relationship than largerprojects that in some cases might not be properly classed as dedications at all. Their epigraphic needs and opportunities will have differed accordingly.86Indeed, throughout the Hellenistic period, major temples sponsored by state funds, the god's own resources,or widely based subscriptions apparently continue to remain uninscribed, in sharp contrast to the abundant contemporaryexamples of inscribedbuildings dedicated by individuals.In light of this clear distinction in later periods, and the importance of the contractual nature of votive inscriptions in general, we should expect that individuals would always have been the most frequent inscribers of dedications, even in the more poorly attested early periods. As we have seen, literary evidence is not lacking for architectural projectscarriedout by individualsin the Classicalperiod.DidThemistokles, Peisianax,Telemachos, Konon, or Xenophon inscribe their buildings? In none of these instances are there physical remains sufficient to indicate whether or how the structurewas inscribed, but in Plutarch'sday, at least the Treasury of Brasidas and the Akanthians at Delphi carried a prominent inscription naming the dedicators.87Others may have been similarly adorned as well, in accordancewith the limits and potential of materials and setting in each case.
DATE AND DISTRIBUTION EXAMPLES
OF EXTANT
In the absence of firm information about the appearanceor inscribing of the architecturaldedications now attested only in literary sources, it has seemed natural to many to interpret the later, better-preserved examples as signs of a new Hellenistic ethos or the result of private citizens aping the actions of kings.The extant,datable,4th-century examplesfrom Greece do fall, for the most part, within the time of Macedonian hegemony. It is my contention, however, that the broad range of types of buildings and of individual dedicators using inscriptions in the final third of the 4th century are more consistent with the florescence of an existing and widespread tradition than with the sudden introduction of a new one. The following selection illustrates this geographical breadth and diversity of context. The Leonidaion at Olympia was a guest hostel of impressive scale and expense, sponsored apparently exclusively by a single individual, 86. The lack of dedicatoryinscriptions on majortemples is often explained by assertingthat religious buildingswere inscribedonly in cases where there was risk of ambiguity,for exampletreasurybuildings in international sanctuaries,but not within the home community,where identity of both divinity and dedicatorsalike would be well known.This position assumesthat it would have been technicallypossible to write, for example,
"theAthenians dedicated..." on most of the majortemples in the city of Athens, and that the only reasonthis was not done was that it was unnecessaryand otiose. I believe, however,that, far from being otiose, such a formulation would have been strictlyaccurate in only the rarestof circumstances; most temples lacked dedicatoryinscriptionsbecause no single group,not even the polis, could rightly claim sole financialresponsibilityfor them.
87. Plut. Lys. 1.1: 'O'AxavOicov 06loaupo6;&vAeXqpo;SmYOCypa(pv EXEl oas' TotLaur'v "Bpaoiatx xCal'AxCavOot
Plutarchcontinueswith the 'A0,lvaxiov." observationthat, becauseof the inscription,many think that the statue inside is of Brasidas,whereasactuallyit representsLysander.He also mentions the inscription(though not its connection to the building) at Mor. 401C, and the building at Mor.400F.
GRETCHEN
284
UMHOLTZ
The surviving architraveblocks indicate that probably in the 330s B.C.88 the donor's name was inscribed not just once, but on at least two sides of the building: A[z]Ovi8[xY]gActCooo NaLog;
o({[loae xoca a&vE0xxEArl.) 'Ovo7urti].89
L[e]onid[e]s, son of Leotos, a Naxian, ma[de and dedicated (this) to Zeus Olympios]. Two monuments at Thasos dating to the later 4th century carry architravaldedicatory inscriptions. A grand, two-storied facade, featuring Ionic half-columns above Doric pillars,was erected on the interior side of the Gate of Zeus and Hera by one Pythippos, probablyson of Paiestratos. The surviving fragment reads: H60rC7roS;Ila[zCoOTp&Too].90
Pythippos, son of Pa[iestratos]. A large square hall with columned porch near the Herakleion was dedicated by Thersilochos (probablyson of Orthomenes) in the final third of the 4th century.Its architravalinscription reads: Tolt~6]t0.)]d[V?60]XEv].91 O{pota7o[Xo;'Ope00o,VOO? Thersilo[chos, son of Orthomenes] d[edicated this to ... ].
88. For bibliographyand discussion, see Svenson-Evers1996, pp. 380-387; for Pausanias's(5.15.2) misreadingof the dedicatoryinscription(taking NA-IOS as HAEIOZ), see SvensonEvers 1996, p. 381. 89. IvO 651. Svenson-Evers(1996, p. 383) points out that other restorations arepossible,for example,?Toi6[YlO? Ex TCOVi&G)v ALl'OXOUM[Cp].
In the Amphiaraion at Oropos a large stoa dating to the middle of the 4th century B.C. carriedan impressivededicatoryinscription on its Doric frieze course, with one letter per metope.92The only letters preserved are 0, I O (two metopes on one block), and a final N (followed by a blank metope); it is therefore not possible to say whether the building was dedicated by an individual or a group.93 Within the city of Athens, choregic monuments constitute an important body of 4th-century architraval dedicatory inscriptions.94 These are not, for the most part, fully functional buildings, but rather architectural structures used as bases for the tripods dedicated to Dionysos by victorious producers in the dithyrambic competitions. These monuments entail some of the same problems as naiskos-type funerary monuments: it is often difficult to match superstructuresto foundations or even sometimes to other elements of the same structure;these problems can, in turn, make dating difficult as well. One such "floating" epistyle fragment from about the middle of the 4th century95recordsthe dedication of Ktesippos: KsxpootCi;7Tcai8[v ?vtXa]. KTnlyt7aToXocpSO[ooALt)ovebg] EXOpiyEL.v v v A _[- - Y _jEt?].96
(The tribe) Kekropiswon in the boy[s' competition]. Ktesippos, son of Chabr[ias, of Aixone], was producer.Da[- was flute-player].
90. Etudes thasiennes III 21. A
Pythippos,son of Paiestratos,appears in a list of archonsand theoroi dating ca. 315-285 B.c. For architectureand date, see Martin 1968, esp. p. 173. 91. Etudes thasiennes III 22.
92. For detailed description,bibliography,and discussion,see Petrakos 1997, pp. 259-260, no. 339 (with fig. 13). The frieze blocks are of poros;metope height: 0.48 m, letter height: 0.21 m. Some of the letters have small holes for the attachment of gilded bronze letters ([sLxpe6octS yLa Tr]vTrpoolXco,orypatqXji-c&va7rr X6AXLvo i7riXpuao EXCaxoa:Petrakos
1997, p. 259). 93. For attemptsat restorationof the text, see Petrakos1997, p. 260. 94. Most of the securelydatable examplesthus far known belong to the final third of the century,but new researchsuggeststhe existenceof some examplesseveraldecadesearlier.Goette (1999) has arguedthat impressivefullscale architecturalchoregicmonuments can be dated back to the 360s at least. 95. Goette 1989, p. 97. 96. IG II2 3040.
ARCHITRAVAL
ARROGANCE?
285
Figure8. Inscribedarchitraveof choregicmonumentdedicatedby Lysikrates,Athens. Courtesy Deutsches ArchaologischesInstitut, Athens
97. For full discussionsee Wilson 2000, pp. 219-226, including consideration of Bauer'sproposal(1977) that the front intercolumniationwas left open to reveala statue of Dionysos displayedwithin. 98. IG II23042. 99. In emphasizingthe difficulties in readingthe inscriptionand frieze, Wilson (2000, p. 222) may underestimate the clarityof fresh stone and painted letters;I believe he is too pessimisticin sayingof the inscription that "undersuch conditions it was surelyillegible."Although Macedonian featuresarerecognizablein the architecturalstyle of the monument, and seem to fit well with Lysikrates'family connections and the political and militarycontext of 335/4 B.C. (McCredie 1984), the architraval position of the inscriptionhas good precedentsin Attica and should not be taken as "Macedonian."
Some of the later choregic monuments arebetter preservedand better known. The innovative and impressive cylindrical monument dedicated by Lysikrates (335/4 B.C.) is a monopteros with dark blue screen walls closing off most or all of the spaces between its white Corinthian columns.97The inscription was carved in three lines on the three fascias of the epistyle (Fig. 8): Auotxpod-n AoaOtei0oo Ktxovvebi EXooryet. 'Axaxiocav-iT; OLriSV evixao. O&covYlXit. AocLatr8g'A0vaci;og Ei86tocxe. EOa(xV?Tzo;Xis.98 Lysikrates, son of Lysitheides, of Kikynna, was producer. (The tribe) Akamantis won in the boys' competition. Theon was flute-player. Lysiades the Athenian was trainer.Euainetos was archon. This inscription provides an indirect indication that the architravalposition was by this time quite canonical. A lower location, such as the stylobate of the monopteros or the crown of the base, might have offered greater legibility, but established aesthetic expectations and conventions favoring the architravalposition seem alreadyto have been strong enough to outweigh this practical consideration.99 Other roughly contemporary examples include a monument of uncertain type (ca. 335-320 B.C.),with an architraveadorned with a series of crowns and with brief inscriptions alluding to the service for which they
286
GRETCHEN
UMHOLTZ
were awarded;'00the choregic monument of Euagides (328/7 B.C.), inand Lysikles'monument, inscribed on its scribed on its Ionic architrave;'0? Doric architrave.l02The slightly later monument of Nikias, son of Nikodemos, commemorating a victory in the dithyrambiccompetition for boys in 320/19 B.C., also carried its dedicatory inscription on its architrave.103 Located just to the northwest of the sanctuary of Dionysos, this monument was a full-sized (16.68 x 11.79 m) building similar to a temple, with a prostyle porch of six columns in the Doric order.104 Most of the datable,inscribed choregic monuments belong to the period during or after the reign of Alexander. Do their prominent inscriptions represent the influence of Macedonian royal practice?'05A look at the visual landscape of Athens in the 4th century strongly suggests otherwise. As we have seen, Attic funerary reliefs, from the late 5th century onward, frequently bore the name of the deceased on the architraveor an equivalent flat upper fascia; likewise, votive plaques were often thus inscribed. In view of these abundant local precedents, it is difficult (and unnecessary) to demonstrate any significant role for externalinfluence in the form or placement of the inscriptionson 4th-century Attic choregic monuments. The proliferation of these prominently inscribed monuments does not indicate any introduction of foreign habits, breakdown in the social structureof the polis, or easing of (putative) restrictions on private architectural activity; it is simply a function of continuing evolution in a wellestablished medium of competitive display.'06 Similarly, these geographically diverse examples-not only from Athens, but from Olympia, Thasos, and perhaps Oropos as well-represent a range of building types, none of which seems particularlylikely to have been affected by any direct influence from Macedonian (or Hekatomnid) monuments, or even from each other. Given the numbers, geographical range, and approximate contemporaneity of these monuments, it is more efficient to see in them a natural manifestation of longstanding, fundamental Greek social and religious habits, rather than to assume external influence or to posit that foreign royal practice was so quickly adopted and imitated by privateindividuals in such widely varying contexts. In sum, a combination of archaeological, epigraphical, and literary evidence suggests a strong possibility that dedicatory inscriptions were inscribed or painted on at least some of the votive structuresnow attested solely (or primarily) in ancient literature. This evidence also points to the architraveas a logical location for such inscriptions, on both aesthetic and practicalgrounds, with good comparandaknown in other media. The lack of inscriptions on buildings not sponsored by individuals is in no way predictive of epigraphical practice on buildings dedicated by individuals. Accordingly, the increase in the number of extant inscriptions in the second half of the 4th century need not represent anything other than the results of the ongoing trend, in both public and private architecture, toward more extensive use of permanent materials (especially marble).107
100. IG I2 3206; Goette 1989, pp. 94-95. 101. IG II23052; SEGXLV 696. 102. IG II23054; SEG XXXIX 206. 103. IG II23055. 104. Hintzen-Bohlen 1997, pp. 6061. Plutarch(Nic. 3.3) seems to erroneously associatethis monumentwith the 5th-centurygeneraland statesman Nikias, son of Nikeratos,who is said to have dedicatedseveraltripodswith his brothersin 415 B.C. (cf. P1. Grg.472a; Hintzen-Bohlen 1997, p. 61, with n. 460). For other large-scalechoregic monumentsof the late 4th century,see the annotatedlists providedby Goette (1989, p. 97) and Hintzen-Bohlen (1997, pp. 145-147). 105. In the case of the Lysikrates Monument,JamesMcCredie (1984, pp. 182-183) has arguedfor Macedonian influence in variousaspects of style and design of the building, but the inscriptionis not one of the elements for which he claims specifically Macedonianprecedents. 106. Wilson (2000, p. 221) notes, with referenceto the monument of Lysikrates:"The stylisticchange should not deceive us into imagining that khoregicdisplayhad necessarily begun to operatewith a completely differentdynamicin this period.Even at the height of city-state'corporatism,' there was alwaysa tension between the collectiveideals of the polis and the more or less acknowledgedpreeminence of individuals."See also Wilson 2000, p. 235. 107. As, for example,in the case of the propylonto the Athenian Asklepieion,originallywooden, but rebuiltin stone, as mentioned above.
ARCHITRAVAL
ARROGANCE?
287
ANCIENT ANECDOTES
108. So, for example,Botermann (1994, p. 182) on Alexanderand Ephesos:"Die ganze Episode ist eine Anekdote. Aber eine signifikante,deren historischerKerndas Befremdender griechischenOffentlichkeitfiber AlexandersAnmassungist." 109. Them.22.1-2 (Loeb, B. Perrin, trans.).Garland(1992, p. 76) suggests that the offense might lie in upstagingthe public cult of Artemis Boulaia,worshippedin the Agora. 110. Aischin. 3.183-185. This formulationand emphasisare no doubt shaped to some degreeby Aischines'desire to arguethat Ktesiphonhad acted improperly in proposinghonors for Demosthenes. 111. This emphasisis clearalso in Plutarch'sreferenceto these herms (Cim. 7). The same principleis illustrated at Cim. 8.1, where Miltiades is said to have been refusedan olive crown in part due to the words of Sophanes of Dekeleia:"whenyou have fought alone, then demand to be honored alone." 112. Plut. Per.14.2 (Loeb, B. Perrin, trans.).
Several ancient anecdotes appear to reveal a deep antipathy for inscriptions of the sort that I argue could have been almost commonplace. Despite the fact that these stories-concerningThemistokles and the Temple of Artemis Aristoboule, Perikles and the Parthenon, and Alexander the Great and the Artemision at Ephesos-are often rejected (or ignored) as apocryphalor anachronistic,assumptionsbased on them nevertheless continue to color modern views.'08In the following discussion I do not seek to prove or disprove authenticity, but rather to point out that in no case do these stories, even if authentic, imply any Classical aversion to architectural dedicatory inscriptions. In each of these anecdotes the locus of the conflict described lies not in the proposed dedicatory inscription per se, but rather in the specific circumstances of the particular building in question. For example, in a list of occasions of public discontent with Themistokles, Plutarch reports:"He offended the multitude also by building the temple of Artemis, whom he surnamed Aristoboule,or Best Counsellor, intimating thus that it was he who had given the best counsel to the city and to the Hellenes."'09If this building did carryan inscribed (or painted) dedication, it would have been the implications of the epithet of the goddess, not the name of Themistokles, or the fact that he had dedicated a temple, that would have rendered the inscription offensive. A roughly contemporary,nonarchitecturalexample, taken by many as evidence for state-enforced anonymity of dedications, concerns Kimon and the herms dedicated after his victories at Eion on the Strymon in 476/5 B.C. Aischines claims that the victorious generals asked the demos for permission to dedicate the herms, but were allowed to do so only on condition that they not inscribe their own names.110The point of contention here is not the inscribing of dedications, but ratherthe fair apportionment of credit for a particular achievement. This story does not indicate that individual dedications would never be allowed, only that in this case the demos was not willing for this particularhonor to be considered an exclusively personal one.lll Similar issues are raised in Plutarch'saccount of Perikles'building activities and the financing of the Parthenon.According to Plutarch,Perikles responded to his opponents' criticism by asking the Assembly whether he was spending too much; when they answeredwith a strong affirmativehe announced that he himself would fund the construction and would inscribe his own name as dedicator."When Perikles had said this, whether it was that they admired his magnanimity or vied with his ambition to get the glory of his works, they cried out with a loud voice and bade him take freely from the public funds for his outlays, and to spare naught whatsoever."12
This story is problematic in many respects and may indeed deserve to be dismissed as an outright fiction, but such rejection should be based on grounds other than an a priori assumption that a proposal of personal
288
GRETCHEN
UMHOLTZ
financing for public buildings or the use of personal dedicatory inscriptions would have been inconceivable in Classical Athens.ll3 Such an offer of private funding for a public project may not be as far-fetched as it is often assumed to be. The Stoa Poikile (or Peisianakteios) may well have been sponsored in this fashion. Few Greek cities at any period could undertake major projects without some special source of revenue being devoted to the project;"matchinggrants"from individuals may have helped to provide that revenue more often than we realize. For example, a fragmentary inscription (the Springhouse Decree) from the 430s seems to include the response of the demos to an offer of funding from Perikles and membersof his family for some kind of waterworks.114 The offerwas apparwith the but demos, thanks, by ently declined, nothing in the remaining fragments of the decree suggests that the offer was considered inappropriate or unusual. Of course this decree, even assuming it is correctlyunderstood, does not prove that Perikles would or could have used the threat described by Plutarch, but it does suggest that neither the idea of private architecturalsponsorshipnor the tensions and rivalrybetween public, communal achievement and private, individual accomplishment would have been wholly out of place in 5th-century Athens.115The point here is not to claim that Plutarch is accuratelydescribing an actual debate, but rather to make clear that, whatever its status, his account cannot be taken as evidence for a Classical aversion to architectural dedicatory inscriptions. Within the context of Plutarch's narrative, as in the other anecdotes discussed so far, the issues at stake are the proper apportionment of responsibility, credit, and glory, not excess in personal display or the appropriateness of inscribing architecturaldedications.16 The story most frequently referredto in scholarly discussions of Late Classicalarchitecturalsponsorshipis that of Alexanderand Ephesos. Strabo, following Artemidoros, reports that Alexander promised to pay all the expenses for rebuilding the Temple of Artemis at Ephesos (destroyed by fire on the night of Alexander's birth) if he could have a dedicatory inscription,but the Ephesians were unwilling.17 Straboremarksthat Artemi113. Ameling (1985, esp. pp. 59-61) presentsa very skepticalview of Plutarch'saccount,emphasizingparallels with 2nd-centuryA.c. Imperialbuilding practice;see also Stadter1989, pp. 181-183. For funding of the Parthenonsee Giovannini 1997; Samons 1993; and Kallet-Marx1989. 114. IG 3 49 (= IG I254); SEGX 47, XII 19, XIX 12. The text is badly damagedand most of the document is missing, but what remainsincludes partsof two amendments.A conservativeversionof the text is presented by Woodhead (1973-1974, p. 761). The matterstreatedinclude water supply,a desireto minimize expense, referenceto the sons of Periklesin the dative case, and probablya motion of
thanks.Scholarsdiffer on whether the thanks accompanya refusal(Connor 1971, p. 127, n. 69) or an acceptance (e.g., Mattingly 1961, p. 164) of the offer.Ameling (1985, p. 59) believes that modernunderstandingof this text has been prejudicedby Plutarch(Per. 14), and that the inscriptionis so fragmentarythat no conclusionsshould be based on it at all. 115. Periklescould not, in any case, have paid for the entire cost of the Parthenonas built, but this need not affect the possibilitythat privateand public financingwould have been consideredand balancedand, in some cases,playedoff againsteach other. 116. The chargesagainst,and exile of, Pheidias suggest that the Parthenon
was a projectin which the boundaries between personaland public spheres were quite problematicand the potential for abusea sourceof real anxiety. The chargesagainstPeriklesand Pheidias all seem to have involvedexcessive or appropriationin one "personalizing" way or another:the inclusion of Perikles'face on the shield of Athena, and the embezzlementof funds. Plut. Per. 31.2-5; Diod. Sic. 12.39.1-2. Suggestion of an inscriptionon the Parthenon would be a yet more extremeversionof this same kind of personalappropriation of the monument. 117. Strab.14.1.22 (C 640). Botermann(1994, p. 182; see above, n. 108).
ARCHITRAVAL
118. IEphesos1519; 29 fragments from at least eight (and probablymore) differentcolumns.The most complete of these (1519a, in four fragments) reads:[ ] [ ] SCap8o6vl 'ApTc[aiL]S6 Hicks (1890, -[6]v o686[v &v06Yxe?v]. pp. 173-174, no. 510a) notes that the letter forms "preciselysuit the date of the restorationof the temple in the second half of the fourth centuryB.C." and he connects the inscriptionswith severalancient testimonia concerning the temple'sfinancing:Strabo(14.1.22 [C 640]) notes that the jewelry and personalpossessionsof the Ephesian women were sold to raise money; the pseudo-AristotelianOecon.2.20 reports that the Ephesianspassed a law requiringwomen to surrenderto the city what gold they had, and allowing any who gave money sufficientfor a column to have their names inscribedas dedicators;and Pliny (HN 16.40 [213]) informsus that "allAsia"contributedto the rebuilding.
ARROGANCE?
289
doros praised the Ephesian who told Alexander that it was not fitting for a god to make dedications to the gods. Does this response reflect deepseated antipathy toward dedicatory inscriptions or even toward dedication of temples by individuals?By no means. As with Plutarch'saccount concerning Perikles and the Parthenon, the Ephesians' objection is not based on the inappropriatenessof the idea of Alexander's building (or inscribing) a temple; they object only to his building and inscribing theirtemple. Alexander was not claiming a privilege of an outrageous or unprecedented type, but in this case it was a privilege the Ephesians considered their own and did not wish to relinquish or share.They did not want to be deprived of the credit, both with the goddess and their fellow Greeks, for the work they themselves had already done, and planned to do, in rebuilding her temple. Inscribed fragments from the base moldings of a number of columns show that individual contributions were in fact epigraphicallycommemorated on the building itself.l18 In both of these stories, Plutarch'son Perikles and Strabo'son Alexander, the opposition between public monument and private appropriation is drawn in the clearest possible terms. But they by no means suggest that the only way an individualcan have a monument inscribedis by usurping a public right. On the contrary,stories in which cities are represented as unwilling to give over into private hands the religious and social benefits of communal responsibility for great temples of their patron deities have little bearing on attitudes toward the distinct class of buildings (such as those of Themistokles, Telemachos, Konon, and Xenophon) that are more truly and appropriatelypersonal in their origins. Individuals could make, and claim credit for, architecturaldedications in the Classical period, but gifts (or other monuments) that were truly corporate in nature were not to be subsumed or obscured under the name of any individual. These anecdotes describing resistance to the appropriationby an individual of something that should be public (Themistokles' claim to "best counsel,"Perikles'claim to the Parthenon, Alexander'sto the Artemision) emphasize the importance of giving credit where credit is due, but none of them implies any restriction on inscribing monuments that properly fall within a person's own responsibility. Indeed, the numerous attested examples of structuresbuilt by individuals in the 5th and early 4th centuries make it clear that, in their proper context, private architecturaldedications were acceptable and appropriateand were in no way expected to be made anonymously.
CONCLUSIONS I have argued that architravalinscriptions carrying the names of dedicators of buildings should not be considered the product of major social or religious changes in Late Classical Greece, but rather an expression of Greek votive and epigraphical traditions of long standing. Literary evidence shows that architecture(including temples) fell within the range of votive offerings that individuals could and did make throughout the Classical period. From the Archaic period on, the architravewas among the
290
GRETCHEN
UMHOLTZ
favored locations for inscriptions, with particularlyfrequent use in votive and funerarycontexts from the late 5th century onward.Apart from practical and aesthetic considerations, the primary issue governing whether a building could carry a dedicatory inscription was not arrogancevs. modesty,not state vs. individual,but financial responsibility.The structuremust represent a true gift to a god from a distinct and nameable individual or group;few large-scale, corporatelyfunded projectsfully met those criteria. Availabilityof, and prioritiesfor,both public and privatecapitalwould certainly have affected the frequency,scale, and permanence of architecturaldedications by individuals;these factorswould have varied from place to place and decade to decade. For Athens in particular,the second half of the 5th century was a time of extraordinarycollective self-assertiveness on the part of the demos, which reveled in many of the activities previously availableonly to wealthy individuals.119 The current state of the evidence, however, does not justify the supposition that Athenian democratic ideology (much less, universal Greek morality) required that the inscribing of architecturaldedications be governed by rules significantly different from those governing religious dedications of other types. Epigraphical anonymity (voluntaryor otherwise) should no longer be assumed to have been a governing principle for personal architecturaldedications in Classical Greece.
119. See, e.g., Ober 1989; KalletMarx 1993.
ARCHITRAVAL
ARROGANCE?
291
REFERENCES AgoraIII = R. E. Wycherley,Literary andEpigraphicalTestimonia(Agora III), Princeton 1957. AgoraXIV = H. A. Thompson and R. E. Wycherley,TheAgoraof Athens:TheHistory,Shape,and Uses of an AncientCity Center(Agora XIV), Princeton 1972. Aleshire, S. 1989. TheAthenianAsklepieion:ThePeople,TheirDedications, and theInventories,Amsterdam. Aloni, A. 2000. "Anacreontea Atene: Datazione e significatodi alcune iscrizionitiranniche,"ZPE 130, pp. 81-94. Amandry,P. 1967-1968. "Th6mistocle a Melite," in Xcapcn'Tpoveli Av. K. 'Opiavaov4, Athens, pp. 265279. .1986. "Chios and Delphi,"in Chios:A Conference at theHomereion in Chios,1984, J. Boardmanand C. E. Vaphopoulou-Richardson,eds., Oxford,pp. 205-232. .1988. "Aproposde monuments de Delphes: Questions de chronologie (I),"BCH 112, pp. 591610. Amandry,P., and Y. Fomine. 1953. La ColonnedesNaxienset le Portiquedes Athenians(FdD 11.4),Paris. Ameling, W. 1985. "Plutarch,Perikles 12-14," Historia34, pp. 47-63. Bauer,H. 1977. "Lysikratesdenkmal: Baubestandund Rekonstruktion," AM 92, pp. 197-227. Bean, G. E. 1966. "The Inscriptions," in TheSanctuaryofHemitheaat Kastabos,J. M. Cook and W. H. Plommer,eds., Cambridge,pp. 5865. Boersma,J. S. 1970. AthenianBuilding Policyfrom 561/0 to 405/4 B.C.,
Groningen. Bommelaer,J.-F. 1991. Guidede Delphes:Le site (Sites et monuments 7), Paris. Botermann,H. 1994. "Werbaute das neue Priene?Zur Interpretationder Inschriftenvon Priene Nr. 1 und 156,"Hermes122, pp. 163-187. Bousquet,J. 1970. "L'inscriptiondu tresorde Corinthe a Delphes,"BCH 94, pp. 669-673.
Burkert,W. 1975. "Apellaiund Apollon,"RhM 118, pp. 1-21. .1988. "The Meaning and Function of the Temple in Classical Greece,"in Templein Society, M. V. Fox, ed., Winona Lake, pp. 27-47. Clairmont,C. 1993. ClassicalAttic Tombstones: Volume, Introductory Kilchberg. Connor,W. R. 1971. TheNew Politiciansof Fifth-CenturyAthens, Princeton. Daux, G. 1958. "LeTr6sorde Marseille a Delphes,"BCH 82, pp. 358-367. Davies,J. K. 1971. AthenianPropertied Familes,Oxford. Develin, R. 1989. AthenianOfficials, 684-321 B.C.,Cambridge. Dillon, M. P.J. 1995. "The Lakedaimonian Dedication to Olympian Zeus:The Date of Meiggs & Lewis 22 (SEG 11,1203A)," ZPE 107, pp. 60-68. Dunbabin,T. J. 1948. The Western Greeks: TheHistoryof Sicilyand SouthItalyfromtheFoundationof the GreekColoniesto 480 B.C.,Oxford. Engelmann,H. 1981. "Die Bauinschriftam Apollonion von Syrakus,"ZPE 44, pp. 91-94. Flaceli6re,R. 1937. Plutarque,surles oraclesde la Pythie.Texteet traduction avecuneintroductionet desnotes, Paris. Funke,P. 1983. "KononsRiickkehr nach Athen im Spiegel epigraphischerZeugnisse,"ZPE 53, pp.149-189. Gadbery,L. 1992. "The Sanctuaryof the Twelve Gods in the Athenian Agora:A RevisedView,"Hesperia 61,pp.447-489. Garland,R. 1992. IntroducingNew Gods:ThePoliticsofAthenian Religion,London. Giovannini,A. 1997. "Laparticipation des allies au financement du Parthenon:Aparcheou tribut?"Historia46, pp. 145157. Goette, H. R. 1989. "Ein dorischer Architravim Kerameikosvon Athen,"AM 104, pp. 83-103.
GRETCHEN
292
. 1999. "Mausoleumor Choregic Momument? On the Self-Representationof the Athenian Upper Class under the Democracy" (paper,Toronto 1999). Guarducci,M. 1967. EpigrafiagrecaI, Rome. . 1985. "Nuoviriflessioni sull'epigrafedel tempio di Apollo a Siracusa,"RendLinc40, pp. 15-17. . 1987. "Il tempio arcaicodi Apollo a Siracusa:Riflessi nuove," in Saggiin onoredi GuglielmoDe Angelisd'Ossat,S. Benedetti, G. MiarelliMariani,and L. Marcucci, eds., Rome, pp. 43-45. Gunter,A. C. 1985. "Lookingat Hecatomnid Patronagefrom Labraunda,"REA 87, pp. 113-124. Harris,D. 1995. TheTreasures of the ParthenonandErechtheion,Oxford. Hellstrom, P. 1981. "The Androns at Labraunda:A PreliminaryAccount of Their Architecture,"MedMusB 16, pp. 58-74. 1989. "FormalBanquetingat and Labraunda,"in Architecture Societyin HecatomnidCaria(Boreas 17), T. Lindersand P. Hellstrom, eds., Uppsala,pp. 99-104. . 1996a. "HecatomnidDisplay of Power at the LabrayndaSanctuary,"in Religionand Powerin the AncientGreekWorld:Proceedings of the UppsalaSymposium,1993 (Boreas24), P. Hellstrom and B. Alroth, eds., Uppsala,pp. 133138. 1996b. "The Andrones at Labraynda:Dining Halls for ProtohellenisticKings,"in Basileia: Die Palastederhellenistischen Konige, W. Hoepfner and G. Brands,eds., Mainz, pp. 164-169. Hesberg, H. von. 1994. Formenprivater in derBaukunstdes2. Reprdsentation und 1. Jahrhunderts v. Chr.(Arbeiten zur Archaologie),Cologne. Hicks, E. L. 1890. TheCollectionof AncientGreekInscriptionsin the BritishMuseumIII, Oxford. Hintzen-Bohlen, B. 1997. Die KulturpolitikdesEubulosunddes Lykurg,Berlin. Holloway,R. R. 1991. TheArchaeology ofAncientSicily,London. Homolle, T. 1893. "Seancedu 15/17 decembre1893,"pp. 611-615 in
UMHOLTZ
"Institutde correspondance hellenique,"BCH 17, pp. 611-623. . 1896. "Inscriptiondu grand autel,"pp. 617-633 in "Institutde correspondancehellenique,"BCH 20, pp. 581-732. Hornblower,S. 1982. Mausolus, Oxford. Immerwahr,H. R. 1990. Attic Script: A Survey(Oxford Monographson ClassicalArchaeology),Oxford. Jeffery,L. H. 1990. TheLocalScripts ofArchaicGreece,rev.ed. (Oxford Monographson ClassicalArchaeology), Oxford. Kallet-Marx,L. 1989. "Did Tribute Fund the Parthenon?"ClAnt8, pp. 252-266. . 1993. Money,Expense,and Naval Powerin Thucydides' History 1-5.24, Berkeley. Kirchner,J. 1948. ImaginesinscriptionumAtticarum:Ein Bilderatlas DenkmilerAttikas, epigraphischer 2nd ed. (G. Klaffenbach),Berlin. Kron,U. 1996. "Priesthoods,Dedications, and Euergetism:What Part Did Religion Play in the Politicaland Social Status of Greek Women?"in Religionand Power in theAncientGreekWorld:Proceed1993 ingsof the UppsalaSymposium, (Boreas24), P. Hellstrom and B. Alroth, eds., Uppsala,pp. 139-182. Labraunda1.1 = K.Jeppesen, ThePropylaea(LabraundaI.1), Stockholm 1955. Labraunda1.2 = A. Westholm, The Architecture of theHieron(Labraunda 1.2), Lund 1963. Labraunda1.3 = P. Hellstrbm and T. Thieme, TheTempleof Zeus (Labraunda1.3), Stockholm 1982. LabraundaIII.2 = J. Crampa,TheGreek Inscriptions,Pt. 2: 13-133 (LabraundaIII.2), Lund 1972. Lawton, C. 1995. Attic Document Reliefs:Art and Politicsin Ancient Athens,Oxford. Mark,I. 1993. TheSanctuaryofAthena Nike in Athens:Architectural Stages and Chronology (HesperiaSuppl.26), Princeton. Martin, R. 1968. "Sculptureet peinture dans les facadesmonumentalesau IVesiecle av.J.-C.,"R4 1968, pp. 171-184. Mattingly,H. B. 1961. "The Athenian
Coinage Decree,"Historia10, pp. 148-188. McCredie,J. R. 1984. "The 'Lantern of Demosthenes' and Lysikrates, Son of Lysitheides,of Kikynna," in StudiesPresentedto SterlingDow on His EightiethBirthday(GRBM 10), A. Boegehold et al., eds., Durham, pp. 181-183. Mitroupoulou,E. 1975. A New InterMonupretationof the Telemachos ment,Athens. Morgan, C. 1993. "The Origins of Pan-Hellenism,"in GreekSanctuaries:New Approaches, N. Marinatos and R. Hagg, eds., London, pp. 18-44. Ober,J. 1989. Mass andElite in DemocraticAthens:Rhetoric,Ideology,and thePowerof thePeople,Princeton. Parker,R. 1996. AthenianReligion: A History,Oxford. at Partida,E. 2000. TheTreasuries Delphi An Architectural Study, Jonsered. Petrakos,B. 1997. OtEmypacqC-roo Athens. Q2pouro6, Picard,C. 1965. "Surles dedicaces monumentalesapposeesen Gr&ce aux entablementsde fagades d'edificessacresou civils,"in Xapt1, aocptovEqiAv. K. 'OpAtdvsov Athens, pp. 91-107. Pouilloux,J.,and G. Roux. 1963. Enigmesa Delphes,Paris. Raubitschek,A. E. 1949. Dedications from theAthenianAkropolis: A Catalogueof theInscriptionsof the Sixth and Fifth Centuries B.C.,
Cambridge,Mass. Ridgway,B. S. 1997. FourthCentury Stylesin GreekSculpture,Madison. Robert,J., and L. Robert. 1983. Fouilles dAmyzonen CarieI: Exploration, histoire,monnaies,et inscriptions, Paris. Salviat,F. 1977. "Ladedicacedu tresor de Cnide a Delphes,"in Etudes (BCH Suppl. 4), Athens, delphiques pp. 23-36. Samons, L. J., II. 1993. "Athenian Finance and the Treasuryof Athena,"Historia42, pp. 129-138. Samothrace 4.2 = K. Lehmann and D. Spittle, TheAltarCourt(Samothrace4.2), New York1964. Schaber,W. 1982. Die archaischen TempelderArtemisvon Ephesos:
ARCHITRAVAL
293
ARROGANCE?
Entwurfsprinzipienund Waldsassen. Rekonstruktion, Scholl, A. 1994. "HoXmotaXavTa Zur literarischenund vvLqji?ca: monumentalenUberlieferung aufwendigerGrabmiler im spatklassischenAthen,"JdI109, pp. 239-271. .1996. Die attischenBildfeldstelendes4.Jhs. v. Chr.:Untersuchungenzu denkleinformatigen Athen im spdtklassischen Grabreliefs (AM-BH 17), Berlin. Schroder,S. 1990. PlutarchsSchriftDe Pythiaeoraculis.Text,Einleitung, undKommentar, Stuttgart. Sourvinou-Inwood,C. 1992. "Further Aspects of Polis Religion,"Annali dell'Istitutouniversitarioorientaledi Napoli,Dipartimentodi Studi delmondoclassicoe delMediterraneo e storia antico,Sezionedi archeologia antica10, pp. 259-274. on Stadter,P. A. 1989. A Commentary Plutarch's Pericles,Chapel Hill. Stewart,A. 1990. GreekSculpture:An Exploration,New Haven. Svenson-Evers,H. 1996. Die Architektenarchaischer griechischen undklassischer Zeit, Frankfurt.
GretchenUmholtz TRINITY COLLEGE DEPARTMENT 300
SUMMIT
HARTFORD,
OF CLASSICS STREET CONNECTICUT
[email protected]
06I06
Vanderpool,E., andJ. Threpsiades. 1966. "Themistokles'Sanctuaryof Artemis Aristoboule,"ArchDelt19, 1964, A'[1966], pp. 26-36. van Straten,F. T. 1981. "Gifts for the Gods,"in Faith,Hope,and Worship: Aspectsof ReligiousMentalityin the AncientWorld,H. S. Versnel,ed., Leiden, pp. 65-151. 1992. "Votivesand Votariesin Greek Sanctuaries,"in Le sanctuaire grec(Entretienssur l'antiquite classique37), A. Schachter,ed., Geneva, pp. 247-289. surles Viviers,D. 1992. Recherches et la cite ateliersdesculpteurs d'Athenesa l'epoquearchaique (Academie Royalede Belgique, Memoire de la Classe des BeauxArts), Brussels. Walsh,J. 1986. "The Date of the Athenian Stoa at Delphi,"AJA90, pp.319-336. Wilson, P. 2000. TheAthenianInstitution of the Khoregia:TheChorus,the City,and the Stage,Cambridge. Woodhead,A. G. 1973-1974. "The Date of the 'SpringhouseDecree' (IG 1254),"ArchCl25-26, pp. 751761.
HESPERIA
71 (2002)
Pages 29-324
EAR
AT
EXCAVATIONS
LY
PERGAMON AND
THE
CH
RHODIAN
RONOLOGY OF
AMPHORA
STAM/PS
ABSTRACT The chronologyof Rhodian amphorastampsdependsheavilyon a collection of roughly 900 stamps found at Pergamonin 1886, known as the Pergamon Deposit. Most of the Rhodian eponyms in this group are dated to ca. 210175 B.C.Two points of historicalinterpretationarefindamental to these dates: good relations between Rhodes and Pergamon at that time, and Rhodian garrisoning of Knidos between 188 and 167. Neither interpretation,however,withstandsscrutiny.The archaeologicaland topographiccontexts ofthe PergamonDeposit, hitherto ignored, areused here to arguefor a closing date in the late 160s or early 150s, and the wider implications for Hellenistic ceramic chronologies are explored. The recent, revised publication of the so-called Pergamon Deposit, an influential collection of Hellenistic stamped amphora handles, prompted the present reconsideration of the deposit's role in the study of amphora stamps, their chronology, and the study of ancient trade. Christoph Borker and Johannes Burow's Die hellenistischenAmphorenstempelaus Pergamon (PF 11) included, as Borker'shalf of the volume, a new publication of the collection, which was first published by Carl Schuchhardt in 1895.1 Soon after Schuchhardt'spublication, the deposit proved to be of considerable importance for developing the chronology, first, for Rhodian stamped 1. B6rker1998; Schuchhardt1895. The presentarticleexpandson an idea first articulatedin my reviewof PF 11 for BonnerJahrbiicher (Lawall2002). An importantsourceof researchhas been the unpublishednotes of Virginia Grace.These files are currentlystored at the Agora excavationoffices in the Stoa of Attalos. I thank Carolyn Koehlerfor permittingunlimited access to these files and for fruitfuldiscussions about their contents. Susan Rotroff graciouslyread and critiquedearlier
drafts,and I hope this final product repaysher efforts.I also thank Gerald Finkielsztejnfor allowing me use of his unpublisheddissertationfor the preparationof this articlewhile his 2001 book was in preparation.I am particularlygratefulto Kathleen Slane and Hesperia'sanonymousreviewersfor savingthis work from many errors.The Facultyof Arts at the Universityof Manitoba providedgenerousresearch support. The following reconsiderationof
the PergamonDeposit is based on a reviewof publishedreportsand on personalobservationof the topography in July 2000. 1 thank my traveling companions,William Aylwardand Andrea Berlin,who enduredthe entire argumentwhile we hiked up the Pergamonacropolis.All referencesto stampshere indicate stampson the handles of amphorasratherthan tiles, loomweights,black-glazefinewares,or other coarsewares.
296
MARK
L. LAWALL
amphora handles and, later, for Knidian stamped handles. These stamp chronologies have become fundamental building blocks for most other Hellenistic pottery chronologies.2 Furthermore, historians of ancient politics and trade often cite this deposit as part of a larger discussion. For example, M. I. Rostovtzeff in 1941 wrote: The [Pergamon Deposit] testifies in all probability to close commercial relations between Pergamon and Rhodes in the years between 220 and 180 B.c. (approximately).... After 180 B.c. Pergamon probably emancipated itself from Rhodes and may have organized its commerce on different lines.3 Rostovtzeff's influence is clear in R. M. Berthold's more recent use of the deposit: It also appearsthat trade between the two countries broke off at this time, as Rhodian handles from the period 220-180 are found in overwhelming numbers in Pergamon [i.e., the Pergamon Deposit], while those dated after about 180 are completely absent from the deposits. The reasons behind this breakwere undoubtedly basically economic and probably represent an emancipation of Pergamene commerce from Rhodian domination, but the deterioration of economic and political relations between the two states is clearly more than coincidental.4 While Berthold's reference to an absence of Rhodian stamps dating after 180 is incorrect,5his comments make clear that the implications of this deposit go far beyond either the narrower field of Hellenistic amphora chronologies or the archaeology of Pergamon per se. Borker's new publication reviews the stamps themselves in the deposit in considerabledetail.6Farless attention is paid to the deposit's findspot (Fig. 1), its role in Hellenistic economic history, or to the ways in which the deposit has figured in archaeological discourse over the past century. Borker begins by largely repeating Schuchhardt'svery brief description of where the stamped handles were found.7After further discussion of the composition of the deposit (giving ranges of dates for different types, noting unusually early or late pieces, and assessing the preservation of the fragments), Borker makes a very important and tantalizing observation. He proposes that the deposit might have resulted from the clearing of a storeroom. After some unknown period, the debris was then dumped into the areawhere it was excavatedin 1886.8 Exploration of this scenario leads, here, to a complete reconsideration of the scholarly history of the deposit, the historical context of Pergamon in the late 3rd and 2nd centuries, and the archaeologicalcontext of the deposit itself. Detailed attention to one group of amphorahandles and their small findspot on the Pergamene acropolisis justified by the immense analyticalweight placed on this deposit by scholars of Hellenistic pottery, architecture,and economic history. My reconsideration of this deposit has three parts. First, scholars'use of the deposit over the last century helps explain the current status of the
2. See, for example,AgoraXXII, pp. 96-110, andAgoraXXIX, pp. 431473, where the dates of the deposits underlyingthe Athenian Agora chronologyarebased in largepart on the stampedhandles. Similarly,see CorinthVII.3, pp. 206,225,230,234. Schafer(PF2, p. 26) also notes the importanceof stampedhandles for Hellenistic potterychronologies. 3. Rostovtzeff 1941, p. 1479, n. 68. 4. Berthold 1984, pp. 173-174, and more recently,Gabrielsen1997, p. 67. 5. See Burow 1998, nos. 43, 44, 46, 47, 57-59, 90-92, 101, 108, 115, 116, etc. 6. Bbrker(1998, pp. 8-9, 13-14) largelyupholds the traditionalRhodian chronology.He does (p. 14, n. 39) recognize the problemof period IV being 10-12 yearstoo long but does not consider the possibilitythat period III stops too early(see below). 7. B6rker1998, p. 5. 8. Borker1998, p. 9. Here, too, he notes that Rhodian importsto Pergamon did not simply cease at the close of the PergamonDeposit (cf. Berthold 1984, pp. 173-174).
EARLY
EXCAVATIONS
AT
PERGAMON
297
Figure1. Findspotof the Pergamon Deposit, withTurkishworkman holdingamphorahandle.AfterAvP V.1, text pl. 20
9. Finkielsztejn1995, pp. 281-282; 2001, p. 175.
deposit in amphora studies (see "History of Research,"below). Since 1907 the consensus for the range of dates for the Rhodian eponyms in the deposit has been ca. 220-180 or 210-175 B.C. This consensus has recently been challenged in a series of works by Gerald Finkielsztejn, who proposes a revised date of 193/0-163/1 B.c.9These competing theories can be evaluated by closely considering the two historical points on which the traditional chronology depended: 1) the existence of close and friendly Rhodian relations with Pergamon between ca. 220 and 180 B.C.; and 2) Rhodian control of Knidos between 188 and 167 B.C.Closer examination of the historical evidence (see "Rhodes, Pergamon, Knidos," below) reveals serious weaknesses in these two long-accepted interpretations of the historical sources.Therefore, in the final stage of this reconsideration, I return to the archaeologicaland topographic setting of the deposit itself (see "Archaeological Context," below), and propose a closing date in the 160s or 150s B.C., without dependence on these problematic historical conclusions.
298
MARK
L. LAWALL
Figure2. Plan of the citadelat Pergamon,showingthe locationof the depositterrace.AfterDreyfus and Schraudolph1997, foldout plate
HISTORY OF RESEARCH The Pergamon Deposit was excavated in September 1886 (Fig. 1), and Schuchhardt published the stamped handles in 1895.10No other artifacts from the area have ever been mentioned or published.1l Schuchhardt described the findspot, marked by a red cross on the plan in his publication, as lying between an earlierand later course of the city wall in the southeast corner of the acropolis (Figs. 2-3). The handles themselves were found as fill within the foundations of a stone building (referred to hereafter as 10. Schuchhardtreceiveda stipend from the GermanArchaeologicalInstitute in 1886 (Grinert 1987, p. 104). In Februaryof that year,he arrivedat Pergamon,where he seems to have met universalapproval(Schulte 1963; Karl and Dorner 1989, pp. 91-97), distinguishing himself with his study of the
watersystemof the cityandthe rehe alsoworkedat gionaltopography;
the nearbysite of Aegae. After working with Dorpfeld in Athens in 1887, he went on to a distinguishedcareermore focused on Europeanprehistory(see Griinert1987). 11. The discoveryis mentioned
twice in an earlypreliminaryreport from the Pergamoncampaign(Humann, Bohn, and Frankel1888), once by Carl Humann (p. 57: "aneiner andererStelle fanden sich bis zu tausendabgebrochenerAmphorenhenkel mit Stempeln. Schuchhardt kopiertesie samtlich";also quoted by Karland Dorner 1989, p. 92), and in a more extended descriptionby Richard Bohn (p. 67, see below).The find is not mentioned in the laterpreliminary reportpublishedin 1899, also covering the year of discovery(Conze and Schuchhardt1899).
EARLY
I
0
Palace V
Figure 3. Detail of the area of the deposit and surrounding structures. After AvP V.1, pl. 6
AT
EXCAVATIONS
I
I
10
PERGAMON
299
MARK
300
L. LAWALL
the "deposit building").12The building was dated "spatestens aus dem 2. Jahrhundertv. Chr.""3Schuchhardt suggested that the reigns of Attalos I (241-197
B.C.) and Eumenes II (197-159
B.C.) were periods of close
ties between Rhodes and Pergamon and, hence, could be expected to witness the intense trade attested by the deposit.14 References to the Pergamon Deposit that appeared soon after this initial publication add nothing further to our understanding of the archaeological context of the material but do begin to refine the range of dates suggested by Schuchhardt. In this regard, the first scholarly use of Schuchhardt'spublication is somewhat puzzling.15H. van Gelder, in his 1899 publication of inscriptions illustratingthe Rhodian dialect, describes the deposit stamps as resulting largely from a single discovery and as belonging to the 2nd or 1st century B.C.16He does not elaborate upon or defend the proposal to extend the dates of the stamps into the 1st century. In a later article, however, he considers the date of the deposit in more detail (see below). Schuchhardt'spublication was also quickly taken up by scholars with a secondary interest in the amphora stamps themselves. Karl Regling, in 1901, published a brief comment on the minting eponyms of Rhodes, in which he argues that these eponyms must not be priests of Helios since they never match the names on the amphora stamps.'7The eponyms on the amphorastampswere assumedto be those of priests of Helios. Schuchhardt's work is cited as only one of a few major publications of the amphora stamps. Ferdinand Bleckmann, in a dissertation on Rhodian stamped amphoras published in 1907, followed Schuchhardt'slead in dating the collection, repeating that the building in which the stamps were found dated to the 2nd century at the latest.l8 Bleckmann proposed that the 40 eponyms 12. Schuchhardt(1895, p. 423) describedthe deposit as follows:"In der sudostlichenEcke der Hochburg (GebaudegruppeVI), zwischen der alten urspriinglichenund der spateren weiter vorgeriicktenBurgmauerfand sich namlich der ganze Innenraume zwischen den Grundmauerneines Gemachesvom Felsen her ausgefiillt mit Scherbenvon Amphoren."Later (p. 434), Schuchhardtrefersto the findspot as a "Hausfundamentein GebaudegruppeVI."Bohn'sdescription of the area,publishedin 1888, is similar but more focused on the surrounding architecture(Humann, Bohn, and Frankel1888, p. 67). The implications of Bohn'sreportare consideredbelow under"Archaeologyof the Deposit." 13. Schuchhardt(1895, p. 432) refersto Bohn, who servedas the head architectof the German excavations at the time, as the sourcefor this
date. Bohn himself did not discussthe date of this building in his preliminary report(Humann, Bohn, and Frankel 1888,pp.62-81). 14. Schuchhardt1895, p. 432. 15. Gelder 1899 is the earliest considerationof Schuchhardt'swork I have found apartfrom reviewsof the volume.The most extended contemporaryreviewis by Keil (1896). Although raisingnumerousquestionsas to what to make of these amphorastamps,Keil adds nothing to the question of the date or archaeologicalcontext of the PergamonDeposit. An anonymous reviewer(identifiedonly as A.H., see H. 1896) notes the importanceof the materialpublishedby Schuchhardtfor the study of Pergamenetrade,especially in terms of Rhodianwine. De Sanctis (1896-1897) devotes two sentencesto Schuchhardt'ssection of the volumegiving fulsome praisewithout substan-
tive comment. Reinach (1896, p. 223) is similarlyfavorableand concise in reviewingSchuchhardt'scontribution. Hiller von Gaertringen(1896, p. 59) and Pridik (1896, pp. 127-128) both note the appearanceof the volume,but neitherwriter drawsout the chronological utility of the material. 16. Gelder 1899, p. 570: "Alle,bis auf 17, stammenaus einem einheitlichen grossen Funde (Sept. 1886). Sie gehoren dem 2. oder erstenvorchristlichen Jahrhundertan." 17. Regling 1901, pp. 109-114. Nilsson (1909, p. 31, n. 1, citing Regling) notes the problemsdetermining the stamp sequencesince eponyms on coins, which might be placedin sequenceby die links, areof no assistance. 18. Bleckmann1907, p. 15: "domum, unde effossae sunt, saeculiII vel etiam aetatispaulo antiquiorisesse affirmant viri harumrerumperiti."
EARLY
EXCAVATIONS
AT
PERGAMON
30I
present in Schuchhardt'spublication should indicate a more specific range of absolute dates for the Rhodian eponyms in the Pergamon Deposit, 220 to 180 B.C.Bleckmann,too, emphasized the good relationsbetween Rhodes and Pergamon in this period.19 While this chronological range for the Pergamon Deposit was generally accepted, an early challenge went largely unnoticed. Gelder, in an article published in 1915, proposed a closing date of ca. 165 B.C.based primarily on epigraphic evidence.20He began by noting that Damokles, an eponym from the deposit, is named as priest of Helios on an inscription from the reign of either Antiochus III (223-187) or Antiochus IV (175163). As Antiochus III was a constant enemy of Rhodes, Gelder preferred the latter's reign, thus giving Damokles the priesthood before 163 B.C.21 Gelder also pointed out that another Rhodian magistrate (head of the prytany) mentioned in the same inscription, Astymedes, is attested as active in relations with Rome in 171,167, 164, and 153; Gelder argued that Astymedes held the priesthood of Helios "in 153 or thereabout"during the Cretan War (154-151 B.C.). The stamp name Astymedes does not appearin the Pergamon Deposit.22A third epigraphicallyattested eponym, Pratophanes, appears on an inscription concerning Rhodian arbitration of the conflict between Samos and Priene. Gelder believed that the inscription (and also the priesthood of Pratophanes) should date to ca. 165 since the other Rhodians named areattested as navalcommandersin events of 190 and 170. Presumably Gelder assumed that their naval service was a necessaryprecursorto their service as arbitrators.In any event, Pratophanes appearsthree times in the deposit at Pergamon and more commonly at Carthage, so Gelder saw him as a very late eponym for the Pergamon Deposit.23 On the basis of these and similar,though less directly relevant, arguments,24Gelder proposed a date of ca. 165 for the closing of the deposit. As noted above, however,Bleckmann'ssuggested range, 220-180 B.C., came to be most widely accepted. In 1922, for example, A. Maiuri published another major Rhodian stamp assemblage, the Villanova deposit from Rhodes itself; he followed Bleckmann's dating and did not refer 19. Bleckmann1907, pp. 24-
25. Nilsson'sstudyof Rhodian
amphorahandles (1909) took notice of both Bleckmannand Schuchhardt's work but did not discuss the dates of the handles or other ramificationsof the PergamonDeposit. In a later article, Bleckmann(1912, passim) continued to use a range of 220-180 for the deposit. 20. Gelder 1915. This articlewas publishedin Dutch in a journalthat did not commonly publish archaeological topics (MarianneStern,pers. comm., 24 May 2001). VirginiaGrace receivedan offprintof the articlefrom Peter Fraserin 1950 and commissioned
a translationfrom an unnamednative speaker.The originaloffprint,now in poor condition, and the typescriptof the translationare currentlystoredin Grace'sfiles in the Stoa of Attalos. For Grace'sdiscussionof this work, see DelosXXVII, p. 290, n. 5; Grace 1985, pp. 12,44. 21. Gelder 1915, p. 199. Dittenberger(as publishedin 1917, Syll.3644) also prefersAntiochus IV, noting that he was a known benefactorto Rhodes. Dittenbergerdoes not referto Gelder and seems to prefera date of ca. 172 for this inscription. 22. Gelder 1915, pp. 200-202. 23. Finkielsztejn(2001, pp. 120-
121, and table 19, p. 192) notes the associationbetween Pratophanesand fabricantsof the Villanovadeposit (see below). As a result,he dates Pratophanesmuch earlierthan Gelder argued.Finkielsztejndoes not refer to Gelder'sargumentconcerning Pratophanes. 24. The other epigraphically attestedeponyms (Eukles,pp. 203204; Sosikles,p. 206) fall more clearly, accordingto Gelder (1915), outside the span of the PergamonDeposit. While interestingin themselves,they do not particularlystrengthenGelder's case.
MARK
302
L. LAWALL
to Gelder.25F. Hiller von Gaertringen (1931) also adopted Bleckmann's dates.26 In 1934, Virginia Grace brought together Schuchhardt'sand Bleckmann'sarguments as follows: [The Rhodian chronology] is based primarilyon the large closed deposit in Pergamon published by Schuchhardt,which according to the agreement of archaeologicalwith traditional probability is to be dated at the end of the 3rd and the beginning of the 2nd century B.C.The forty-odd eponyms ... indicate about four decades of almost continuous trade.This period may be placed ca. 220-180 B.C., immediately before and closely after which Rhodes and Pergamon were not in friendly relations with one another.27 Grace'spublicationsof the 1950s and 1960s continued to employ this chronology.28E. Levi, publishing stamps from Olbia in 1964, follows the same chronology.29A later closing date, "before 167 B.C.," is suggested in Fraser and Bean'sRhodianPeraea,but their referenceis to Grace'searlierpublications, in which she never suggested such a date.30 As early as February1961, however,it is clear from Grace'snotes that, in preparingher contribution to Ziegenaus and de Luca'sfirst volume on the Pergamene Asklepieion,31she had begun to reconsider aspects of the deposit. Over February and March 1961, Grace and her colleagues restudied the names appearingin the deposit and came to a figure of"under" or "about"35 accuratelyread eponyms. This correctionwas first published in 1970.32Through the 1960s, too, Grace turned her attention to Gelder's challenge to the 220-180 chronology. In a manuscript dated 25 May 1968, entitled "Dating of the Rhodian Eponym Damokles and of the Original Pergamon Deposit," she summarized Gelder's argument and responded that perhaps Antiochus III, late in life, was seeking to win back 25. Maiuri 1924, pp. 264-265. 26. RE Suppl.V, 1931, cols. 834840, s.v. Rhodos(F. Hiller von Gaertringen). 27. Grace 1934, p. 215. 28. Grace (1950), publishing materialfromTarsus,cites a rangeof 220-180 B.c. for the deposit;similarly, Grace 1952, p. 525, for Delos; 1953, p. 119, in a new list of Rhodianeponyms;and 1956, p. 176, for the Pnyx. Both editions of the AgoraPictureBook on amphoras(Grace 1961 and 1979, with figs. 32-33) also cite 220-180, even though Grace no longer favored this chronologyafter 1970. 29. Levi 1964, p. 227. 30. Fraserand Bean 1954, p. 12; Grace'snotes (7 March 1961) on this publicationdrawattention to the differencebetween her view of the
deposit'sclosing date and the opinion attributedto her. KathleenSlane pointed out to me that Fraserhad given the offprintof Gelder 1915 to Grace and so may have been familiar with Gelder'sproposedlater closing date. Perhaps,Slane suggested,Fraser abbreviatedboth Virginia Grace and H. van Gelder as V.G. in his notes and made a simple mistakein the text of RhodianPeraea. 31.AvPXI.1, pp. 175-178. 32. DelosXXVII, p. 291. Andreas Dimoulinis assistedGrace in reconsidering the eponyms.The resultsof their work were recordedin a seriesof handwrittenand annotatedlists of eponyms from the PergamonDeposit (especiallynotes dated 24 February 1961 and 11 March 1961).
EARLY
EXCAVATIONS
AT
PERGAMON
303
the support of Greek cities, such as Rhodes, whose support he had lost in the preceding hostilities. She went on to suggest that if the inscription dated to the reign of Antiochus IV, then the stamp eponyms might not have been the priests of Helios as had been so long assumed. A revised version of this typescript appeared as an appendix to Grace's 1985 article on the Middle Stoa in the Athenian Agora: If the date of the inscription cannot be put earlier,we must accept the fact that the eponym AcJioxXi; 2nd in the stamps is not the of the inscription, and then same person as the eponym AaiooxXiA; apparentlythat the stamp eponyms (often called priests in the stamps) are not the priests of Halios who date other Rhodian documents.33 33. Grace 1985, p. 44. 34. Grace 1985, pp. 5, 7, 9, 13. That the eponyms arepriests of Helios is an assumptionstill found commonly in the literature,e.g., Empereurand Hesnard 1987, p. 15; Lund 1999, p. 188. Finkielsztejn (2001, pp. 42, 176, n. 53) also favorsthis attribution,though he does note that it is uncertain. 35. On secondarystamps,see Badal'yants1973; Palaczyk1999; Finkielsztejn2001, pp. 116-120, 177-178. 36. DdlosXXVII, p. 291. 37. Grace (1974, p. 196) summarizes the connectionsbetween Pergamon, Rhodes, and Knidos only very briefly;the argumentwas expanded upon considerablyin 1985. 38. Grace 1974, p. 196. 39. Grace 1985, p. 15, where she notes that the latest Knidianeponyms in the fill for the Stoa of Attalos "are perhapsnot later than 157."It should be noted that the constructionfills of the Stoa of Attalos have yet to be reconsideredfor publication.Townsend (AgoraXXVII) has amplydemonstrated the potential for extremelyuseful stratigraphyin this region of the Agora. Attalos II's reign was, for the purposes of clarifyingthe constructiondate of the Stoa, unfortunatelylong (159-138). 40. The Rhodian ethnicity of these phrourarchswas never suggested:Grace felt that the nameswere "nomore Rhodian than Knidian"(DelosXXVII, p. 318). 41. DelosXXVII, pp. 318-319; Grace 1985, p. 14; 1974, p. 196. For the situationat Priene, see IPriene19.6 and 37.66.
Indeed, while Grace does discuss priesthoods of Athena Lindia and Helios in the same article, the eponyms on the stamps are referred to more generically as "datingofficials."34 Grace'srejection of such a late date for Damokles depended largely on the fact that a small, square, secondary stamp never accompanies an eponym stamp of Damokles.35In 1970, Grace was the first to draw attention to the following: 12 or 13 [Rhodian eponyms in the Pergamon Deposit] sometimes occur on handles bearing also secondary stamps ... a feature which it now seems, may have begun to be used in 188 B.c., at the time of chief Rhodian expansion on the continent. Taking then as our end date 13 years after 188 B.c., and allowing a period of about 35 years, we would come out with approximately210-175 B.C.36 Damokles' term as eponym should fall before 188, and the closing date of the Pergamon Deposit was thus worked out to be 175 B.c. In 1974 and 1985, Grace elaboratedon the connection between Rhodian control of Caria ("the continent," see above) and a closing date of 175 for the deposit.37The argumentspresented in these two articles make clear for the first time the importance of the Knidian stamp chronology to the date of the deposit and the Rhodian stamps. One new element to the issue is Grace'sinterpretation of the Knidian title (poo6oaoXo;. She noted that the title appears on Knidian handles at Corinth (sacked 146) and beneath the Stoa of Attalos (built, according to Grace [1974], "somewhereabout 150 B.C.,"38and later [1985], "rightafter [Attalos] became king, in 159"39),so the title must have been in use before the middle of the 2nd century. Grace proposed that these phrourarchs were mercenarycommandersfor Rhodian garrisons(or Rhodian-employed garrisons) controlling Knidos while Caria was a Rhodian possession between 188 and 167 B.c.40Since more than one phrourarchname can appear with a single eponym, Grace proposed that each phrourarch'sterm was perhapsfour months long (phrourarchsattestedepigraphicallyat Priene held terms of four months).41The second element now taking on a greater role in the discussion is the Middle Stoa building fill (hereafter,MSBF),
3o4
MARK
L. LAWALL
long under study by Grace, referredto briefly in Delos XXVII in reference to the date of the Pergamon Deposit,42 and published in 1985.43 Grace brought the two elements together as follows. The Knidian stamps in the MSBF include nine names that are, elsewhere, labeled as phrourarch;and two full-year eponyms, one of whom served alongside two phrourarchsincluded on stamps in the fill. Thus, subtracting two full years plus two and one-third years (the terms of the remaining seven phrourarchs)from 188 results in a closing date for the MSBF roughly five The MSBF and the associated Knidian amyears later, i.e., ca. 183 B.C.44 are linked to the phoras Pergamon Deposit by the fact that seven Rhodian at are not found under the Middle Stoa. If the final eponyms Pergamon of date the Pergamon Deposit is ca. 175, then counting back seven years results in the Middle Stoa dating to 183 or 182 B.C.45Grace'searlierargument (in DelosXXVII) concerningthe first appearanceof secondarystamps around 188 B.C.adds further support to this sequence of points. Six MSBF Rhodian eponyms are also attested with secondary stamps; counting six years from 188 (and including that year) also yields a date of 183 B.C. for the Stoa. Subsequent surveys of the Rhodian (and Knidian) chronologies have tended to follow Grace's 1974 article and the conclusions put forward in the publicationof the MSBF. Roland Etienne, publishing finds fromTenos in 1986, adopted Grace's dates of 210-175 for the Pergamon Deposit.46 J.-Y. Empereur,summarizing the Rhodian chronology in 1987 and 1990, gave the same range.47Borker, too, suggested "around3 decades"48as an appropriatelength for Rhodian period III (= the Pergamon Deposit), querying whether the least commonly representednames in the deposit might be better considered late period II (i.e., preceding the apparentlycontinuous series of eponyms within the deposit). While Borker suggested that the core of the deposit comprises only 26 names, he retained the essential closing date of ca. 175 B.c.49
The only significant change in the dates of the Pergamon Deposit has appeared as part of Finkielsztejn'sproposal that the Rhodian chronology for periods I-III be lowered roughly 11 years and that Grace'speriod IVA be shortened from ca. 174-156 to ca. 163/1-156.50 This lower chronology was reached from a number of different directions: 1) counting eponyms backward from destructions of Samaria (108 B.c.) and Marisa (110 B.C.);
2) reconsidering the assignation of names within periods IV and V; and, most importantly,3) not allowing for any missing eponyms in the period from 174 to 146. Grace had been working forwardfrom a terminus of 175 for the Pergamon Deposit, and Finkielsztejn noted that at least 11 eponyms were missing between that time and the destruction of Corinth in 146. He recognized, however, that "l'incompatibiliteentre la chronologie basse et l'interpretationhistorique du dep6t de Pergame demeure problematique,"51referring to the much earlier argument that the date of the deposit was tied to good relations between Rhodes and Pergamon. He suggests, by way of resolving the problem, that relations between the cities seem best between 201 and 180 B.C., so his dates of 199-164 fit just as well. The terminus date, for Finkielsztejn, can then be tied to Rhodian difficulties after the creation of the free port at Delos instead of particular conflicts between Rhodes and Pergamon.
42. DdlosXXVII, p. 291. 43. The earliestnotes on the
MSBFin Grace'sfilesaredated22 August 1949 and consist of a typed list of the stampspresent.Furtherdetails and correctionswere made between 1953 and 1955. Although the contents of the MSBF do not play a significant role in Grace'spublicationsuntil 1970 and 1974, her detailedstudy of the fill preceded,and clearlyinfluenced,her studies of the PergamonDeposit in the 1960s. A list from 11 March 1961, for example,lists those eponymspresentin the PergamonDeposit, but not in the MSBF. 44. Grace 1985, p. 15. 45. DelosXXVII, p. 291; Grace 1985, pp. 7-8. 46. Etienne 1986, p. 235. Badal'yants(1983, p. 93; Badal'janc 1999, p. 249) uses the earlierdates of 220-180 for the deposit. 47. Empereurand Hesnard 1987, p. 19; Empereur1990. 48. Borker1998, p. 6. 49. B6rker1998, pp. 9, 14. 50. This revisedchronologywas developedby Finkielsztejnover several yearsand presentedin his dissertation (1993, esp. pp. 230-269) and subsequent publications(1995, pp. 280-281; 2000, pp. 215-218; 2001). 51. Finkielsztejn1995, p. 280.
EARLY
EXCAVATIONS
AT
PERGAMON
305
In one sense then, Finkielsztejnhas replacedone "historical" argument for anotherby suggestinga laterperiodfor good relationssufficient for intensivetradebetweenPergamonandRhodes.Yetthis is essentiallya minor element of his revisions.Finkielsztejnmakesa sharpbreakwith past practicesby startingwith later contextsand workingbackwardtowardthe Pergamongroup.GraceexplicitlyusedPergamonas the starting point and so requiredthe varioushistoricalargumentsin orderto have someabsolutelaunchingpoint forthe chronology.52 Nevertheless,Grace's that be there some might eponymsmissing (despitethe fact suggestion thatno neweponymshaveappearedforsometime)orunrecognizedhomonyms between 175 and 146 B.C. opens a loophole that Finkielsztejn'sar-
gumentalone cannotclose.53Regardlessof the numberof extantstamps between the close of the PergamonDeposit and the sack of Corinth, Carthage,Samaria,and othermajorcenters,the stampsthemselvescannot necessarilydefinethe periodof time in question-missing namesor misinterpretedhomonymsremainpossibilities.If Grace'shistoricalpins hold up to scrutiny,then Finkielsztejn'scase is weakened.If, however, the historicalevidenceis foundto be lackinganda lowerdatecanbe substantiatedfor the deposit'sclosure on the basis of other externalevirevisedchronologywouldbe strengthened.Suchexdence,Finkielsztejn's ternalevidencewould also removethe need for Finkielsztejnto find an forgood relationsbetweenthe two cities. alternatehistoricalcircumstance Gelder,in that case,wouldhavebeen essentiallycorrect(albeitfor different reasons).
RHODES,
52. SeeespeciallyGrace1974, p. 196, where she notes the fundamental importanceof the deposit as the startingpoint for her chronological studies. 53. For the possibilityof missing eponyms or missed homonyms, see Grace 1974, pp. 197-198. 54. Gruen 1984, p. 545, and likewise pp. 536 and 538; other recent studies of Rhodian history,especially with discussionof relationswith Pergamon,include Berthold 1984; Gabrielsen1997; and Reger 1999. 55. Polyb.4.47-52; Gabrielsen 1997, pp. 44-46.
PERGAMON,
KNIDOS
Grace and others based their absolute dates for the Pergamon Deposit on two historical points: first, that the years 220-180 (or 210-175) marked a period of good relations between Rhodes and Pergamon, and second, that Rhodes occupied Knidos with a garrisonbetween 188 and 167. A detailed review of the literary and epigraphic evidence highlights problems with both conclusions. Erich Gruen sums up Rhodian relations with Pergamon as follows: "Cooperation between Rhodes and Pergamon ... derived from expediency, aims that temporarilyconvergedwhile enduring interests diverged."54 A review of events of the late 3rd and early 2nd centuries calls into question the idea of any extended period of close relations between Rhodes and Pergamon. At the same time, it becomes difficult to determine which political events should be connected with good commercial relations and which would have had no effect on trade.The period of ca. 220-150 B.C. included occasional times of close cooperation between Rhodes and Pergamon, and other moments of outright hostility. Other events and actions lend themselves to varying interpretationsof underlying motivation. Within this span, the last years of the 3rd century include the only noteworthy period of cooperation;the earlieryears produce little evidence for particularfriendship between the two cities. In 220 B.C., for example, Byzantium began to charge a toll on access to the Black Sea; Pergamon supported Byzantium; Rhodes joined with Bithynia against Byzantium.55
306
MARK
L. LAWALL
Early on in Philip V's aggressions in the Aegean, Rhodians were more active in diplomacy to end the hostilities than they were in siding with any particular major power.56In 202, however, Rhodes joined Chios, Cos, Byzantium, Pergamon, and others against Philip. This alliance may have been related to demonstrable Rhodian interest in free access to the Black Sea.57The Rhodians, by entering this alliance, may have also hoped indirectly to check Attalos I's interests in the Aegean.58Cooperation between Rhodes and Pergamon is clear in three events that took place in 202 and 201: theirjoint effort against Philip near Chios; the subsequent meeting at Athens involving the Romans, Pergamenes, and Rhodians, all agreeing on a position against Philip; and the presence of both Rhodian and Pergamene troops at the defense of Abydos.59 After 201, cooperation between Rhodes and Pergamon seems more the exception than the rule. In 200, when Rhodes and Pergamon gained help from Rome against Philip, Rhodes hesitated to help in united efforts against Philip.60In 199, Pergamene and Roman interests diverged from those of Rhodes when Rome assisted Pergamon in exerting control over Andros, an island that had been the object of Rhodian interest.61Rhodes and Pergamoncooperatedagain,however,in the captureof Eretriain 198,62 and in the same year both Rhodian and Pergamene ambassadorsissued peace-settlement demands at the conferencewith Philip near Nicaea overseen by Flamininus.63While Pergamon requested restitution for specific damage inflicted on Pergamene territory,Rhodes asked not only for the return of Rhodian territory but also for the freedom of "all emporia and harbors of Asia."64There is scant evidence for or against friendship between Rhodes and Pergamon between 197 and 190 B.C. (i.e., eight of the years postulated as being a period of good relations).65 In the two years immediately preceding the Peace of Apamea, however, relations between Rhodes and Pergamon were clearly poor. In 190, when Antiochus III explored possibilities of peace, Eumenes II spoke against Rhodes in rejecting the possibility.66At Rome in 189, during discussions leading up to the Peace of Apamea, Eumenes claimed that the Rhodian envoys were bound to speak against Pergamon'sinterests.67The 56. Livy 27.30.4-10; Polyb. 11.4.16.10 (207 B.c.;hostility againstPhilip is considered"thenominal pretext
of thewar," W. R. Paton, [rxpoo-a'xc]
trans.,New York1927) and 15.22.423.6 (203 B.c.; the Rhodianswere activein diplomacywith Philip but were so offended by his treatmentof Cius that they sworeenmity against Philip). See also Walbank1967, pp. 274-275,476; Ager 1991. 57. Polyb.4.47; Livy 27.30.4-10; Gruen 1984, p. 533. 58. Gruen 1984, pp. 535-537; Berthold 1984, p. 115. The existence of an ad hoc allianceof the sort attested between Rhodes and Pergamonneed not imply that the "allies"were in full
agreementon all matters,and an alliance might be hoped to resultin the eventualweakeningof one of the allies. 59. See Polyb. 16.2-9 (battle near Chios); 16.26 (meeting at Athens in 201); 16.30.7 (siege of Abydos). Livy (31.2.1) also attests to a joint meeting in Rome in 201; see Gruen 1984, p. 534. 60. Polyb. 16.35. An Achaean delegation offeredto arbitratea separate peace between Rhodes and Philip; the Romans successfullyrequestedthat the Rhodiansnot do so. Gruen (1984, p. 536) emphasizesthe possibilitythat Rhodes might have agreedwith the Achaeans. 61. For the captureof Andros, see
Livy 31.45.1-8; for Andros as a target of Rhodianinterest,see Gruen 1984, p. 536, with referenceto Livy 31.15.8. 62. Livy 32.16.6-7. 63. Polyb. 18.1.4-18.2.5. 64. Polyb. 18.2.4: xoa T'rov Jl'TopL&O -Tr 'AoLOC xao Xi,eiv(ovTc'v xocaTa ocTracvTov.
65. Gruen (1984, pp. 544-545) notes that both Rhodes and Pergamon worked againstthe SpartanNabis in 195, but Gruen emphasizesthe differing aims of Rhodes and Pergamonat this point. 66. Polyb.21.10; Livy 37.19.1-6. See Walbank1979, p. 101. 67. Livy 37.53.2-3; Polyb.21.19.312.
EARLY
EXCAVATIONS
AT
PERGAMON
307
Rhodians noted their friendly relations with Pergamon, but proceeded to argue against Eumenes' territorialrequests.68During Pergamon'swar with Pharnakes of Pontus between 183 and 179, Rhodes prevented Eumenes from blockading the Hellespont.69 Rhodian support for Sinope against Pergamon'senemy,Pharnakes,maybe seen as Rhodes helping an old friend rather than as an attestation of Rhodian friendship with Pergamon.70In 178, Pergamon encouraged and assisted a Lycian revolt against Rhodes with attacksagainst Rhodian borderpositions.7 Late in the 170s, Rhodian judges voted along with others to strip Eumenes of honors granted to him by the Achaean League.72 In subsequent crises involving Perseus of Macedon between 171 and 167, Rhodes and Pergamon operated from quite different perspectives; although they did not cooperate closely, they were not antagonistic toward one another.73In 168-167 B.C.,when Rome began to settle its conflict with Perseus, the Pergamene and Rhodian deputations made none of the hostile claims against one another that marked the settlement leading up to Apamea.74 In short, Rhodes and Pergamon were never consistent allies, and there is certainly no period during which the two powers were so friendly that intensive trade can be assumed to have been the result. Indeed, an important element in the idea that intensive trade was facilitated by political friendship remains unaddressed:What sorts of political relations in antiquity would or could have encouraged trade?75If Rhodian troops assisted Pergamenes at Abydos, would Rhodian merchantsor other merchantscarrying Rhodian products be any more inclined to stop at Pergamon?Would a Rhodianjudge'svote againstEumenes II influence such merchants?These questions cannot be answered as long as the only diachronic evidence for Rhodian-Pergamene trade-Rhodian amphoras found at Pergamon-is structuredby the very assumption that good political relationscreatedgood commercial relations. In order to begin to address such questions, the archaeological evidence for trade must be datable independently of such assumptions. This need is even clearerin the particularcase of the Pergamon Deposit, for which the assumption of good political relations at the time of the deposit is so poorly supported by the textual sources. If, however, the phrourarchsin Knidos can be placed with certainty between 188 and 167, it is no longer necessary to depend on the assump68. Polyb.21.22.6-12. 69. Polyb.27.7.5. 70. Polyb.23.9.2; Walbank1957, pp. 511-513; 1979, p. 227. Berthold (1984, pp. 172-173) notes that Rhodian favortowardSinope did not translateinto substantiveaction by Rhodes in aiding either Sinope or Pergamon. 71. Polyb.27.7.5-6; cf. 24.15.13 for Eumenes assistingRhodes in Lycia;Livy 42.14.8. 72. Polyb.28.7.9-10; for discussion and referencesfor the date of and
participantsin this event, see Berthold 1984, p. 180, n.2. 73. For events and motivations duringthis period, see Berthold 1984, pp. 181-194; Gruen 1984, pp. 556568. 74. Polyb.29.29; 30.1-5; Livy 45.19. 75. Gabrielsen(1997, p. 67) might overstatethe assumptionthat lay,in part,behind the dating of the Pergamon deposit when he writes that "no Rhodian amphoras... were allowed into a city that currentlydid not stand
in a good political relationshipwith Rhodes."Instead,I imagine that Schuchhardt,Bleckmann,Grace,and others assumedfar fewer goods would be importedin periods of hostility. The core of Gabrielsen'sobjectionto the dating being tied to assumptions about how politics affectedtradeseems, however,eminently sensible.In general, on difficultiesreconcilingthe archaeological recordwith changes in political climate, see MacDonald 1982; Kracht 1990; Adams 1979.
308
MARK
L. LAWALL
tion that good political relations create intense trade relations.The argument that Grace put forth linking the MSBF, Knidian eponyms and phrourarchs,and Rhodian eponyms in the Pergamon Deposit would still yield a time span of ca. 210-175 B.C.for the deposit. The status of Knidos after the treaty was signed at Apamea in 188 is unclear.The status of specific cities has become a central point of debate concerning Apamea. On the one hand, the settlement of Apamea gave Rhodes control of Caria and Lycia south of the Maeander River, and this territorialsettlement did translateinto some real territorialgains.76Nevertheless, Lycian complaints to Rome against Rhodes as early as 188 B.C. show the ambiguity of the terms of the settlement even in an areawhere the Rhodians had established the office ofhagemon.77 Knidos's assistance to Rome against Antiochus may have guaranteed its freedom after 188,78 but any degree of certainty as to the city's status seems nearlyimpossible to achieve. The most direct evidence for control-the very amphora stamps in question-cannot outweigh the uncertainties raised by other sorts of evidence relating to Knidian autonomy.79 Indeed, the likelihood that Knidos was placed under Rhodian control and that Rhodes would have felt the need for establishing a garrison there is lessened in view of the close, uncoerced connections between Rhodes and Knidos both before 188 and after 167.80Obverses of earlier Knidian coins show Helios, a potential referenceto Rhodes;81Knidians cooperated with Rhodes in diplomatic efforts in 196;82and in 164, Knidos collaborated with Rhodes in the Calyndian revolt against Caunos that resulted in a resurrectionof Rhodian control of part of the mainland.83This last act hardly seems the action of a newly independent city that had just ejected a Rhodian garrison. Finally, as Gary Reger points out, the appearanceof an autonomous mint at Knidos in precisely the years following the Apamea settlement is further evidence arguing against Rhodian control.84 Nikolai Jefremow has recently arguedthat a phrourarchneed not have led a foreign, occupying force.85Local or friendly garrison commanders had the same title. Phrourarchs at Syracuse, Alexandria Troas, Miletos, and Priene all commanded friendly garrisons.86Furthermore,phrourarchs, 76. For the settlement of Apamea, see Polyb.21.17,24.7-8, 43, 46; Livy 37.45.12-16, 38.38, 39.7-17, 55.4-7, 56. 77. Most recently,see Bresson 1999, pp. 106-118. 78. Livy 37.16.2. 79. Those favoringKnidianautonomy point to the city'srole in mediating disputes,minting coins, and engaging in militaryactivity;see Magie 1950, pp. 108-109, 952, n. 61,958, n. 75; Reger 1999, pp. 89-90; Walbank1967, p. 615; cf. Walbank1979, p. 171, highlighting the uncertaintyof the case of Knidos.Mediation is rejectedas an indication of autonomyby Gruen (1984,
p. 540, n. 49). Fraserand Bean (1954, pp. 93-94) also favorRhodiancontrol of Knidos (but their argumentdepends heavilyon Grace'sinterpretationof the amphorastamps).They also cite Rhodian roses on the reversesof Knidian coins of this period as suggestiveof Rhodian control.Rhodianiconography need not imply Rhodian control:Knidos included Helios on their coins before 188 (Head 1967, p. 616). 80. Cf. Fraserand Bean 1954, "Rhodesmay have garrisoned 93: p. her newly acquiredcities more rigorously than the old cities, since the were more highly civilised cO6XEL; likely to give trouble."
81. Head 1967, p. 616. 82. Syll.3588. 83. Polyb.31.4.1; Berthold 1984, pp.219-220. 84. Reger 1999, p. 89, n. 47, with furtherreferences. 85. Efremov 1992, pp. 256-257; Jefremow1995, pp. 50-58. 86. For examplesof"friendly" phrourarchs,see, for Syracuse, Plut. Dion 11 (368/7 B.c.);AlexandriaTroas,SEG IV 671 5; Miletos, Milet 1.3, nos. 22.17-33, 37d.65; and Priene,IPriene4.50. In general on phrourarchs,see RE XX, 1921, cols. 773-781, s.v. Phrurarchos (H. Kortenbeutel).
EARLY
EXCAVATIONS
AT
PERGAMON
309
in rare cases, played nonmilitary roles in their cities.87The rarity of such cases means that it is safe to associate the Knidian phrourarchswith garrisons. Whether the phrourarchs at Knidos were Rhodian, employed by Rhodes, or even part of an occupying force is entirely unclear from the evidence. Therefore, any further conclusions about these phrourarchsare interesting possibilities, but they do not bear the weight of an entire amphora chronology. Both the proposed period of close relations between Pergamon and Rhodes and the existence of Rhodian garrisons at Knidos are thus called into question for the late 3rd and early 2nd centuries B.C. As noted earlier, the use of the Pergamon Deposit as a linchpin for the Rhodian amphora chronology has long depended heavily on precisely these two points. If these points are now set aside as insufficiently supported, there is still one aspect of the deposit that can assist in determining its date of deposition and, hence, the date of its contents: the archaeological and topographic context of the deposit itself.
ARCHAEOLOGICAL
87. For duties of phrourarchsbeyond militaryactivity,see Syll.341.12 (phrourarchsat Erythraiservingto administerthe city council).In RE XX, 1921, cols. 777-778, s.v.Phrurarchos, H. Kortenbeutelnotes that nonmilitary roles are especiallyseen in Ptolemaic Egypt in the 2nd centuryB.C. 88. Gelder 1915, p. 198. 89. DelosXXVII, p. 290; also Grace 1985, p. 3. 90. For layersof amphoradebris raisingthe floor level of a building after its initial construction,see Williams 1978, pp. 17,20; 1979, p. 112, on the Punic Amphora Building. 91. A note in Grace'sfiles dated 6 July 1958 recordsa conversationwith J. Schaferindicatingthat Schaferwas planning to excavatenearthe Pergamon Deposit building. Grace expressed the hope that new study of the relevant walls might resultin evidencefor dating the deposit.There is, however, no follow-up to this exchangeand I find no publishedreferenceto such excavations.
CONTEXT
Only Schuchhardt's 1895 comments on the findspot of the deposit have been cited in subsequent amphoraresearch,and even these have been subject to gradually changing interpretation. Schuchhardt described the deposit as fill within the foundations of a building. A reasonable interpretation of his description is that the fill belonged to the construction phase of the building, the scenario favoredby Gelder (1915) when he described the deposit as containing "882 stamps, all found together as rubbish to support a house on sloping ground, and apparentlyall deposited at the same time."88Grace (1970) described the deposit as having been "dumped in ancient times between the foundations of a room set between the earlier and later city walls, the filling no doubt put in to support a raised floor level."89She seems to have envisioned either a renovation with the fill lifting the level of a new floor or a "raisedfloor level" being part of the original construction.90For the most part, the above descriptions place the deposit earlier than the two-room building in which Schuchhardt places the deposit itself. Only Grace's description admits the possibility of the deposit having been added during a later phase of renovation. Other descriptionsof excavationsin this areaof the Pergameneacropolis suggest that these reconstructions of the relative sequence of events are entirely incorrect. Only new excavation might resolve the uncertainties surrounding the relative sequence of the deposit building and the deposit itself.91Published descriptions and a consideration of the surroundingtopography,however, indicate that the deposit accumulated nearby,perhaps during the period of use of the building, and that the latest material was dumped in after the building itself was abandoned. In 1930, Georg Kawerauand Theodore Wiegand published the palaces and other buildings of the Pergamene acropolis, including the building where the deposit was found. Of primary importance for the present discussion are three areas (Fig. 2): 1) Palace V, a large, elaborately
MARK
3IO
L. LAWALL
I
I
\ Deposit Terrace
I
\
Pre-Attalid fortification
I
I
I
I
I ?I
I
\ 01 Line of Byzantine CityWall (over Hellenisticcity wall)
ftfa
0000
10m
Figure4. Phase 1: Pre-Hellenistic constructionin the areaof the deposit. AfterAvPV.1,pl. 6
decorated building focused on a peristyle court, perched on a terraceabove the area of the deposit;922) the deposit terrace and the deposit building;93 and 3) building group VI, a series of storage buildings whose preserved contents and plan suggest that they were used for military storage and barracks.94Kawerau'sdescription, Carl Humann and Richard Bohn's earlier preliminary report on the excavations of 1886, and more recent research by Wolfgang Radt and others allow the following revised reconstruction of phases of activity on and around the terrace. The earliest constructions near the terrace appear as scant masonry facing of the bedrock bluff (Figs. 3-4). Kawerau describes this facing as the "oldwall,"and this is generally associated with the pre-Hellenistic city fortifications.95Regardlessof its date or even precisefunction, this masonry's relative chronological position is clear from the fact that the south wall of the deposit building covers this pre-Hellenistic masonry (see Fig. 3). The construction of the building required as a prerequisite that its terrace be incorporated into the fortified circuit of the acropolis (Figs. 3, 5). A short 92. AvP V.1, pp. 30-39; Radt 1998, pp. 8-10; 1999, pp. 68-74. 93.AvP V.l, p. 40. Kawerauand others have treatedboth the terraceand building as part of buildinggroupVI. For the presentstudy it is necessaryto distinguishthese two areas.When I referto buildinggroup VI below,I am referringto the structuresforming the
southernborderof the deposit terrace and the complex of buildingsfurther south. 94.AvP V.1, pp. 41-46; for arsenal buildingsat the north end of the acropolis,see AvP X. 95.AvP V.1, p. 40. Radt (1999, fig. 8, and pp. 55-56) illustratesthis stretchof pre-Attalidfortification
in his plan of the phases of the fortificationbefore Eumenes II, but does not discussit further.Conze (AvP 1.2, pp. 164-165) describesthe remainsof pre-Attalidfortifications north of the deposit area,but does not discussthe tracesof walls nearerthe deposit.
EARLY
,
_ -
\
EXCAVATIONS
AT
PERGAMON
3II
Palace V
/TH II
%
\
I
I % Deposit Terrace
Deposit Buildi
\
\ \
Deposit Terrace
Citywall
I
I
(Philetairos) I
I I I
BuildingGroupVI
0.00 000 .00* .0-\ 000 0#0 I01-00 10m
_
_
10 m
Figure5 (above,left).Phase2: Constructionin the areaof the deposit, PhiletairosthroughAttalosI.
AfterAvP V.1,pl. 6
Figure6 (above,right).Phase3: Constructionin the areaof the deposit,EumenesII andAttalosII. After AvP V.1, pl. 6
96. Conze (AvP 1.2, pp. 177, 183) and Radt (1999, pp. 56-57) describe fortificationsattributedto Philetairos. 97. AvP V.1, p. 40. Kaweraualso notes here that the westernmostcrosswall is comprisedof two drastically differentmasonrystyles, irregular polygonal masonryat the north and large squaredblocks to the south.This
stretch of polygonal-masonry wall core under the current Byzantine city wall is the only evidence for this event, and this wall is currentlyassociated with the reign of Philetairos.96Once this outer wall appeared,the building itself could be constructed. Kaweraudescribed the deposit building in considerable detail, noting its position on what must have been an extension of the acropolis area created by the outer (Philetairan) wall.97He describes the fill against this outer, later wall as being rich in pottery, including "morethan a thousand handles," and he points out connections between the stamps on these handles and many workshops.98That Kawerauis describing the Pergamon Deposit or something very closely related is indicated both by his reference to Schuchhardt'spublication and by his illustration of the areawith a Turkish workman holding an amphora handle while standing inside the deposit building (Fig. 1). Bohn, in 1888, used precisely the same description for fill against the Philetairan wall.99In Bohn's case, however,the description comes within a description of the (later) long, east-west building, which divides the main part of building group VI from the terrace (Figs. 3, 6). The back wall of differencein masonrymay raisethe possibilityof a phase of renovationfor the deposit building. 98. AvP V.l, p. 40: "In dem als Hinterfiillungder vorgeschobenen Mauer [the 'latercity wall'] benutzen Schutt fand sich eine auffallendgroge Menge von ScherbeneinfacherTongefaiie, darunternamentlichmehr als
tausendHenkel, welche die Stempel sehrverschiedenerWerkstattentragen. (Bohn, vgl. A.v.P.VIII 2, Inschriften, S. 423ff., nr.766-1322)." 99. The lines cited in the previous note from Kawerau'sdescriptionare a direct quotationof Bohn'spreliminary report(Humann, Bohn, and Frankel 1888, p. 67).
3I2
MARK
L. LAWALL
this set of rooms abuts the acropolis wall. Bohn does not discuss the deposit terraceor building, nor does he specify the extent of the pottery-rich fill. The only other published reference to this fill appears earlier in the same preliminaryreport,where Humann notes the discovery of the many stamped handles that were then copied carefully by Schuchhardt.'00Despite Schuchhardt'stestimony that the published stamps filled rooms of the deposit building, Bohn's and Kawerau'sdescriptions of the fill lying against the acropolis wall indicate that the fill extended over much of the terrace. Why Schuchhardt's description of the context of the deposit is more restricted than Bohn's (or Kawerau's,which was clearly taken from Bohn) remains unclear,but the extension of the deposit beyond the confines of the building itself seems very likely, given these published reports. The spatial and chronological relationship between the building and this amphora-rich fill thus becomes more complicated. Before addressing this problem, I continue with the relative sequence of buildings. At roughly the same time as the deposit building'sconstruction, other storage or perhaps barracksbuildings appeared both north and south of the terrace (Figs. 3, 5). To the north are scanty remains of these buildings beneath Palace V; their masonry technique is generally described as similar to that of the deposit building.'10To the south, similar masonry is poorly preserved in the area of later arsenal and barracksbuildings just within the circuit of the city wall and near a major gateway to the acropolis.102These walls, too, are considered roughly contemporarywith those of the deposit building. All of these structures,including the deposit building, have been interpreted as a series of storage buildings spread over the south end of the acropolis. The construction of Palace V over the storage buildings north of the terrace and of new arsenals and barracksof building group VI together mark the final phase of Hellenistic building activity in this area (Figs. 3, 6).103Once these buildings were in place, the terrace became strikingly isolated. Construction of the palace removed the storage buildings that had overlooked the terrace.The south and west walls of the new palace are perched on the same rocky escarpment that had supported the preHellenistic city wall, and these new walls would have backed against and towered over the deposit terrace.Likewise, the back wall of the east-west building of building group VI effectively blocks access to the terracefrom
100.Humann,Bohn,andFrainkel 1888,p. 57. 101.Radt1999,p. 74;AvPV.1, pp. 30, 34, 40; AvP X, p. 56. 102. AvP V.1, pp. 42-44; Radt 1999, 74. p. 103. For the approximatecontemporaneityof these new buildingswith the constructionof PalaceV, see Radt 1999, p. 74, where the authoralso
of PalaceV notesthatthe construction wouldhavereducedthe numberof militarybuildingson the citadel,
and p. 76, where he suggeststhat, as a result,the arsenalbuildingsat the north end of the citadelwere expanded. The masonrystyle of the walls of the long east-west building resembles the masonryof the latest Hellenistic acropoliswall, which has been attrib-
utedto EumenesII;see Radt1999, figs.10 (atright)and26;the southernpartof buildinggroupVI appears to date to the same phase as the Great Altar (p. 91, fig. 39d). Kawerau (AvP V.1, pp. 41-45) considersthe many phases of constructionin this area,but does not suggest specific dates.
EARLY
EXCAVATIONS
AT
PERGAMON
313
the south. The heavy,long south wall of the deposit building would have furtherdiscouragedcommunication with building group VI; any door from the north side of building group VI (and none is preserved) would have opened only onto a narrow space, 2.5-4 m wide, if the deposit building was still standing and in use.104 As for the west side of the terrace,Radt's recent reconstruction places an entrance to Palace V along its southern facade approached by stairs from the southwest.105Near this entry route it might seem as though there is continuing access to the terracebetween the palace'ssouthernmost wall and the north side of building group VI. Here too, however, access would have been made difficult by the bedrock escarpment.The precise height of this escarpment over the original ground level of the terraceis not certain, but observation of the site today and photographs from the early excavations (e.g., Fig. 1) confirm the difficulty of access to the terrace from the west.106It is possible, therefore, that the deposit building, now located behind this heavy wall to the south and down a rocky slope from the buildings to the north, went out of use once these buildings were completed. There is no indication of access between the palace and the lower terrace; instead the palace appears to back up against the rocky drop down to the deposit terrace.'07Thus, the terracewas clearly isolated by this final phase of construction, and it seems most unlikely that the deposit building remained in use. Where, then, does the Pergamon Deposit fit into this sequence of buildings? The broader extent of the deposit, now apparent from Bohn's and Kawerau'sdescriptions, should exclude from consideration any interpretation that is limited to the interior of the deposit building itself. The deposit did not serve simply to raise the floor level of the building itself.'08 Other possibilities include the following: 1) the deposit representsthe construction fill for the building and its accompanying terrace;2) the deposit representsgradual fill that accumulated over the area as a result of the use of the deposit building; or 3) the fill covered the terrace after the building 104. The parallelorientationof the southernwall of the deposit building and the northernwall of the eastwest building in building groupVI might imply their contemporaneity (I owe this observationto K. Slane). Kawerauthought that the masonry styles of the two walls were too different to be contemporary(see Fig. 3, afterAvP V.1, pl. 6), and Radt (1999, fig. 15) modified Hans Schlief's plan of the acropolis(1932) to separatethese walls into differentbuilding phases (see Radt 1999, fig. 14). 105. Radt 1999, p. 68, fig. 15. 106. Radt (1999, pp. 74-76) notes the separationbetween the deposit
terraceand the higher areato the north:"Erwar an den nach Norden steil anstiegenden Fels gelehnt. Diese natiirliche,hohe Felsstufe scheidet das Plateauvon PalastV von der niedrigergelegenen BaugruppeVI."Radt is using the standard definition of buildinggroup VI as including the terrace.In addition to Fig. 1 here (fromAvP V.1, text pl. 20), see Radt 1999, p. 75, fig. 26, in the background,where the upperpart of this escarpmentis visible at roughly the height of the Byzantinetower door; andAvP V.1, text pl. 19.1, showing the height of the bedrockover the remains of the deposit building.The drop to
the lower terraceis also described by Kawerau(AvP V.1, p. 40). 107. The doorwaymarkedat the back of PalaceV (See Radt 1999, fig. 15), opening roughlytowardthe deposit terrace,opens onto an area between the palaceand the fortification wall still far abovethe deposit terrace. 108. Grace'soriginalintention in her phrasingis hardto ascertain.For the presentpurposes,the possibilities of interpretation-regardlessof the risk of misconstruingGrace'sintended meaning-are what must be addressed and, in this case, excluded.
314
MARK
L. LAWALL
went out of use (or as the building was going out of use). One point should be emphasized as a preliminary caution: there is no need to imagine that the deposit comes from a single, stratigraphicallyuniform fill. During excavation, fill over the ancient activity surface might well have been accidentally mixed with the earliest construction fill in the area; similarly, destruction-phase fill may have been combined with use-period fill. Such mixing occurs in excavation today and there is no reason to exclude the possibility in 1886. From these three possible circumstances for the accumulation of the deposit come further expectations and implications. If the deposit represents (in large part) construction fill, we might expect to find a concentration of eponyms around the time of the building's construction and then a significant decrease in the appearanceof later eponyms. In that case, the date of the construction of the deposit building should establish the latest date for the bulk of the Rhodian eponyms. For the other two scenariosaccumulation during the use of the building and a dumped fill following its abandonment-the significant event for the chronology of the deposit is the abandonment of the building. This event would mark the latest date of the bulk of the deposit. If the debris accumulated through the use-life of the building, then the eponyms should be spread(not necessarilyevenly) across the use-period of the building. Even if all or some of the material accumulated in a nearby areawhile the deposit terracewas in use and was only dumped in after the terrace was abandoned, a broad spread of eponyms would still be expected. We areleft with two fundamental points to choose between: either the Pergamon Deposit dates to the construction phase of the deposit building or its terminal date corresponds with the abandonment of the terrace. Had the debris been terrace fill for the construction of the deposit building, as Kawerauand Gelder envisioned,it is hardto imagine the source of so much pottery,particularlyamphoras.The sheer size of the Pergamon Deposit deserves attention. Nearly 900 amphora stamps were published from this area,which covers at most 400 m2.A minimum density in terms of surface area (calculated on the basis of the horizontal spread of the deposit and excavations,without taking the depth of the fills into account) would be 2.3 stamped amphora handles (sah) per square meter (sah/m2). If the ca. 900 stamps published by Schuchhardt were indeed found only within the building (ca. 10 x 10 m), this density rises to roughly 9 sah/m2. Even at 2.3 sah/m2,few other accumulationsof stamps areas denselypacked or as numerous. The construction fill for the Middle Stoa in Athens, for example, far exceeds the Pergamon Deposit in terms of number of stamps, but the density of stamps is significantly lower (0.6 sah/m2).109Moreover, the MSBF accumulatedin an areawhere commercial activities took place that would have generated much debris.10 That some unusually rich source of amphoras is necessary to explain such a high density of amphora stamps in the Pergamon Deposit is demonstrated by further examples. A late-second-century building from Hellenistic Ilion, covering ca. 109 m2,included only 11 stamps in its construction fill (0.1 sah/m2); the building is clearly separated from the Late Hellenistic agora of Ilion.1" Fill over the floor of the 1,225-m2 Square
109. Nearly 1,500 stampsare reportedfrom the MSBF. There are other,uncatalogued,stampsfrom this fill, most of which arebarelylegible, storedin the Agora context pottery tins; I have seen no more than 30 of these. The Middle Stoa coversan area of ca. 2,572 m2,but the entire areahas not been excavated.Thus, the minimum density of stampsin this areais ca. 0.6 stampsper squaremeter.Even though the fill was most substantialin the westernhalf of the area(S. Rotroff, pers. comm., 4 April 2002; and see Grace 1985, pp. 21-24; Rotroff 1988), the density there does not approachthe figurefor the PergamonDeposit. 110. Earlier(5th-4th centuryB.C.) evidencejust east of the Middle Stoa areafor commerceinvolving amphoras is highlighted in Lawall2000. 111. Panasand Pontes (1998) publish many of these stamps (among many others from excavationsat Troy); the building in question is reportedas the "NorthBuilding"in Rose 1993, pp. 100-104; 1994, pp. 76-80.
EARLY
EXCAVATIONS
AT
PERGAMON
3I5
PeristyleBuilding, which lay under the Stoa ofAttalos, included 103 stamps (0.08 sah/m2).The building fill for the Stoa of Attalos, covering an areaof 2,300 m2, only contained ca. 135 amphora stamps (ca. 0.06 sah/m2).12 The density of stamps in terms of surface area for the Pergamon Deposit is clearly unusual and indicates a substantial source of amphora storage and use. At this early stage, there is no candidate for a nearbysource of the debris (Fig. 4). The poverty of evidence for such a source preceding the construction of the deposit building contrasts sharply with the plentiful evidence for potential sources after the construction of the building. The construction date of the deposit building is difficult to know with precision or certainty.The terminus post quem should be placed within the reign of Philetairos (283-263 B.C.),after his construction of the wider course of the city wall encompassing the deposit terrace. The terminus ante quem is the construction of Palace V and the roughly contemporary embellishment of building group VI. As noted above, these constructions isolated the terrace, making it unlikely that any subsequent construction took place on the terrace.The sixteen Thasian amphora stamps published as part of the deposit may provide the construction date of the deposit building."3 These Thasian stamps tend to date near or shortly after the middle of the 3rd century,with a group of four all with the same magistrate and a fifth with a different name necessarily close in date to the The clustering of Thasian magistrates near the middle of group of four.114 the century might indicate a construction date during the time of Eumenes I (263-241 B.C.) for the deposit building and the other neighboring storage buildings. While these few Thasian stamps might date near the time of the deposit building's construction, the vast bulk of the deposit, as noted above, requiresthe presence of large-scale amphora storage and use in the vicinity. For this reason, the period of use of the deposit building and the other storage buildings in the area should provide a chronological context for the accumulationof the Pergamon Deposit. The frequentlyrepeating eponyms (and fabricants)among the Rhodian stamps attest to large-scaleaccumulation in the vicinity (Table 1). Had this accumulation occurredfarther away on the site, the chronological unity of the fill, with so much repetition of names, is less likely to have been maintained. The point at which accumulation stopped should correspond to the destruction of these older storage buildings. These storage buildings overlooking the deposit terrace were destroyed as part of the project to build Palace V. The contents of 112. Grace 1985, p. 24, n. 63; see alsoAgoraXXII, p. 106: "morethan 100." 113. Borker1998, nos. 537-552. The Thasian amphorachronologyfor the 3rd centuryB.C.has been developed independentlyof that for the Pergamon Deposit; see Debidour 1979; 1986; 1998a; 1998b; and Histria VIII.1. 114. Borker1998: nos. 539-542 all carrythe eponym Aristomedes,while anotherThasian stamp (no. 546) carries
the eponym Herakleides.A stamp of this eponym was recutto show the eponym Aristomedes,so presumably these two eponyms are close to one anotherin date (Debidour 1986, p. 333; 1979, pp. 298-299). Avram (HistriaVIII.1) only includes a Herakleides stamp that Debidour considers to be an earlierhomonym to the example at Pergamon.Avramdoes not include Aristomedesin his year-by-year lists of Thasian magistratesfrom
specific complexes.He does propose dates of 256 B.C. (Borker1998, no. 538, Antianax) and 255 B.C. (B6rker 1998,
no. 543, Diagoras) for two earlierThasian stampsin the PergamonDeposit. The only Thasian stamp from the deposit that has been given a date so far earlierthan 263 B.C. (the beginning of the reign of Eumenes I) is Borker1998, no. 550, SatyrosI, which Avramplaces at 270 B.C.
MARK
316
L. LAWALL
TABLE 1. RHODIAN EPONYMS ASSIGNED TO PERIOD III FROM THE PERGAMON DEPOSIT Eponym
Number ofExamples
Eponym
AopxuXi8ag 'AyXo6p3ippozTo aocpaoioXL;
1 1 -A
'Apoxis8aog Aivulocax8[ogII AlvzTcop
StapoloJ;
--
Scho7TpaTo;
-
KaXXtxpdTrr II AaotoxXh; II
KAX?Lt6oXoS Oeatco
1 5 2 1 8 3 8 13 17 9 11 14 30
Aaol606?tS; 'Ia7Lxpar'lS E?vo(pavrl; IIpaTccoqo(pa KpoatA'oc 'Ipcov I II 'ApxoxpTYCrg; Ttaiococy6opa; (IoX68aLo;SII KX?c6vuJto; 'AyitXaXog;
KaXXixpa-ocia; II KXA?xpCa-q;I E6.Jt aXoS NLxaaocy6pocaI 0?aOCrTrog; 'AOav6ootzo I 'ApaTroopvYq; 'ApLtoaceSc;II II 'ApaCTcov 'Apto6oa(ioog II 'ApxXaL'oac; ^Evoqc)v
'Ay?aTcpaTogII
Number ofExamples 16
19 16 8 21
26 13 18 10 12 14 8 10 20 23 6 5 3
are order,followingFinkielsztejn 2001,p. 192.Frequencies Eponymsarelistedin chronological takenfromBorker1998,pp.6-7.
these buildings could have been easily tipped downhill onto the now-abandoned terrace and surely could have been mixed with debris from the deposit building itself."5 This proposal for an initial accumulation of the deposit material both under Palace V and on the terrace, followed by its incomplete redeposition on the terrace,is supported by Borker'sobservation of the rarity of matching pairs of handles in the deposit.ll6 The bulk of the fills described by Bohn, Schuchhardt, and Kawerauthus must have overlain the construction fills for the deposit building. The construction date of Palace V marks the closing date for the Pergamon Deposit. The best, and most often cited, evidence for the construction date of Palace V comes from the fact that a coffering block intended for the Great Altar was built into the palace wall;"' construction thus must have occurred sometime after the beginning of work on the altar. Fragments of relief decoration from the palace show scenes from a gigantomachy and from the life of Telephos, directly echoing the iconography on the Great Altar and further implying the chronological proximity of the palace and the altar.l8 Study of the mosaic decoration of Palace V and the pottery beneath the floor has led to a suggestion of a mid-2nd century date.l9 The pottery in question, however,has not been published (or even described in general terms),120so it is necessary to consider what indications the Great Altar itself might provide for the date of Palace V. The date of the construction of the Great Altar is debated, but various argumentspoint to the late 170s or the 160s B.C.A series of articlesby
115. It is difficultto imagine the debriscoming from earlierbuildings within the southernpart of building group VI. Such debriswould have had to be moved upslope and may have been useful as leveling fill in the immediatearea. 116. Borker1998, p. 8. Some portion of the debrisfrom the storage buildingsbelow PalaceV would have been used in the constructionfill for the palace. 117. Radt 1999, pp. 72-74, figs. 2425; AvP V.1, pp. 32-33. 118. Hoepfner 1997, p. 39, fig. 14. 119. Salzmann1995, esp. pp. 109110. 120. The chronologicalsequence for Hellenistic pottery at Pergamon until very recentlydependedin parton a closing date of ca. 180 for the Pergamon Deposit, i.e., the traditionalRhodian stamp chronology,and on a similar date for the constructionof the Great Altar (PF 2, p. 26; and now see de Luca in PF 12, pp. 126-127).
EARLY
EXCAVATIONS
AT
PERGAMON
317
P.J. Callaghan offered an argument for a date in the late 160s. Callaghan proposed that Megarian bowl fragments found within the foundations of the altar carrywreath forms that first appear on Pergamene coins in 172; the use of this motif on pottery thus occurred sometime after 172. Other bowl fragments carry a "long-petal" motif that, according to Callaghan, first appearsshortly before 165 B.C.121Likewise, from an iconographic and historical perspective, Bernard Andreae has suggested that the construction of the Great Altar fits best with Eumenes II's successes against the Gauls, culminating in the battle of Mt. Tmolos in 166 B.C.122 Callaghan's argument has been criticized for having been based on More recent excavationsand publications have now contoo few sherds.123 siderablyexpanded the availableevidence. A series of further trenches investigating the Great Altar's foundations were excavated in 1994.124 Gioia de Luca and Radt note that, even on the basis of this wider sample of pottery,a construction date in the 170s cannot be securelydistinguished from one in the 160s. While raising the possibility that certain fineware types began to be produced earlierat Pergamon than elsewhere, they reject the earlier date of ca. 180 for the altar'sconstruction. Using arguments of historical probability and elements of the altar'siconography (i.e., nothing from the new excavations!), they conclude that construction began Susan Rotroff, reviewing de Luca and Radt's publication, ca. 172 B.C.125 notes that, if that date is correct, many associated pottery forms must be given surprisinglyearlier dates than at other sites.126For the present purposes, ascertaining a construction date late in the 170s or within the 160s is sufficiently precise. At some point after construction of the Great Altar began, it became apparentthat coffering blocks-already fully carved and intended for the altar-were no longer needed.127The state of completion of the altar is a matter of debate. For those who consider it to have been completed as planned, the unused coffer blocks (and there are others, in addition to the one built into a wall of Palace V) can only be explained as extras, ordered for the project but never used. In such a scenario, the point at which it was determined that these blocks were no longer needed would seem impossible to reconstruct. If, on the other hand, construction was brought to a close before the altarwas completed, then the date of this interruption is likely to provide a terminus post quem for the reuse of the unused coffer blocks in other buildings. Kastner and Radt both argue for this latter 121. See Callaghan1980 on the invention of certainmoldmadebowl types in the areaof Pergamonin the second quarterof the 2nd century;also 1981 (wreathforms);1982 (long-petal bowls);Rotroff 1988. 122. Andreae 1997. Grace,too, seemed to lean towarda date for the Great Altar in the 160s. On 3 March 1960, she wrote (unpublished notes) that the altarwas dated "180 or 160 B.C., of which 160 goes
better stylistically." 123. PF 12, p. 123; Radt 1998, p.20. 124. Radt 1995. 125. PF 12; Radt 1999, p. 169; cf. Radt 1988, p. 190 (publishedbefore the renewedattention to the altar foundationsin 1994). 126. Rotroff2001. 127. Kawerau(AvPV.l, p. 33) notes the uncertaintyas to why the cofferblocks were not used on the altar
as intended. Kastner(1997, pp. 72-73) explainsthe move of blocks from the altarto other buildingson the citadel (includingPalaceV) by positing an interruptionto the constructionof the altar;similarly,Radt 1999, p. 178; see also Hoepfner (1997, p. 64), who proposesthat the altarwas completed without change of plan, but does not offer an explanationfor the unused cofferblocks.
3i8
MARK
L. LAWALL
scenario, and Radt suggests that work on the altar ceased with Attalos II's ascension to the throne in 159, following the death of his brother Eumenes II. Had the brothers begun work on Palace V during the 160s, which remains possible, that work had not progressed far before the coffer block became availablefor use just above the foundation level.128A date for the construction of the palace must fall very late in the 160s B.C., or even in the early 150s.129This projectwould have requiredthe leveling of buildings on the site of the palace. A reasonableplace to have dumped the debris from the buildings would have been the small, now-isolated terraceto the south. In sum, three factors seem to rule out a closing date for the Pergamon Deposit in the 180s or 170s: 1) the date of the Great Altar (no earlierthan 172 B.C.);2) the construction date of Palace V (near or after 159 B.C.); and
3) the topographic situation of the deposit terrace(cut off by the construction of Palace V and the east-west building of building group VI). The construction or early period of use of storage buildings under Palace V or the deposit building itself could have provided the scatter on the terraceof independently datable "early"stamps, mostly Thasian, from the decades around250 B.C. The concentratedseries of Rhodian stamps,with its nearly continuous series of eponyms, would date to the main period of use of the deposit building and other nearby storage buildings. The end of the series should coincide with the clearance of the buildings under Palace V. The most likely closing date, therefore,for the Pergamon Deposit is during the very late 160s or early 150s.130
FOR ARCHAEOLOGY IMPLICATIONS ECONOMIC HISTORY
AND
A closing date of the late 160s or early 150s for the Pergamon Deposit fits very well with Finkielsztejn'srevision of the Rhodian chronology by 11 or more years. He has placed the latest Rhodian eponym in the deposit as late as ca. 161, instead of 175 B.C. Since Finkielsztejn worked backward from the known dates of the destructionsof Carthage,Marisa,and Samaria, his revision provides independent support of the later date proposed here for the Pergamon Deposit.13' This lower date for the close of the deposit has ramifications for other amphora stamp chronologies, as well as for determining the construction dates for major Hellenistic buildings outside Pergamon. The two latest Knidian amphora stamps in the Pergamon Deposit, Borker no. 559 with and B6rker no. 554 with the eponym AOJql( the eponym KXlqv6orokXt Oe(, both fall within the period of the phrourarchs,and neither is present in the MSBF. For these reasons, Grace placed them after 183 B.C.but, since they Lowering the closing date at Pergaappearat Pergamon,before 175 B.C.132 mon to the late 160s makes it possible that these two Knidians, and indeed the other eponyms with phrourarchs,might also date later than had been proposed by Grace. The earlier dates for Knidian eponyms associated with terms of phrourarchsdepended very heavily on an interpretation of the historical record that now appears far from certain. Lowering these Knidian dates
128. See AvP V.1, pp. 32-33. 129. Radt (1999, p. 178) places the constructionof PalaceV ca. 159 or shortlythereafter;on p. 74 he suggests that constructionof the palacecould have begun in the 160s or 150s. 130. I thank Susan Rotrofffor drawingmy attention to Strabo's unfortunatelybrief summaryof the constructionactivitiesof Eumenes II and his brotherAttalos (13.4.2). Although the text is debated(see H. L. Jones, trans.,London 1970, p. 167, n. 2), it would seem that Attalos was farbetter known for his actualconstructionprojects,while Eumenes was more involvedin landscapingprojects. Whether Attalos'sprojectswere carried out in his brother'sreign or in his own reign is not clarifiedin the text. 131. Finkielsztejn(2001, passim) cites the manuscriptof my reviewof PF 11 (Lawall2002), but his revised chronologywas developedwell before I wrote that reviewand is based on a sequenceof argumentsthat stand independentlyof the PergamonDeposit dating. 132. Grace 1985, p. 15, n. 32.
EARLY
EXCAVATIONS
AT
PERGAMON
319
leads in a very similar direction as Finkielsztejn'sRhodian revision. Grace's Knidian eponym list published in 1985 did not account for at least 12 years between 146 and 108 (period V).133Lowering the 146 boundary on the basis of a lower date for the Pergamon Deposit could fill this gap. A point of difficulty in doing so, it might be argued, is the need to place any Knidian eponyms at Corinth before the city's destruction in 146 B.C. While there is as yet no publication of Knidian (or other non-Corinthian) amphora stamps from Corinth with information about their findspots, 42 Knidian stamps, assigned by Grace to the period after 146 down to 80 B.C. (i.e., securely before the resettlement of the city by Julius Caesar in 44 B.C.), do appear at Corinth. This figure represents 24% of the Knidian stamps found at Corinth as of 1990.134If nearly a quarter of that city's corpus of Knidian stamps is to be dated after its destruction, then the security of 146 B.C.as a terminus ante quem needs careful evaluation, context by context.135 Returning briefly to the MSBF, the lower date proposed for the Pergamon Deposit necessitates assigning a later date for the construction of both the Middle Stoa and the Stoa of Attalos. Such a shift is not as problematic as it might seem. A closing date near 161 B.C. for the deposit brings the MSBF to 169 or later. According to the list of Knidian eponyms published by Grace in 1985, there are 25 or 26 period IV eponyms found in the Stoa of Attalos building fill that are not found in the MSBF. This difference might imply a ca. 25-year gap between the two periods of construction (probablyfewer years if some of the period IVA Knidian eponyms under the Stoa of Attalos are earlier than the latest in the MSBF but simply not represented there). A date near 144 B.C. for the construction of the Stoa of Attalos is certainlypossible in terms of the long reign of the donor.136 133. Grace 1985, pp. 32-35, providesthe most up-to-date published list of Knidianeponymswith her assignationsof periods.In this list, 22 names arepresentedas certain membersof period V, while another four arelisted as possibilities.Grace never mentions the discrepancy between these numbersof eponyms and the ca. 38 yearscoveredby period N. Efremov(1992, pp. 257-258,264) proposesthat the periodbefore the startof the duovirion the Knidian stamps should end ca. 114 B.C. (his period VI = Grace'speriod V). He takes this date from the fact that no duovirstampshave been found in the region of Chersonessosand that region was attackedin 114 B.C.,therebyproviding a possible cause for the interruptionof tradewith Knidos. Even so, Efremovallots 32 yearsto his period VI-still not short enough to be filled with Grace'speriod V eponyms.
134. As of 1990, a total of 176 Knidianstampshad been found at Corinth;see C. G. Koehlerand P. M. W. Matheson, "Importsof KnidianWine at Athens and Corinth," http://www.chass.utoronto.ca/ for amphoras/aia90.htm#imp-kor, a slightly modified versionof a paper deliveredat the 92nd Annual Meeting of the ArchaeologicalInstitute of America (Koehlerand Matheson 1991). 135. On the problematic"interim" phase at Corinth, between 146 and 44 B.C., see Romano 1994;Williams 1978, esp. pp. 21-22; CorinthVII.3, p. 90. 136. K. Slane (pers.comm.) comments that there should be 16 years, as accountedfor by Knidianphrourarchsand eponyms,between the final date of the PergamonDeposit and that of the Stoa of Attalos building fill. Using Finkielsztejn'sdate of
161 B.C., we come to ca. 145 for the
Stoa of Attalos. Rotrofftentatively suggesteda date in the 140s B.C. for the Stoa of Attalos on the basis of the similarityof the stamped amphorahandles there to examples found at Corinth (AgoraXXII, p. 106). More recently,that date was changed to ca. 157 on the basis of Grace'spublicationof the MSBF in 1985 (Grace 1985, pp. 14-15; followed, for example,by Finkielsztejn 2001, p. 41; cf. AgoraXXIX, p. 468, "to ca. 150").Neither the coins nor the moldmadebowls in the construction fill of the Stoa of Attalos prove a date of 157 as opposed to a date in the 140s. The date of 157 B.C.rests solely on Grace'sinterpretationof the Rhodian and Knidianstamp chronologies; this date is not a historically"fixed point"(see, too, AgoraXXVII, pp. 104, 111).
320
MARK
L. LAWALL
Several implications thus follow from a revision of the closing date of the Pergamon Deposit. An accurate accounting of the relationship between the Pergamon Deposit and other contexts, however,and full understanding of the effects of the proposed lower date on other chronologies, will depend on the completion of certain monumental tasks. Assembling correct lists of stamps from the Stoa of Attalos and South Square (South Stoa II, East Building, and Middle Stoa) in Athens will require a thorough reexamination of the stratigraphy of those building fills and the The frequent referencesmade above to the Knifindspots of the stamps.137 dian chronology highlight the need for a publication of that class of amphora stamp with particularattention to which eponyms arepresent in the various majorbuilding fills and other useful contexts.138Publication of the Knidian and Rhodian stamps from Corinth, with particularattention to distinguishing finds assignable to before and after 146, would provide a useful test of the implications described above for the Knidian chronology. Finally, this separation of the dates of Rhodian stamps at Pergamon from assumptions about the relationship between trade and politics frees us to explore more fully the nature of such a relationship.The early connection drawn between the Pergamon amphora stamps and political relations has long colored historians' consideration of Rhodian trade with As noted though recentlyscholarshave been more skeptical.140 Pergamon,139 it is difficult to the at which above, imagine point decaying political relations would stop or slow Rhodian-Pergamene trade, especially when only a part of that trade might have involved Rhodian merchants themselves as opposed to merchants of other states. Even so, our ability to separate Rhodian chronology from assumptions connecting trade and politics, a possibility first introduced by Finkielsztejn'srevisions to the Rhodian chronology and strengthened by the present study, allows us to consider the decline or expansion of Rhodian (and other cities') trade with Pergamon in light of historical events. Finkielsztejn has taken this process quite far alreadyby noting the potential impact of the creation of the "free port" at Delos on Rhodian trade Yet other questions remain. Do other amphoraimports with Pergamon.141 decline at Pergamonafterthe 160s?The events of 166 B.c. certainlychanged the commercial topography of the Aegean, but the impact of these events on Rhodes and other commercial centers is still much debated.142 A more thorough study of import patterns before and after 166 B.c. at Pergamon and elsewhere would begin to clarify who benefited from the new status of Delos (or if that had any impact at all). Is there any indication that local events nearerPergamon had a greater impact on what was imported to that city?What was the economic impact of Pergamon'sconflict with Gallic mercenariesin 168-166 B.c., or with PrusiasII of Bithynia If some portion of Pergamon's imported amphoras in the 150s B.C.?143 arrivedindirectly through a series of overlappingregional spheres of trade, rather than coming directly from Rhodes, then the local environment of Pergamon should have had a significant impact on the attractiveness of the city to merchants. In that case, too, as with the question of Delos's
137. See, e.g., Rotroff 1988 and Townsend'swork in AgoraXXVII. 138. Cf. Jefremow'sproposed Knidianchronology(1995), which does not, unfortunately,take sufficient accountof Aegean contexts. 139. See above,p. 296. 140. E.g., Gabrielsen1997, p. 67. 141. Finkielsztejn1995, p. 280. 142. See, e.g., Gabrielsen1997; Berthold 1984. Reger (1994, p. 270) placesgreateremphasison the destructionof Corinth as the turning point in the Delian and Cycladic economies. 143. For the events of 168-166, see Polyb.29.22,30.1-3; Livy 20.1, 34.1014, 44.21, 45.19.3, 12; for Pergamon's conflict with Bithynia,see Polyb. 32.15.1-14; App. 12.1.3-7; Strab. 13.4.2.
EARLY
EXCAVATIONS
AT
PERGAMON
32I
impact, study is needed not only of the Rhodian finds at Pergamon, and not only of the stamped handles, but of all diagnostic amphora fragments from carefully selected datable contexts. No doubt other historical events around the Aegean (e.g., the destruction of Corinth in 146 B.c. and the bequest of Pergamon to Rome in 133 B.c.) could also be consideredagainstthe amphorarecordat Pergamon. Now that the Rhodian and Knidian chronologies, and any furtherceramic chronologies built with reference to these two stamp classes, have been separatedfrom an assumed connection between politics and trade,we can begin to explore that problematic connection with less danger of circular reasoning.
REFERENCES Adams,W. Y. 1979. "On the Argument from Ceramicsto History: A Challenge Based on Evidence from Medieval Nubia,"CurrAnthr 20,pp.727-744. Ager, S. L. 1991. "Rhodes:The Rise and Fall of a Neutral Diplomat," Historia60, pp. 10-41. AgoraXXII = S. I. Rotroff,Hellenistic Pottery:AthenianandImported MoldmadeBowls(AgoraXXII), Princeton 1982. AgoraXXVII = R. Townsend, The East Sideof theAgora:TheRemains beneaththe StoaofAttalos(Agora XXVII), Princeton 1995. AgoraXXIX = S. I. Rotroff,Hellenistic Pottery:AthenianandImported Wheelmade TableWareand Related Material(AgoraXXIX), Princeton 1997. Andreae,B. 1997. "Dating and Significanceof the Telephos Frieze in Relation to the Other Dedications of the Attalids of Pergamon,"in Dreyfus and Schraudolph1997, pp. 121126. AvP 1.2 = A. Conze, Stadtund Landschaft(AvP 1.2), Berlin 1913. AvP V.1 = G. Kawerau and
T. Wiegand, Die PalastederHochburg(AvP V.1), Berlin 1930. AvP VIII.2 = M. Frankel,ed. Die Inschriftenvon Pergamon: RomischerZeit. Inschriftenauf Thon(AvP VIII.2), Berlin 1895.
AvPX = A. Von Szalayand E. Boehringer,Die hellenistischen Arsenale(AvP X), Berlin 1937. AvPXI.1 = O. Ziegenaus and G. de Luca, DasAsklepieion1: Der siidlicheTemenosbezirk in hellenistischenundfriihromischer Zeit (AvP XI.1), Berlin 1968. Badal'janc,Ju.S. [= Badal'yants] 1999. "La Rhodes hellenistique et le nord de la Mer Noire (les relations6conomiquesd'apres l'6pigraphieceramique),"in Productionet commerce desamphoresanciennesen Mer Noire, Y. Garlan,ed., Aix-en-Provence, pp. 247-253. Badal'yants,Yu. S. 1973. "Dopolnitelye i 'kursivye'kleima na amforakh ellinisticheskogoRodosa,"VDI 126, pp. 48-64. . 1983. "Razvitiyerodosskoy keramicheskoyepigrafiki,"Klio 65, pp. 81-105. Berthold, R. M. 1984. Rhodesin the HellenisticAge,Ithaca. Bleckmann,F. 1907. De inscriptionibus quaelegunturin vasculisRhodiis, Gottingen. . 1912. "Zu den rhodischen Eponymen Heliospriestern,"Klio 12, pp. 249-258. Borker,C. 1998. "Der PergamonKomplex,"in PF 11, pp. 3-69. Bresson,A. 1999. "Rhodesand Lyciain Hellenistic Times," in Gabrielsenet al. 1999, pp. 98-131.
322
Burow,J. 1998. "Die tibrigenstempel aus Pergamon,"in PF 11, pp. 73138. Callaghan,P.J. 1980. "TheTrefoil Style and Second CenturyHadraVases," BSA 75, pp. 33-47. . 1981. "On the Date of the GreatAltar of Zeus at Pergamon," BICS 28, pp. 115-121. . 1982. "On the Origin of the Long Petal Bowl,"BICS 29, pp. 6368. Conze, A., and C. Schuchhardt.1899. "Die Arbeiten zu Pergamon,18861898,"AM24, pp. 97-240. CorinthVII.3 = G. R. Edwards, CorinthianHellenisticPottery (CorinthVII.3), Princeton1975. Debidour,M. 1979. "Reflexionssur les timbresamphoriquesthasiens," in Thasiaca(BCH Suppl.5), pp.269-314. .1986. "En classantles timbres surles thasiens,"in Recherches amphores grecques(BCH Suppl. 13), J.-Y. Empereurand Y. Garlan,eds., pp.311-334. .1998a. Rev.of HistriaVIII.1, in Topoi8, pp. 397-406. . 1998b. "Latutelle de la cite sur la productiondes amphoresa Thasos a la lumiered'un example: Le 'groupeau rhyton,"'Topoi8, pp.591-606. DelosXXVII = V. R. Grace and M. Savvatianou-Petropoulakou, "Les timbresamphoriquesgrec," in Explorationarcheologique de DelosXXVII: L'zlotde la Maison desCome'diens, P. Bruneau,ed., Paris 1970, pp. 277-382. De Sanctis,G. 1896-1897. Rev. of AvP VIII.2, in Rivista di Storia anticae scienzaAffini2, pp. 103104. Dreyfus, R., and E. Schraudolph,eds. Frieze 1997. Pergamon:TheTelephos from the GreatAltar2, Austin. Efremov,N. V. [= Jefremow]1992. "Kistorii torgovikhsvyazaiKnidas SevernimPrichernomor'em[On the history of tradebetween Knidos and the North Black Sea],"in Grecheskie amfory[Greek amphoras],V. I. Kats and S. Yu. Monakhov,eds., Saratov, pp.254-264.
MARK
L. LAWALL
Empereur,J.-Y. 1990. "HXpovoX6-ocr oT'v TcoVpoStaxc)v ai.q(popov eXXryvLctx'lTcepioao,"in B'EimaOT7ovixr 2Uovdv7o7qyea n7
KcpaCyixwx. E2Avrireomx XpovoAoylxdnpoopAucara v7q E22rlvtartxq, KEpayIUexIc. HpaxTcxd. Pod6o,22-25 MxapToo1989, Athens, pp. 199-209. Empereur,J.-Y.,and A. Hesnard.1987. "Les amphoreshellenistiquesdu monde Egeen,"in Ceramiques et romaines(Centre de hellenistiques recherchesd'histoireancienne70), P. LevequeandJ.-P.Morel, eds., Paris,pp. 10-54. Etienne, R. 1986. TenosI: Le sanctuaire dePoseidonet d5Amphitrite (BEFAR 263), Athens. Finkielsztejn,G. 1993. "Amphoreset timbresd'amphoresimporteesen Palestinea l'fpoque hellenistique: Etudes de chronologieet d'histoire" (These de doctoratde l'Universite ParisI, Pantheon-Sorbonne). . 1995. "Chronologiebasse des timbresamphoriquesrhodienset evaluationdes exportations 6, d'amphores,"Acta Hyperborea pp. 279-296. . 2000. "Amphoresimportees au Levant Sud a l'epoquehellenistique,"in E' Etmol7ovtxv Zvovdvrqi77 yta r7v E2ATavtoTrx KEpacytx7. XpovoAoytxdHpopAluaxraKAeLtcad Z6voAa-EpyaoTrpta. PIpaxrTIi, Athens, pp. 207-220. 2001. Chronologie detailleeet rhoreviseedeseponymsamphoriques diens,de270 a 108 av.J-C. environ:Premierbilan (BAR-IS990), Oxford. Fraser,P. M., and G. E. Bean. 1954. TheRhodianPeraeaandIslands, London. Gabrielsen,V. 1997. TheNavalAristocracyofHellenisticRhodes, Aarhus. Gabrielsen,V., P. Bilde, T. EngbergPedersen,L. Hannestad,and J. Zahle, eds., 1999. Hellenistic Rhodes:Politics,Culture,and Society(Studies in Hellenistic Civilization 9), Aarhus. Gelder,H. van. 1899. "Rhodos,"in DialektSammlungdergriechischen
EARLY
Inschriften3, H. Collitz and F. Bechtel, eds., Gottingen, pp.411-686. . 1915. "OverRhodische Kruikstempelsen hun Belang voor onze Kennisvan den Rhodischen en Mededeelingen Handel," Verslagen derKoninklijkeAkademie van WetenLetterkunde 5(1), schappen, Afdeeling pp.186-222. Grace,V. R. 1934. "StampedAmphora Handles Found in 1931-1932," Hesperia3, pp. 197-310. 1950. "The StampedAmphora Handles,"in TarsusI: Excavationsat GozliiKule,H. Goldman, ed., Princeton,pp. 135-148. 1952. "Timbresamphoriques trouvesa Delos,"BCH 76, pp. 514540. .1953. "The Eponyms Named on RhodianAmphora Stamps," Hesperia22, pp. 116-128. .1956. "StampedWine Jar Fragments,"in L. Talcott et al., SmallObjectsfrom thePnyx II (HesperiaSuppl. 10), Princeton, pp. 113-189. 1961. Amphorasand theAncientWineTrade(AgoraPicBk 6), Princeton. . 1974. "Revisionsin Early Hellenistic Chronology,"AM 89, pp. 193-200. . 1979. Amphorasand theAncient WineTrade(AgoraPicBk 6), 2nd ed., Princeton. .1985. "The Middle Stoa Dated by Amphora Stamps," Hesperia54, pp. 1-54. Gruen, E. S. 1984. TheHellenisticWorld and the Comingof Rome,Berkeley. Grfinert,H. 1987. "VonPergamonbis Garz:Carl Schuchhardt,Begriinder der prahistorischenurgenarchaologie in Mitteleuropa,"DasAltertum 33,pp.104-113. H., A. 1896. Rev. of AvP VIII.2, in Literarisches CentralblattfurDeutschland 17, pp. 624-625. Head, B. V. [1911] 1967. Historia Numorum:AManual of Greek Numismatics,2nd ed., repr.Chicago. Hiller von Gaertringen,F. 1896. "Inschriftenaus Rhodos,"AM 21, pp.39-66.
EXCAVATIONS
AT
PERGAMON
Histria VIII.1 = A. Avram,Lestimbres 1: Thasos(Histria amphoriques VIII.1), Bucharest1996. Hoepfner,W. 1997. "The Architecture of Pergamon,"in Dreyfus and Schraudolph1997, pp. 23-67. Humann, C., R. Bohn, and M. Frankel. 1888. "Die Ergebnisseder Ausgrabungenzu Pergamon,18831886: VorlaufigerBericht,"JPKS9, pp. 40-93. Jefremow,N. 1995. DieAmphorenKnidos stempeldeshellenistischen (Quellen und Forschungenzur antikenWelt 19), Munich. Karl,F., and E. D6rner, 1989. Von Pergamonzum NemrudDag: Die archiologischen EntdeckungenCarl Humanns(Kulturgeschichteder antikenWelt 40, Schriftender Hermann-Brockelschen-Stiftung 8), Mainz. Kastner,V. 1997. "The Architecture of the Great Altar and the Telephos Frieze,"in Dreyfus and Schraudolph 1997, pp. 68-82. Keil, B. 1896. Rev.of AvP VIII.2, in BPW 15, pp. 1606-1614. Koehler,C. G., and P. M. W. Matheson. 1991. "Popularityof Knidian Wine at Corinth,"AJA95, p. 336 (abstract). Kracht,P. 1990. "Uberlegungenzur Problemedes attischenHandels wahrenddes peloponnesischen Beitrage Krieges,"Miinstersche zur antikenHandelsgeschichte 9, pp. 95-98. Lawall,M. L. 2000. "Graffiti,Wine Selling, and the Reuse of Amphorasin the Athenian Agora, ca. 430 to 400 B.C.," Hesperia 69,
pp.3-90. .2002. Rev.of PF 11, in BJb 199, 1999 [2002], pp. 578-580. Levi, E. I. 1964. "Keramicheskiy kompleksIII-II v. do n.e. iz raskopokOl'viyskoy-Agori,"in Ol'via:TemenosiAgora,V. F. Gaidukevich, ed., Moscow, pp. 225-280. Lund,J. 1999. "RhodianAmphoraein Rhodes and Alexandriaas Evidence of Trade,"in Gabrielsenet al. 1999, pp. 187-204. MacDonald, B. 1982. "The Import of Attic Potteryto Corinth and the
323
Question of Tradeduringthe PeloponnesianWar,"JHS102, pp. 113-123. Magie, D. 1950. RomanRule in Asia Minor:TotheEnd of the Third CenturyafterChrist,Princeton. Maiuri,A. 1924. "Unafabbricadi anforeRodie,"ASAtene4-5, 19211922 [1924], pp. 249-269. Nilsson, M. P. 1909. TimbresamphoriquesdeLindos(Exploration archeologiquede Rhodes 5, Bulletin de l'Academieroyaledes sciences et des lettres de Danemarke),Copenhagen. Palaczyk,M. 1999. "Die Zusatzstempel ('SecondaryStamps')der rhodischenAmphoren,"OJh68, pp. 59-104. Panas,C. I., and H. R. Pontes. 1998. "StampedAmphora and Lagynos Handles from the 1989-1995 Seasons,"StudiaTroica8, pp.223-262. PF2 = J. Schifer, Hellenistische KeramikausPergamon(PF 2), Berlin 1968. PF 11 = C. B6rkerandJ. Burow,Die hellenistischen aus Amphorenstempel Pergamon(PF 11), Berlin 1998. PF 12 = G. de Luca and W. Radt. 1999. Sondagenim Fundamentdes grossenAltars(PF 12), Berlin 1999. Pridik,E. 1896. "Amphorenstempelaus Athen,"AM21, pp. 127-187. Radt,W. 1988. Pergamon:Geschichte undBauten,FundeundErforschung einerantikenMetropole.Cologne. .1995. "Pergamon:Vorbericht tiberdie Kampagne1994,"AA1995, pp.575-595. . 1998. "RecentResearchin and about Pergamon:A Survey (ca. 1987-1997)," in Pergamon. Citadelof the Gods.Archaeological Record,LiteraryDescription,and ReligiousDevelopment,H. Koester, ed., Harrisburg,pp. 1-40. . 1999. Pergamon:Geschichte undBauteneinerantikenMetropole, Darmstadt. Reger,G. 1994. Regionalismand Changein theEconomyofIndependentDelos,314-167 B.C. (Hellenistic Cultureand Society 14), Berkeley.
MARK
324 .1999. "The Relationsbetween Rhodes and Cariafrom 246 to 167 B.C.," in Gabrielsen et al. 1999,
pp. 76-97. Regling, K. 1901. "Zurgriechischen Miinzkunde,"ZJN23, pp. 107-116. Reinach,S. 1896. Rev.ofAvPVIII.2, in Revuecritiqued'histoireet de literature41, pp. 221-224. Romano,I. B. 1994. "AHellenistic Deposit from Corinth:Evidence for InterimPeriodActivity," Hesperia63, pp. 57-104. Rose, C. B. 1993. "The 1992 PostBronzeAge Excavationsat Troia," StudiaTroica3, pp. 97-116. . 1994. "The 1993 Post-Bronze Age Excavationsat Troia,"Studia Troica4, pp. 75-104. Rostovtzeff,M. I. 1941. TheSocialand EconomicHistoryof theHellenistic World,3 vols., Oxford. Rotroff,S. I. 1988. "The Long-Petal Bowl from the Pithos Settling Basin,"Hesperia57, pp. 87-93. .2001. Rev.of PF 12, inAJA 105,pp.129-130.
MarkL. Lawall UNIVERSITY DEPARTMENT 220
DYSART
WINNIPEG,
OF MANITOBA OF CLASSICS ROAD MANITOBA
CANADA
[email protected]
R3T
2M8
L. LAWALL
Salzmann,D. 1995. "Zu den Mosaiken in den PalastenIV und V von Pergamon,"in Studienzum antiken Kleinasien3 (Asia Minor Studien 16), pp. 101-112. Schuchhardt,C. 1895. "Amphorenstempel,"in AvP VIII.2, pp. 423499. Schulte,E. 1963. ChronikderAusgrabungvon Pergamon,1871-1886. Aus BerichtenundBriefendesHumann-Kreises(Schriftender Hermann-Brockelschen-Stiftung2), Dortmund. Walbank,F. W. 1957.A Historical on PolybiusI, Oxford. Commentary .1967. A HistoricalCommentary on PolybiusII, Oxford. .1979. A HistoricalCommentary on PolybiusIII, Oxford. Williams, C. K., II. 1978. "Corinth, 1977: Forum Southwest,"Hesperia 47, pp. 1-39. .1979. "Corinth,1978: Forum Southwest,"Hesperia48, pp. 105144.
CO
R R I G
E N
DA
Wine Selling,andthe Reuseof Amphoras In MarkL. Lawall,"Graffiti, 69.1 (2000), the in the AthenianAgora,ca. 430 to 400 B.C.," Hesperia numberfor28 onpage35 shouldreadP 33416;for30 on Agorainventory page36, readP 33420;andfor87 on page59, readP 33421. qF
+
q!
q!
andProvisionsfor In Alan S. Henry,"TheAthenianStateSecretariat Hesperia 71.1 (2002), the citationin PublishingandErectingDecrees," of theSecretary themiddleof page104 fromIGII257, lines1-9, concerns andshouldbe movedto page99 the Council,not the PrytanySecretary, V-S-O, O-S-V, sequences: andinsertedas the firstof threealternative andS-V.Thus,on page99, read: we alsofindthe following sequences In additionto thesepredominating threesequences: grant,. . . 1) V-S-O, in a proxeny 2) O-S-V, so IGII243, . . . 3) S-V, so IGII276, . . .
CO
R R I G
E N
DA
Wine Selling,andthe Reuseof Amphoras In MarkL. Lawall,"Graffiti, 69.1 (2000), the in the AthenianAgora,ca. 430 to 400 B.C.," Hesperia numberfor28 onpage35 shouldreadP 33416;for30 on Agorainventory page36, readP 33420;andfor87 on page59, readP 33421. qF
+
q!
q!
andProvisionsfor In Alan S. Henry,"TheAthenianStateSecretariat Hesperia 71.1 (2002), the citationin PublishingandErectingDecrees," of theSecretary themiddleof page104 fromIGII257, lines1-9, concerns andshouldbe movedto page99 the Council,not the PrytanySecretary, V-S-O, O-S-V, sequences: andinsertedas the firstof threealternative andS-V.Thus,on page99, read: we alsofindthe following sequences In additionto thesepredominating threesequences: grant,. . . 1) V-S-O, in a proxeny 2) O-S-V, so IGII243, . . . 3) S-V, so IGII276, . . .
Back Matter Source: Hesperia: The Journal of the American School of Classical Studies at Athens, Vol. 71, No. 3 (Jul. - Sep., 2002) Published by: The American School of Classical Studies at Athens Stable URL: http://www.jstor.org/stable/3182032 . Accessed: 09/09/2011 06:01 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
The American School of Classical Studies at Athens is collaborating with JSTOR to digitize, preserve and extend access to Hesperia: The Journal of the American School of Classical Studies at Athens.
http://www.jstor.org
:
JOSEPH W. SHAW, ALEYDIS VAN DE MOORTEL, PETER M. DAY, VASSI LIS Kl LlKOGLOU
RecentlyPublished
A
LM
IA
Ceramic
Kiln
South-Central and
175pages,66 figures,16 tables HesperiaSupplement 30.
ISBN0-87661-530-2. July2001.Paper$35.00
in
Crete:
Pottery
Function
Production
An in-depthstudyof theLateMinoanIA cross-draft kilnfoundin excavationat Kommos,Crete.The kilnis of a typepopularduringthe Neopalatialperiod,andits goodstateof preservation hasallowedtheauthors to speculate aboutits originalinternallayoutanduseaswellasaboutthe roofthatcoveredit. Muchof thelargequantityof obviously locallyproducedpotteryfoundassociated withthekilnis analyzed in detail,allowing forthefirsttimethestudyof theshapes,decoration, andtechnical characteristicsof vasesknownto havebeenfiredin a specificLM IA kiln.The bookpresentsanintegrated program of analytical techniques usedto illustratethe rangeof firingtemperatures, the compositional similarities anddifferences in theclayused,andaspectsof thefiringprocessandthe upperkilnstructure. Offeredhereis avaluable contribution to ourknowledgeof thetechnology andorganization of ceramicproduction atthebeginningof theLateMinoanperiod,whichwillforma basisforstudiesof potteryprovenience andexchange.
--
be
.f
.o
*
w
b
CO
t.
#
" '- .# '; .
*.
.,..
a.
v <
s.
: .
o
< -''.''s,
,
-.
,,,
°
*---
. D°,
",'-,, ,, * :
'
s
'
.'
°
§
t
a
,",''d
'.
@
.
h
7
.
6
2
, * ,'-,
%
9
{
\
*
.--
,
.
vt -
.
; -
. ;:s < .
.
S
. .- S
-.
@
49
d
. -.
CHARLESK. Wl LLIAMSII AN D NANCY BOOKIDIS, EDITORS
Corinth,
the
Cen
ten
ary:
I89
6
I99
Forthcoming
6
Thisbookpublishes twenty-five of thepaperspresented attheDecember Ca. 500 pages,400 illustrations 1996symposium heldin Athensto celebrate theone-hundredth anniver- CorinthE saryof theAmericanSchoolof ClassicalStudiesexcavations at Ancient ISBN0-87661-020-8 2002. Corinth.The papersareintendedto illustrate therangein subjectmatter December Clothca. $75 / £57.50 of research currently beingundertaken by scholarsof AncientCorinth, andtheirinclusionin onevolumewillserveasa usefulreference workfor nonspecialists. Eachof the topics whichvarywidelyfromCorinthian geologyto religiouspracticesto Byzantinepottery is presented by the acknowledged expertin thatarea. Thebookincludesa fullgeneralbibliography of articlesandvolumes concerning material excavated at Corinth.As a summary of onehundred yearsof research, it willbe usefulto generations of scholarsto come.
HESPERIA SUPPLEMENTS 13 MarcusAurelius:Aspects of CivicandCultural Policyin theEast, byJamesH. Oliver(1970).S15.00 14 ThePoliticalOrganization ofAttica,byJohnS.Traill(1975).S15.00 16 A Sanctuary ofZeusonMountHymettos, byMerleK.Langdon(1976). S15.00
17 KalliasofSphettos andtheRevoltofAthensin 286 B.C., byT. LeslieShearJr.(1978).S15.00 19 StudiesinAtticEpigraphy, History,andTopography Presented toEugene Vanderpool (1982).S15.00 20 StudiesinAthenian Architecture, Sculpture, andTopography Presented to HomerA.Thompson (1982).S15.00 21 Enscavations at Pylosin Elis,byJohnE. Coleman(1986).S25.00 22 AtticGraveReliefsThatRepresent Women in theDressofIsis, by ElizabethJ.Walters(1988).S40.00 23 Hellenistic ReliefMolds promtheAthenian Agora,by ClaireveGrandjouan (1989).S25.00 24 ThePrepalatial Cemeteries at Mochlos andGournia andtheHouseTombs of BronzeAgeCrete,byJeffreyS. Soles(1992).S35.00 25 Debrisfrom a PublicDiningPlacein theAthenianAgora, by SusanI. RotroffandJohnH. Oakley(1992).S35.00 26 TheSanctuary ofAthenaNikeinAthens:Architectural StagesandChronology,byIraS.Mark(1993).S50.00 27 Proceedings oftheInternational Conference on GreekArchitectural Terracottas oftheClassical andHellenistic Periods, December 12-15, 1991, editedby NancyA. Winter(1994).S120.00 28 StudiesinArchaicCorinthian Vase Painting,by D. A. AmyxandPatricia Lawrence(1996).S65.00 29 TheAthenian Grain-TancLaw of374/3 B.C., by RonaldS. Stroud(1998). S35.00 30 A LMIA Ceramic Kilnin South-Central Crete: Function andPottery Production, byJosephW. Shaw,AleydisVande Moortel,PeterM. Day, andVassilisKilikoglou(2001).S35.00
[H THE OF
E S P E R I A JOURNAL
OF
CLASSICAL
THE
1l
AMERICAN S C H O O L
S T U D I E S AT
Vo LU M E 7I 2002
A
T HEN S
Submissions:Manuscriptsand communicationsshould be addressedto Dr.TraceyCullen,Editor,Hesperia,AmericanSchoolof ClassicalStudiesat NewJersey08540; tel.609-683-0800; Athens,6-8 CharltonStreet,Princeton, andphotocopiesof ilManuscripts fax609-924-0578; [email protected]. originalartworkandphotographs lustrations mustbe submittedin triplicate; aremadewith the Editor.A shouldnot be sent unlesspriorarrangements themajorconclusionsof the articleshouldalsobe shortabstractsummarizing reviewprocessandauthors included.Articlesaresubmittedto a double-blind withouttheirnameor accordingly, arerequestedto preparetheirmanuscripts preparation, notes,bibliograThe stylefor manuscript affiliationappearing. canbe foundin theGuidelinesfor on submissions phy,andotherinformation Authorspublishedin Hesperia62,1993, pp. i-xvi; on the School'swebsite at the aboveaddress. orbywritingto ASCSAPublications (www.ascsa.org);
CopyrightC)2002 The AmericanSchoolof ClassicalStudiesatAthens
TheAmericanSchoolof ClassicalStudiesatAthenswillnotknowinglyprint orinitialscholthe announcement in Hesperiaor anyof its otherpublications afterDecember30,1970, byanymeans of anyobjectacquired arlypresentation otherthanthroughan officiallysanctionedexcavationor survey,unlessthe existingcollectionorwaslegallyexportedfrom objectwaspartof a previously the countryof origin.
MA 020214529, U.S.A.
price The annualsubscription Hesperiais publishedquarterly. Subscriptions: X70forinstitutions(X80,internais X60forindividuals(X70,international), Paymentmust tional),andX33forstudents(proofof studentstatusrequired). beinU.S.dollars,drawnon aU.S.bankorbymoneyorder,andsenttoHesperia, SubscriberServices,P.O. Box 529, Canton,MA 02021-0529;tel. (U.S.) 800-821-7823; (outsideU.S.) 781-828-8450;fax 781-828-8915; e-mail of whenavailable) Singleissues(currentandbacknumbers [email protected]. for $15 eachpluspostagefromthe DavidBrownBook Hesperiaareavailable Company,P.O.Box 511, Oakville,CT 06779;tel. 800-791-9354,860-9459329;fax 860-945-9468;or (outsideNorthAmerica)OxbowBooks,Park End Place,OxfordOX1 1HN, U.K.;tel. +44 (0) 1865-241249;fax +44 (0) from 13 and19-30 arealsoavailable 1865-794449.IndexII andSupplements 1-12 and18,andearly DavidBrownBooks.Reprintsof IndexI, Supplements b.v.,P.O.Box issuesof Hesperiashouldbe orderedfromSwetsandZeitlinger, [email protected]. 10,2160 SZ Lisse,Netherlands;
Producedat EdwardsBrothers, Inc.,AnnArbor,Michigan. Designby EllenMcKie. postagepaidat Periodicals Princeton,NewJersey,andat additionalmailingoffices. Sendaddresschanges Postmaster: to Hesperia,P.O.Box529, Canton, ISSN 0018-098X
VOLUME 7I (2002)
CO
PUB LI CATI O N S STAFF
VIRGINIAR. ANDERSON-STOJANOVIC AND J. ELLIS JONES
EDITOR-IN-CHIEF KerriCox EDITOR,Hesperia TraceyCullen EDITOR,MONOGRAPHS Michael Fitzgerald PRODUCTION MANAGER SarahGeorge Figueira MANUSCRIPT EDITOR CarolA. Stein CREATIVE COORDINATOR JordanPeled
AncientBeehivesfromIsthmia
PUBLICATIONS C OMMITTEE Thomas G. Palaima(Chairman) Universityof Texas at Austin Darice Birge Loyola Universityof Chicago Carol C. Mattusch George Mason University JeremyMcInerney Universityof Pennsylvania MargaretMiller Universityof Toronto JeniferNeils Case Western ReserveUniversity James P. Sickinger Florida State University KathleenW. Slane Universityof Missouri-Columbia Rhys F.Townsend (exofficio) ClarkUniversity Stephen V. Tracy (exofficio) Ohio State University
NTE
NTS
345-376
NORADIMITROVA Inscriptions andIconography in theMonuments of the ThracianRider
209-229
BRICEL. ERICKSON AphratiandKatoSyme:Pottery,Continuity, andCultin LateArchaicandClassicalCrete
41-90
A. J. GRAHAM The Colonization of Samothrace
231-260
ALAN S . HENRY The AthenianStateSecretariat andProvisions for PublishingandErectingDecrees
91-118
JEFFREYM. HURWIT ReadingtheChigiVase
1-22
PETERKRENTZ FightingbytheRules:TheInventionof theHopliteAgon
23-39
MARKL. LAWALL EarlyExcavations atPergamon andtheChronology of RhodianAmphoraStamps
295-324
BY MARKL. LAWALL(WITHA CONTRIBUTION M. LYNCH, AUDREYJAWANDO),KATHLEEN AND SUSANI. ROTROFF JOHNK. PAPADOPOULOS, in the Stoaof Attalos NotesfromtheTins2: Research
415-433
MARYR. LEFKOWITZ "Predatory" Goddesses
325-344
YANNISA. LOLOS Quarry A PublicColumnDrumfroma Corinthian
201-207
JEANNETTE MARCHAND The Orneai A New BronzeAge Sitein theCorinthia: of StraboandHomer?
119-148
AND JOHNK. PAPADOPOULOS EVELYNLORDSMITHSONt Amphora: of a CycladicGeometric The CulturalBiography 149-199 of Metics Islanders in AthensandthePrehistory IRENEPOLINSKAYA Funerary MonumentfromAigina A NewInscribed
399-413
MARTHAC. TAYLOR OneHundredHeroesof Phyle?
377-397
GRETCHENUMHOLTZ in Greek Dedicatory Inscriptions Architraval Arrogance? Architecture of theClassicalPeriod
261-293
INDEX GREEKEPIGRAPHICAL
435-436
es_ I
I
THE
JOURNAL OF
O FC L A S S l C A
THE
SCHOOL
AMERICAN
L S TUDI
E
S AT
H
AT UNI I E 7 I: Vo
N
E3E-Et-19 0CTO
XZerZDsEZ
6b$SiaZ
2002
S
E N
UAI ECEEt
4
B E Et 200 EeC E-S1LI
ScbooZ
SttfzEs
of
At i]tbeJZs
2
The AmericanSchoolof ClassicalStudiesatAthensis a research and teachinginstitution dedicated to advanced studyof thearchaeology, art,history,philosophy, language, andliterature of Greeceandthe Greekworld.Establishedinl881byaconsortiumofnineAmerican universities, the Schoolnowservesgraduate studentsandscholars frommorethan150affiliated collegesanduniversities, actingasa baseforresearch andstudyin Greece.The mainbuildingsof the Schoolandits libraryarelocatedin Athens,withadministrative and publications officesin Princeton, NewJersey.As partof its mission, the Schooldirectsongoingexcavations in theAthenianAgoraandat CorinthandsponsorsallotherAmerican-led excavations andsurveys on Greeksoil. It is theofficiallinkbetweenAmericanarchaeologists andclassicists andtheArchaeological Serviceof the GreekMinistry of Cultureand,assuch,is dedicated to thewisemanagement of cultural resources andto thedissemination of knowledge of the classical world.Inquiries aboutmembership in the Schoolorparticipationin the SummerSessionsshouldbe sentto theAmerican Schoolof ClassicalStudiesatAthens,6-8 CharltonStreet, Princeton, NewJersey08540-5232. Hesperiais published quarterly bytheAmericanSchoolof Classical
StudiesatAthens.Foundedin 1932anddevotedprimarily to the timelypublication of reportson School-sponsored andSchooldirectedprojects, Hesperiawelcomessubmissions fromallscholars workingin thefieldsof Greekarchaeology, art,epigraphy, history, andliterature, fromearliestprehistoric timesonward. Hesperiais a refereed journal.
VOLUME 7I: NUMBER 4 OCTOBER-DECEMBER 2002
I |:--I
THEJOURNAL OFTHEAMERICAN SCHOOL OFCLASSICAL STUDIESATATHENS
PUB LI CATI ON S STAFF EDITOR-IN-CHIEF Kerri
325
Cox
E TDOIR,
Hesperia
Tracey
Cullen
EDITOR, MONOGRAPHS Michael
Fitzgerald
PRODUCTION MANAGER Sarah George Figueira MANUSCRIPT EDITOR Carol
MARY R. LEFKOWITZ "Predatory" Goddesses
A. Stein
VIRGINIA R. J.LELI S J ONEANDERSON-STOJANOVIC AND S Ancient BeehivesfromIsthmia 345
MARTHA C. TAYLOR One HundredHeroesof Phyle? 377
PUBLICATIONS COMMITTEE Thomas G. Palaima (Chairman) University of TexasatAustin Darice Birge Loyola Universityof Chicago Jeremy McInerney University of Pennsylvania Margaret Miller University of Toronto Jenifer Neils Case WesternReserveUniversity Kathleen W. Slane University of Rhys F. Missouri-Columbia Townsend bex ofYicio) Clark University
IRENE POLINSKAYA ANew Inscribed Funerary MonumentfromAigina MARK LAWALL(WITHA CONTRIBUTION BY AUDREY JAWANDO),KATHLEEN M. JOHN LYNCH, K. PAPADOPOULOS, AND SUSANI. ROTROFF Notes fromtheTins2: Research in the Stoaof Attalos
399
415
GREEK EPIGRAPHICAL INDEX 435
Submissions:Manuscriptsand communicationsshould be addressedto AmericanSchoolof ClassicalStudiesat Dr.TraceyCullen,Editor,Hesperia, NewJersey08540;tel.609-683-0800; Athens,6-8 CharltonStreet,Princeton, andphotocopiesof ilManuscripts fax609-924-0578;[email protected]. originalartworkandphotographs mustbe submittedin triplicate; lustrations aremadewith the editor.A shouldnot be sent unlesspriorarrangements of the articleshouldalsobe themajorconclusions summarizing shortabstract reviewprocessandauthors included.Articlesaresubmittedto a double-blind withouttheirnameor manuscripts accordingly, arerequestedto preparetheir notes,bibliograpreparation, The stylefor manuscript affiliationappearing. on submissions canbe foundin the Guidelinesfor phy,andotherinformation 62, 1993,pp. i-xvi;on the School'sWeb site Authorspublishedin Hesperia at the aboveaddress. orbywritingto ASCSAPublications (www.ascsa.org);
Copyright(C)2002 The AmericanSchoolof ClassicalStudiesat Athens
TheAmericanSchoolof ClassicalStudiesatAthenswillnotknowinglyprint orinitialscholthe announcement oranyof its otherpublications in Hesperia afterDecember30,1970,byanymeans of anyobjectacquired arlypresentation otherthanthroughan officiallysanctionedexcavationor survey,unlessthe existingcollectionorwaslegallyexportedfrom objectwaspartof a previously the countryof origin.
postagepaidat Periodicals Princeton,NewJersey,andat additionalmailingoffices.
Producedat EdwardsBrothers, Inc.,AnnArbor,Michigan. Designby EllenMcKie. Lekythosby the Coverillustration: Painter (Madrid11158). Oionokles Eos carryingoff Kephalos.CourMuseo tesyArchivofotografico, nacional,Madrid. arqueologico
Sendaddresschanges Postmaster: P.O.Box529, Canton, to Hesperia, MA 02021-0529,U.S.A.
price The annualsubscription Hesperia is publishedquarterly. Subscriptions: $70 forinstitutions($80,interna- ISSN 0018-098X is $60 forindividuals($70,international), Paymentmust tional),and$33forstudents(proofof studentstatusrequired). beinU.S.dollars,drawnon aU.S.bankorbymoneyorder,andsentto Hesperia, SubscriberServices,P.O. Box 529, Canton,MA 02021-0529;tel. (U.S.) 800-821-7823; (outsideU.S.) 781-828-8450;fax 781-828-8915; e-mail of Singleissues(currentandbacknumberswhenavailable) [email protected]. for $15 eachpluspostagefromthe DavidBrownBook Hesperia areavailable Company,P.O.Box 511, Oakville,CT 06779;tel. 800-791-9354,860-9459329;fax 860-945-9468;or (outsideNorthAmerica)OxbowBooks,Park End Place,OxfordOX1 1HN, U.K.;tel. +44 (0) 1865-241249;fax +44 (0) 13-17 and19-30 arealsoavailable 1865-794449.IndexII andSupplements fromDavidBrownBooks.Reprintsof IndexI, Supplements1-12 and 18, b.v., shouldbe orderedfromSwetsandZeitlinger, andearlyissuesof Hesperia [email protected]. P.O.Box 810,2160 SZ Lisse,Netherlands;
HESPERIA
71 (2002)
Pages 325-344
"PREDATORY" GODDESSES
ABSTRACT It is oftenassumedthatdepictionsonAtticvasesof thegoddessEoscarrying offyoungmortalsweremeantto conveya strongnegativemessageaboutthe Butcanwebe surethatthe mythsaboutEos and dangersof femalesexuality. her lovers,like those of abductionsof mortalsby other gods, were intended as commentarieson humansexualconduct?Goddesses (unlikewomen) areimmortal, ageless, and powerful. Evidence from the ancient sources suggests instead that depictions of abductions by Eos were meant to representboth the romance and anguish of divine interventions into mortal life, and to remind their users of the inexorablepower of the gods.
ABDUCTIONS OF MORTALS OTHER GODDESSES
BY EOS AND
The male gods of ancientGreeceareknownfor their interestin mortal women.1But goddessesalsowereactivein seekingout mortalconsorts,at leastin the erabeforetheTrojanWar,when the godswerestilldiningwith mortals([Hes.] Cat.,fr.1.1-10 MW; Hes. Theog.585-587). Hesiod'sTheogony,in the formin whichit hascomedownto us,endswith a catalogueof Zeus'sconsortsand children,followedby those of the other gods (Hes. Theog.886-962). To this catalogueis appendeda list of "theimmortal 1. Comparethe Hebrew tradition about the origin of the giants known as the Nephelim: "thegods saw that the daughtersof humanswere attractive, and they took wives from whomever they chose"(Genesis 6.2); West 1997, p. 117. My thanks especiallyto D. von Bothmer,M. Kilmer,and also to the refereesand editors of Hesperiafor specific improvementsand corrections; to E. R. Knauer,B. Ridgway,and
C. Sourvinou-Inwoodfor adviceand encouragement;and to C. Gentilesco for editorialassistance.Acknowledgment is made also to the following: Antikensammlung,StaatlicheMuseen zu Berlin, PreussischerKulturbesitz (Figs. 1, 14); BaltimoreMuseum of Art (Fig. 2); WaltersArt Museum, Baltimore (Fig. 3); Museum of Fine Arts, Boston (Fig. 4); MetropolitanMuseum of Art, New York(Fig. 5); Bowdoin College Museum of Art, Brunswick,
Maine (Fig. 6); Museo archeologico nazionaledi Ferrara(Fig. 7); Madrid, Museo arqueol6giconacional(Fig. 8); Hermitage Museum, St. Petersburg (Fig. 9); Musei Vaticani,Archivo fotografico(Fig. 10); Soprintendenza archeologicaper Etruriameridionale (Fig. 11); Musee du Louvre,Paris (Fig. 12); British Museum, London (Fig. 13). Translations,unless otherwise noted, are my own.
326
MARY
R. LEFKOWITZ
goddesses who bedded with mortal men and bore children who resembled the gods" (Hes. Theog. 967-968): Demeter and Iasion, Harmonia and Kadmos, Kallirhoe and Chrysaor, Eos and Tithonos, Eos and Kephalos, Medeia and Iason, Psamathe and Aiakos, Thetis and Peleus, Aphrodite and Anchises, Kirkeand Odysseus, and Kalypsoand Odysseus (Hes. Theog. 969-1020).2 When in the OdysseyKalypso complains that the gods are jealous of the goddesses who sleep with mortal men, she alludes to the story of Eos and Orion along with that of Demeter and Iasion (Od.5.118128). Homer also mentions Eos and Kleitos (Od. 15.572-575), and Sappho alludes to the story of Aphrodite and Adonis (fr. 140 Voigt).3The Hesiodic Catalogueof Women,although mostly about the liaisons of gods with mortal women, appearsto have included accounts of Thetis's marriageto Peleus (fr.210-211 MW) and of Selene'sromancewith Endymion (fr.245 MW).4 In view of the frequency with which these stories turn up in Archaic poetry,it is not surprisingthat the gods' liaisons with mortals arefrequently portrayedon vases by Athenian painters, or that Eos and her lovers appear on a relativelylarge proportion of them. In the vases catalogued by Sophia Kaempf-Dimitriadou there are more scenes depicting Eos and her lovers than scenes portraying Zeus, either with female mortals or with Ganymedes.5The myth of Eos and Kephalos had special appeal for an Athenian audience because Kephalos was a local boy; so did vases depicting the abduction of the Athenian princess Oreithyia by the wind-god Boreas.6 The Eos vases had a wide appeal for Athenian audiences, as well as for the Italian market:37 of 210 red-figuredvases listed by CarinaWeiss were found in mainland Italy or Sicily, and 38 are Nolan amphoras.7KaempfDimitriadou concluded that scenes in which gods abducted mortals might have served as reminders of the precariousnessof human existence.8 She also suggested that scenes in which Eos carried off young men might in some cases have served as consolation in time of death.9 In recent years, however,scholars have sought to extrapolateinformation from these scenes about Athenian attitudes toward human male and female sexuality.Andrew Stewart,in a detailed discussion of the Eos vases, rightly makes a careful distinction between ordinary rape of mortals by 2. On the traditionthat Medeia was immortal,cf. Pind. Pyth. 4.11; Braswell 1988, p. 76. The myth of Kirkemay have its origins in Near Easternepic: Ishtarin the Gilgamesh epic turnstwo of her loversinto animals(Gilg. 3.7, 6.2.26); see West 1997, p. 408, n. 14. The myths about Eos and her lovers may have influencedthe characterization of Kalypso,Kirke,and Aphrodite; cf. Boedeker 1974, pp. 64-84; Friedrich 1978,pp.39-43. 3. The myth of Aphrodite/Adonis came to Cyprusand Asia Minor from the cult of IshtarandThammuz in Syria;see Page 1955, p. 127. On the identity of Aphroditewith Ishtar,cf. Burkert1992, pp. 20, 97-98. The union
of Demeter/Iasion takes place in a plowed field, like the ritualmarriageof the goddess Inannato the king of Sumer.One king is even called Dumuzi, the name of her divine spouse;cf. Frymer-Kensky1992, pp. 50-57. 4. On Selene and Endymion,see Ap. Rhod. 4.57; the story may come from Musaeus'sTheogony(cf. Epimenides 3B2 I 33 DK; see FGrHist IIIb Suppl.I, p. 575). Another of Selene'smortalloverswas the Eleusinian rulerand hierophantEumolpos, by her the fatherof Musaeus (Philochorus 328, FGrHistF208 = schol. Ar. Ran. 1033). 5. In the catalogueof vases in Kaempf-Dimitriadou1969, pp. 76-
109, there are 147 sceneswith Eos and her loversas opposed to 116 depicting Zeus and his lovers (60 with women and 56 with Ganymedes).Boreasand Oreithyiais the next most popular subjectwith 56, followed by Poseidon and his loverswith 40. 6. Both myths were portrayedin the akroteriaof the Athenians'temple at Delos; Robertson1975, p. 356. 7. LIMC III, 1986, pp. 759-775, s.v. Eros (C. Weiss). 8. Kaempf-Dimitriadou1969, pp. 47-53. 9. Kaempf-Dimitriadou1969, p. 57; Vermeule1979, pp. 162-178; LIMC III, 1986, p. 779, s.v. Eos (C. Weiss).
"PREDATORY)
GODDESSES
327
mortalsand the eroticinterventionsinto humanlife madeby gods.?1But he accountsfor the popularityof the Eos vasesby suggestingthat they might have been understoodas a means of defining and justifyingthe dominanceof malesexuality:"Mythologicalpursuitsand abductionsrepresentnothing moreor less than the projectionof Athenianmale desire firstupon the heroicworldandthen upon the divineone."11 of Athenianmale What roledo the Eos vasesplayin this "projection desire"?Stewartsuggeststhatthe discourseaboutEos andherboyvictims mayultimatelybe intendedas a warning:"notonly do these pictureshint and easycapitulationto at the evils offemaledominance(gynaikokrateia) desire,but nervouslyevoketheir appallingconsequences:femalecontrol RobinOsborneagreeswith Stewartthatfemalepursuers of the phallus."12 In his view,the vasesprovide invertthe normalcodesof sexualbehavior.l3 a displayof what nevercould or shouldhappenin Atheniansociety,becauseof the closerestrictionson femaledesire,at leastamongrespectable women:"in limiting scenesof femalepursuitto the case of the winged Eos, pot painterswere able both to suggestthat women did desiremen andthat femaledesirecouldnot be activein the realworld."14 In the presentarticleI arguethat Kaempf-Dimitriadou's understandthe is more of of the Eos vases to be correct: the primary ing likely purpose purposeof the vaseswas to remindviewersof the powerof the gods to alterthe courseof humanlife, whetherfor betteror worse.Painterswho wishedto commenton the sexualityof mortalfemalesdid not needto use depictionsof goddessesto do so. In anycase,thereis no reasonto believe thatAthenianmen wishedto discouragefemalepassion,evenwithin the contextof marriage.Rather,what the Eos vases portrayare the various stagesin the usesof persuasionandconstraintin the pursuitof the goddess's desire.None of the Eos vasesdepictssexualactivity,other than eye conin her approachesto mortalsthan gods tact. Eos is no more"predatory" like Zeus or Boreas.15
Why do we supposethat vases depictingabductionsof mortalsby Cergodswereintendedas commentarieson mortalsexualrelationships? human their own characteristics onto the tainly beings project gods, as observed: "Homer and the Hesiod ascribed to Xenophanes gods all the actionsthat among men causereproachand blame,lies, adulteries,and deceptionsof one another"(fr.21 B12 I 132 DK = 160 KRS).But it is anothermatterto assumethat the reverseis true,that all of the actionsof the godscanin turnbe mirroredin the livesof humanbeings.16 The Greeks neverforgotaboutthe limitationsimposedon humanactionby the fact of mortality.7They made a cleardistinctionbetweenhumanrapesand 10. Stewart1995, pp. 74-77; cf. Zeitlin 1986, p. 150, quoting Dover 1978, p. 88: "pursuitis the role prescribedfor the male, flight for the female,"but these categoriesapply only to mortals. 11. Stewart 1995, p. 86; cf. also Zeitlin 1986, p. 150. 12. Stewart 1995, p. 86.
13. Osborne 1996, pp. 65-77. 14. Osborne 1996, pp. 67-68, 76; cf. also Frontisi-Ducroux1996, p. 83: "aterrifyingscenarioto Greek men." 15. West 1966, p. 486: "[Eos] was one of the most predatoryof goddesses";cf. Friedrich1978, p. 41: "Dawnis rapaciousin her way." 16. E.g., Osborne 1996, p. 72: "just
as behind every man there is a satyr restrainedonly by social protocols,so behind everywoman there is an Eos: Female sexualdesirecan be rampant, too." 17. See especiallythe perceptive remarksof Vermeule1979, p. 121; cf. also Stehle 1990, p. 94; SourvinouInwood 1991, p. 49.
328
MARY
R. LEFKOWITZ
abductions by gods.18Abduction by a god was something to boast about, even after the woman's death; but rape by a mortal was a disgrace,punishable in Athens by divorce and other social restrictions. Gods cannot serve as role models for mortal men, because gods can abduct a woman with impunity.It is essential to rememberthis important distinction when talking about divine abductions.19Gods do not (paceEva Keuls) go on "raping expeditions,"but choose their female consorts deliberately and carefully.20 A similardistinction applies also to goddesses who abduct mortal men; a mortal woman cannot carryout with impunity either abduction or adultery. Klytemnestrais killed by her son for murderinghis father because she wanted to marry Aigisthos. Phaidra (who was only thinking about committing adultery) hangs herself when she believes that she is going to be disgraced. In his discussion of these relationships, Osborne fails to distinguish between goddesses and mortal women. He includes Phaidra along with the goddesses Eos, Aphrodite, and Selene in a list of "females in pursuit."21But Phaidra is mortal; she cannot carry Hippolytos away by persuasion,let alone by force. Her passion cannot do anything positive for him, such as guaranteeing that he will have a famous son or that he could become immortal. Only a goddess could bestow some or all of these benefits on a mortal man. A god could assume the form of an animal to abduct a mortal woman. But when a mortal woman mates with an animal, her lust has dreadful consequences: Pasiphae's passion for the bull produced the Minotaur.22 Our present-day sensitivity to the treatment of women has made it difficult to discuss with equanimity ancient myths of divine abductions of mortals, even when we can separate them in our minds from ordinary rape.The ancients themselves, however,responded to these stories in more nuanced ways.23When depicting divine abductions, artists avoided representing the moment of sexual union or direct manifestation of sexual arousal.24Like the poets, they concentrated on the process of seduction, showing how the gods persuade their mortal lovers to welcome and cooperate with their advances.As Kheiron advises Apollo, when he is planning to abduct Kyrene:"hidden are the keys of wise Persuasion of holy loves; this makes gods and men alike hesitant openly to approacha sweet bed for the first time" (Pind. Pyth. 9.39-41).25 Male gods always employ persuasion (or enchantment) to gain the cooperation of the females, even though they could easily compel mortal women by force. But a goddess cannot seduce a mortal man without his active cooperation. In the HomericHymn to Aphrodite,the goddess disguises herself as a mortal and tells an elaborate story about herself in order to reassure Anchises and encourage him to make love to her; the poet states that her power is universal, and shows the effect her presence has even on animals.26But the male gods, who could easily force mortal boys or women to give into them, prefer to draw their lovers by the power of their glance. Homosexual coupling is shown only in a few cases where a winged young male god has frontal intercruralintercourse with a young human male. But even here, as Martin Kilmer has shown, "thereis no hint of violence."The mortal seems neither to consent nor object. His passivity brings out the disparity in power between himself and the god.27
18. For furtherdiscussion,see Lefkowitz 1993, pp. 20-21; cf. also Harrison1997, p. 188. 19. Cf. Keuls 1985, p. 50, who speaksof Zeus and Hermes going on a "rapingexpedition." 20. Lefkowitz 1986, pp. 35-36. 21. Osborne 1996, p. 67. 22. Robson 1997, p. 81, who also cites the myth of Artemis'scompanion Polyphonte(Ant. Lib. 21), who fell in love with a bear:her two huge sons became savagecannibals. 23. Cf. esp. Ridgway1999, pp. 9394. 24. Shapiro1992, p. 61; Calame 1999, p. 67. Even Danae'slap is coveredby her cloak and dresswhen she receivesthe showerof gold from Zeus, as shown on a kalyxkrater from Cerveteriin St. Petersburg, inv. ST 1723:ARV2 360, no. 1, 1648; LIMC III, 1986, p. 327, no. 1, pl. 1, s.v.Danae (J.-J.Maffre).Nonetheless, Stewart(1995, p. 87) suggeststhat by mid-century"inpolite society, these imageswere simply no longer acceptable;in effect they had all but become pornography." 25. Clay 1989, p. 159: "notforce but persuasionand guile constitute the secrettools of successfulseduction." Dougherty (1993, pp. 140-141; 1998, p. 270), in her discussionof Apollo's abductionof Kyrene,emphasizesthe "violence"of Pindar'sagricultural metaphors,but cf. Bremmer2000, pp. 102-103. 26. Cf. Clay 1989, pp. 158, 175. 27. Shapiro1992, p. 66; Kilmer 1997, p. 132; and in general,Kilmer 1990, p. 270; 1993, p. 200.
"PREDATORY"
GODDESSES
329
Figure1. Neck amphoraby the NausicaaPainter(BerlinF 2352). Eos approachingKephalos.Courtesy Museenzu Staatliche Antikensammlung, Kulturbesitz Berlin,Preussischer
VASE PAINTINGS
OF PURSUITS
BY EOS
In general, the goddess Eos's approaches to the young men she seeks to abduct do not differ significantly from those of male gods abducting boys or women. Painters show her (1) approaching the mortal, (2) catching hold of him by his hand or arm, or (3) carryinghim away.An example of the first stage is provided by a neck amphorawith twisted handles by the Nausicaa Painter (Fig. 1), which shows on the left a winged Eos walking with arms outstretched toward a young man, whose hunting garb identifies him as Kephalos.28He is walking awayfrom her, holding his right arm with his palm facing toward her, as if warding her off. His dog jumps toward her in alarm. But at the same time Kephalos has turned his head back in orderto look at Eos, and we can see that the power of the goddess's glance has begun to keep him from getting away.29 In a bell kraterby the Christie Painter (Fig. 2), a winged Eos strides from the left towarda retreatingKephalos (again recognizableby his hunting gear). He is about a head shorter than the goddess, and is gazing back 28. Berlin F 2352, from Nola:ARV2 1107, no. 1; LIMC III, 1986, p. 762, no. 104, s.v. Eos (C. Weiss); KaempfDimitriadou 1969, p. 85, no. 104. 29. On the powerof the gaze, see (in general)Robertson1975, p. 214;
Buxton 1982, pp. 51,112-113,214, n. 86; Sourvinou-Inwood1991, p. 69; Padel 1992, pp. 62-63. On the erotic gaze, see Reeder 1995, pp. 125-126; Frontisi-Ducroux1996, pp. 82-84; Sutton 1997-1998, p. 35.
330
MARY
R. LEFKOWITZ
Figure2 (left).Bell kraterby the ChristiePainter(BaltimoreMuseum of Art 1951.486).Eos approaching Museumof Kephalos. CourtesyBaltimore Art,bequestof SadieA. May
Figure3 (above).Stamnos(Walters Art Museum48.2034).Eos aptoward her, while a comrade behind the goddess moves off to the left.30 Hunters areout at dawn, and schoolboys also rise early.When Eos is shown in pursuit of the Trojan prince Tithonos, he is often shown holding a lyre, as if on his way to his lessons. On a skyphos by the Pantoxena Painter, a winged Eos reaches out to a young man identified by an inscription as Tithonos.31 His right arm is bent up and back, and in his right hand he holds a lyre with which he tries to strike Eos; but again their eyes have already met, and we know that his resistance will be ineffective. When painters depict mortal couples gazing at one another, they often show a winged Eros between or near them.32But the gaze of the god or goddess can be effective even without the help of Eros. In the next stage of the pursuit Eos takes hold of the young man; on a stamnos in the Walters Art Museum (Fig. 3), Eos, here without wings, is shown between two young men, striding from the left toward Tithonos, who is moving away from her toward the right.33She has placed her left 30. Bell kraterby the Christie Painter,BaltimoreMuseum of Art 1951.486:ARV2 1048, no. 27; LIMC III, 1986, p. 762, no. 99, s.v. Eos (C. Weiss); Kaempf-Dimitriadou1969, p. 85, no. 110; Reeder 1995, pp. 401402, no. 131 (C. Benson). Comparethe bell kraterfrom Cumae (Paris,Cab. Med. 423: ARV2 1055, no. 72; LIMC III, 1986, p. 762, no. 100, s.v. Eos [C. Weiss]; Kaempf-Dimitriadou1969, p. 185, no. 112), where a winged Eos,
approachingfrom the left, reachesout towarda fleeing Kephalos(identified by an inscription)who holds two spears over his left shoulder;a companion, behind Eos, is runningaway. 31. Paris,Cab. Med. 846:ARV2 1050, no. 1; LIMC III, 1986, p. 768, no. 182, s.v. Eos (C. Weiss); KaempfDimitriadou1969, p. 90, no. 173. 32. For examples,see esp. Sutton 1992, p. 27; Sutton 1997-1998, pp. 3239; also the scene on the fragmentof a
Walters proaching Tithonos. Courtesy ArtMuseum,Baltimore
loutrophorosin the Ashmolean Museum, OxfordUniversity,inv. 1966.888: LIMC III, 1986, p. 905, no. 639c, pl. 646, s.v. Eros (A. Hermaryet al.); Reeder 1995, pp. 168169, no. 25. 33. Baltimore,WaltersArt Museum 48.2034: ARV2509, no. 1657; LIMC III, 1986, p. 765, no. 139, s.v. Eos (C. Weiss); Kaempf-Dimitriadou 1969, p. 87, no. 130; Reeder1995, pp. 399-401, no. 130 (C. Benson).
"PREDATORY
GODDESSES
33I
Figure4. Kylixby the Telephos Painter(Boston95.28). Eos taking hold of Tithonos. CourtesyMuseumof FineArts,Boston,Catharine PagePerkins Fund
arm acrosshis back and her left hand rests on his right shoulder.Tithonos has dropped his lyre from his right hand and it falls to the ground. Behind Eos, on the left, a comrade is looking back toward Tithonos, but at the same time moving out of the way.34 On a kylix by the Telephos Painter in Boston (Fig. 4), a winged Eos approachesTithonos from the left, while he moves away from her toward the right.35She grasps his right wrist with her right hand, has placed her left arm around his back, and has her left hand on his left shoulder. Tithonos's left arm is bent toward her at the elbow, with his hand outstretched in her direction, but not touching her. He turns around to look at her,but she is gazing upward,over his head. She does not need to attract him by her gaze, because she is holding him by the wrist, xs'p'Er xoCprTc, as a bridegroom might grasp the wrist of a bride.36 34. CompareEos catching Kephalos on a Nolan amphoraby the Niobid Painterin the Rijksmuseumin Leiden (PC 78:ARV2 605, no. 58; LIMC III, 1986, p. 761, no. 77, pl. 77, s.v. Eos [C. Weiss]; Kaempf-Dimitriadou1969, p. 83, no. 83), and on a pelike by the Niobid Painter,formerlyin K6nigsberg (Univ.F 162:ARV2 603, no. 44; LIMC III, 1986, p. 761, no. 76, pl. 76,
s.v. Eos [C. Weiss]; Kaempf-Dimitriadou1969, p. 83, no. 81). 35. Boston 95.28, from Vulci: ARV2482, no. 816; LIMC III, 1986, p. 769, no. 201, pl. 201, s.v.Eos (C. Weiss); Kaempf-Dimitriadou 1969, p. 90, no. 179. 36. On the gesture,see Jenkins 1983, pp. 139-140; Oakley and Sinos 1993,p.32.
332
MARY
R. LEFKOWITZ
Figure 5 (left). Polychrome bobbin by the Penthesilea Painter (New York 28.167). Zephyros taking hold of Hyakinthos. CourtesyMetropolitan Museumof Art,FletcherFund Figure 6 (opposite,left). Hydria by the Niobid Painter. Boreas taking hold of Oreithyia (Brunswick 1908.003). CourtesyBowdoinCollegeMuseumof Art, Maine.Gift of EdwardPerry Brunswick, Warren, Degree,1926. Esq.,Honorary
ANALOGIES WITH ABDUCTIONS BY MALE GODS In approachingher lovers in this way Eos follows what appearsto be one of the customary rules of divine behavior.The tactics she employs are also used by male gods. Virtually the same stage of pursuit is shown on a polychrome bobbin by the Penthesilea Painter in New York (Fig. 5).37Here a winged male figure,possibly Zephyros, stridesfrom the left towarda young male (Hyakinthos?) who holds a lyre in his left hand. The god holds his left arm behind the young man and is reaching down toward him with his right hand. The young man is about to step off to the right, but he gazes back into the eyes of the god. The stance of the god/mortal pair is similar to that seen on a hydria by the Niobid Painter in the Bowdoin College Museum of Art: a winged Boreas with winged boots approachesfrom the left through the air (Fig. 6).38The god puts his right arm aroundOreithyia's waist and his left around her shoulders. Oreithyia turns toward him, but does not embrace him in return. She holds both arms in the air and is still moving toward the right, away from him. The goddess Athena, holding spear and shield, stands behind Oreithyia, looking over her head toward Boreas. She does nothing to help Oreithyia or oppose Boreas;rather,she is acting as Boreas'spompos: according to Herodotos (7.189.1), he made Oreithyia his wife.39 In the vases discussed above, the gods do not use force to compel their intended lovers to accompany them. But ancient Greek painters did not hesitate to show how women protest and suffer when they are carriedoff, even by a god. The exquisite wall painting in the tomb of Persephone at Verginadepicts Persephone'sanguish as Hades carriesher off on his chariot, Hades may have needed while her Oceanid friend raisesher armin horror.40
37. New York,MMA 28.167: ARV2 890, no. 175,1673; LIMC V, 1990, p. 549, no. 44, pl. 44, s.v. Hyakinthos(L. and F. Villard); Kaempf-Dimitriadou1969, p. 81, no. 53; Shapiro1992, pp. 70-72, fig. 3.10. 38. Brunswick,Maine 1908.003, from Athens:ARV2 606, no. 68; LIMC III, 1986, p. 137, no. 56, pl. 56, s.v.Boreas(S. Kaempf-Dimitriadou); Kaempf-Dimitriadou1969, p. 107, no. 364. 39. On Athena aspompos,see Sinos 1993, pp. 80-83. On Herodotos 7.189, cf. How andWells 1912, vol. 2, p. 215; LIMC III, 1986, p. 134, s.v.Boreas (S. Kaempf-Dimitriadou).There is a rationalizedversionof the myth in Plato'sPhaedrus(229c-d). 40. LIMC VIII, 1997, p. 969, no. 213, s.v.Persephone(G. Gtintner); Andronikos1994, pp. 59-69, pls. 1926. Cf. Cohen 1996, pp. 119-121.
"PREDATORY
GODDESSES
333
Figure 7 (above,right). Kylix by the
PenthesileaPainter(Ferrara9351). Zeus takinghold of Ganymedes.
CourtesyMuseo archeologiconazionaledi Ferrara
41. The narcissusin this myth is a "miraculousflower";see Richardson 1974, p. 144. 42. Athens 13119, from Athens: ARV2 1656; LIMC III, 1986, p. 135, no. 9, pl. 9, s.v.Boreas(S. KaempfDimitriadou);Kaempf-Dimitriadou 1969, p. 105, no.343. 43. Ferrara9351, from Spina: ARV2 880, no. 12; LIMC IV, 1988, p. 157, no. 44, pl. 44, s.v. Ganymedes (H. Sichtermann);Kaempf-Dimitriadou1969, p. 79, no. 34. For similar scenes, cf. a lekythos from Novoli in Taranto,Mus. Naz. 54383: ARV2 556, no. 108; LIMC IV, 1988, p. 156, no. 30, s.v. Ganymedes(H. Sichtermann);Kaempf-Dimitriadou1969, p. 77, no. 8; and an oinochoe (with Zeus approachingGanymedesfrom the right) in Basel, H. A. Cahn 9: ARV2 874, no. 3; LIMC IV, 1988, p. 157, no. 41, pl. 41, s.v. Ganymedes (H. Sichtermann);Kaempf-Dimitriadou1969, p. 79, no. 31.
to use both deception and violence to abduct Persephone,because she, like himself, is a divinity, powerful, immortal, and ageless. As the story is related in the HomericHymn to Demeter,Zeus tells Gaia to produce an intoxicating narcissus,which Persephone wants to pick; Hades then takes her by surprise,by making the earth open beneath her feet, as he rushes out to carryher off on his chariot (Hymn.Hom. 2.8-18).41 But when gods approach mortals they do not need to rely on such violence. Vase painters show male gods employing their superiorstrength, and allow the mortals whom they have chosen to display surpriseand initial reluctance.In a scene on a hydria in Athens, Boreas runs from the left after Oreithyia,who is fleeing to the right with her left hand raised in the air. But Boreas has caught her right hand by the wrist in his right hand and is pulling her from behind with his left hand, and she has turned her head around to look into his eyes.42In a kylix by the Penthesilea Painter (Fig. 7), Zeus has put down his thunderbolt and scepter and is using both his arms to pull Ganymedes toward him.43His left foot is braced against a rock on the lower right side of the vase, and with his right hand he grasps Ganymedes' right arm; Ganymedes' head is turned toward him, but he is walking away,as if he wanted to returnto his friends, holding the fighting cock Zeus has given him in his left hand. This painting appears on the cover of Keuls's TheReign of the Phallus, as if it were representativeof the sexual violence endemic in Athenian society. But the vase does not describe sexual life in Athens. If a human male behaved that way toward a young man, he would be harshly treated, if not by the young man's relatives, then by the young man himself. Rather, the painting depicts the conflict between a god's desire and a mortal'swish to retain his independence. The onlooker knows whose power is greater,and understands that the gods alwayswin.
334
MARY
R. LEFKOWITZ
Figure8 (left).Lekythosby the OionoklesPainter(Madrid11158). Eos carryingoff Kephalos.Courtesy
Archivofotografico, Museoarqueol6gico Madrid nacional,
VASE PAINTINGS
OF ABDUCTIONS
BY EOS
Figure9 (above).Bird-rhyton (St. PetersburgB 682). Eos carrying off a boy. CourtesyHermitage Museum
The final phase of the pursuit is the abduction itself. In a lekythos by the Oionokles Painter in Madrid (Fig. 8), a winged Eos flies holding a young male (identified in an inscription as Kephalos)who is cradledin her arms.44 The goddess's torso is turned toward the boy, but she looks up to the left, while the boy looks down to the right, away from her. His right arm is extended in front of her face, and his left hand still clutches his lyre. The composition brings out the tension between the goddess's desire to abduct him and his eagerness to rejoin his comrades.There is also a vivid contrast between Eos and the boy she is abducting in a bird-rhyton in St. Petersburg (Fig. 9).45Eos again is looking away from the boy toward the left, while he looks down to the right, and stretches his left arm downwardas if reaching out to someone on the ground. Vermeule'scaption for this vase is "Eos the dawn-goddess carries a boy to a better world."46But life in the palace of the Dawn would only be better for a boy if he could somehow be ageless and immortal. In the myths only a few mortals achieve this 44. Madrid 11158:ARV2649, no. 45; LIMC III, 1986, p. 773, no. 268, pl. 268, s.v. Eos (C. Weiss); Kaempf-Dimitriadou1969, p. 92, no. 194.
45. St. PetersburgB 682: ARV2 391, a (Kephalos);LIMC III, 1986,
46. Vermeule1979, p. 166. If the paintinghad some relevanceto death,
Kaempf-Dimitriadou1969, p. 91, no. 193; Vermeule1979, p. 167, fig. 17.
thos ratherthan a rhyton;cf. KaempfDimitriadou1969, p. 57.
p. 773,no.267, s.v.Eos (C.Weiss);
onewouldexpectto findit on a leky-
"PREDATORY"
GODDESSES
335
Figure10. Etruscanbronzeand silvermirror(Vatican12241). Thesan (Eos) carryingoff Kephalos. Musei CourtesyArchiviofotografico, Vaticani
status.47There are no inscriptions or literary documents that suggest that Athenians in the 5th century attached a happy ending to any of the Eos myths. They buried the dead just before dawn, for practical reasons, not because they supposed the goddess would be there to carrythe soul away.48 For them a benevolent winged Charon representeda good and easy death.49 At best, scenes of Eos on funerarylekythoi might have conveyed to mourners a vague hope that the deceased, like the heroes of old, might find himself in a brighter place in the lower world.50It is only in later antiquity that abductionsby goddesseswere used to suggest a mode of death and a promise of a future existence in the light, nearerto the gods.s5 Other scenes depict the boy sitting more comfortably in Eos's arms. An Etruscan bronze and silver mirrorfrom Vulci in the Vatican (Fig. 10) shows Kephaloslooking directlyinto the eyes of the goddess Thesan (Eos's 47. Zeus makes Ganymedes immortal,and also his sons Herakles and Polydeukes,who generouslyshares his immortalitywith his half-brother Kastor.Aphrodite makes Kephalos's son Phaethon a "brightdivinity" (68aliovca8ov, Hes. Theog.991), but this is a statusmore like that of a hero, becausehe is confined to one place as "thesecretkeeperof her temple"(Nagy 1979, p. 192); no sexualrelationshipis implied (Stehle 1990, p. 96). 48. Kurtzand Boardman1971, pp. 144-145; cf. LIMC III, 1986, p. 779, s.v. Eros (C. Weiss). Vermeule (1979, p. 163) cites only Heraclitus
(below,n. 51) in supportof her claim that "Eos the Dawn goddess carried off the dead on 'the wings of the morning' or to motivatethe event by simple sexualattractionor love."The notion of flying to God on the wings of the morningderivesfrom Ps. 139.9-10 (KingJamesVersion). 49. Sourvinou-Inwood1995, pp. 337-346; cf. Garland1985, pp. 5459. In Etruriain the first half of the 5th century,mirrorsdepictingthe goddessThesan (Eos) carryingher dead son Memnon were popular.As De Puma (1994, p. 187) suggests,"she, like her mortaldevotees,has had to
bow to a power strongerthan the king of the gods." 50. LIMC III, 1986, p. 779, s.v. Eos (C. Weiss). 51. As Athenaeushas Plutarchsay at his learnedsymposium:"Whom do the goddessesabduct?[&avapTr6ooouv; on the translation,cf. Bremmer2000, p. 103] Is it not the most beautiful ones?And these are the ones they live with: Eos with Kephalosand Kleitos andTithonos, Demeter with Iasion, Aphroditewith Anchises and Adonis" (Symp.566d). See also Heraclitus, Allegoriae(Quaest.Hom.) 68; KaempfDimitriadou1969, p. 57.
336
MARY
R. LEFKOWITZ
Etruscan counterpart) as she carrieshim away.52She is moving to the left, holding the smaller figure of Kephalos lightly in her arms. Her left arm is beneath his shoulders,and her right hand supportshis knees. He has placed his right hand on her right shoulder,and his left arm hangs straight down; he is not looking back toward his friends or his hunting. Similarly, in a neck amphora by the Achilles Painter from Vulci, the young man has his arm around Eos's back, and looks to the left along with her.53
METHODS OF PERSUASION ABDUCTORS
USED BY DIVINE
We now need to ask what the gods do to make abducted lovers compliant, or even actively cooperative.On a kylix by Douris in Paris Ganymedes lies back comfortably in Zeus's arms.54Zeus is able to hold his scepter in his left hand, because he does not need to restrain the boy who, resting comfortably,faces him. In the terracottastatue group at Olympia, Ganymedes calmly holds a cock with his left hand, his present from Zeus, while Zeus carrieshim under his right arm.55As we have seen (above, note 41), when Hades wanted to put Persephone off her guard, Gaia grew a narcissusthat astounded both gods and men, and its sweet scent caused heaven, earth, and sea to smile (Hymn.Horn.2.8-14). When in the Hesiodic Catalogueof WomenEurope was gathering flowers, Zeus changed himself into a bull and "breathedfrom his mouth the scent of saffron [ocro ToOaoT6[acroo xpoxov 7ivetv].After enticing her he picked her up, carried her over the sea to Crete, and had intercourse with her" (fr. 140 MW). When in a fragment of Aeschylus's Karesor Europe,Europe gives a brief account of what happened, she has no unpleasant memories: "and a friendly lush meadow was there for the bull; in that way, by waiting there, Zeus managed to take me, an effortlesstheft, from my aged father"(fr.99.1-3 Radt).56 Scenes in 6th- and early-5th-century vase paintings also show a contented and relaxedEurope. She is most often representedon the bull'sback, riding away acrossthe sea to Crete. In these scenes she often holds the bull'shorn or neck as she sits on him, and places her other hand on the bull'sback, to steady herself, as the bull rushes along.57Some painters show her turning around to call back to her friends, as she does in the 2nd-century B.C. poem of Moschus, which drawson these earlierversions (lines 111-112).58 She is not only completely unafraid of the bull, but fascinated by him, 52. Vatican12241: LIMC III, 1986, p. 795, no. 30, pl. 30, s.v. Eos/Thesan (R. Bloch); De Puma 1994, pp. 181182, pl. 16.7. 53. Once Rhodes Mus.:ARV2 987, no. 5; LIMC III, 1986, p. 773, no. 269, pl. 269, s.v. Eos (C. Weiss); KaempfDimitriadou 1969, p. 92, no. 199. 54. ParisG 123:ARV2 435, no. 94; LIMC IV, 1988, p. 157, no. 52, pl. 52, s.v. Ganymedes(H. Sichtermann);
Kaempf-Dimitriadou1969, p. 79, no. 39. 55. LIMC IV, 1988, p. 157, no. 56, pl. 66, s.v.Ganymedes (H. Sichtermann);Kaempf-Dimitriadou 1969, p. 79, no. 40; Dover 1978, p. 92, pl. 1; cf. Robertson1975, p. 277, pl. 60: "the Greeks saw honour,not shame,in the youth'shaving a god for a lover." 56. On the fragment,see Lloyd-
Jones 1971, pp. 414-417; Lefkowitz 1993, p. 25. 57. Cf. LIMC IV, 1988, pp. 78-80, nos. 22-75, s.v. Europe (M. Robertson), for a list of red-figurevases showing Europe riding:"bullgenerallygalloping, often over sea." 58. Europe'sstorywas told in poems (now lost) by Eumelus, Stesichorus,and Bacchylides;see Campbell 1991, p. 3; Buhler 1968, pp. 25-26.
"PREDATORY"
GODDESSES
337
Figure11. Bell kraterby the Berlin Painter(TarquiniaRC 7456). Europerunningalongsideof Zeus. archeologica, CourtesySoprintendenza Ministeroperi Benie le attivitaculturali per l'Etruria meridionale
and eager to touch and embrace him.59She does not hesitate to climb on his back.60 Europe is shown contentedly running alongside the bull in a bell krater by the Berlin Painter (Fig. 11).61Her feet are spreadwide apart,and with her left hand she holds on to the tip of one of his horns. The other side of the vase shows a girl running. Europe and the bull have not yet reached the sea, since the folds of her dress fall straight down and do not billow out in the wind; her friend is still running after her. Kilmer observes that "the 'rape'of Europe is atypical in showing contact between god and mortal; but the contact is never overtly sexual and Zeus is shown as a bull... what we have here is pursuit, rather than the abduction that will come from it."62By contrast, Keuls emphasizes the potential sexuality of this scene by stating that Europe is grasping the bull by his "phallus-horn."63 Perhaps that is possible, but there is a simpler explanation. By having Europe hold the bull's horn and keep pace with him, the painter shows that the bull is tame and approachable. 59. In Moschus'spoem Europeis readyfor marriage,and respondswithout hesitationto the bull'sapproaches. Similarly,in 4th-centuryvase paintings, Danae is shown looking upward,with her breastsbare,collecting the golden rainin her lap, as on a bell kraterfrom Boiotia in the Louvre(ParisCA 925: CVA,Paris 17 [France46], pls. 44 [1167]:1;45 [1168]:3;LIMC III, 1986, p. 328, no. 9, pl. 9, p. 335, s.v. Danae [J.-J.Maffre]). 60. To speakof this abductionas "bestialrape"seems extreme;cf. Robson 1997, pp. 74, 77. There is no sug-
gestion that Zeus remainedin the form of a bull afterhe took Europeto Crete. Even in Moschus'spoem, where the bull is extraordinarilyaffectionate,they leave before they have any erotic contact,with Europe (as in manyvase paintings)sitting on the bull'sback, holding the bull'slong horn in one hand, and graspingthe hem of her chiton in her other hand, to keep it from falling in the sea, while the cloth on her shouldersbillows out like a sail (Moschus,Europa,lines 108-130). 61. Bell kraterby the Berlin Painter (TarquiniaRC 7456), fromTarquinia:
ARV2206, no. 126; LIMC IV, 1988, p. 77, no. 2, pl. 2, s.v. Europe (M. Robertson). Kurtz(1983, p. 102) observes that this vase "representsan earlier stage in the story"than the Berlin Painter'sdepictionof Europe on the bull'sback (Oxford 1927.4502:ARV2 210, no. 172; LIMC IV, 1988, p. 79, no. 41, s.v. Europe [M. Robertson]). Bihler 1968, p. 54: "Europemit weit ausschreitendenFiissen neben dem Stierherlauft." 62. Kilmer1997, p. 128. 63. Keuls 1985, p. 51.
338
MARY
POSSIBLE SCENES
MEANINGS
R. LEFKOWITZ
OF THE ABDUCTION
If only a relativelyfew of the Eos vases show her abducted lovers in such a state of contentment, it is because the vase painters are more interested in the earlier stages of the pursuit, her first approach to the young man, and his initial resistance and longing to return to his friends and previous life. It has been suggested that the painters of these vases concentrate on the negative aspects of abduction in order to send a covert message about women's sexuality: Woman at the mercy of the male. Women availableto be taken, to be raped.The proof: the inverted motif of the rape of Tithonos or Kephalos by Eos, during which, as we have seen, looking back in "consent"is rare.64 But why suppose that these scenes have any bearing on the behavior of mortal women? If the Eos painters have concentrated on the negative, it is because the goddess's attentions bring at best mixed blessings to her lovers. Eos asked Zeus to make Tithonos immortal, but she forgot to ask Zeus to make him ageless;when he became too old to move or get up, she shut him up in a chamber,and locked the doors (Hymn.Hom. 5.220-236). Eos allows Kephalos to return to Athens, but when he is out hunting, he accidently kills his wife, Prokris.65By contrast, painters allow Europe to appear content because her fate is relatively good, given that mortals, at best, can never have a completely happy life.66She becomes the mother of three sons, each of them famous: Minos, Rhadamanthys, and Sarpedon (Aesch. fr. 50.11-16 Radt). Abduction almost always causes trouble for the mortals involved, and sometimes even leads to their deaths.67Semele is consumed by flames as a result of her liaison with Zeus. So, in later sources,is Demeter's lover Iasion (a brother of Dardanos) because he desired Demeter (Konon 26, FGrHist F21). Anchises was struck by lightning because he boasted about his relations with Aphrodite (Soph. fr. 373.2-3 Radt).68In most cases, as it is for Anchises, the most positive aspect of an abduction by a god is the child that it produces.69 Only a few mortals have no regrets. Ganymedes is made immortal and ageless, though his father Tros grieves for him until Zeus tells him what he has conferred on his son, and gives Tros a pair of immortal horses in compensationfor his loss (Hymn.Horn.5.210-217). Oreithyia(see above, note 38) becomes the consort of Boreas and lives with him in Thrace. Amymone is rescued by Poseidon from the unwelcome attentions of satyrs, and given the spring of water that she was searching for, hydria in hand. Some vase painters show her looking back into the god's eyes as he approaches her, holding his trident.70Others show her about to let him embrace her or depict them standing together in the garden of Aphrodite, like bride and bridegroom.71 Like vase painters, poets are willing to represent both the negative and positive aspects of divine abduction.72The authorof the HomericHymn
64. Frontisi-Ducroux1996, p. 88. 65. The storywas told in the lost
cyclicepicEpigonoi(fr.5 Bernabe); fordetails,see Kearns1989,p. 177. 66. Hom. II. 24.525-526: "sothe
godshaveallottedforunfortunate mortals,to live in sorrow;but they are free from cares."Cf. Macleod 1982, p. 133: "thegods, who in this book show their pity for men and demand that men pity each other,also will human sufferingand neversharein it." 67. Cf. esp. Sourvinou-Inwood 1991, pp. 47-50. 68. Cf. Verg.Aen.2.648-649. Accordingto Hyginus, Fab.94, the lightning bolt killed him. On other variantsof the story,see Austin 1964, pp. 247-248. 69. Cf. Calame 1999, p. 67. 70. For examples,see Reeder 1995, pp. 352-357, nos. 111-113; and in general,Kaempf-Dimitriadou1969, pp. 28-30.
71. Forexamples,see Reeder1995,
pp. 358-361, nos. 114-116; Burn 1987, p. 52. 72. The authorof the Homeric HymntoApollosets aside obscurelocal myths of Apollo's abductionsof mortal women and concentratesinstead on how the god founded his shrine at Delphi; see Clay 1989, p. 56. But comparethe emphasisin Stewart1995, p. 76, "the Homeric hymn to Apollo recountedthe god's progressfrom insatiablewomanizerto panhellenic deity."
"PREDATORY"
73. Cohen (1996, p. 130) sees Helios's statement as a "dubious consolation,"and the promiseof marriageas an authorizationof violence. 74. Cf. Dougherty 1993, pp. 144145. 75. E.g., Rabinowitz 1993, pp. 197201; cf. Stewart 1995, p. 80, fig. 1. 76. Lefkowitz 1993, pp. 26-29, which Cohen (1996, p. 134, n. 39) sees as "anotherauthorization."Cf. Rabinowitz 1993, pp. 201-202. 77. Danae says to Zeus in Aesch., fr.47a, lines 784-785 Radt, "youhave the [responsibility]for the greaterfault, but I have paid the whole [penalty]." Comparethe fate of Antiope, who was abductedby Zeus (schol. Ap. Rhod. 4.1090, pp. 304-305 Wendel), but forced to expose her twin sons and then was imprisonedand tortured.In Eur., Antiopefr.208 Nauck (= fr. 33 Kambitsis), she observes:"if the gods have abandonedme and my twin sons, there is a reasonfor it; for some people are unfortunate,and others fortunate." Cf. n. 66, above. 78. Cf. Clay 1989, p. 171. 79. Cf. Od. 10.341, where Odysseus is afraidthat intercoursewith Kirkewill make him "weakand unmanned";Clay 1989, pp. 182-183.
GODDESSES
339
to Demeterdescribes Persephone'sanguish and Demeter's grief, but he also shows Hades treating her kindly. Helios points out to Demeter that Hades is "not an unworthy bridegroom among the immortal gods," because he is her own brother and lord of a third of the universe (Hymn. Horn. 2.82-87).73 Kyrene,after Apollo comes to her as a "bridegroom,"becomes the founder of the city of Kyrene and mother of a son, Aristaios, who will be made immortal (Pind. Pyth. 9.51-65).74 In Euripides'Ion Kreousacomplains that she was seduced and abandoned by Apollo, and that the child she bore and abandoned was lost. Her painful narrativehas struck some modern critics as emblematic of "rapes"of mortal women by the gods.75 But at this point in the drama Kreousadoes not know that in fact the god has not forgotten her after all, and is about to return to her the child she thought she had lost forever.Meanwhile, in her ignorance, anger,and confusion, Kreousahad tried to poison her child, Ion; she was jealous because she thought he was her husbandXouthos's son by another mother.76In the end, she is grateful to the god (Eur.Ion 1609-1614), even though for many years he has done nothing to mitigate her suffering.77 As in the case of Kreousa,mortals often make things worse for themselves through their lack of understanding. In the Homeric Hymn to Aphrodite,Anchises rushesinto bed with Aphrodite,strippingoff the clothes and jewelry with which she had so carefullyadorned herself (Hymn. Hom. 5.162-166, cf. 58-66).78The poet does not describetheir copulation;rather, with great conciseness, he comments on Anchises' ignorance: Then with the will of the gods and in accordancewith fate, he, a mortal, slept with an immortal goddess, without knowing clearly what he did (o0 o(acocgiSto;, Hymn. Hom. 5.166-167). When he realizes what he has done, he begs the goddess not to make him "strengthless among men" (Hymn. Horn. 5.188).79 As the author of the HomericHymn to Demeterhas the goddess observe, "mortalsare ignorant (vi'tj&) and without the knowledge to recognize fate beforehand,whether of coming good or of evil" (Hymn. Horn.2.256-257). The poets use the myths to remind their audiences that nothing in mortal life is an unqualified good. That is the meaning they attach to the story of Thetis and her mortal lover Peleus (briefly mentioned in the catalogue of goddesses in the Theogony;see above, note 2). Thetis, unlike Eos, did not seek to capture a mortal lover. It was the will of Zeus that she marryPeleus, and she did her best to prevent him from catching her. She changed into fire, and took on the form of a savage lion that attacked him with tooth and claw. For Peleus, marrying a goddess was the pinnacle of human achievement, a voyage to the outer limits of the known world, the straits of Gibraltar: he married one of the high-throned Nereids, and he saw the circled throne, where the kings of heaven and of seas were seated, and they showed him their [wedding] gifts and revealed to him the power of his descendants; but a man cannot cross the dark boundary of Gadeira (Pind. Nem. 4.65-69).
340
MARY
R. LEFKOWITZ
Figure12 (left).Pelike(Louvre G 373). PeleusholdingThetis. CourtesyMusee du Louvre,photo H. Lewandowski
Figure13 (above).Lebesgamikos (BritishMuseumB 298 [XXXIXC13]). PeleusholdingThetis.
CourtesyTrusteesof the BritishMuseum
Even though his wedding to Thetis was the high point of Peleus'slife, his goddess wife Thetis was miserable in the marriage (II. 18.430-435), and soon left him. His son Achilles was the greatest Greek hero, but he died at Troy. Marriages with goddesses are emblematic of the difficulties involved in extended relationships between immortals and mortals. Peleus wrestling with Thetis was a popular subject on Athenian vases from 520 to 460 B.C.80Reeder suggests that these scenes are intended to demonstrate the need for male control of the female's animal nature, on the grounds that hunting metaphors are linked with eroticism.81But Thetis is a goddess, not a mortal woman. If she is shown with snakes and even small lions (e.g., Fig. 12), it is not because her natureresembles an animal's,but rather because in the myth she turned herself into many different animals in order to avoid being caught by Peleus.82The vase painter has no other way of alluding to this aspect of the story,and it is also a way to identify her. No mortalwoman has the power to metamorphoseherselfinto any other form. Whatever these scenes were intended to mean, there is little to suggest that their primary intention was to remind women of their subordinate role in society. For example, Rush Rehm argues that a scene with Peleus and Thetis on the pedestal of a lebes gamikos, beneath a picture of a wedding procession, represents"aviolent abduction"(Fig. 13).83Thetis is raising her arms to express her resistance, like Persephone on the wall
80. LIMC VII, 1994, p. 269, s.v.Peleus (R. Vollkommer). 81. Reeder1995, pp. 299-300; cf. Sourvinou-Inwood1991, p. 66. 82. LouvreG 373:ARV2573, no. 9; LIMC III, 1986, p. 239, no. 29, pl. 189, s.v.Cheiron (M. Gisler-Huwiler). 83. Rehm 1994, p. 20. British Museum B 298: LIMC VII, 1994, p. 260, no. 117, s.v.Peleus (R. Vollkammer);Oakleyand Sinos 1993, p. 87, fig. 67.
"PREDATORY
A
) GODDESSES
34I
B
Figure14. Kylixby Peithinos(Berlin F 2279). SidesA-B: couplescourting andembracing;tondo:PeleusholdAntikensammlung, ing Thetis. Courtesy StaatlicheMuseen zu Berlin,Preussischer Kulturbesitz.Photo JohannesLaurentius (sides A-B); JuttaTietz-Glagow (tondo).
84. As Sourvinou-Inwood(1991, 66) observes,the iconographyof Pep. leus'scaptureof Thetis differsradically from scenes of other abductions. 85. Berlin F 2279, from Vulci:ARV2 115, no. 2, andARV21626; LIMCVIII, 1997, p. 8, no. 13, pl. 13 (R. Vollkommer);Reeder 1995, no. 106, pp. 341343. On the wrestlinghold, cf. Poliakoff 1987, p. 40. 86. Dover 1978, pp. 95-96, pl. R 196:a-b; Stewart1995, pp. 8182; Sutton 1997-1998, p. 30. The name Peithinos does not occurelsewhere;see Pape and Benseler1959, p. 1154; LGPN2, p. 365.
painting in Vergina (above, note 40). But there is no explicit violence. The scene with Peleus holding Thetis shows a standstill, a moment of acquisition: both figures are static.84On the interior of the Peithinos kylix (Fig. 14), the mortal hero has caught the goddess in a waistlock.85Both are fully dressed, and neither is looking at the other; there is no erotic contact. The outside of the cup offers a striking contrast:it portraysfully dressedheteroand homosexual mortal couples, negotiating with and embracing one another. Persuasion alone is enough in the human world, as the artist'snom de plume Peithinos ("Persuasion-man") suggests.86But it cannot help Peleus captureThetis; he can gain his prize only by patient pertinacity,and with the cooperation of gods who are even more powerful than Thetis-Zeus and Poseidon. The scene tells us little or nothing about the sexuality of Athenian women, but a great deal about the power of the gods.
MARY
342
R. LEFKOWITZ
CONCLUSION Many of the vases depicting scenes of Eos would have been used at festive occasions, such as weddings or symposia.87On vases given as wedding gifts, abduction scenes might even convey a sense of romantic affection; abduction by a god or goddess was an honor, despite the problems that it could introduce into a mortal'slife.88But ultimately,vase paintings of divine abductions display recognition of the power of the gods and of mortal vulnerability.They are expressions of what Euripides says about divine intervention at the close of five of his dramas: Many are the forms of divinity; the gods bring many things to pass unexpectedly.And what we thought would happen did not come to pass, but the god found a means to bring about what we did not imagine. That is how this action went. These lines celebrate the ways in which gods can change their appearance; they applaud the gods' ingenuity, and their ability always to surprise and confound mortal expectations. Their interventions can work to the advantageof the protagonists (as in the Alcestis,Andromache,and Helen) or against them (as in the Medea and Bacchae).89Since we know that the ancient Athenians held these views about the role of gods in human life, it is a mistake not to assume that they have some bearing on the encounters in vase paintings between gods and mortals. In assuming that abduction scenes were primarily meant to convey the complex messages about sexuality and gender that we now attribute to them, we may be preventing ourselves from seeing other meanings that were more important to the people for whom the vases were made, who still believed in the powers of their own gods.
87. Of the Eos vases listed in Kaempf-Dimitriadou1969, there are 15 lekythoi,but 44 amphoras, 30 kraters,28 drinkingcups, 16 hydriai,6 stamnoi, and 6 oinochoai. 88. Sutton 1997-1998, p. 31. Also see above,n. 6, on the placementof the Eos-Kephalosand Boreas-Oreithyia myths on the akroteriaof the Athenians'temple at Delos; the terracotta statueof Zeus and Ganymedesat Olympia may have been an akroterion; cf. Robertson1975, p. 277, cited in n. 55 above. 89. Eur.Alc. 1149-1153;Andr. 1284-1288; Hel. 1688-1692; Bacch. 1388-1392, and a differentfirst line, Med. 1415-1419. If the lines were addedlaterby actors,it was in order to please a publicwho liked such sententiae;Barrett1964, p. 417. Accordingto Roberts(1987, p. 56) and Dunn (1996, pp. 24-25), the lines are proformaritual;but cf. Lefkowitz 1989, pp. 80-82.
"PREDATORY
GODDESSES
343
REFERENCES Andronikos,M. 1994. VerginaII: The Tombof Persephone, Athens. Austin, R. G. 1964. P VergiliMaronis AeneidosLiberSecundus,Oxford. Barrett,W. S. 1964. Euripides:Hippolytos,Oxford. Entry Boedeker,D. D. 1974. Aphrodite's into GreekEpic (MnemosyneSuppl. 32), Leiden. on Braswell,B. K. 1988. A Commentary theFourthPythianOdeof Pindar, Berlin. Bremmer,J. 2000. "Foundinga City: The Case of Cyrene,"in Lafundacidnde la ciudad:Mitosy ritosen el mundoantiguo(Arquitext12), P.Azara, R. Mar, E. Riu, and E. Subias,eds., Barcelona,pp. 101109. Biihler,W. 1968. Europa,Munich. Burkert,W. 1992. TheOrientalizing Revolution:NearEasternInfluencein theEarlyArchaicAge,Cambridge, Mass. Burn,L. 1987. TheMeidiasPainter, Oxford. Buxton, R. G. A. 1982. Persuasionin GreekTragedy:AStudyofPeitho, Cambridge. Calame,C. 1999. ThePoeticsof Erosin AncientGreece,Princeton. Campbell,M. 1991. Moschus,Europa Texte (Altertumswissenschaftliche und Studien 19), Wiesbaden. Clay,J. S. 1989. ThePoliticsof Olympus: FormandMeaningin theMajor HomericHymns,Princeton. Cohen, A. 1996. "Portrayalsof Abduction in Greek Art: Rape or Metaphor,"in Kampen 1996, pp. 117135. Deacy, S., and K. F. Pierce,eds. 1997. Rapein Antiquity,London. De Puma, R. D. 1994. "Eos and Memnon in EtruscanMirrors,"in Murlo and theEtruscans: Art and Societyin AncientEtruria,R. D. De Puma andJ. P. Small, eds., Madison, pp. 180-189. Dougherty,C. 1993. ThePoeticsof Colonization:FromCityto Textin ArchaicGreece,New York. . 1998. "Sowingthe Seeds of Violence: Rape,Women, and the Land,"in Parchments of Gender:
DecipheringtheBodiesofAntiquity, M. Wyke, ed., Oxford,pp. 267284. Dover, K.J. 1978. GreekHomosexuality, Cambridge,Mass. End: Closure Dunn, F. 1996. Tragedy's and Innovationin Euripidean Drama,New York. Friedrich,P. 1978. TheMeaningof Aphrodite,Chicago. Frontisi-Ducroux,F. 1996. "Eros, Desire, and the Gaze,"in Kampen 1996,pp.81-100. Frymer-Kensky,T. S. 1992. In the Wake Women,Culture, of the Goddesses: and theBiblicalTransformation of PaganMyth, New York. Garland,R. 1985. The GreekWayof Death, London. Harrison,T. 1997. "Herodotusand the Ancient Greek Idea of Rape,"in Deacy and Pierce 1997, pp. 185208. How, W., andJ. Wells. 1912. A Commentaryon Herodotus,Oxford. Jenkins,I. 1983. "IsThere Life after Marriage?A Study of the Abduction Motif in VasePaintingsof the Athenian Wedding Ceremony," BICS 30, pp. 137-145. Kaempf-Dimitriadou,S. 1969. Die Liebeder Gotterin derattischen Kunstdes5.Jahrhunderts v. Chr. (AntK-BH 11), Bern. Kampen,N. B., ed. 1996. Sexualityin AncientArt:NearEast, Egypt, Greece,andItaly, Cambridge. Kearns,E. 1989. TheHeroesofAttica (BICS Suppl. 57), London. Keuls,E. 1985. TheReign of the Phallus:SexualPoliticsin Ancient Athens,Berkeley. Kilmer,M. F. 1990. "SexualViolence: ArchaicAthens and the Recent Past,"in "OwlstoAthens".Essays on ClassicalSubjectsPresentedto Sir KennethDover,E. M. Craik, ed., Oxford, pp. 261-277. . 1993. GreekEroticaonAttic Red-FiguredVases,London. . 1997. "Rapein Early RedFiguredPottery:Violence and Threat in Homo-Erotic and Hetero-Erotic Contexts,"in Deacy and Pierce 1997, pp. 123-141.
MARY
344 Kurtz,D. C. 1983. TheBerlinPainter, Oxford. Kurtz,D. C., andJ. Boardman.1971. GreekBurialCustoms,London. Lefkowitz,M. R. 1986. Womenin Greek Myth,Baltimore. . 1989. "'Impiety'and Atheism' in Euripides'Dramas,"CQ 39, pp. 70-82. . 1993. "Seductionand Rape in Greek Myth,"in Consentand Coercionto Sexand Marriagein Ancient and MedievalSocieties,A. E. Laiou, ed., Cambridge,Mass., pp. 17-37. LGPN = P. M. Fraserand E. Matthews, eds., A Lexiconof GreekPersonal Names,4 vols., Oxford 1987-2000. Lloyd-Jones,H. 1971. "Appendix," in AeschylusII (Loeb Classical Library),H. Weir Smyth, ed., Cambridge,Mass., pp. 325-603. Macleod, C. W. 1982. Homer,Iliad BookXXIV,Cambridge. Nagy, G. 1979. TheBestoftheAchaeans, Baltimore. Oakley,J.H., and R. H. Sinos. 1993. The Weddingin AncientAthens, Madison. Osborne,R. 1996. "DesiringWomen on Athenian Pottery,"in Kampen 1996,pp.65-80. Padel, R. 1992. In and Outof theMind: GreekImagesof theTragicSelf Princeton. Page, D. 1955. SapphoandAlcaeus, Oxford.
Pape,W., and G. Benseler.1959. Worterbuch dergriechischen Eigennamen,3rd ed., Graz. Poliakoff,M. B. 1987. CombatSportsin theAncientWorld,New Haven. Rabinowitz,N. S. 1993. Anxiety Veiled: Euripidesand theTrafficin Women, Ithaca. Reeder,E. D. 1995. Pandora:Womenin ClassicalGreece,Princeton. Rehm, R. 1994. Marriageto Death: TheConflationof Weddingand FuneralRitualsin GreekTragedy, Princeton. Richardson,N. 1974. TheHomeric Hymnto Demeter,Oxford. Ridgway,B. S. 1999. Prayersin Stone (SatherClassicalLectures63), Berkeley. Roberts,D. 1987. "PartingWords: Final Lines in Sophocles and Euripides,"CQ 37, pp. 51-64. Robertson,M. 1975. A Historyof Greek Art, Cambridge. Robson,J. E. 1997. "Bestialityand Bestial Rape in GreekMyth," in Deacy and Pierce 1997, pp. 6596. Shapiro,H. A. 1992. "Erosin Love: Pederastyand Pornographyin and Greece,"in Pornography in A. Greece, Richlin, Representation ed., New York,pp. 53-72. Sinos, R. H. 1993. "Divine Selections,"in CulturalPoeticsin Archaic Greece:Cult,Performance, Politics,
Mary R. Lefkowitz WELLESLEYCOLLEGE DEPARTMENT io6
CENTRAL
WELLESLEY,
OF CLASSICAL
STUDIES
STREET MASSACHUSETTS
mlefkowitz@wellesley.
edu
R. LEFKOWITZ
02481
C. Dougherty and L. Kurke,eds., Cambridge,pp. 73-91. Sourvinou-Inwood,C. 1991. "Reading" GreekCulture:TextsandImages, RitualsandMyths,Oxford. .1995. "Reading" GreekDeath: TotheEnd of the ClassicalPeriod, Oxford. Stehle, E. 1990. "Sappho'sGaze: Fantasies of a Goddess and Young Man,"Differences2.1, pp. 88-125. Stewart,A. 1995. "Rape?"in Reeder 1995, pp. 74-90. Sutton, R. F.,Jr. 1992. "Pornography and Persuasionon Attic Pottery," in Pornography and Representation in GreeceandRome,A. Richlin, ed., New York,pp. 3-35. .1997-1998. "NuptialEros: The Visual Discourse of Marriage in ClassicalAthens,"JWalt55-56, pp.27-48. Vermeule,E. 1979. Aspectsof Death in Early GreekArt and Poetry (SatherClassicalLectures46), Berkeley. West, M. L. 1966. Hesiod,Theogony, Oxford. .1997. TheEast Faceof Helicon: WestAsiaticElementsin GreekPoetry andMyth, Oxford. Zeitlin, F. 1986. "Configurationsof Rape in Greek Myth,"in Rape, S. Tomaselliand R. Porter,eds., Oxford,pp. 122-151.
HESPERIA
71 (2002)
Pages 345-376
ANCIENT FROM
BEEHIVES
ISTHMIA
ABSTRACT Ceramicbeehivesareknownto havebeenutilizedin ancientGreece.Excavationsby the Universityof Chicagoat Isthmiahaveproducedmanyfragof thisspecialvessel.In mentswithaninteriorcombedsurfacecharacteristic additionto fouralmostcompletebeehivesof horizontaltype,fourrestored hivesanda numberof fragmentsof anuprightstyleof hivefromthe Sanctuaryof Poseidonandthe Rachisettlementarepublishedhere.They provide the firstsubstantialevidenceforthe existenceof the uprighthivein ancient Greece. There is a black and white thyme, and it flowers very freely: it is in bloom about the summer solstice. It is from this flower that the bee gets the honey, and by it the beekeepers say that it is made known whether they have a good yield of honey or not; for if the thyme flowers abundantly,they have a good yield, but the bloom is injured or even destroyed if it is rained upon.... They say that it (thyme) cannot be grown or become established where a breeze from the sea does not reach (Theophr. Caus.pl. 6.2.2-4).
1. See Gebhard,Hemans, and Hayes 1998 for recentbibliography. A summaryof this article ("Isthmia Hives and Others")was presentedby J. Ellis Jones at the conference"BeeKeepingin the Graeco-RomanWorld," held at Lady MargaretHall, Oxford, on 7 November2000. 2. For type 2 see Broneer1958, no. 42; Kardara1961, pp. 26432, p. 265, fig. 6, pl. 81; Crane 1983, pp. 200202.
Several curiously shaped ceramic vessels were discovered in the course of excavations by the University of Chicago at the Sanctuaryof Poseidon at the Isthmus (Fig. 1).1They fall into two types. Vessels of type 1 (Fig. 2) are long and slightly tapering in shape, described when first published as resembling an umbrella stand.The vessels are about two-thirds of a meter in length, and usually have a rounded bottom. Type 2 (Fig. 3) vessels are about half this size and have a flat base; in shape they are not unlike a basket (kalathos)for wool or olives, often with handles on either side and an upper diameter significantly greater than that of the base.2 A characteristicfeature of both vessel types is the roughening of the interior surface from bottom to top with a comb or stick(s) before firing; the area covered varies from a quarter to a third of the circumference in
346
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
Figure1. Plan of Isthmia,showing locationswherebeehivefragments havebeen found.J.EllisJones,after Anderson-Stojanovic1996, p. 58, fig. 1 (F. P. Hemans and W. B. Dinsmoor Jr.)
some cases, to two-thirds in others. It was primarily the treatment of the interior that suggested to D. I. Pallas that combed pots of type 1, found in a ruined tower of the Hexamilion fortress at the Isthmus, were beehives.3 Subsequent analysis of Attic examples of the "umbrellastand"type from the Varifarmhouse revealedthe presence of beeswax, confirming the identification. Ethnographic parallelsfrom the Cycladic islands and Crete had led earlier to the same conclusion.4 The type 1 hive was used in a horizontal position, with the combed portion at the top for the attachment of honeycombs that hung parallelto the mouth of the hive in a long series; a lid, with a small bee-entry hole (flight hole) in its lower edge, was placed over the opening and could be removed at need for harvesting the nearest combs (Fig. 2). Sometimes, detachable extension or expansion rings were placed at the front end, between hive and lid, to make the harvesting easier. Cylindrical beehives open at both ends were used in Spain during the 3rd and 2nd centuries B.C.5 The horizontal hive is the most commonly documented type in the archaeologicalrecord,and appearsearly in Egyptian tomb paintings.6The
3. Broneer1959,p. 337. 4.Jones,Graham,andSackett1973, pp.397-414;Jones1976;Ruttner1979, pp.221-225;Jones1990,pp.63-71; Crane1983,pp.35-76;Ifantides1983, p. 79. Hives,extensionrings,andlid foundat Knossosare fragments publishedin Catlinget al.1981,p. 95, V 325,fig.6; Sackett1992,pp.20,27, 93, 176-177,189,203,225,241,254, 480,ns. 83-85, pls.96, 100,134,221. Fourcompletebeehiveextensionrings havebeenfoundat Eleutherna (Vogt 2000, p. 95, figs. 47-48). 5. Bonet Rosado and Mata Parrefio 1997, pp. 35-36. Dimensions varyfrom 0.24 to 0.29 for diameterand 0.53 to 0.58 for length. Interiorscoringis horizontal,coveringthe entire surface. 6. Crane 1983, pp. 35-39.
ANCIENT
?\--tr? I
(
''
' 'i-+
'II
: : S: ir ?,
iii
.t ?? 1;
i
i:1. i!J :?
ji
::
? i
? 6?? 1.
if; ;? :i i j iji?
?.
:i i i .? -. -1 i
r;-jI
Y:.i.i?c?.;-
i
i: :: ir
.i ;1..
;..;?; ;
--
Figure2 (left).Ancienthorizontal hives:Late Romanexample(1) from Isthmia;andHellenistichivesfrom Trachones,Attica, shownwith and withoutextensionring.J.EllisJones, afterJones, Graham,and Sackett 1973, p. 447, fig. 19
Figure3 (right).Three uprighthives fromIsthmia(5, 6, 7) comparedto a modernhive at Palaiochorio,Crete, andsketchesof threePalaiochorio hivesin use.J.EllisJones
7. Jones, Graham,and Sackett 1973, pp. 399-401, ns. 24-27; Jones 1990, p. 69, n. 9; Crane 1983, pp. 196203. An unusualhive from Knossos with a domed end and possible entry holes is discussedin Hayes 1983, pp. 108-109, 132, no. 177, 134, fig. 15.
347
ISTHMIA
ti.?;. t
I r?.
i..i.. 1 :;: :; irj1
?:? :?r? i
FROM
BEEHIVES
H -i; H I
I
I
I
I cm
shape is still used in countries aroundthe Mediterranean and also in many other areasof the world. Such hives can be used singly,but are more commonly found stacked in clusters, large or small, often in the shade of a shed or roof. Excavations at Isthmia at the Sanctuaryof Poseidon and at the Hellenistic settlement on the ridge known as the Rachi, south of the sanctuary, have produced additional examples of coarse ware with internal combing, several identifiable as the shorter type 2 vessel (Fig. 3) or parts of such a vessel. The aim of this article is to show that the type 2 vase from Isthmia was a beehive set upright on its flat bottom with honeycombs hanging down from above into the body of the vessel. An upright hive of this sort could have been used either as a movable-nest hive with the whole mass of honeycombs adhering to some form of removablelid fitting over the open top, or as a top-bar hive, a refinement in which combs hang from individual wooden bars set fairly close together across the open mouth, allowing separateremovalof individual combs in harvesting the honey. A lid of some kind was used to hold down and shade the bars.Wicker and ceramic bar hives can still be seen in Crete at Palaiochorio, near Katalagari,south of Knossos (Fig. 3). A characteristicof these upright hives is that the bees are provided with a small entry hole low on the wall of the hive itself.7 Such hives are set out in the open or in the shade of an overhanging rock, well spaced or in loose groups.
348
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
0 .I
-
a
a
E
I
l.
4
l-
60 --- .... I ll)
TYPE 1 BEEHIVES
Figure 4. Horizontal hives 1-4. J. EllisJones
Four horizontal beehives of type 1 from tower 7 of the Hexamilion fortress have been restored (Fig. 4).8 The fabric of these vessels is reddish yellow (7.5YR 6/6), gritty, and fired rather hard;surfacesvary from red to reddish brown (2.5YR 5/4-6/4) with some patches of burning. Heights vary from 0.58 to 0.64 m and the body tapers from a diameter of 0.250.27 m at the mouth (D. rim 0.29-0.30 m) to ca. 0.17-0.19 m at the exterior of the base, which is rounded or domed to a greater or lesser degree (see Table 1 below). Rims tend to have a simple projecting profile and to be flat on top. The exterior of the lid (Fig. 5) and body shows traces of wheel ridging while the interior was incised vertically with a comblike instrument;the incisions coverbetween a quarterand a third of the vessel's circumference.Perpendicularto these marks and just below the rim is a band of horizontal combing, and frequentlyseveralother similarbands are found at intervals below. All four examples display marks or inscriptions added before firing just below the rim on the exterior (Fig. 4).9 It seems likely that these marks, which were incised in still-soft clay, denote the potter himself, or an owner who had ordered a number of hives in advance;if they had been scratchedon after firing, they would be more likely to representthe owner's own identification mark.
Figure 5. Beehive lid 34, upper surface
8. No. 3 is illustratedin Broneer 1959, p. 337, no. 17 (IP 2139), fig. 11, pl. 72c. See also IsthmiaV, p. 114, n. 5(c), where anotherrim fragment
(IPR 68-40, pl. 40d from tower 2, in a 5th-century A.c. context) is described.
9. Liidorf (1998-1999, p. 51, n. 71) discussesinscriptionsas owner'smarks.
ANCIENT
BEEHIVES
FROM ISTHMIA
349
TYPE 2 BEEHIVES The first example of a type 2 vessel was found in the 1955 season of excavation at the Rachi settlement (Fig. 3:5; see below, Figs. 8-9). In an article about the settlement, C. Kardarasuggested that this container was a beehive.10It too has interior roughening, and just below the rim a series of letters, OPEETAAA,perhaps the owner's name, has been incised after firing. The vessel is ca. 0.30 m in height and has a small rectangularopening at the bottom of the side where there is no combing on the interior. Three similar vessels (6, 7, 8) have been restored from recovered fragments, and other rim and base fragments with comparable features are also presented here. The fabric of the type 2 vessels is in almost all cases that used for the typical Corinthian coarse ware or plain ware of the Greek and Early Roman periods. Vessel 18, however, is made of cooking fabric.l1The fabric normallyfires to a range of colors between reddishyellow (5YR 7/6, 7.5YR 7/6) and pink (7.5YR 7/4) or very pale brown (10YR 8/3).12All of these vessels appearto be wheelmade. Two types of rims may be distinguished: the more complex projecting rim with flat upper surface and prominent ridge beneath (5, 6, 9, 10, 11, 13, 15, 16, 18) and a plain projecting rim without the ridge (7, 8, 12, 14, 17, 19, 20, 21).13A small flight hole, either rectangularor round, at the bottom of the wall or in the floor close to it, would allow the bees to go in and out of the hive. Round holes vary from 0.026 to 0.030 m in diameter, rectangular slits from 0.025 to 0.045 by 0.005 to 0.015 m. The upright beehive may have interior combing from bottom to top, and a few examples (e.g., 20) display a band of horizontal combing near the rim. In the Isthmia sample, the preserved fragments suggest that combing was omitted on the part of the interiorwall on either side of the flight holes. The placement of the combing in this arrangement is analogous to that seen in the horizontal hive. It is possible that it was meant to encourage the bees to attach their combs to those parts of the wall with the roughened surface and not to the area near the opening. Details of combing vary from regular,closely spaced, deep incisions to a ratherirregular,shallow treatment.There is some evidence to suggest that not all hives were roughened on the interior, although one cannot be certain unless most of the circumference is preserved. Hive 7, for example, has no combing on its surviving parts.'4 10. Kardara1961, pp. 264-265. 11. In addition to its use for containersplaced on or in the fire, cooking fabric,becauseof its lightness,was used for pitchersand kraters,many of which have been found in the Rachi settlement. 12. Whitbread 1995, pp. 293-308, esp. pp. 301-302; CorinthVII.2, pp. 94-95; CorinthVII.3, p. 104. This Corinthiancoarseware,with greateror lesser amountsof grit and grog, was
used until the Hellenistic period for roof tiles, other terracottaarchitecture, perrirhanteria,type A' amphoras,basins, pitchers,and bathtubs.By the Late Roman period,light-bodied coarsewares were no longer used.Thin sections of the Isthmia combed sherdswere made, and we await a report. 13. Liidorf 1998-1999, figs. 32 and 33, aregood examplesof complex rim treatment;figs. 16 and 17 depict the simple form.
14. Other examplesof beehives without characteristicopenings in the floor or base, or diagnosticrims and lips, may have escapedour attention since the interiorcombing is the only feature that distinguishesbody sherdsof beehives from other coarsewares (such as lekanai,for example,noted by Lohmann [1993, p. 39]). Severalof the pieces published by Lohmann arewithout internal combing and, indeed, he comments that it is not necessary.
350
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
Only one (5) of the three pots restored included handles among its surviving fragments.These handles could be confidently "replaced"in position; they are oval in section and set horizontally on either side of the vessel, opposite each other. The other restored pots did not have handle fragments or handle attachments securely associated with them. Several of the rim and wall pieces (9, 10, 13) do preserve a horizontal groove at the handle zone; on 18 the handle stub is placed directly upon such a groove. In fact, a small part of a similar groove can be seen on 5 about midway between the handles (see below, Fig. 9). This technique, used by Roman and modern potters, facilitates the placement of handles at the same level.15 Since the discovery of the so-called Orestada vase (5), there has been a certain amount of controversyregardingits identification as an example of an ancient upright hive. Arguments also have been made for and against the probable use of the top-bar hive in ancient Greece.16The Orestada vase, in particular,was considered very small (estimated volume 17 liters) in comparison to modern hives of a similar type; it was too heavy, the interior combing was neither uniform nor did it cover the entire surface, and, in looking at the vase itself, it was difficult to determine how much of the flight hole was plaster and how much was an original feature.17This opening (0.050 m in width and ca. 0.015 m in height), although partly restored,is an original feature of the vase:while the outside surfaceis broken, the interior retains the original form of the opening; the restorationis on the outside surface only. A program of absorbed organic residue analysis of samples from combed/beehive sherds has been carried out under the direction of Richard P. Evershed of Bristol University to provide some scientific evidence for the use of the Isthmia pieces as beehives.18Sherds from three areas(the fortress,Rachi settlement, and sanctuaryand environs) were selected to be tested for the presence of beeswax. It was not possible, however, to submit inventoried, restored rim or base samples for this destructive analysis, and as a result all samples chosen were wall fragments. The presence of beeswax, found in sixteen of forty-one samples of combed sherds from Isthmia, supports the identification of many samples as beehives, but since types 1 15. We would like to thank one of the anonymousHesperiareviewersfor these observations.Roman examples of the use of groovesto align handles may be seen in CorinthXVIII.2, p. 63, no. 138, pl. 7; p. 81, no. 172, pl. 10; p. 123, no. 270, pl. 16; alsoAgoraV, pl. 3, F84-85, and pl. 7, G190. 16. Crane 1983, pp. 200-202; Crane and Graham 1985, pp. 30-31. 17. Crane 1983, pp. 17, 201. When Kardaraand Papadopoulou(1984, pp. 2-3) studied the vessel in the Isthmia Museum, they were not able to obtain accurateinformation. 18. See Evershedand AndersonStojanovic1995; Evershedet al., forthcoming.The chemicalinvestigations
were performedon residuesof beeswax absorbedwithin the fabricof the vessels. High-temperaturegas chromatography(HT-GC) and HT-GC/mass spectrometry(HT-GC/MS) were used to screenlipid extractsfor the presence of compoundscharacteristicof beeswax. Samplesof beeswaxtaken from a 19th-centuryceramicbeehivewere used as referencematerials.Potsherds from ten pithoi recoveredfrom the Rachi excavationsservedas controls. None of the pithoi containedbeeswax residues.The presenceof beeswaxwas deducedfrom the distributionsof n-alkanes(C 23 to C 33), wax esters (C 40 to C 52), and long-chain alcohols (C 24 to C 34) releasedfrom the
wax estersduringburial.As in a recent investigationof beeswaxin Late Minoan lamps and conical cups from Crete (Evershedet al. 1997), 13Cvalueswere determinedfor the individualcomponentsof the lipid extractsby means of GC-combustionisotope ratio MS. The valuesobtained for the n-alkanes(-24.5 to -26.9%) and n-alkanols(-25.0 to -25.7%)were comparableto the characteristicvalues obtainedfor the referencebeeswax. Thus, on the basis of molecularstructures,carbonnumberdistributions, and 3C values,a significantproportion of beehive sherdswere shown, unambiguously,to contain beeswax.
BEEHIVES
ANCIENT
TABLE
FROM
1. DIMENSIONS
ISTHMIA
35I
OF TYPES 1 AND 2
Height
MouthDiam.
Rim Diam.
BaseDiam.
Type 1(1-4) Type 2 (5-8)
0.58-0.64 0.29-0.33
0.25-0.27 0.31-0.38
0.29-0.30 0.36-0.42
0.17-0.19 0.18-0.27
Rims (9-23) Bases (24-32)
-
0.31-0.40 -
0.35-0.43
Sample
0.15-0.31
All measurementsare in meters.
19. A full listing of the sampleswill appearin the publicationof the study by Evershedet al. (forthcoming).Of the ten samplestaken from the fortress,two were positive;five of twelve samplesfrom the Rachi, and nine of nineteen from the sanctuary,were also positive.All positive samplesfrom the fortressand sanctuaryhad been washed in an acid bath after excavation,while the five samplesfrom the Rachi had not. There is no simple correlation apparentbetween the extent of beeswax degradationand whether or not acid washing took place. 20. A flat base with a hole in the center and no combing might have been a flowerpot.See Thompson 1963, p. 11, fig. 13, for an illustrationof a flowerpotfrom the Hephaisteion garden. 21. Fabricis not, on its own, a reliabledistinguishingfactor. 22. Jones, Graham,and Sackett 1973, pp. 400,446-447, pls. 77, 84; Jones 1976, pp. 82-84. It was not possible to test the lid fragmentsfor the presenceof beeswax.Similarlids areillustratedin Munn 1993, p. 77, figs. 30-31. 23. Jones, Graham,and Sackett 1973: see appendixI by M. I. Geroulanos,p. 446. 24. Any additionalfragmentsthat might be recoveredfrom going through the lots from the Sanctuaryof Poseidon would not significantlychange the discussionabove,although they would give us a better idea of the numbersof beehives in the area.
and 2 cannot be distinguished in the small fragments sampled, this analysis cannot be used to support the identification of type 2 hives.19 In summary,the upright hive at Isthmia is characterized by the following characteristics:1) a flight hole for the bees in the lower wall or in the base of the vessel; 2) a height of 0.29-0.33 m; 3) a mouth diameter of 0.31-0.38; 4) a flat base; 5) commonly, a partiallycombed interior; and 6) less commonly, a pair of horizontal handles below the rim. Of the six attributes, a combination of several of them, or the flight hole on its own, would be a fairlygood indication that the vessel was intended to be used as an uprighthive. Thus, those flat base fragments with flight holes (25, 26, and 30) are likely to be from type 2 hives.20In order to distinguish the upright hive from the horizontal hive, one should ideally have a complete profile, or a combinationof certain attributes that are not, unfortunately, ordinarilypresent in a single sherd.21 One fragmentary ceramic lid (34) has been included in this report because it is not unlike examples typically associated with horizontal beehives. It is flat in profile (Fig. 5), with curved ridges on the upper surface and a smooth underside.22It has been suggested that such ridges provided reinforcement for the added pressure exerted by the cords that were used to attach the lid to the hive (Fig. 2).23A flight hole, which would be expected if used with a horizontal hive, is not preserved in this lid. But the curved ridges, arguablylogical for the horizontal hive, are of no specific use in the vertical type, for which no cords were required.
CATALOGUE The following catalogue is designed to illustrate the range of variation present in the beehives from Isthmia. It includes those vessels that could be restored from the fortress (1-4), the Sanctuary of Poseidon (6, 7), and the Rachi settlement, including the Orestada vessel (5) and one other (8), as well as inventoried rims and bases.24One wall fragment (33) is included to show an unusual combing pattern on the interior. The dimensions of the vessels are summarized above in Table 1. Since the focus of this article is the use of the vessels as containers for honeycombs, the diameter of the vessel mouth ratherthan of the rim is listed in the catalogue. All measurements are in meters. Despite the fact that type 1 hives were used in a horizontal position, it has been traditional to use the term height rather than length. We record here essential measurements only; more precise determinations would reflect the variation that frequently exists from one side
V. R. ANDERSON-STOJANOVIC
352
AND
J. E. JONES
of the rim to another, or other irregularitiesof pots that have not been centered on the wheel. Corinthian clay is known to be too plastic to sustain the weight of large vessels. The Isthmia inventory number follows the catalogue number. For fragments inventoried after 1985, the lot number indicates context; for others, the field notebook reference is given.25More specific context information appears below in Table 2. Munsell readings are given only for those pieces that appear different from the typical Corinthian fabric. NEARLY
TYPE
1
COMPLETE
AND
RESTORED
VESSELS
I
IP 2216 Figs. 2, 4,6 5 Horizontal beehive
Hexamilionfortress,tower7, notebookDP, 1958, p. 85. Rest.H. 0.620-0.645; est. Diam. mouth 0.254;Diam. base0.170, end only veryslightlydomed.Restoredfrom manyfragmentswith plasterinfill. Hard-firedfabric,reddishyellow (7.5YR6/6) fabricwith interiorfired redtoreddishbrown(2.5YR5/46/4), discoloredgrayandblackon exterior. Flat rim,outwardlythickened. Taperingprofilewith roundedbottom.Wheel-ridgedexterior,incised horizontalgroovesjust belowrim on exterior.Interior:Overlappingbands of verticalcombingon a thirdof interiorcircumference; perpendicular bandsjust belowrim and abouta thirdof the distancedowninterior wall.IIO incisedbeforefiring,just belowrim exterior. Date: 6th century A.C.
downinteriorwall.On exteriorbelow rimis a markmadewith a comb, in the shapeof a roughT risingfrom an O. Date: 6th century A.c. 3
IP 2139 Horizontal beehive
Figs. 4, 7
Samecontextas 1. Broneer 1959,pp.320-321,337, no. 17, fig. 11, pl. 72:c.Rest.H. 0.60 (with heightto startof dome at 0.55, makingthe domeprominentand somewhatpointed);est. Diam. mouth0.275;Diam. base0.17. Restoredwith someplasterinfills. Fabricas 1 and2. Thick rim,slightlydownturned. Taperingprofilewith slightlyconcave wall.Stronglywheel-ridgedexterior. Interiorhas deepverticalcombing with threebroadcross-bandsat regularintervalson interiorwall. Justbelowrimexteriorarethe letters IIO, incisedbeforefiring. Date: 6th century A.C.
2
IP 2217 Horizontal beehive
Figs. 4, 6
Samecontextas 1. Rest.H. 0.58; H. to startof dome 0.54; est. Diam. mouth0.26; Diam. base0.19. Largesectionof rim andwall missing;restored.Fabricas 1. Similarin shapeto 1 but rim moreflared,exteriorwall less ridged, andend moredomedandpointed. Exteriorwheel ridgessmoothed. Deep, regular,verticalcombingon interior;horizontalcross-bandsjust belowrim and a quarterof the way
4
IP 2140 Horizontal beehive
Figs. 4, 7
Same context as 1. Rest. H. 0.63; H. to start of dome 0.60; est. Diam. mouth 0.255; Diam. base 0.170. Domed bottom similar to 2. Mended and restored with plaster infills. Fabric, exterior ridging, and interior combing as 3. Similar in shape to 1. W or upside-down M incised with a comb, below rim on exterior. Date: 6th century A.C.
25. These are storedin the Isthmia Museum along with all documentation from the Universityof Chicago Excavationsat the site.
ANCIENT
BEEHIVES
FROM
ISTHMIA
353
2
1
2
1
Figure 6. Horizontal hives 1 and 2
2
354
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
3
4
4
3
Figure7. Horizontalhives3 and4
ANCIENT
TYPE
5
BEEHIVES
FROM
2
IP 586 Figs. 3, 8, 9 "Orestadavase,"upright beehive
Rachi, tr. IV, notebook 8, pp. 62-63, house XV. Kardara1961, fig. 6, pl. 81; Kardaraand Papadopoulou 1984. H. 0.301; est. Diam. mouth 0.340; Diam. base 0.23. Flight hole just above slab of base measures not more than 0.050 wide, with the original bottom and right edges preserved; easier to see on the interior than on the exterior of the vessel. The original width could be a millimeter smaller since the left side is chipped. Restored as rectangularon the right, with the height of the hole ca. 0.010; plaster fills the upper part for two-thirds of the length on the right. On the upper left, the preserved shape might indicate that the opening was arched on the top, rising to a total height of ca. 0.015. Restored from numerous fragments, with plaster infill; small part of rim and one handle missing but wall preserves position for handle attachment. Typical Corinthian coarse ware fabric. Exterior surface worn. Kalathosshape with wall flaring outward from flat base to outward thickened rim with prominent ridge or flange below. Originally two handles, oval in section, placed horizontally. Handle on left preserved;right handle restored. Shallow vertical combing covers nearly four-fifths of interior from lip to bottom of wall. Inscription just below rim: OPESTAAA scratched on after firing (SEG
XVII 134:'Opeoro'coc).
Date: late 3rd century B.C. 6
355
ISTHMIA
IP 2215 Upright beehive
Figs. 3, 8, 9
Sanctuary,tr.D, West Stoa Reservoir,notebookJGH,baskets17, 30, pp. 164,214. Est. H. 0.33;est. Diam. mouth0.335; Diam. base0.26. Rectangularflighthole in floornearjoin of basewith wall;partlyrestoredbut definitely0.03 x 0.05. Slightlyless thanhalf of the vesselwas restored with fragmentsandplaster.Fabric variesfromreddishbrownto reddish yellow(7.5YR6/4-6/6), ratherfine
andwithoutvisibleinclusions. Slightlyconvexwallwith little flare.Outwardthickenedrimwith flangebeneath.Exteriorhas random scrapelines createdwhile the vessel was turningon the wheel.Roughly a thirdof interiorhaswidelyspaced verticalscoringmadewith a fivetooth comb. Date: ca. 180-160
7
IP 2512 Uprightbeehive
B.C.
Figs. 3, 8,9
Sanctuary,tr.GW2, notebook JGH, basket3, p. 352. H. 0.29;rest. Diam. mouth0.38; Diam. base0.270 (slightlyovoid).Rectangularflight hole, 0.03 wide x 0.02 high at bottom of wall,just aboveinteriorfloorlevel. Restoredfrommanyfragmentswith muchplasterinfill;missingsections of rim andbase.PaleCorinthian coarsefabricwith few apparentinclusions(restorationdoes not allow for inspectionof the cross-section). Shapehas flaringwall fromfoot to broadopening;rim similarto but less pronouncedthan5 or 6. No combingon survivingpartsof interior.
Date:Hellenistic. 8
IP 7753
Fig. 8
Upright beehive
Rachi,lot 89-116, houseI, room A. Uppersection:P.H. 0.22; est. Diam. mouth0.315;est. Diam. base 0.180. Rectangularflighthole with roundedcornersat bottomof wall, 0.02 x 0.025. About two-thirdsof upperpartpreserved,mendedfrom manyfragments;threejoining fragmentsof a baseprobablycome fromthe samehive.Upperandlower partsdo notjoin. Corinthianplain warefabric. Flaringwallwith convexcurve towardbottom;plain,flat outturned rim.Flight hole is inclinedso that, insidethe pot, the floorriseshigher thanthe outerloweredge.No combing preservedon interiorof surviving fragments. Date: late 3rd century B.C.
V. R. ANDERSON-STOJANOVIC
356
AND
J. E. JONES
r TArAA
Or
5
---_
7
8 0
10
- H H H-H' LI _
_ _
_ _.
30
20 _
_
_
--i l
;
40 1 cm
Figure 8. Four upright hives: 5 and 8 from the Rachi settlement, 6 and 7 from the sanctuary
ANCIENT
BEEHIVES
FROM
ISTHMIA
357
5
5 5
6
6 6
7 Figure 9. Orestada vessel (5): exterior, interior, and detail of entry hole; hive 6: exterior, interior, and detail of entry hole; hive 7: interior, exterior
7
V. R. ANDERSON-STOJANOVIC
358
AND
J. E. JONES
RIMS
9
IP 6712 Figs. 10, 11 Rim and upper wall
Rachi, lot 1052, housseXVII. P.H. 0.17; Diam. mouth( ).365. One large fragment. Corinthia.n coarse ware. Heavy rim with outvvardly thickened lip and ridge beslow. Smooth exterior;single g]roove 0.115 down from rim, perhaps tto align handles. If so, this was pr(obably an upright hive. Cf. 10, 13, and 18. Combing on interior is ve;ryfine, closely spaced, slightly de,eper at bottom with traces of fing,er ridging. Date: 3rd century B.C-. 10 IP 7388 Rim and wall R-ark; 11k.All.. I.Vdll UUC
Figs. 10, 11 pDate:
Q ,d.,-,-1, h,~,,~ 0Q* TNT 1 v11VnntJIUDUUU. OU,
p. 39. P.H. 0.21; est. Diam. mouth 0.332. Three joining fragments. Corinthian coarse ware. Outwardly thickened rim with ridge below. Single groove (not aligned with wheel marks) on exterior wall, 0.137 down from rim, probably for handles. Cf. 9 above. Handles suggest an upright hive. Overlapping, vertical combing covers entire preserved interior surface,becoming deeper and more closely spaced at bottom. Probably from same vessel as 11. Date: 3rd century B.C. 11 IP 7390
Fig. 11
Rim
Rachi,houseIV, notebook8, p. 39. P.H. 0.108;est. Diam. mouth 0.32. Twojoiningfragments.Corinthiancoarseware. Similarin shapeto 10. Vertical combingon interior.Probablybelongswith 10. Date: 3rd century B.C.
12 IP 7759
Figs. 10, 11
Rim and wall Rachi, lot 89-491, street 5, house XI. P.H. 0.145; est. Diam. mouth 0.375. Two joining fragments. Small worn areajust below rim interior. Corinthian coarse ware. Plain rim, slightly concave below. Interior displays deep, messy
combing,overlappingat top. Graffito scratchedon wall appearsto be either a backwardN or an M. Date: late 3rd century B.C.
13 IP 7780
Figs. 10, 11
Rim and wall Rachi, lot 89-492, street 5, house XI. P.H. 0.13; est. Diam. mouth 0.34. Three joining pieces. Burned and discolored Corinthian coarse ware. Rather large grit visible at break. Heavy, double-ridged rim. Groove just above break at bottom edge, for handles? Cf. 9, 10, and 18. Closely set vertical combing on interior, made with alternating deep and shallow strokes. Perhaps an upright hive. late 3rd century B.C.
14 IP 7841
Figs. 10, 11
Rim and wall Rachi, lot 89-508, house VII. P.H. 0.15; est. Diam mouth 0.33. Two joining fragments. Corinthian coarse ware with small amount of grit. Projecting rim with shallow wheel ridging beneath. One narrow band of rather deep vertical combing on interior with traces of shallow combing to either side. Date: late 3rd century B.C.
15 IP 7391 Rim
Fig. 11
Rachi,houseXVI, notebook8, 66. P.H.0.08; est. Diam. mouth p. 0.33.Two fragments,mended.Discoloredgraysurfaces;probably Corinthianfabric. Outwardlythickenedrimwith prominentridgebeneath.Three broadbandsof verticalcombingon interior. Date: 3rd century B.C.
16 IP7758a Rim
Fig. 11
Rachi, lot 89-114, house I, room B. PH. 0.067; est. Diam. mouth 0.32. Two joining fragments. Corinthian coarse fabric. Outwardly thickened, doubleridged rim above incurving wall. Shallow, vertical, interior scoring,
ANCIENT
BEEHIVES
FROM
ISTHMIA
359
9
12
ir,
13
13
14
Figure 10. Hives 9, 10, 13,14, 20: exterior, interior; hive 12: exterior
20
20
V. R. ANDERSON-STOJANOVIC
360
AND
J. E. JONES
15
'' ' '6 1 16
17
4
~~~
\^ .
?a-1.
.;.'
.
11/
18.
-}--- Vs
-:
\.
18
11.
mI
?
mI
( ..I./ll/ 1.t.I ..i... I
.
? 1;--? -?
, ,
:"~j ..1
'
I .
19 .
I
1-1 t
0-./,, ,,;. . ,,-2.:.
-
, --I
,:;
20
21
40
0
-
-.
.
HHH .-
..
....
.
.
.
.
. II
Ik
....
H I II
-I
Icm
Figure 11. Rims 9-21.J. Ellis Jones
BEEHIVES
ANCIENT
FROM
36i
ISTHMIA
i`i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 22
>
r
Figure 12. Hives 22 and 23.
23
J. W. Hayes
alternatelydeep and shallowwith overlap. Date: late 3rd century B.C.
17 IP 7392 Rim
Fig. 11
Rachi,tr.T, notebook2, p. 15. P.H.0.075;est. Diam. mouth0.35. One fragment.Corinthiancoarse ware. Plainrim.Regularlyspaced verticalcombingon interior. Date: after 200
18 IP 6509 Rim
B.C.
Fig. 11
Rachi,lot 1068, South Slope cistern,upperfill. P.H.0.10; est. Diam. mouth0.37. Sevenjoining fragments.Hard,thin, cookingfabric, grayto reddishbrownon surface. Rim roundedat exteriorwith ridgebelow.Traceof handlerooton exterior,set alonga faintgrooveat 0.09 belowlip. On the evidenceof the handle,this rim is likelyto be froman uprighthive. Date: late 3rd century B.C.
1997, p. 16). (Anderson-Stojanovic 19 IP7760
Rim
Fig. 11
Rachi,lot 89-507, houseXIV. P.H. 0.062. One fragment.Smooth Corinthianplainware. Plainrim.Widely spaced combingon interior.Cf. 20 and21 for similarprofiles. Date: late 3rd century B.C.
20 IP 1546 Rim
Figs. 10, 11 lot 1408, tr.H, cisSanctuary, tern.P.H. 0.049;est. Diam. mouth 0.38. Two fragments,mendedwith
plaster.Corinthianplainware. Plainrim.Interiorpreserves verticalcombing,overlappingwith horizontalbandjust belowrim interior.Cf. 19 and21 fromRachi with similarrimprofiles. Date:Mixed Hellenisticand Romancontext. 21
IP 7791
Rim
Fig. 11
Rachi,lot 89-497, houseX, roomC. P.H. 0.158;est. Diam. mouth0.37.Twojoiningsherds. Spallingon outersurface.Corinthian coarseware. Plainrim.Close-setvertical combingon interior.Cf. 19 and20 for similarprofiles. Date: late 3rd century B.C.
22 IP6416
Rim
Fig. 12
lot 161, tr.Vc, noteSanctuary, book 7, circularcistern,pp. 20-21. Layerin which rimwas found:0.701.20 m. P.H. 0.033;est. Diam. mouth 0.40. Singlesherd.Soft localfabric, ratherfine in texture. Simpleevertedrim.Tracesof verticalandhorizontalscoringon interior. Date: early 2nd century A.c.
23 IP 6938 Rim
Fig. 12
lot 720, tr.NTD, secSanctuary, tion III, notebook10, pp. 167-170. P.H. 0.041;est. Diam mouth.0.30. Singlesherd.Corinthianplainware. Evertedrim. Segmentof vertical combingon interior. Date:Late Hellenistic/Early Roman.
362
V. R. ANDERSON-STOJANOVIC
( . * -/.j . j ' ,,...
JI I. r_:
.
AND
J. E. JONES
, :j .
..
24
\--
,rr
v
29
' p,
?
i
:'t:.: 25
wS
30
'
'' --t.d
V
. -3
"
. -
.
31
26
1 _ 27
32
.______--_ 34 34
28
0 IT Lrir I I' ...iI
40 H
1 =-1-f
r rI " I
F4
I I I
I I
.-
I
cm
Figure 13. Base fragments 24-32 and lid 34. J. EllisJonesandV.R. AndersonStojanovic
BASES
24
IP 628 Fig. 13 Part of base and lower wall
Rachi, lot 1034, well, layer 13. P.H. 0.035; est. Diam. ext. 0.20. One fragment. Corinthian coarse ware. Flat base with inturned wall. Faint traces of interior combing. Date: late 4th century B.C. (Anderson-Stojanovic 1993, p. 297).
25
IP 562 Figs. 13, 14 Base and wall with flight hole
Rachi, lot 138, N. Building, rooms 3, 4. P.H. 0.130; est. Diam. int. 0.021; Diam. flight hole 0.030. One fragment. Corinthian coarse ware. Flat base with convex, flaring wall. Exterior smoothed over but worn in some areas;wheel ridges on
ANCIENT
BEEHIVES
FROM
363
ISTHMIA
26
25
25
28 Figure 14. Base fragments 25,26,28, and 30
interior.Flight hole at bottomof wall.No combingpreservedon interior. Date: 3rd century B.C.
26 IP 7372 Figs. 13, 14 Basewith rectangular flighthole Rachi,lot 546, houseXVII, P.H.0.020;est. Diam. ext. 198. p. 0.240.W. flighthole 0.025;H. flight hole 0.015. One smallfragmentwith outerwallpreserved.Grittyfabric, reddishyellowat core(5YR6/6) with pinksurfaces(7.5YR7/4). Flat basewith rectangular open0.017 frominneredge of wall. ing Date: 3rd century B.C.
27 IP7389 Fig. 13 Partof baseandlowerwall Rachi,lot 312, N. Building, rooms3, 4. P.H. 0.035;est. Diam. ext.0.31. One fragment.Corinthian plainware. Flat base.Closelyspaced,rather deeplycut combingon interior. Date: 3rd century B.C. 28
IP 7776 Figs. 13, 14 Base and lower wall _1 1__1_: ..1 . norm -.......
Kacni, cleaning unit,
siae,
notebook 58, basket 63, p. 125. P.H. 0.040; est. Diam. ext. 0.30. One fragment. Corinthian plain ware. Flat base. Combing on floor at various angles. Date: after 200 B.C. 29
IP 8430 Base and lower wall
Fig. 13
Rachi, lot 89-492, street 5, house XI. P.H. 0.019; est. Diam. ext. 0.21. Slightly more than half preserved. Corinthian coarse ware. Flat base. On interior floor a row of tiny circularimpressions appear to be the imprint from the comb used to make the characteristic inner surface. Date: late 3rd century B.C. 30 IP 8449 Figs. 13, 14 Base and lower wall with rectangularflight hole Rachi, lot 89-122, house II, room B. P.H. 0.016; est. Diam. ext. 0.205; W. flight hole 0.036; H. flight hole 0.011. One piece, preserving bottom edge and floor. Corinthian coarse ware. Flat base and flaring wall. Smoothed interior surface. Rectangu-
364
V. R. ANDERSON-STOJANOVIC
laropeningwith smootheded, floor,immediatelyinsidewall, fromedge of pot. Date: late 3rd century B.C
31 P 8128 Baseandlowerwall
Fig. 13
Rachi,lot 89-519, houseI P.H.0.056;Diam. base0.145; holes0.016-0.020. One fragm Most of flatbasewith two hol one in wall,the otherin floor; punchedthroughfromabove. OTHER
BEEHIVE
33 IP 8678
Wall
AND
J. E. JONES
bly a flowerpotratherthana beehive?Corinthianplainware. Flat basewith curvedsides. Date: late 3rd century B.C.
32 IP8111 Baseandlowerwall
Fig. 13
Rachi,lot 89-519, houseIX. P.H. 0.122;Diam. ext. 0.22. One fragment.Corinthiancoarseware. Coarse,pittedandburnedsurface. Concavebasewith flaringwall. Date: late 3rd century B.C.
FRAGMENTS
Fig. 15
Rachi, lot 89-507, house P.H. 0.074. One fragment ofv Corinthian coarse ware. Straight profile. CombinE two parts:deep horizontal con in lower half, three groups of diagonal shallow grooves abov Date: late 3rd century B.C
34 IP 8129 Lid
Figs. 5, 13
Rachi,lot 89-518, houseIX. Est. Diam. 0.330;Th. 0.020. One fragment,restoredwith plasterinfills, ca. two-thirdspreserved.Corinthian coarseware,burned. Undersideplain,uppersurface with semicircular ridgesas atTrachones.Cf.Jones1976, p. 83. Date: late 3rd century B.C.
DISTRIBUTION
Figure 15. Wall fragment 33, interior
AND DATE OF THE BEEHIVES
It is not our intention to present an in-depth review of recent work on beehives or beekeeping, but rather to set forth the evidence for beehives, especially the upright hive, as it exists from Isthmia, and to encourage others to use this information to advance the study of ancient beekeeping. The distribution and date of the beehive fragments found at Isthmia are described in Table 2. The four securely identifiable examples of horizontal beehives (1-4) of type 1 from the Hexamilion fortresswere found in association with Late Roman pottery in the interior fill of tower 7.26 We can only speculate as to whether they were in use at the fortress. A rim from tower 2 in a 6th-century A.C. context has an estimated diameter of
0.34 and could have belonged to either a type 1 or type 2 hive (Table 1).27 Other body sherds occurredin contexts of Roman and Byzantine date. On the other hand, current evidence indicates that the type 2 beehives were used at Isthmia in Hellenistic times and into the Early Roman period. Given this wide range of dates, the sample from Isthmia does not include a sufficient number of securely identified horizontal and upright hives to determine which variations in profile, fabric, and surface treatments (variables typically identifiable on fragments) might be characteristicof a particular type or origin. The majority of beehive fragments from the Rachi settlement were found in destruction debris within the houses or on the streets of the settlement (Table 2). With the destruction dated to ca. 200 B.C., we have secure
26. IsthmiaV, p. 127: "This material, however,does not appearto have been stratified,and Middle Byzantinematerial,an earlyRoman lamp, and Late Roman potterywere all encounteredin the lowest level of the tower.The examplesof nearlycomplete sixthcenturyvessels appearto have been scatteredthroughoutthe fill, from top to bottom, and so it is unlikelythat the amphorasand beehiveswere used in the tower.Rather,they must have been dumpedinto the tower at some later date, perhapseven in the fourteenthor fifteenth century." 27. IsthmiaV, p. 114, n. 5.
ANCIENT
TABLE 2. DISTRIBUTION
OF BEEHIVE
Area
Feature
Fortress Fortress Fortress Fortress Fortress Fortress
fill fill
Rachi
tower 2 tower 7 tower 14 NE gate NW S gate N bldg H I, II
Rachi
H IA
destr
basement
destr destr destr destr wash destr destr destr
3rd B.C. 3rd B.C. 3rd B.C. 3rd B.C. post 200 basement 3rd B.C. basement 3rd B.C. press,wkrm 3rd B.C.
Rachi
Rachi Rachi Rachi Rachi Rachi Rachi Rachi Rachi
HIB H II B H IV H VII H VIII H IX HX C HXI Rachi H XII Rachi H XIV Rachi H XV Rachi H XVI Rachi H XVII Rachi street 1 Rachi street 5 Rachi SS cist Rachi well Rachi N side Rachi tr T Sanctuary W stoa Sanctuary tr GW2 Sanctuary tr H Sanctuary circ cist Sanctuary NTD Sanctuary NTD Sanctuary SEA Sanctuary ET Sanctuary ET 7 Sanctuary ET Sanctuary NE Sanctuary NE Sanctuary ST Theater N/NE Stadium STE-5 Stadium STW-1 Stadium STE-3E Stadium STW-2S Stadium STE-1 Stadium SS-1, 1A TOTAL
Context
BEEHIVES
Location
Date
FROM
FRAGMENTS Cat.No.
Type1
1-4
4**
within walls 5th A.c. interior Roman-Byz Roman
ISTHMIA
365
AT ISTHMIA Type2
Rim
Base
1
3rd B.C.
surface
post 200
2* 5* 1* 2*
destr destr
3rd B.C.
3rd
6 (5*)
B.C.
8
basem, ctyd basement press basement
destr
2**
25, 27
16 30
1**
10,11 14
2** 1**
31, 32, 34 1 lid(?)** 21
1**
1**
B.C.
2**
3rd B.C.
19, 33
ctyd
3rd
5
B.C.
1**
destr
3rd B.C.
15
1**
destr
3rd
9, 26
1**
surface destr destr dump wash
ctyd
upperfill layer 13
fill
reservoir
fill fill
cistern
B.C.
post 200 B.C. 3rd B.C. 12,13, 29 late 3rd B.C. 18 late 4th B.C. 24 post 200 B.C. 28 post 200 B.C. 17 2nd B.C. 6 Hell 7 Hell-Rom 20 early2nd A.C. 22 Hell-Rom 23
2**
1** 1**
surface fill fill surface surface surface dump
road road E caves
1 2 (1*) 1 2
3
1*
1
3 (1*)
6 1 1 1 1 1 1 1 1
1** 1** 1**
1** 1**
early2nd A.C. Hell-lst A.C. 2nd, 3rd A.C.
water chan Roman w chan tun Roman startline Byz Byz Roman N tunnel Roman 5
4
16
5 3 2 2 2 4 3
1*
1**
surface
fill fill fill surface
4(1*) 2
4 2 5 1 2 2 6 1
3 (2*) 3 1 1 3 (1**) 4 1 1
B.C.
storage
Total
1
Roman-Byz Hell-Rom 13th A.C. destr
Wall
9
1* 1*
1 1
2*
2
1* 1* 2* 2* 1* 1* 2* 1* 1* 1* 1* 1* 59
1 1 2 2 1 1 2 1 1 1 1 1 93
*Samplefor analysis.**Includedin catalogue.Abbreviations:destr = destructiondebris;wash = earthwashed down from deposits above;basem = basement;ctyd = courtyard;wkrm = workroom;chan = channel;tun = tunnel.
366
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
dates for those pieces.28These include two restored examples (5, 8) of upright hives and the fragments of others (9-16, 18,19, 21, 25, 27, 29-34) possibly or probably of that type. We cannot be sure if they were in storage, or in use in courtyards and on roofs within the settlement during a late phase, when the inhabitants may have been largely confined to the hill for security (Fig. 1).29 Other pieces come from contexts associated with deposits close to the surface,or from sediments washed down from above. Some may belong to a period when habitation at the site had mostly ended, and when abandoned dwellings were reused as convenient shelters by local beekeepers.Many beehive fragments have recentlybeen documented from the Athenian Agora, which suggests that bees did not require an isolated location for their hives.30 Beehive fragments from the Sanctuaryof Poseidon are primarilyfrom secondary contexts (Table 2). Two restored examples (6, 7) of the upright hive as well as some rim fragments (20,22,23) are from the temenos area, while several wall sherds were found in the northeast caves and the theater.According to John Hayes, these examples belong either to the Hellenistic or Early Roman period.31Were they once in use there or did they arriveas votives? Other wall sherds found in the vicinity of the Hellenistic stadium (Fig. 1, "laterstadium"),southeast of the sanctuary,were selected for chemical analysis. Beehives may originally have been placed in the fields near the stadium or on the northeast side of the Rachi, which rises above the stadium. It is possible that the upright type of beehive was in use elsewhere in the region but has simply not been identified because of the fragmentary state of the remains. Similarly,the horizontal hive may also have existed in the Corinthia during Classical and Hellenistic times. No beehives have been published from Ancient Corinth, and we are aware of only a few others from the Corinthia.32Both type 1 and type 2 hives may have been made of wicker and other perishable materials,which have not survivedin the archaeologicalrecord.33There is no reason why several different types of hives in a varietyof materialsmight not have been in use simultaneously, as recently observed on Crete, Cyprus, and Siphnos.34 28. See Anderson-Stojanovic1996, pp. 93-94, on the demise of the settlement, perhapsas a resultof military activityin 198 B.c. A detaileddiscussion of the contexts in which the combed sherdswere found will appear in the forthcomingvolume on the Rachi settlement in the Isthmiaseries. 29. Bonet Rosado and Mata Parrefio(1997, p. 42) discussbeehives in settlementcontexts,in particular mentioningbeehives on rooftops. See also generallythe informationposted in 2000 on the SphakiaSurveyWeb site: L. Nixon, J. Moody, S. Price, and O. Rackham,"SphakiaSurvey:The Internet Edition,"http://sphakia. classics.ox.ac.uk.Abstractsof"BeeKeeping in the Graeco-RomanWorld," a conferenceorganizedby Simon Price
and Lucia Nixon at Lady Margaret Hall, Oxford,on 7 November2000, may also be found on this Web site. The Akrotiricentralcorridorof Crete is discussedin Raab2001, pp. 134 (Hellenistic), 137 (EarlyRoman), and 160 (generaldiscussion). 30. Rotroff2002. I am gratefulto Susan Rotroff,who generouslyprovided us with a copy of her paper. 31. To be publishedin a forthcoming volume in the Isthmiaseries by Hayes on the Hellenistic and Roman pottery from the Sanctuary of Poseidon. 32. Severalbody sherdsare describedin Gregory1985 (p. 428, nos. 37-39, pl. 111), dated Late Roman to Early Byzantine.See the comments in Jones,Graham,and
Sackett 1973, p. 397;Jones 1976, p. 83; and Salmon 1984, p. 24. 33. Crane 1983, pp. 51-54. Davis (1996, p. 461) suggests that wicker hives were typical of apiculturein the Peloponnese,althoughthis cannot, of course,be proved. 34. Jones, Graham,and Sackett 1973, pl. 78d, shows terracottaand wicker hives in use at the same time and in the same location on Crete. See Graham1975, p. 75, for cylindrical terracottahives along with wooden box hives in use simultaneouslyon Cyprus (as photographedby Jones in 1972). As recentlyas 1987, Jones saw horizontal pot hives and wooden hives in use on a field terracenearthe tower of Aspropyrgoson the south coast of Siphnos.
ANCIENT
BEEHIVES
FROM
ISTHMIA
367
Figure16. Greekbeehive.
FromWheler 1682, p. 11. Courtesy Founders'Library,Universityof Wales, Lampeter.
BEEHIVES
FROM ELSEWHERE
IN GREECE
Upright bar hives of wickerwork are documented in 17th-century Attica by Sir George Wheler in A Journey into Greece(Fig. 16).35Attica, where type 1 was in use in the Classical and Hellenistic periods, is well known as a source of numerous examples of ancient beehives.36Even today,Mt. Hymettos is famous for its honey. Examples from the excavationsin the Athenian Agora and Eretria, although taller than the Isthmian type 2 hives, have a similar rim diameter as well as flat bases.37A number of the rims found during Lohmann's surveyof Attica have diameterslarge enough for upright hives, but bases with interior combing and entry holes have not been published.38A hive from the Kerameikosexcavationsis quite similar to the Isthmian type 2, with a height of 0.355 m and exteriorrim diameter of 0.380 m.39Ltidorf, in her classification of the three major types of beehives in Attica, classifies both the Agora and Kerameikoshives as type Al, the kalathosshape.40Type Al varies in height from 0.35 to 0.55 m, in rim diameter from 0.32 to 0.40 m, and in base diameter from 0.17 to 0.27 m. Finally, from excavations on Chios in the 1950s come several rim fragments whose dimensions may qualify them as examples of type 2.41 Although the larger diameter is a feature of the type 2 hive, the flat base is also characteristicof many hives of type 1. Furthermore,none of 35. Wheler 1682, pp. 412-413. See below,p. 369, for the full text. 36. See the comprehensivestudy of Liidorf (1998-1999) and the earlier studyof Lohmann (1993). 37. AgoraXII, p. 366, no. 1853, pl. 88; and Liidorf 1998-1999, p. 84, B1, pl. 11, are the same vessel:H. 0.437, Diam. rim 0.390; EretriaVIII,
p. 112, no. 76, H. 0.418, Diam. rim 0.363. For severalbeehivesfrom the western tombs of Oropos,see Pologiorge 1998, pp. 62-63, 127-128. 38. Lohmann 1993, p. 355, A5318, with est. H. 0.46-0.48, Diam. rim 0.363. 39. Liidorf 1998-1999, p. 84, B3, dated 340-300 B.c. 13, pl.
40. Liidorf 1998-1999, pp. 56-57. 41. E.g., Anderson 1954, p. 137, no. 28, p. 173, fig. 5, Diam. rim 0.47; p. 142, no. 79, p. 173, fig. 5, Diam. rim 0.40. Both have horizontalcombing on the interiorwalls and both rims correspondto the complexprofile of type 2.
368
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
the beehives mentioned just above has a flight hole in the lower wall or floor of the vessel, suggesting that the hive stood upright. A recently published fragmentary,flat beehive base from the late-5th-century Agora does have a flight hole in the center of the floor; although only 0.148 m of the height is preserved, the diameter of 0.29 m at the top of the fragment must have been greater at the original rim.42The presence of the flight hole, combing on the interior,and a rim diameter of probablyat least 0.35 m suggests that this is the first known example of an upright hive in Attica. Just as modern beehives vary in size, ancient Greek beehives too will have varied, as is clear from Ludorf's typology of four types (Al, A2, B1, B2).43 Few beehives have been reported from surveys in Greece. Almost no beehive fragments have turned up in the Pylos region or in the Nemea valley, while in the southern Argolid and on Methana, only a handfiulof fragments have been identified.44Some areas of Greece, however, have produced more fragments of ceramic beehives than the Peloponnese: Kea, About eighsouthern Euboia, Boiotia, and western Crete, in particular.45 teen beehive fragments spanning the ancient to medieval periods are cited in Raab'swork (2001) on the Akrotiri peninsula in Crete. An unusual ring base fragment with internal combing and a small hole (D. ca. 0.025 m) in the floor near the vessel wall was found in the Kathiani district and may have been part of a type 2 hive.46Columella (Rust. 9.14.19), citing Celsus, also mentions Euboia, Akhaia, the Cyclades, Skyros, and Sicily as regions where bees are moved in order to be near favorableplants such as thyme, marjoram,and savory. No ancient Greek or Roman visual representations of beehives exist as far as we know. Ancient literaturehas extensive references to beekeeping, the most substantial of which are in Aristotle (Hist. an. 5.21.553a554b; 9.40.623b-627b; Gen.an. 3.10.759a-761a), Varro(Rust. 3.16), and Columella (Rust. 9.2-16). Varro and Columella give practical accounts from the point of view of the beekeeper, but the question of whether a movable-barhive existed in ancient Greece or Rome cannot be satisfactorily or conclusively answered from any of the sources.47The description of 42. Rotroff2001. 43. Liidorf 1998-1999, pp. 55-76, 84-85. Crane (1983, p. 17, table 2) providesa table of modernhive variations. 44. Davis (1996, p. 461) considers the evidenceprovidedby recent surveys for apiculturein ancient Greece. Wright et al. (1990) mention no beehivesin the Nemea survey;see also Davis 1996, p. 461, for the Pylos area. One site in the southernArgolid produced fragmentsof ceramicbeehives, andJameson,Runnels,and van Andel (1994, pp. 289-290) suggest that hives of wood or strawmust also have been used. One beehive fragmentfrom the Roman period is noted from the
Methana survey(Mee and Forbes1997, p. 122). 45. Two rims from Kea picturedin Sutton 1991 (pp. 260-263, no. 11.6, fig. 5.9 and no. 57.4, fig. 5.10) have a markedlyslantedprofilebut, with rim diametersof under0.20, aretoo small to be from uprighthives. Davis (1996, p. 461) also mentions Boiotia and southernEuboia as sourcesof hives, but Bintliff and Snodgrass(1985) report no beehives,nor are many noted in unpublishedreportson the Southern Euboia ExplorationProject.D. R. Keller(pers.comm., Oct. 2002) excavated a nearlycomplete exampleof a hive in a Classicalcistern (with later reuse)on the Paximadhipeninsulain
southernEuboia,but he comments that fragmentsfound duringsurveyof the areaaregenerallyrare.Beekeepingin western Crete is the subjectof Francis 1999, and is also discussedon the SphakiaSurveyWeb site (see above, n. 29). 46. Raab2001, p. 97, no. 85, p. 204, fig. 41. 47. Klek and Armbruster(19191921) compiled surveysof the ancient literatureon the topic. See also Fraser1951, pp. 14-17; Pritchett 1956, pp. 260-261; Jones, Graham,and Sackett 1973, pp. 403-412, n. 46; Crane 1983, pp. 200-202; Graham 1975, pp. 74-75; Liidorf 1998-1999, pp. 47-51.
ANCIENT
BEEHIVES
FROM
ISTHMIA
369
the top-barhive by Wheler is the earliestaccountof the appearanceand use of hivesthat he comparesto modernwastebasketsand that resemble type 2 beehivesin shape(Fig. 16). As this workis not availablein many universitylibraries,the passageis quotedherein full:
48. FromA. Brothers' textof Wheler (1682), pp. 412-413. We acknowledgeour thanks to Tony Brothersof the Departmentof Classics, Universityof Wales, Lampeter,and to the Founders'Libraryof the University of Wales, where a copy of this rarework exists.
The Hives they keep their Bees in, are made of Willows, or Osiers, fashioned like our common Dust-Baskets, wide at the Top, and narrow at the Bottom; and plaister'dwith Clay, or Loam, within and without. They are set the wide end upwards, as you see here. (A, B,). The Tops being covered with broad flat Sticks, (as at C.C.C.) are also plaister'dwith Clay at the Top; and to secure them from the Weather, they cover them with a Tuft of Straw, as we do. Along each of these Sticks, the Bees fasten their Combs; so that a Comb may be taken out whole, without the least bruising, and with the greatest ease imaginable. To increase them in Springtime, that is, in Marchor April, until the beginning of May, they divide them; first separating the Sticks, on which the Combs and Bees are fastened, from one another with a Knife: so taking out the first Combs and Bees together, on each side, they put them into another Basket in the same Order as they were taken out, until they have equally divided them. After this, when they are both again accommodated with Sticks and Plaister, they set the new Basket in the Place of the old one, and the old one in some new Place. And all this they do in the middle of the day, at such time as the greatest part of the Bees are abroad;who at their coming home, without much difficulty,by this means divide themselves equally. This device hinders them from swarming, and flying away.In August they take out their Honey; which they do in the day-time also, while they are abroad;the Bees being thereby,they say,disturbed least. At which time they take out the Combs laden with Honey, as before; that is, beginning at each out-side, and so taking away,until they have left only such a [p. 413] quantity of Combs in the middle, as they judge will be sufficient to maintain the Bees in Winter; sweeping those Bees, that are on the Combs they take out, into the Basket again, and again covering it with new Combs and Plaister. All that I doubt concerning the Practice of this here in England, is, that perhaps they gather a less quantity of Honey; and that, should they take the like quantity of Honey from the Bees here in England, they would not leave enough to preserve them in Winter. But this hinders not much: For by being less covetous, and not taking so much Honey from the poor Bees, the great encrease and multiplying of them would soon equalize, and far exceed the little Profit we make by destroying of them. This is done without Smoak; wherefore the Antients call this Honey UnsmokenHoney:and I believe the Smoak of Sulphur, AXOCrvtLOV, which we use, takes away very much of the Fragrancyof the Wax; and sure I am, the Honey can receive neither good Taste, nor good Smell from it.48
370
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
DISCUSSION Regardless of the type or orientation of the hive, bees always attach their combs to the top of the hive (Columella, Rust. 9.15.7-8). The shape of the combs is determined by the angle of the beehive wall, and the honeycombs hang either facing the opening or parallel with its axis.49The former arrangement would be ideal for harvesting combs from type 1 hives, but bees are annoyingly inconsistent, according to Columella (see above).The beekeeper might try to control the orientation of the combs by placing wax on the combed surface of the interior to encourage attachment of the combs in particularplaces. In a movable-nest hive, all the combs are attached to the underside of the lid and removed as one mass when the lid is lifted. They then must be cut off the lid. In a top-bar hive (Figs. 3, 17), the combs are stuck underneath individual bars set across the mouth of the hive, between the hive and the lid. Only with the latter arrangementwould the beekeeper be able to control the orientation of the comb by the arrangement of the bars or sticks but, since all combs would be equally easy to access, the orientation would not matter. If the upright hives from Isthmia were used as movable-nest hives with lids, or as top-bar hives, it would appearthat the combing on the inside served no purpose;as we noted earlier,hives 7 and 8 have no combing preservedon what remains of the interior.Perhapsthe tradition of scoring the interior was simply a holdover from the technique used with the horizontal hive. If the combs were stuck at their edges to the sides of the pot, then the beekeeper would have had the extra work of cutting the comb from the side of the pot with a knife.50Some modern studies have suggested that sloping side walls discourage the attachment of combs.51In one experiment, however,it was found that combs were indeed attached to hive walls in some cases, particularlyif the angle of slope was less than 50?.52Further work on identifying upright hives in the existing ancient beehive corpuswill be necessaryto solve the relationship between beehive shape or orientation and comb attachment to upright hive walls. An essential element of type 2 beehives is the flight hole located in the lower wall or base of the hive. It is typical for bees to check carefullya new nest site for the right amount of space, a south-facing view, sufficient distance from the ground, and an entrance no more than ca. 0.04 m in diameterwith proximity to the floor.53By not attaching combs to the floor, the bees leave a space in the lowest part of the hive for circulation and the removal of debris. One may well say that the opening in the wall seems sensible, but a hole in the bottom of the floor somewhat impractical because it would requireplacement on an open stand or frameworkto allow the bees access.To simplify matters, piles of stones, bricks, or wood might also serve the same purpose. Columella (Rust. 9.7.5-6) emphasizes the need for the opening to be as small as possible to keep out the cold, as well as beetles, cockroaches,lizards, and butterflies, and that it is useful to have more than one opening. Perhaps an opening in the floor is less obvious to predators,and protected from the wind. The top-bar beehive made beekeeping and the acquisition of honey much easier, as the combs could be lifted out of the hive for inspection or
49. Ifantides(1983, pp. 79, 84, fig. 4) illustratesthe combs parallelto the axis of the opening. See alsoJones, Graham,and Sackett 1973, p. 447, pl. 85;Jones 1976, pp. 84-86. 50. Jones, Graham,and Sackett (1973, p. 448, n. 246) note that, in Cyprus,long kniveswere used to cut out combs from the rearend of the long-pipe horizontalhive. Ratherless effortwould be required-only at the edges of the dangling comb-if the comb were stuck on a bar.Columella (Rust.9.15.7) says that "thecombs adherevery little to the sides."On the other hand, Seeley (1995, p. 34) writes: "Theythen build their beeswax combs downward,attachingthem to the cavity'sroof and walls and leaving small passagewaysalong the edges." 51. E.g., Ifantides 1983, p. 80. 52. Free and Williams 1981, p. 240. 53. Seeley 1995, p. 34, where 25 liters is mentioned as the minimum amount of spacerequired;see also http://www.nbb.cornell.edu/neurobio/ seeley/seeley.htmlfor informationon the behaviorof bees.
ANCIENT
a
b
Figure17. a) Emptyuprighttop-bar hive at Palaiochorio,Crete,1973; b) samehive in use;c) samehive in use, andone emptyhive, readyfor use. J. Ellis Jones
BEEHIVES
FROM
ISTHMIA
37I
372
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
for harvesting rather than cut away from the top of the hive, which would have been necessary in type 1 hives.54Although this arrangementmade it easier to harvest the honeycombs, the number of combs in the upright hive must have been smallerthan in the horizontal hive, inasmuch as examples 0.58-0.64 m in height would allow as many as fifteen combs, given the naturaldistance of 0.038 m from the middle of one comb to another.55 Those at the smaller end would be smaller than those at the larger,front end. The upright hive, however,with a mouth of ca. 0.31 to 0.38 m, would have held no more than seven to ten combs, those at the center being the largest and those at the edges very small (see Figs. 16, 17:b-c). Ifantides reconstructs the possible steps in the development of the movable-nest or top-bar hive: 1) the typical horizontal hive with cover is placed in an upright position; 2) as a result of this orientation, bees attach the combs to the lid; 3) the (wooden?) lid is removed in sections to free a smaller number of honeycombs at once; and 4) the hive is made shorter and sticks or bars are placed across the mouth for the attachment of the combs, making it possible to remove each comb individually.56The upright beehives at Isthmia may illustrate a late stage of step 3 or step 4. These ancienthives were presumablyproducedby pottersfor the general market, and not by beekeepers themselves. The beekeeper could make adjustments to his purchased hives as needed. For example, an excessively large hole could have been made smaller through the use of clay or mud filling (Columella, Rust. 9.14.13-15). Individuals certainlymust have purchased or acquired used hives and lids whenever possible and made do with what they had. We should not expect any particulardegree of standardization. A variety of beehive pots of upright type were seen by Ellis Jones in a potter'syard located on the coast west of Heraklion, Crete. Moreover, the examples of upright hives still in use at Palaiochorio, mentioned earlier, illustrate combs hanging down from a rack or mesh of sticks or sticks, wire, and twigs (Figs. 3, 17). The hives themselves are slightly larger than the restoredupright hives from Isthmia (5-8), with curving sides and small bee-flight holes just above the flat bases. The Cretan ceramic hives generally also have projecting lugs or handles on the upper part of the vessel, one on each side, to allow hives to be lifted and moved. There are examples of lids with downturned edges and lug handles, some of which, at least, were made to fit individual hives closely (as was tested with empty hives). Generally, however, the lids are used simply as umbrellas over the bars or rack, to shade the hives from sun or rain, and no particularcare is taken to use a close-fitting lid. These modern Palaiochorio hives have no internal combing, and in that respect they resemble the horizontal hives in use in the Cycladic islands, and vessels 7 and 8 from Isthmia. 54. Crane 1983, pp. 196-212, esp. p. 196, fig. 221, p. 198, fig. 225, and p. 210, table 8. 55. Jones, Graham,and Sackett
1973, p. 411; Crane 1983, p. 210. 56. Ifantides 1983, pp. 81-85. We have brieflysummarizedthe sequence here.
ANCIENT
BEEHIVES
FROM
ISTHMIA
373
CONCLUSION
57. Crane 1983, pp. 238-240;
Forbes1966,pp.35, 90-96;Fraser 1951, pp. 109-121.
58. Anderson-Stojanovic 1996,
pp. 90-93; 1998. 59. Anacreontea(D. A. Campbell, trans.,Cambridge,Mass., 1988), frs. 3 and 16, pp. 164, 165, 182-185, provides
the earliestreferences to encaustic painting. 60.The evidenceforthe casting of smallbronzesat the Isthmusis
presentedin Rostokerand Gebhard 1980, pp. 350-354,361; and Rostoker, McNallan, and Gebhard 1983.
As many have alreadynoted, beekeeping was an important economic activity in the ancient world.57Honey was the only available sweetening agent, a component in medicines and in fixing scents, and also served as a suitable offering to the gods. An equally valuable commodity of apiculture was beeswax, useful for writing, sealing items, lighting, and for lining vessels. From the Isthmus of Corinth we have presented four almost complete examples of ancient beehives of the upright variety, together with fragments of others, including three bases with flight holes (25, 26, 30) and a rim with one handle preserved (18). Three other rims with a requisite mouth diameter and grooves for placing handles (9, 10, 13) are also good candidates for upright hives. Of the two restoredhives with combing on the walls of the interior, neither preserves this treatment on the inner surfaces close to the flight hole, nor on the floor. Thus, the absence of interior combing on fragments is not a criterion for rejecting them as beehives. The other base fragments (27, 28) display a combination of combing and a flat base, the estimated exterior diameters of which (0.31 and 0.30 m, respectively) are very large for horizontal hives (cf. Table 1). We do not know, however, whether the Isthmian upright hives were of the movable-nest or top-bar variety. Regardless of the specific type of beehive represented, these new examples of upright hives document a clear advance in the practice of ancient apicultureby providing greaterefficiency in the extraction of combs, portability,and ease of use in small spaces.What is particularlysignificant about the beehives from the Rachi settlement is their securely dated context in a village setting, where economic activities included the production of olive oil and textiles on a scale beyond domestic subsistence.58This evidence, together with the ethnographic parallels that we have adduced, provides a compelling argument for the use of the upright hive in the Corinthia by the 3rd century B.C. The Isthmus of Corinth was a busy port and a natural crossroads,an international gathering place and market. Beekeeping would have provided a significant source of honey and wax not only to sell, but also to use or dedicate at the Sanctuaryof Poseidon, the site of the Isthmian Games. In addition to the uses of honey and wax that were mentioned above, there were two purposes for which artists and artisans employed beeswax that have a special connection with votives at the Sanctuaryof Poseidon. First, painters needed wax for encaustic painting on votive panels, tablets, and walls. One of the events in the Isthmian Games in the Roman period was indeed devoted to a painting competition (Plin. HN21.83-84; 35.49,149).59 Second, beeswax was essential for the manufacture of bronze objects by means of the lost-wax process. Debris from the casting of small bronzes has been found at many of the major Greek sanctuaries,including Nemea, Olympia, and the Isthmus.60Ancient Corinth and the Isthmus are mentioned a number of times in myth and other testimonia in connection with
374
V. R. ANDERSON-STOJANOVIC
AND
J. E. JONES
the Greek words meaning hive (see LSJ, s.v. xunorkiX)and honey (see LSJ, s.v. [lJtX).61 The Isthmus of Corinth may have had a long tradition of apiculture, in keeping with the association of Corinth with advances in craftsmanship.62When in use, these beehives were probablyplaced in availablespots in the settlement, or scatteredjust as they are today on the slopes of the Rachi, which were then, as now, covered with fragrant purple flowering thyme in late spring and early summer.
ACKNOWLEDGMENTS We express here our thanks to Elizabeth R. Gebhard, Director of the University of Chicago Excavations at Isthmia, for permission to publish the beehives from Isthmia. Photographs are the work of I. Ioannidou, L. Bartzioti, M. Huysing, and I. Heystek, with the exception of Figures 4 and 17, which were taken byJ. Ellis Jones. The majority of drawings used in this article are by Jones. Figure 16 is reproducedby permission of Tony Brothers of the Department of Classics and the Founders'Library of the University of Wales, Lampeter.Figure 12 was drawnbyJohn W. Hayes, to whom we are also grateful for information on the beehive fragments from the Sanctuaryof Poseidon. We also thank Richard Evershed and Sarah Vaughan for their work on the analysis of fabric and residue of samples of the beehive fragments from Isthmia, and William Morrison of Shippensburg University for information on the habits of bees. Anderson-Stojanovic owes thanks to Wilson College for sabbaticalsupport and for funds from the Drusilla Stevens Mazur Professorship,which made possible her work at Isthmia with Jones during March of 1999. Mary Skarbelis, Faculty Secretary,was most patient in typing a difficult manuscript. We also would like to express our thanks to the anonymous Hesperiareviewersfor their helpful suggestions.
61. The CorinthiantyrantKypselos was savedfrom death when his mother hid him in a chest (xurCoieXv) (Nic. Dam., FGrHist 90, F 57; Polyaen. 5.31). The relationshipof this term to the name of the Corinthiantyrant Kypselosand to the Thraciantown of Kypselahas alreadybeen remarked upon by Kardara(1961, p. 265), who comparesthe Orestadavase (5) with images of containerson the coins of Kypsela;the handles depicted on the coins, however,areplacedvertically ratherthan horizontallyas on 5. It is possible that the orientationof the handleswas changed for the sake of the clarityof the design;Ure 1922,
pp. 198-214. Herodotos (5.92) tells us that the name of the wife of Kypselos's son Perianderwas Melissa, or honey/ bee maiden. See RE XXIX, 1931, cols. 524-528, s.v.Melissa (Van der Kolf); Larson1995, pp. 352-353. Priestessesof Demeter and Korewere also called [EzXIooTx (Apollod., FGrHist 244, F 89; Kallimachos,Hymn2.110), and honey was consideredan appropriate offeringfor these two goddesses, who had a shrineon the slopes of Acrocorinth(CorinthXVIII.3), as well as on the summit of the Rachi (Anderson-Stojanovic,forthcoming). We know from Servius(Aen.1.430) that Melissa was the name of the
priestessof Demeter who took part in the Thesmophoriaat the Isthmus. There was also a town in the Corinthia named Melissos, home of a famous Corinthianfamily (RE XXIX, 1931, cols. 529-530, s.v.Melissos [W. Nestle]; Wiseman 1978, p. 10). Finally,the name Melikertes,the boy hero in whose honor the Isthmian Games were established,means "honey-cutter,"and honey,as a symbol of immortality,was often associatedwith heroes (RE XXIX, 1931, cols. 514-520, s.v.Melikertes [A. Lesky]; Gebhard1992, pp. 73-74). 62. Wiseman 1978, p. 13.
ANCIENT
BEEHIVES
FROM
ISTHMIA
375
REFERENCES AgoraV = H. Robinson,Potteryof theRomanPeriod:Chronology (AgoraV), Princeton 1959. AgoraXII = B. A. Sparkesand L. Talcott, Blackand Plain Potteryof the 6th, 5th, and 4th CenturiesB.c. (AgoraXII), Princeton 1970. Anderson,J. K. 1954. "Excavations on the Kofina Ridge, Chios,"BSA 49, pp. 123-182. Anderson-Stojanovic,V. R. 1993. "AWell in the Rachi Settlement at Isthmia,"Hesperia62, pp. 257302. .1996. "The Universityof Chicago Excavationsin the Rachi Settlement at Isthmia, 1989," Hesperia65, pp. 57-98. . 1997. "AThird-Century B.C. Deposit from the South Slope Cistern in the Rachi Settlement at o SZvIsthmia,"in A'E7rmac^oveix dvcrqrno ytcarq EAa7vtcrtx Kepaytxy:XpovoioytxdI1po/piaara-KAEtoirzd S6voAa--Epyaoacpta,
S. Drougou et al., eds., Athens, pp. 13-19. .1998. "Dye Works or Olive Press:A Reconsiderationof Installations in the Rachi Settlement at Isthmia,"AJA102, p. 323 (abstract). . Forthcoming."The Cult of Demeter and Kore at the Isthmus of Corinth,"in Peloponnesian Sanctuariesand Cults:Proceedings of the 9th InternationalSymposium at the SwedishInstituteatAthens, 11-13 June 1994, R. Hagg, ed., Stockholm. Bintliff,J. L., and A. M. Snodgrass. 1985. "The Cambridge/Bradford Boeotian Expedition:The First FourYears,"JFA12, pp. 123-161. Bonet Rosado,H., and C. Mata Parreio. 1997. "The Archaeology of Beekeeping in Pre-Roman Iberia,"JMA10, pp. 33-47. Broneer,0. 1958. "Excavationsat Isthmia,Third Campaign,19551956,"Hesperia27, pp. 1-37. .1959. "Excavationsat Isthmia, FourthCampaign,19571958,"Hesperia28, pp. 298-343.
Catling, H. W., E. A. Catling, P. Callaghan,and D. Smyth. 1981. "Knossos,1975: Minoan Paralipomenaand Post-Minoan Remains,"BSA 76, pp. 83-108. CorinthVII.2 = D. A. Amyx and P. Lawrence,ArchaicCorinthian Potteryand theAnaplogaWell (CorinthVII.2), Princeton 1975. CorinthVII.3 = G. R. Edwards, CorinthianHellenisticPottery (CorinthVII.3), Princeton 1975. CorinthXVIII.2 = K. W. Slane, The SanctuaryofDemeterand Kore:The RomanPotteryandLamps(Corinth XVIII.2), Princeton 1990. CorinthXVIII.3 = N. Bookidis and R. S. Stroud, TheSanctuaryof Demeterand Kore:Topography and Architecture (CorinthXVIII.3), Princeton 1997. Crane,E. 1983. Archaeology of Beekeeping,Ithaca. Crane,E., and A.J. Graham.1985. "Beehivesof the Ancient World," Bee World66, pp. 25-41, 148-170. Davis,J. L. 1996. "APageTurnsin the History of Greek Regional Studies," rev.ofJameson, Runnels,and van Andel 1994, inJRA 9, pp. 458-465. EretriaVIII = P. Ducrey,I. R. Metzger, and K. Reber,Le quartierde la Maisonaux mosaiques(EretriaVIII), Lausanne1993. Evershed,R. P., and V. R. AndersonStojanovic.1995. "Identificationof Beeswaxfrom Ancient Beehives by Means of GC,GC/MS Analysis of Absorbed Organic Residues,"AJA 99, p. 323 (abstract). Evershed,R. P., V. R. AndersonStojanovic,S. N. Dudd, and E. R. Gebhard.Forthcoming."New Chemical Evidence for the Use of Combed Ware in PotteryVesselsas Beehives in Ancient Greece,"JAS. Evershed,R. P., S.J. Vaughan,S. N. Dudd, andJ. S. Soles. 1997. "Fuel for Thought? Beeswaxin Lamps and Conical Cups from Late Minoan Crete,"Antiquity71, pp. 979-985. Forbes,R. J. 1966. Studiesin Ancient 5, Leiden. Technology
Francis,J. 1999. "Beehives,Honey, and the Whole Ball of Wax,"AJA 103, p. 268 (abstract). Fraser,H. M. 1951. Beekeepingin Antiquity,London. Free,J.B., and I. H. Williams. 1981. "The Attachment of Honeybee Combs to Sloping Hive Sides and Side-Barsof Frames,"Journalof ApiculturalResearch20, pp. 239242. Gebhard,E. R. 1992. "The Early Stadiumat Isthmia and the Founding of the Isthmian Games,"in of an International Proceedings Symposiumon the OlympicGames, 5-9 September 1988, W. Coulson and H. Kyrieleis,eds., Athens, pp. 73-79. Gebhard,E. R., F. P. Hemans, and J. W. Hayes. 1998. "Universityof Chicago Excavationsat Isthmia, 1989: III,"Hesperia67, pp. 405456. Graham,A. J. 1975. "Beehivesfrom Ancient Greece,"Bee World56, pp. 64-75. Gregory,T. E. 1985. "AnEarlyByzantine Complex at Akra Sophia," Hesperia54, pp. 411-428. Hayes,J. W. 1983. "The Villa Dionysos Excavations,Knossos:The Pottery," BSA 78, pp. 97-169. Ifantides,M. D. 1983. "The MovableNest Hive: A PossibleForerunner to the Movable-Comb Hive,"Bee World64, pp. 79-87. IsthmiaV = T. E. Gregory,TheHexamilionand theFortress(IsthmiaV), Princeton 1993. Jameson,M. H., C. N. Runnels,and T. H. van Andel. 1994. A Greek The SouthernArgolid Countryside: from Prehistoryto thePresentDay, Stanford. Jones,J. E. 1976. "Hivesand Honey of Hymettus:Beekeeping in Ancient Greece,"Archaeology 29, pp. 80-91. .1990. "AncientBeehives at Thorikos:Combed Pots from the Velatouri,"in ThorikosIX: 1977/ 1982, H. F. Mussche,J. Bingen, J. E. Jones, and M. Waelkens,eds., Ghent, pp. 61-71.
V. R. ANDERSON-STOJANOVIC
376
Jones,J. E., A.J. Graham,and L. H. Sackett. 1973. "AnAttic Country House below the Cave of Pan at Vari,"BSA 68, pp. 355-452. Kardara,C. 1961. "Dyeing and WeavingWorks at Isthmia,"AJA65, pp.261-266. Kardara,C., and P. Papadopoulou. 1984. "Oaliqpopeo6OPEETAAA," ArchEph1984, pp. 1-4. Klek,J., and L. Armbruster.19191921. Die BienenkundedesAltertums (Archivfir Bienenkunde1.6,2.7, 3.8), 3 vols., Leipzig. Larson,J. 1995. "The Corycian Nymphs and the Bee Maidens of the Homeric Hymn to Hermes," GRBS36, pp. 341-357. Lohmann, H. 1993. Atene.Forschungen zu Siedlungs-und Wirtschaftsstruktur desklassischen Attika, Cologne. Ludorf,G. 1998-1999. "Leitformen der attischenGebrauchskeramik: Der Bienenkorb,"Boreas21-22, pp. 41-130. Mee, C., and H. Forbes.1997. A Rough and RockyPlace:TheLandscapeand SettlementHistoryof theMethana Peninsula,Greece,Liverpool. Munn, M. 1993. TheDefensesofAttica: TheDema Walland theBoiotianWar of 378-375 B.C.,Berkeley.
TO Pologiorge,M. I. 1998. MvYrl7pia oVunxo6vexpoTacpCeooTO2Qp&noov (ArchDelt63), Athens. Pritchett,W. K. 1956. "The Attic Stelai, Pt. II,"Hesperia25, pp. 178-317.
WILSON COLLEGE OF FINE
ARTS
PENNSYLVANIA
CHAMBERSBURG,
edu
vstoj anovic@wilson.
J. EllisJones 2 FRONHELOG,
SLING
TREGARTH BANGOR,
GWYNEDD
LL57 4RD NORTH BRITAIN GREAT
john@ell
WALES
isjones.freeserve.co.uk
J. E. JONES
Raab,H. A. 2001. RuralSettlementin Hellenisticand RomanCrete.The AkrotiriPeninsula(BAR-IS 984), London. Rostoker,W. M., and E. R. Gebhard. 1980. "The Sanctuaryof Poseidon at Isthmia:Techniquesof Metal Manufacture,"Hesperia49, pp.347-363. Rostoker,W. M., M. McNallan, and E. Gebhard.1983. "Melting/ Smelting of Bronze at Isthmia, Greece,"Journalof theHistorical MetallurgySociety17, pp. 23-27. Rotroff,S. I. 2001. "ANew Type of Beehive,"in M. Lawall et al., "Notes from the Tins: Researchin the Stoa of Attalos, Summer1999,"Hesperia70, pp.176-177. .2002. "UrbanBees,"AJA 106, p. 297 (abstract). Ruttner,F. 1979. "Minoischeund AltgriechischeImkertechnikauf Kreta,"in Apimondia:Bienenmuseumund Geschichte derBienenzucht.Internationales Symposium iiberBienenwirtschaft, Freiburgi Br.BR Deutschland,16-18August 1977, Bucharest,pp. 218-225. Sackett,L. H., ed. 1992. Knossos: FromGreekCityto RomanColony. Excavationsat the Unexplored MansionII (BSA Suppl.21), 2 vols., London. Salmon,J. B. 1984. WealthyCorinth, Oxford.
VirginiaR. Anderson-Stojanovic DEPARTMENT
AND
17201
Seeley,T. D. 1995. The Wisdomof the Hive: TheSocialPhysiologyofHoney Bee Colonies,Cambridge,Mass. Sutton, R. F.,Jr.1991. "CeramicEvidence for Settlement and Land Use in the Geometric to Hellenistic Periods,"in LandscapeArchaeology as Long-TermHistory:Northern Keosin the CycladicIslandsfrom EarliestSettlementto ModernTimes (MonumentaArchaeologica16), J. F. Cherry,J. L. Davis, and E. Mantzourani,eds., Los Angeles, pp. 245-263. Thompson, D. B. 1963. GardenLoreof AncientAthens(AgoraPicBk 8), Princeton. Ure, P. N. 1922. TheOriginof Tyranny, Cambridge. Vogt, C. 2000. "The EarlyByzantine Pottery,"in IlHporopSoavrtvi E2e6Oepva1.2, P. G. Themelis, ed., Rethymnon,pp. 37-205. Wheler, G. 1682. A Journeyinto Greece, London. Whitbread,I. K. 1995. GreekTransand portAmphorae:A Petrological Athens. Study, Archaeological Wiseman,J. 1978. TheLand of the AncientCorinthians,Goteborg. Wright,J. C.,J. F. Cherry,J.L. Davis, E. Mantzourani,S. B. Sutton, and R. F. SuttonJr. 1990. "The Nemea ValleyArchaeologicalProject: A PreliminaryReport,"Hesperia59, pp. 579-659.
HESPERIA 7I (2002) Pages 377-397
O OF
N
E
H
U
N
D
R E D
H
E RO
E S
PHYLE?
AB STRACT Reconsideration of Archinos'sdecreefor the heroesof Phyleshowsthat it honoredmenwhowithstoodthe siegeof theThirtyverysoonafterThrasyboulos'smentookthestronghold.It is notevidence,asis sometimesclaimed, thatThrasyboulos's forceswereoverwhelmingly foreign.On the contrary, at leasthalfwereAthenians,andthe mostconservativerestorationof the decreesuggeststhatalmostallof Thrasyboulos's troopswereAtheniansat this point.Archinos'sdecree,however,doesnothonoronlyAthenians.Although theirpresenceis often overlooked,one to threeof the menlistedwerecertainlyeithermeticsor Eleutherians. Finally,A. E. Raubitschek's earlypositionthatthe decreehonoredover100mendividedinto twolists (withforty orso foreignersincludedin a second,lost,list)remainsmostlikely. 1. I wishto thankMichaelJameson forinsightfuldiscussions on thisdecree, Treasa Beyerforherresearch assistance, andthe anonymous Hesperiareviewers fortheirnumeroushelpfulsuggestions. Myresearch wassupported by a grant fromLoyolaCollegein Maryland. This article is dedicatedto ToniRaubitschek, fromwhoseteachingI was lucky to benefit. 2. Arginoussai: SeeAr.Ran. 693694with scholion.The scholiast quotes Hellanikos,FGrHistIII 323a fr.25.See alsoAr.Ran. 33-34;Andoc. 2.23,1.149; andDiod. Sic.13.97.1, whomentionsthe naturalization of metics. Osborne1983,pp.33-37, T10,hasa thoroughdiscussion.See also Hunt1998,pp.92-95. Samos: Meiggs-Lewis 94;Tod 1946,no.96. 3. See alsoDem.24.135;Kratippos, FGrHist 64 T2.
SoonafterThrasyboulos andhisarmyof thePeiraieus madetheirtriumphant entryintoanAthensnewlyfreedby theireffortsfromthetyranny oftheThirty,Thrasyboulos triedto rewardhis non-Athenian supporters richly.l He proposed togiveAtheniancitizenship to allhisforeignfighters whether theyhadjoinedhimearlyat Phyleor lateat the Peiraieus, even though "someof themwereclearlyslaves" (Arist.Ath.Pol.40.2).Inthishe followed thegenerous precedents setinthecrisisyears ofthePeloponnesian War, whentheAthenians grantedcitizenship to theslaveswhofoughtfor themat Arginoussai andofferedcitizenshipto theirloyalallies,the Samians.2 Thrasyboulos wasnotsuccessful in hisefforts,however. A certain Archinos of Coele,a fellow-member of thearmyof thePeiraieus, andone whohadbeenwithThrasyboulos at Phyle,indictedThrasyboulos and blocked his measure on a chargeof illegalitybecauseit hadnotfirstgone through the(asyetnonexistent) Boule(Arist.Ath.Pol.40.2).3At probably thesametime,Archinosproposedhis ownbill to honorthosewho had restored the democracy. Thisbill conveyedmuchmoremodestrewards, however, andgavethemnotto allof Thrasyboulos's supporters, butonly tomenwhohadfoughtwithThrasyboulos atPhyle-and notevento all ofthese(Aischin.3.187-190).
os
-
fi
-
n
fi
a
378
N
-
>
MARTHA
b
fi
C. TAYLOR
THE DECREE OF ARCHINOS Fragments of Archinos's decreewerediscovered in the AthenianAgora (Fig.1).4 The documentis an extremely valuablesourceforourunderstanding notonlyof theforeigncomposition ofThrasyboulos's earlyforces, butalsoof Athenianattitudes towardtheirforeignbenefactors. Thedocumentis, however, anambiguous witness.Unjustifiably, recenttreatments havemutedits uncertainties, withscholarsin somecasessuggesting that we knowasfactthatwhichis onlyinterpretation, andin othersasserting pointsthatcontradict whatseemsquiteclearin thedecree.Morerecently, therehasbeena trendto discount, withoutargument, theimportant literaryevidenceaboutthe decree,leadingto markedly differentinterpretationsof thedecreethantheoneproposed byA. E. Raubitschek somesixty yearsago.5Yetkeyelementsof thatoriginalinterpretation arealmostcertainlycorrect.As Raubitschek proposed,Archinoshonoredindividuals whojoinedThrasyboulos earlyat Phyle,not thosewhoparticipated in a laterbattle-and so his bill doesnot offerevidencethatThrasyboulos's armywasoverwhelmingly foreign,asPeterKrentzwould argue.6 Further4.Agorainv.nos.I16,I 16b,I 17, more,although oftenoverlooked, goodevidenceexiststhatArchinos hon- I 18,I 93. SeeMeritt1933,pp.151oredon his decreeat leasta few menwhowerenot Atheniancitizens. 155;Raubitschek 1941,pp.287-295, Finally,Archinosmaywellhavehonoredtheslightlymorethan100for- no.78. 1941. eignandAthenianindividuals alludedtobyAischines, justasRaubitschek 5. Raubitschek 6. Krentz 1982, p. 84. originally proposed. Raubitschek failedto providea detailedargument in 7. Raubitschek 1941,p.294. supportof hisinterpretations, however, andArchinos's important decree Thereis nojoinbetweenthetop of hasnotreceived thefulltreatment it deserves. fragmentB andthe bottomof fragAt the heartof anyinterpretation of the decreeof Archinosaretwo mentC.Theremusthaveoncebeen questions: 1)whateventis commemorated on thestone a siegeatPhyle at leastone additionallinebetween verysoonafterThrasyboulos captured the stronghold or a battleat (or thesetwofragmentsbecausethe topnear)AcharnaiwhenThrasyboulos commandedsome 700 men;and mostpreservedlineson fragmentB recordthe namesof individuals 2) howmanymenarehonoredon the stone-the fifty-fiveor so menin (Fig.2).Theselinesmusthavehad the preserved list aloneor (to followAischines)somewhatover100 di- abovethema headingfortheirtribe videdintotwolists(thepreserved listanda secondlistthatis nowlost). Ksxtoowebos. Theirtribalaffiliationis The fivepreserved fragments of the decreecontainpartsof a list of clearbecausethe fourlinesinscribed namesarranged byAttictribe,andinscribed in twocolumnsabovea four- withnamesarefollowedbythe tribal Isso0X[vCdos],the tribe line epigramandthe textof the decreeitself.Partsof bothcolumnsof heading thatfollowedKekropis in Athens' namesarepreserved butonlythe firstfewlettersof the fourlinesof the officialtribalorder.Furthermore, the epigramandtheveryfirstlinesof the decreearepreserved (Fig.1).A1- namesthatRaubitschek's restoration thoughonlya fewlettersof theepigramarepreserved, theyareenoughto placesabovethem(allon fragmentC) allowidentification with a poemquotedby Aischines(3.187)as having belongclearlyto demesof Oineis, beenpartof Archinos's honorsforthemenof Phyle.Thiscorrespondence the tribebeforeKekropisin the tribal order.The fournamesthatbeginfragthenidentifiestheAgoradecreeasbeingthatof Archinos. According to mentB, therefore,musthaveoncehad Raubitschek's restoration therewereat mostfifty-eightcitizenson the Ksxtoowebos inscribedabovethem,as mainlistofthedecree(Fig.2);thestoneitselfgivesclearevidenceforonly Raubitschek hasrestoredin column2, fifty-fournames.7Aischines (3.187)implies,however, thatover100men line49.Abovethis,however,Raubiwereinvolvedin the honorsArchinosproposed, becausehe saysthatthe tschekhasrestoredan additionalline a namein the tribeOineis,of which 1,000drachmas provided to thegroupforsacrificeanddedicatory offer- for thereappearfromhis reconstruction ingsresultedin anawardof lessthantendrachmas perrecipient(£ypaQ£ andthe publishedphotographto be no O£ ZrO@TOV£V
avTog £tg uvotav zat avanyaTa
TOVT'£aT6V £Raov
oouvat xAtag oRoaXRaq,zat
remainson the stone.I havenot myself
n 8£xa 8RoaXRatzaT' av8Roa). Forthisreason, Raubitschek seenthe stone,butif it is truethatthere
ONE HUNDRED HEROES OF PHYLE?
379
conoriginally to assumethatthe monument arguedthatit is "necessary thatthissecondlistcontainedanother(a second)list,"andhe proposed whoreceivedthesame tainedthenames"ofmorethanfortynon-citizens "may listed."The secondlist, he suggested, honorsbutwereseparately belowthedecree."8 havebeeninscribed thateventhe of thedecreestronglysuggest,however, The fragments Atheniancitionthestonehonoredmenotherthanregular listpreserved undertribal listof namesinscribed zens.At theveryendoftheright-hand This emptyline is followedby a headings,one line is left uninscribed. epsilon,indentedslightlyto theright,and,in thefirstline singlepreserved belowit, the beginningsof an inscribedname.BenjaminMerittunderCitingthe stoodthis indentedepsilonas the beginningof a heading.9 menundertheheading£vyRoafp[o] listthatrecorded ofacasualty precedent attheendof a listof Atheniandeaddividedbytribe(IGI2949,lines7678), Meritttookthis decreealsoto be a casualtylist andproposedthe andMerittapparhere.If, as Raubitschek £[yyRoafpot, samerestoration, thereshouldhavebeen stroveforsymmetry, theinscription entlybelieved, underthisheading(Fig. three)namesinscribed atleasttwo(andprobably wouldindicatethatthe twoorthreenamesthat 1).1°Meritt'srestoration ontheother followtheepsilonin line69 arethoseof metics;Raubitschek, indior'E[Xrv0rRorg], to restorethelineas'E[Xrv0rRooe0rv] hand,proposed catingthatthemenlistedin lines70-72werefromthe non-demevillage of Eleutheraiin northwestAttica.llThis restorationis necessaryto includeda secondlistof nonviewthatthedecreeoriginally Raubitschek's metics, on citizens.If lines70-72 of thelistpreserved thestonerecorded wouldindicate,therewouldbe no needfora sepaasMeritt'srestoration of line69, ratelistfornoncitizenheroes.On eitherproposedrestoration citizens.l2 Athenian more than includes decree Archinos's however, enoughof its headingto thedecreedoesnotpreserve Unfortunately, leeventual subsetofThrasyboulos's of theparticular allowidentification thatwas honoredby Archinos.The firstline of the gion of supporters thatmentionsPhyle.But headingendswithanHN, allowinga restoration thepossiblefieldonlydownto thosemenwho thiswouldseemto narrow areno remainsof thislineon the stone, Becausethe thisline is not required. Raubitschek's stoneis symmetrical, linein the of anunnecessary restoration secondcolumnrequiresone in the first columnaswell.The stone,thatis, seemsto giveclearevidenceforonly fifty-threeAtheniansin the mainlist. andMerittreadthe beRaubitschek ginningIIof one additionalnameat the endof the list undera nontribal heading(seebelowformoreon this heading).Thus,the stonehonoredat leastfifty-fourmen. 1941,p. 294. 8. Raubitschek 9. Meritt1933,pp.154-155. andMeritt 10.As Raubitschek
the fragmentsof the decree, arranged wouldcallforthree exactsymmetry namesbelowthisheading,and (1941,p. 294) restored Raubitschek the decreethus(Figs.1-2). Meritt (1933,pp.153-154)assumedonly two namesunderthe heading,meaning thatthe right-handcolumnwould be one line shorterthanthe column on the left.Thus,althoughthe stone givesclearevidenceforonlyfifty-four (fifty-threeAtheniansand honorands one manlistedunderE[ ]), symmetrysuggeststhattherewasin additionat leastone morenamelisted underthisheading.We canbe reasonablycertainthatthe stoneoriginally
namedat leastfifty-fivehonorands. 11.Meritt1933,pp. 154-155. 1941,pp.288,293.I Raubitschek (1978)in the followVanderpool of places assignmentof identifications in the northwestborderregion: fortandvillage= Gyphtokastro Myoupolis= Oinoe.The Eleutherai; discoveryof an ephebicdecreeat the in the Skourtaplain fortaboveKavasala of thatsite confirmsthe identification anddisprovesthe as Panakton, of sometimesadvanced identification withPanakton.See Gyphtokastro Munn1989. MunnandZimmerman of 12.I discussthe ramifications below. eachof theserestorations
ATHET °A°S/\Y
I
/\ k°ETE E °
IE
E
IM TI
EA E
380
Yk
hE
MARTHA
0
I
T
0
E P E xoH
A
E
A
N I to
k H
C. TAYLOR
A A B oN
A TA M 0
N
AT
T E E + r A CON
H
fiagment
/
)
/
I NH I 80E
O
E
Av
Y /AY
rl C 9
6E
AST
f>/A VSAX
V AP WIE Y S
AP S E YS I ALllE I E P I ° ¢ Y A A f I Of I k P ATO E E Y^AS I O S
r P°hAk
y A 110 + Y hA S I O S >wH \ ° m Y A A E I o t
Q
ggO
A I CH I A o
<
>
E
kE
fragment B
rSal
\
P S EY t
o N z°t
r rA P r
O PA SY B
C4/1 A /NlLop MENIoSy XkYoHP)P
A E rtOT I, A0
E
IEY s oBA \lElot YPPI oUloS
lboACUnEAHO
ANT
I
ox
3 toU
S
t
1°E
t Ag ro+N k N+N
/ / jToPA I 0K1 tHE ) P 0N0E iP E A P P 1 +E O E O k tH S hE H I O SONgEY E j /L I o N Y S 130 A kAMAN T LE O E / HoEAIoAnM0
E
{ rnoo tSTILoS , p x X N E M Y Pn b1 1 A ° E k 1<° I /\ H S ° I 19H I dHE A I ANT I to > A FJAp o
X /
/ r
g
I Lo
9
A P,TT° H ZT I )/
/
K P oP
B( tAotHMAX MEt HS I D E I TEA I fi ATHK > /
/ <
llEHt
M
E YO E PAO E N
/ /
/A P ET HE E N E k A S T E + A N O I E E r E P A I P E r AA A I x OA N HFo E A(3HN A I < N o I r o T ETo > AA I g o I X EA^O I sa P IAN TAS rO/\EnS rP TO I k ATAPAY E N iANk I NLYNONEh/\ASI NAPA/v\ENoI AI Bo/sH I kA IT
A O
I so+
foughtwithThrasyboulos beforehe movedto the Peiraieus. Sinceour sourcessaythatThrasyboulos hadat least1,000menwith him by this time (Xen.Hell. 2.4.10-11),the reference to Phyledoesnot sufficeto identifywhichmenatPhylewerehonoredon thedecree. Raubitschek arguedthatthecitizenandforeignrecipients he located onArchinos's decreeincludedtheseventymenwithwhomThrasyboulos leftThebes,accordingtoXenophon (Hell.2.4.2),andanadditionalgroup of morethanthirty,whojoinedhim almostimmediately at Phyle"even beforetheOligarchs wereableto launchtheirfirstattack.''l3 Raubitschek thuslocatedthecutoffpointforinclusionin Archinos's honorsveryearly inThrasyboulos's timeatPhyle.On Raubitschek's understanding, thedecreedoesnot distinguish betweenthe initialseventyandthe morethan
Figure1. Inscriptionhonoringthe heroesof Phyle.AfterRaubitschek 941,p.289,fig.1
13. Raubitschek 1941,p. 284. Even beforethe discovery of the decree, Cloche(1915,p. 15) hadreconciled in thiswayXenophon's numberof seventyforthe sizeof thebandwith whichThrasyboulos leftThebesand Aischines'implication thatArchinos honorednot seventybutover100 men.
25
nt
[ [[ l6 1?15 14 16 ............... ............... ],urvo5 TI]atav[Crv]g ]o ]fi)V68o KvOrlto[otog] ]vTo ]ppovos M[vtotoe]vootog fIaeov[C8rlg] Hto[oa]Xatos q>[prap06]os
40 45 65B'Av[ [ [ 11 )<[ [[[fIok[v Kv[topxv ..... 17 11 t]20 t r]Bavxe7r7r0 Ev8N,uo ]os'Ioto t1.0 'A]xatovrug Etsat]os 12 ]] 'Aqv]vs Mso]ve[o]8NFo 'Axatov£vs q>vXaseos q>vXaseos OvXaseos Aaxta8Yls - - -]
ONE HUNDRED HEROES OF PHYLE? [o
't
8
X
a
X
a
X
a
X
o
v
T
C
5
@
[T
o
v
N
F
o
v
X
a
X
N
y
a
y
o
v].
10
[
]
[ [ [ [
] ] ] ]
[
]
[ [
] ]
v
B]
N
v
[OLV!t805]
[ Etocx0Ntbo5]
5
38I
[. . . . . . . .17. . . . . . . .]o 'Axatovrug
[
entC
'A]vCztoaTosq>vXaseos
10
[
14
* * .]fi)vo5
[AtyNtAo5]
15
[
]
[
]
[
]
[
]
lOp
[
]
Sagme
50 jt /
E
[
55
]
[. . . . . . . . .18 . . . . . . . .rato]n[=t]o5 [ . . . . . . . .15 . . . . . . rato]TTto5 [TIav8tov68oS]
[Otoavv,BoBo5 Avxo SrO]CrV[5]
Ftxvx
[v-
-
-
-
Evapop[-- - - -]
01
[. . . . . . . , . . . . .26 . . . . . . . . . . . .]5
20
[Kxpo7re80s] [. .]oxB[- - - - -]
60
-]
'I7r7r00X[vt80s] 'Atoxcuo[g MvtoxveAo£)t OLVY16[8Y15 ] Atav[taos] 'Av8too[-- - - -] EvxAs[t- - - - -] [T]tllo[-----]
KoLN5]
[ A9Tt°Xt8°5]
[ [
15
]
t[
...]8o'A7[rN0]\
16
. .]t[-
- -
]
Av[. . . . . . . 1.2. . . . ,13]oBo S[rlsuaxe8Ylg] 30
[A@vtAos]
vacat
[OroxXYls Arv]xto Sove[vs] [ .
35
.
.
. 9.
.
.
.
.]
/\ touvoe
[o-
-
-]
70
['Axa,uavTC]8og [....7...]05 [- - . .xp]av5
tt0800[0-
- -]
TLRO[XA£05
Figure2. Text of the inscription honoringthe heroesof Phyle. AfterRaubitschek 1941,p. 288
14. Krentz1982, pp. 83-84, n. 54.
Et£at]05
'E[XcvOstoa0sv] fI[--] [ ] [ ]
thirtylatervolunteers. It separates citizens(andEleutherians) andnoncitizensinto two lists,but doesnot distinguish betweenThrasyboulos's firstsupporters in Thebesandthosewhojoinedhim onlyafterhe had takenPhyle. Krentz,however, hasproposeda differentunderstanding of thisdecree,which,if correct, wouldindicatethatThrasyboulos's dependence on noncitizen forceswasdramatically greater, andtheAthenians' willingness to acknowledge thatdependence dramatically less,thanRaubitschek suggests.14Like Raubitschek, KrentzfollowsAischines'implicationthat Archinosoriginally honoredover100 men,andalsoacceptsthatthere wereoriginally twolistson the stele.But Krentzbelievesthatbothlists honoredonlyAtheniancitizens,andin themhonoredmostof theAthenianswhoservedwithThrasyboulos duringhisentiretimeat Phyle.
382
MARTHA
C. TAYLOR
THRASYBOULOS AND HIS ARMY A briefsketchof theactivities of thearmyunderThrasyboulos is in order. First,Thrasyboulos and,atmost,seventymenmarched fromThebesand seizedthe stronghold of Phylein northwest Atticawhenwinterhadalreadybegun(Xen.Hell. 2.4.2; Arist.Ath. Pol. 37.1).15 Xenophonreports thattheThirtyquicklymarchedout againsthimwiththeThreeThousandandthecavalry, butfailedintheirfirstattack. TheThirtythenplanned to reducethemenatPhylebysiegebutwerefrustrated in theirattemptby a snowstorm, andso returned to Athensafteronenightandpartof a day (Xen.Hell. 2.4.2-3). Severalmonthslater,according to Xenophon,the ThirtysentoutagainstThrasyboulos a forceconsisting ofthewholeofthe Spartan garrison andtwodivisionsof cavalry, "andstationedthemin the outlyingdistricts aboutfifteenstadesfromPhyle.''16 Bythistime,Thrasyboulos's forcehadincreased to about700men.Withthesevolunteers Thrasyboulosattackedanddefeatedthe menfromAthens,killingmorethan 120 hoplites.Thrasyboulos putup a trophy,collectedthe captured arms, andreturned to Phyle(Xen.Hell.2.4.4-7). Afterthis,when"about athousandmen"hadjoinedhimatPhyle,Thrasyboulos marched onthePeiraieus 15. Different sourcesgive different andseizedMounychia hill(Xen.Hell.2.4.10-11). numbers,rangingfrom thirty to Diodoros,forhispart,recountsanearlyfailedattemptbytheThirty Xenophon'sseventy,for the size of to besiegePhyle(14.32.2-3). He thenreportsthatthe Thirtylaterled Thrasyboulos'sinitial group:Nepos, "manytroops" out againstPhyleandpitchedcampat Acharnai. Thrasy- Thrasyb.2.1 sayshe had "notmore than thirty men";[Arist.] Rh.31. 8 saysfifty; boulos,afterleavinga"sufficientguard" atPhyle,led1,200menoutagainst Aristides (Lenz/Behr) 1.254 says a"few theThirty'sforcesandsurprised androutedthem.He thenmarched im- more than fifty";Paus. 1.29.3 sayshe mediately on thePeiraieus (14.32.6-33.1-2). As PaulClochepointsout, had sixty men;Xen. Hell. 2.4.2, Plut. it is unlikelythatThrasyboulos surprised theThirtytwicein a matterof Mor.345d (= Kratippos,FGrHist64 days,so XenophonandDiodorossurelydescribethe samesurpriseat- T2), and Aristides (Dindorf) 43.556 (vol. 1, p. 822 = Pseudo-Aristides tack.17 Therewere,then,two encounters betweenThrasyboulos andthe 25.65) all say he had seventy.Krentz Thirtybeforethedemocrats movedto thePeiraieus: 1) theThirty'sattack (1982, p. 70) compromisesby saying andattempted siegethatwasfrustrated bysnow,probably verysoonafter that Thrasyboulosleft Thebes with Thrasyboulos tookPhylewithatmostseventymen;and2) Thrasyboulos's about thirty men but reachedPhyle surprise attackon theThirtyseveralmonthslaterwhenhe hadabout700 with about seventy.These figurescertainly suggest that Thrasyboulosdid menwithhim(Xenophon) ormorethan1,200 (Diodoros). not have as many as 100 men with him PREVIOUS INTERPRETATIONS THE DECREE
OF
when he first took Phyle, so Raubitschek'sproposedtwo lists on Archinos'sdecreeshould probablynot honor only men involvedin that very first action.
KrentzarguesthatArchinos's decreecannotcommemorate the firstencounterbetweenthe forcesof Thrasyboulos andthoseof theThirtybecausethatencounterdoesnot accordwithAischines'description of the deedforwhichthesemenof Phylewerehonored.18 Inhisdiscussion ofthe decree,Aischines(3.187) indicatesthatin grantinghis honorsArchinos madeprovision fora ''carefill scrutinybythe Boule,to determine whoof themwasactuallybesiegedat Phylewhenthe Lacedaemonians andthe Thirtyattacked"axptg Trlv ovXrlvox£a,u£vrlv, osotavv £zt oVXn £z0Xt0R0XN0N^aV 0T£ Aax£8aLyovToL xat ot otaxovTa wRooo£aov. Dentz
16. See Krentz1982, p. 151, and Buck 1998, p. 75, for the date. 17. Cloche 1915, pp. 25-26. Krentz (1982, p. 89) and Ostwald (1986, p. 490) concur.Ostwald thinks that Diodoros is right that the battle was at Acharnaibecause"theoccupationof Eleusis will have made it difficultfor the men of Phyle to reachthe Piraeus by a routewest of Mount Aegaleos." 18. Krentz1982, pp. 83-84, n.54.
ONE HUNDRED HEROES OF PHYLE?
383
emphasizes thatin thefirstconfrontation betweenThrasyboulos andthe oligarchs, theThirty'stroops"didnot includeanyLacedaemonians and wenthomeafteronenight."KrentzacceptsthatAischinesfaithfillly representsthetermsofArchinos's decree,andarguesthatArchinoscouldnot havedescribed thisfirstactionasa siegebytheLacedaemonians andthe Thirty.By Krentz's reasoning, the decreemustinsteadcommemorate the secondconfrontation betweenThrasyboulos's menandtheThirtywhen oursourcessaythatThrasyboulos hadat least700 men.19 Archinos's decree,which,according to Aischines, honorsonlysomewhat morethan100 men,clearlycannothonorall the democrats involvedin thisbattle;if it doescommemorate the surpriseattack,it musthonoronlysomespecial portionof the democrats present.Krentzarguesthatthelogicalsubsetis theAtheniansinThrasyboulos's army. On Krentz's reading, then,thehonorands of thedecreerepresent the totalAtheniancontribution toThrasyboulos's armyatthetimeof thesurpriseattack.According to Krentz,thetwolistsdistinguished thecitizens onthebasisofwhentheyjoinedThrasyboulos's movement.The first(partiallypreserved) listof namesabovethe decreehonoredthoseAthenians whohelpedto capture Phyle.Thesecond(assumed) listrecorded theAthenianswhojoinedThrasyboulos atPhylebetweentheoccupation of Phyle andthesurprise attackon theThirty.20 If correct,thisreadingof thedecreewouldshowthatThrasyboulos's dependence on non-Athenian forceswasdramatic: onlysomewhatmore than100of 700 men(usingXenophon's numbers), orroughly15%,were Athenian; in contrast, almost600of 700 men,orroughly85%,wereforeignormercenary. If we useDiodoros's figureof 1,200forthedemocratic forcesin the battle,Archinos's decreewouldindicatethatlessthan10% wereAthenianand90%ormorewereforeign.21 Howlittlethedemocracy owedits rebirthto the Athenians! Indeed,Krentzconcludes,"the overwhelmingmajorityof [Thrasyboulos's] troopsat Phylewereforeign."22 Krentz's viewhasfoundacceptance,23 butit cannotstand.Krentzhasmisinterpreted Aischines,andoverlooked othercrucialevidenceconcerning thenumberofThrasyboulos's foreignsupporters. 19. Krentz(1982,pp.89-90) acceptsXenophon's figureof 700 for the sizeof the forceThrasyboulos led in attackagainsttheThirty.He then (1982,p. 91, n. 3) givesan"easyreconciliation" of Diodoros'sandXenophon's numbersby suggestingthatThrasyboulosleft Phylewith 1,000menand, gaining200 on theway,reachedPeiraieuswith 1,200.This interpretation glossesoverthe 500-mandifferencein the sizeof the forcethatXenophonand DiodorossaythatThrasyboulos hadin his attackon theThirty.It alsoseems to ignorethatDiodoros14.33.1says thatThrasyboulos led out his 1,200
afterhavingleft"asufficientguard" at Phyle,indicatingthathe hadmorethan 1,200menin total. 20. Krentzdoesnot discussthe apparent headingat the endof the triballist thatMeritttookto indicate the presenceof meticson the stone, andRaubitschek tookto heada short list of Eleutherians. SinceKrentz argues(1982,pp.83-84, n. 54) thatthe decreehonorsonlyAtheniancitizens, he presumably wouldpreferRaubitschek'srestoration, in whichthe men belowthe headingare,if not Athenian citizens,at leastmenwithveryclose tiesto Attica.Seebelowforfilrther
discussionof the implications of Raubitschek's restoration. 21.Thiswouldseemto be Ostwald'sunderstanding. Ostwald (1986,pp.489-490)creditsKrentz withthe ideathatonly100 of Thrasyboulos's forcewereAthenians,but arguesthatThrasyboulos's surprise attackon the forcesof theThirty shouldbe placedasDiodoroshasit, whenThrasyboulos is comingdown to the Peiraieus, leading1,200men. 22. Krentz1982,p. 84. 23. See,forexample,Ostwald1986, p. 489;and,mostrecently, Wolpert 2002,pp.25, 44, 109.
384
MARTHA
C. TAYLOR
Krentzbaseshisinterpretation onAischines' paraphrase of theprovisionsof thedecreeandhisbeliefthatArchinoscouldnotdescribe thefirst attackagainstthemenatPhyleasa siegebytheLacedaemonians andthe Thirty.This is partlybecauseKrentzfollowsAristotle's chronology and acceptsthattheThirtyreceiveda garrison fromSpartaonlylate.Thusit wasonlybythetimeof thesecondencounter betweentheThirtyandthe menfromPhylethattheThirtyhadtheuseof Lacedaemonian troops.In addition,Xenophon,whohastheThirtycallfora garrisonfromSparta almostimmediately, nevertheless doesnot saythatanyLacedaemonians wereinvolvedin the firstconfrontation betweenthe forcesof theThirty andThrasyboulos. I amwillingto useAischines' paraphrase asevidenceoftheprovisions ofArchinos's decree, butcannotacceptKrentz's interpretation ofAischines' reference to the Lacedaemonians. First,noteveryoneagreeswithKrentz thatAristotle's chronology shouldbe followed.24 If we followXenophon instead,andacceptthattheThirtycalledin a Spartangarrison early,the Lacedaemonians wouldloomas a forcebehindtheThirtywhetherthey werepresentatanindividual eventornot.ForthisreasonArchinosmight haveincludedthemastheputativeenemyof themenof Phyleevenif he wasdescribing the firstencounter betweentheThirtyandthe defenders ofPhyle.I doubt,therefore, thatthementionoftheLacedaemonians means thatthedecreecannotbedescribing theThirty'sfirstattackagainstPhyle. This is so especiallybecausethe descriptions by Aischinesand(apparently)Archinosof thedecree's honorands presentevengreaterdifficulties forKrentz's interpretation. As notedabove,Aischinessaysthatthedecreerequired theBouleto determine whowasactually besiegedatPhylewhentheLacedaemonians andtheThirtyattacked.25 Krentzfocuseson thepresence of theLacedaemoniansin thisdescription butoverlooks the rest.In particular he gives no attentionto Aischines' indication thatthemenhonoredon thedecree suffereda siegeandanattackat Phyle.In theencounter to whichKrentz wouldhavethisrefer, Thrasyboulos's bandattacked anddefeated theforces of theThirtyandthe entireSpartangarrisoneitherfifteenstadesfrom Phyle(Xen.Hell.2.4.4-7)orasfarawayasAcharnai (Diod.Sic.14.32.6). They werenot besieged;nor did the Thirtyattack.It is unlikelythat Archinoswouldchooseto describesuchmenmerelyas thosewhowith24. See,forexample,Rhodes1993, pp.416-422,455;and,recently, Munn 2000,p. 413, n. 15. 25. Aischines(3.187-190)describes the menhonoredbyArchinos's decree severaltimes.He firstremindsthe Atheniansthatthey"gavea gift to thosefromPhylewhobroughtthe demosbackfromexile"-£'80T£ 8Xp£aV °tS aso 44t)MS cp£i)yovta tov 8NZov xaTayayoi)ocv. This is too imprecise to serveas a chronological indicator becauseit mightreferto anyonewho hadat leastjoinedThrasyboulos before
he arrivedin thePeiraieus. The verses recordedon thedecree,however, indicatethatArchinoshonoredthe men whojoinedThrasyboulos first.They honorthose"whofirstbeganto stop thosewho onceruledthe citywith unjustlaws" ot =0T£ °0g a8tXotS A
_
,
S
W£aZ0tS ap ;avTag
S w°^tog
=p@T°t
Unfortunately, it is hardto knowhowto interpret thisphraseunlessoneknowsexactly whatgavethesemenpriority. The later decreegrantinghonorsto theforeignersin Thrasyboulos's army(IGII210) zaTawai)£ev pE,av.
dividesthehonorands into three groups,andseemsto haveplacedin the firstgroupallwhowerewith Thrasyboulos at Phyle,whetherthey camewithThrasyboulos fromThebes or arrivedmonthslater.So too might Archinos's decreecall"first" allthose whojoinedThrasyboulos beforePeiraieus.Krentzis, therefore, rightto focuson Aischines'moredetailed description of the menhonoredby Archinos,particularly becauseit seemsto describea procedure mandatedby the decreeitself.
ONE EIUNDRED HEROES OF PHYLE?
385
themen Ontheotherhand,in thefirstencounter stooda siegeatPhyle.26 at Phylewereattackedandbesieged if onlyfora nightandpartof the nextday althoughno sourcesaysthatthe enemyincludedthe Spartan garrison.
26. Krentz(1995,p. 140,s.v.2.4.2) callsthe secondencountera siege,but telS1S not a alrdescrlptlon. 27. IG II210,withthe additions of Hereward1952.See Rhodes1993, pp.476-477,andOsborne1982, of the date. pp.29-32, fordiscussions of the first 28.The description groupcontinues(in Osborne's[1981, "orto thosewho D6] restoration) towardsthe returned[gaveassistance returnto the Peiraieus]" °tS ..
.
.
.
.
.
eS rrv zaEA[0ooLsI)veXaD]l[ovTo zaOoAovrnv rLSHrpaLa,. . .]. Krentz
(1980,p.304), on the otherhand, restoresthe sentenceto read"orto thosewho returned[fromPhyle gavehelpby donatingmoneyor supzazA[0oo]l [aso plies]" YiToLs <)t)Y15 rio0Nsav Aours xPYlFata rstrr8zca . . . ].
Yi
29. Osborne1982,pp.41-42. 30. Osborne(1982,pp.32-35) thatthe firstgroup arguesvigorously receivedcitizenshipandthe othertwo, isoteleia. Krentz(1986),on the other hand,insiststhatallthreegroupsreceivedonlyisoteleia, andWhitehead (1978,1984),forhis part,arguesthat allthreegroupsreceivedcitizenship. The secondandthirdgroupsincluded thosewho"foughttogetherthe battle . . . ] oUvrFain Mounichia"-[OCTOl p7sav rr F Fap7vrr F Moulxiacylv
(cf.line 7 of the decreeandOsborne to the 1982,p.27), and(according on the stone)"those headingpreserved withthe demosin Peiwho remained xxl] rF raieus"-oi8r[s]apr[Frvov Ilrparl 8[Nl] (faceB, columnII).
ThewordingthatAischinesindicatesArchinosuseddoesnotexactly betweenthemenof Phyleand fit eitherthefirstorthesecondencounter the Thirty.Therefore,we mustchoosewhetherit is morelikelythat havebeen whomaynotactually wouldmentionLacedaemonians Archinos victoryin battleasmenwho presentor describemenwhowona surprise I believethe formeris morelikely,andthat werebesiegedandattacked. withtheThirty. firstencounter Thrasyboulos's thisdecreecommemorated anda laterdecreefor interpretation betweenKrentz's A discrepancy twoyearsafter thisview.Probably foreigntroopssupports Thrasyboulos's who honorsto the foreigners decree,the Atheniansawarded Archinos's of whichhavebeen The decree,onlyfragments foughtfordemocracy.27 dividedthehonoredmenintothreelistsunderheadcarefully preserved, to theThirty.The ingsthatindicatewhenthe menjoinedthe resistance as possible.It infirstgroupseemsto havebeendefinedas expansively cludes(line 4) all men "whocamedowntogetherfromPhyle" osot orotherassisouvxarikoovaso <>vnas wellasmenwhogavematerial participate did not themselves fromPhylebut tanceto thosewhoreturned layout studyof theprobable In a rigorousandconvincing in thereturn.28 thattherewerebetweenfiftyand116 of thisdecree,Osbornecalculates Theymayhavebeen seventyto ninetymenin thisfirstlist.29 andprobably honoredwithcitizenship.30 does of thesizeof thefirstgroupof honorands calculation Osborne's decree.The firstgroup, of Archinos's interpretation not fit withKrentz's to Osborne,should whichcomprisesno morethan116 men,according atPhyle(orgavehimmawhojoinedThrasyboulos includeallforeigners It includes,thatis, not terialaidat Phyle)beforehis marchto Peiraieus. attack,whenthe forcesof whofoughtin the surprise onlythe foreigners includedin butalsoanyforeigners wereatleast700-strong, Thrasyboulos between joinedThrasyboulos to Xenophon, the300 menwho,according (Xen.Hell.2.4.10).The attackandhismarchto thePeiraieus thesurprise materiwhomerelyhelpedThrasyboulos groupalsoincludesnoncitizens ally,but did not actuallyfight.We cannotknowhow manyof the 300 Xenophonreportsmayhavebeenforeign,norhow volunteers additional onlywithmoneyandnottheirbodmanyrichmeticsaidedThrasyboulos thatnotallofthe (atmost)116menlisted ies.Wecanbecertain,however, in the firstgroupon IG II210 a group,we shouldrecall,thatOsborne moreof a sizeof seventyto ninetymen-werewith wasprobably indicates attack.IG II210 indicatesthatThrasyboulos atthesurprise Thrasyboulos substantially withhimthen(andprobably hadfewerthan116foreigners readingof Krentz's fewerthanseventyto ninety),not the 600 foreigners decreewouldsuggest. Archinos's to the (at most)116 600 foreigners It is difficultto reduceKrentz's menhonoredonIG II210.We couldexplaintheabsenceof somemenby but oursourcesdo not indicatethat appealto deathon the battlefield, victorycosthimmanymen.We mightsuggestthatsome Thrasyboulos's
386
MARTHA
C. TAYLOR
menfailedto producethe witnessesrequired for confirmation; perhaps somesimplydidnotwantcitizenship orisofeleia inAthens(although men whohadfoughtforthedemocracy arelikelyto havewantedto sharein it). Butit is hardto imaginehowa groupnumbering over600waswhittled downto (probably) seventyto ninety,andat most116.31 Thisdiscrepancy anda morecarefulreadingof Aischines' description of theoccasionforArchinos's decreearguethatKrentz's interpretation of thedecreeiswrong.Thisdecreedoesnotcommemorate alltheAthenians involvedin the secondencounterbetweenThrasyboulos's menandthe Thirty,andis not,therefore, evidencethat"theoverwhelming majority of [Thrasyboulos's] troopsatPhylewereforeign."32 Krentz's studyfaltersin partfromhismisinterpretation of Aischines. In otherrecenttreatments of thedecree,scholarshavesimplydiscounted Aischines,andreachedconclusions at oddsnot onlywithAischines,but alsowiththepreserved fragments of thedecree.RobertBuck,forexample, assertswithoutargument thatArchinos's decreehonored"Thrasyboulos' firstbandof attackers," andclaimsthatthedecreecontains"locations for fifty-eightnamesin a spaceforperhapsseventy."33 He thussuggeststhat the decreeclearlyhonoredonlyseventy"first attackers" andthatthe decreeconfirmsXenophon'sfigureof seventyfor the forcewith which Thrasyboulos captured Phyle.Thispictureis tidy,butit is false. Buck's"spacefor perhapsseventy" namesis illusory.As mentioned earlier, Raubitschek locatedfifty-eightnamesonthestone-not seventyandthestonegivesclearevidenceforfifty-fournames.Symmetry argues thattherewereprobably at leastfifty-five.Thereis spaceon the stone belowthe two columnsof namesandabovethe epigram(Fig. 1), and Buckmayhavecalculated thatthis spacewouldhaveallowedan additionaltwelvenamesto be inscribed here,buttheywerenot;the stoneis blankbetweenthebottomof thefirstcolumnof thelistof namesandthe epigram. Therecouldbe morenamesin thelist,however, becausethereis a breakbetweenfragments B andC whereit wouldbe possibleto restore furthernames.Raubitschek, forexample,restoresone namehere(anda corresponding namein thefirstcolumn)to reachhis countof fifty-eight 31. I cannotacceptMunn's(2000, p. 258) contentionthatthe listsof IG II210 are"selective" andhonored onlythoseforeignsupporters of Thrasyboulos who"possessed some measurable fortuneor,at least,. . . had achievedsomemeasureof socialrecognitionamongtrueAthenians." The preserved portionof thelist of thethird group,whichgivesa man'soccupation togetherwithhis name,hasusually beentakento showthelowlyoriginof the honorands. Osborne(1982,p. 42), forexample,remarks thatthe men's "humbleoccupations forthe mostpart [are]matchedonlytoo closelyby the bizarreness of theirnames." We have
gardeners, walnut-sellers, tanners, donkey-tenders, andtub-makers and, as Hereward(1952,p. 117)notes,men named"PhanosandKnips,Mikosand Astyages,Abdes,IdyesandAtys."It is true,asMunnnotes,thatthe men identifiedbytradeon thisdecree"are not necessarily humbleorpoor,"andit maywellbe (asMunnalsonotes)that theprosperous businessman Lysias wouldhavebeenidentifiedas"Lysias, shield-maker" in thistext.Thussome of thesemenmaybe prosperous, butit is hardto believethatwe shouldfinda wealthyandupwardly mobileentrepreneurbehindeverymanlistedby his occupation here the ownerof a
walnutorchard behind"walnut-seller" andthe ownerof a thrivingdonkey stablebehind"donkey-tender." This maybe the caseforsome,butthereis no evidencethatit is so forall.Withoutevidenceto the contrary, it is preferable to takeat leastsomeof the occupations at facevalueandaccept thatthelistsof thisdecreehonored allthoseinvolved.SeeWolpert2002, p. 44, fora recenttreatmenttaking thisdecreeas"yetanotherforceful reminderthatthe civilwardivided theAtheniansalongclasslines." 32. Krentz1982,p. 84. 33. Buck1998,pp.71-72.
ONE HUNDRED
HEROES
OF PHYLE?
387
(Figs.1-2), but onlyfifty-fournamesarestrictlyrequired by the stone. Onecouldrestoreanadditional sixnamesin eachcolumn(toreachBuck's figureof seventy),butnothingotherthana desireto matchXenophon's figureforthesizeof thebandthatleftThebesurgesoneto doso.Furthermore,therearedifficulties withsucha restoration. Addingsix namesat thispointin eachcolumnwouldmakethenumbers of menfromthetribes Erechtheisor AigeisandOineisor Kekropisexcessively andunusually large.34 Thus,the numberof honorands in the firstlist shouldnot rise muchaboveRaubitschek's fifty-eight. It wasthisrealization thatledRaubitschek to proposea secondlistin orderto makethe numberof honorands conformto Aischines' implication thatthe decreehonoredover100 men.Buck,in contrast,rejects Aischines. ApartfromchidingKrentzforrelyingtooheavilyonAischines' "looseremark,"35 Bucknowhereaddresses Aischines'discussionof this decreenordoeshe explainwhywe shoulddiscountit. In thisBuckis not alone.MarkMunn,too,reportswithoutdiscussiononlythatArchinos honored"approximately sixtyAthenians."36 LikeKrentz,MunnbelievesthatArchinoshonoredonlyAthenians. Munnnotesthatthe epigramon the decreeproclaims thatthe menare beingrecognized by"theindigenous demosof theAthenians"-waBaiX0xv dY1tuog'A0Nvaixv (whichBucktranslates as "longin theirland"),andarguesthattheAthenians areconcerned to give"priority . . . in thehonorsof theday"to "native-born Athenians."37 In thishe followsa trendto interpretthe decreeas one of a seriesof whatKrentzcalls"conservative answers" givenafterthe restoration of the democracy to "important questionsregarding theformof Atheniangovernment andthenatureof Attic citizenship."38
34.To reacha countof seventy namesone mustrestoresixadditional namesin eachcolumn.In the second column,thesenamesmusteitherbe attributed to one of twotribes-Oineis andKekropis-orbe sharedbetween thosetribes.Oineisalreadyhasa large presenceon the stone,however, with ninedemesmen. This contingentis largerthanthatof anyothertribe exceptErechtheisandAigeis,whose contingentstogetherfill eighteenlines in RaubitscheWs restoration. Both ErechtheisandAigeisincludednine men butthisis an arbitrary division becausethe tribalheadingthatdivided the twogroupsis a restoration and couldfallanywhere in theselines.The onlycontrolis thatthe cohortfrom Aigeiscomprisedat leasttwo men,if Raubitschek's restorations of the demoticsrap]n[rcl]ogandrapnM]TTlOg arecorrect in lines21-22. ErechtheisandAigeistogetherfielded .
.
eighteenmenat Phyle,according to Raubitschek's restoration. The contributions of the other tribesweremuchsmaller: Pandionis andAntiochissentsixmen,Kekropis andLeontissentfour,Aiantissent three,andAkamantisandHippothontistwo each.The largesizeof the contingentfromOineis(ninemen) is probably dueto the proximityof demesof thistribeto Phyle.Fivemen fromPhyleitselfandthreemenfrom nearbyAcharnai join one manfrom Lakiadesto makeup Oineis'group. Addingsixlinesto column2 between fragmentsB andC musteithermake Oineis'contingentevenlargeror increasethatof Kekropisto ten.In column1, addingsixnameswould meanincreasing the alreadylargecontingentsof ErechtheisandAigeiseven more.Noneof thisis impossible, of course,butone shouldnotethewayin whichaddingsixlineswouldfilrther
imbalance the tribes.This is why Raubitschek declinedto restorethe preserved listwiththe numberof namesonewouldhaveexpectedfrom Aischines'description of the decree. As Raubitschek (1941,p. 294) notes, "itis extremely unlikelythatmembersof the tribesErechtheis, Aigeis, OineisandKekropisaccountedfor almosteightynames,whilethe remainingsixtribesfilrnishedonly twenty-three." 35.Buckl998,p.75,n.22. 36. Munn2000,p. 257. Funke (1980,p. 37, n. 9), too,reportsmerely thattheAgoradecreeprovesthatall thosehonoredbyArchinoswere alreadyAtheniancitizens.He cites Raubitschek andAischinesbutdoes not discussthem. 37. Munn2000,p. 257;Buck1998, p.72. 38. Krentz1982,p. 113.
388
MARTHA
C. TAYLOR
TheAthenians refused, initially, to honorallof theforeignsupporters of Thrasyboulos withcitizenship, andwaitedtwoyearsbefore(possibly) awarding citizenshiponlyto a few andmereisoteleiato the rest.In the sameyearasArchinos's decree(403/2),theAtheniansalsorefusedto give publicsupportto the orphansof foreigners who haddiedforAthenian democracy, although theyofferedpublicburialto thedeadforeignfighters themselves (Lys.2.66).Theozotides' decreegrantedpublicsupportquite pointedlyonlyto the orphans"ofallthoseAthenianswhodieda violent deathin theoligarchy comingto the aidof thedemocracy."39 Andit was notonlyforeigners whofoundthemselves in dangerof beingleftout.Accordingto Dionysiosof Halikarnassos (DeLysia32),a certainPhormisios proposed (perhaps intheinitialperiodofnegotiations betweenthePeiraieus andthe cityparties)thatcitizenshipshouldbe restricted to thosewho werelandowners.40 Dionysios reports thatthiswouldhavemeantthedisenfranchisement of 5,000Athenians(De Lysia32). Archinos's decreefortheheroesof Phyle,if it honoredonlyAthenian citizensandexcluded foreignfighters, wouldseemto fitinwithsuchmeasures.Ostwald,forexample, callsthebilla "safedecree" because"ithonoredonlycitizens," andthus"cannothaveoffendedthe now-important formersupporters of the oligarchy left in the city.''4l Butthe evidenceof the stone,althoughoftenoverlooked, counselscaution.Evenif we are readytodiscount Aischines' implication thattherewereover100honorands, thepreserved fragments of thedecreedemonstrate that,alongwithatleast fifty-three Athenians, Archinoshonoredasmanyasthreemenwhowere eitherEleutherians or metics,depending on thewayin whichwe restore line69.42 PROPOSED RESTORATIONS OF LINE 69 If Meritt'srestoration ofr[yypago] is correct, itwouldshowthatArchinos washappyto honoratleasta fewmeticsalongside Athenianpatriots. His 39. Forthe decree,see Stroud1971, quarrel withThrasyboulos's originalproposal, in thatcase,wouldbe with pp.280-282,no.7. Stroud(pp.285language" of the numbersof the menhonoredor withthe ideaof grantingso many 286) notesthe''colorfill lines 4-7, and that the decree is "exclumencitizenship, notwiththeideaof honoringforeigners perse orrecognizinga debtowedto them.Furthermore, if weweresurethatthebillonly sivelyconcernedwithAfthenians." 40.The oratorLysiasopposedthe honoredthepreserved list,thedecreewouldalsogivepreciseinformation proposal:Lys.34. aboutthesizeandmakeupofThrasyboulos's army.If thedecreeincluded 41. Ostwald1986,p. 506. onlycitizens,onewouldalwayswonderhowmanyunlistednoncitizens 42. Buck(1998,p. 72) assertsthat werelistedon the servedwithThrasyboulos earlyatPhyle,andneverknowexactly howmany threeEleutherians decree. He does not mentionMeritt's democrats withstoodthesiegeof theThirty.Meritt'srestoration, however, proposedrestoration. wouldsuggestthatArchinosincludedallwhowereinvolvedonthedemo43. It is possiblethatno menjoined craticsideinthesiegeatPhyle.This, inturn,would implythatThrasyboulosThrasyboulos betweenthe captureof hadonlya fewmorethanfiftymenwithhimwhentheThirtymarched Phyleandthe siegethere,butI thinkit out againsthim,andwouldfurthersuggestthathe hadfewerthanthat unlikely.The choiceof the siegeas the numberwhenhe marchedfromThebesandcaptured Phyle.43 Farfrom cutoffpointforthesehonorsalsosuggeststhatsomemenjoinedThrasyboubeingan"overwhelmingly foreign" band,Meritt'srestoration on a decree los's bandbetweenthe captureandthe containing onlythepreserved listwouldindicatethatThrasyboulos's army siege otherwisewhyusethe siegeas wasoverwhelmingly Athenianatthetimeof thesiegebytheThirty,with the cutoffpointratherthanthe capture itself) onlyoneto threeforeigners in a bandof fifty-fiveorso.
ONE HUNDRED
44. IG I31162,lines96-97 = Meiggs-Lewis48, lines96-97. 45. Figueira1991,p.155. 46. See,e.g.,Sgora XVII,no.147 line 35;no.17, lines5-9,25-29; no.22, lines152-155.Camp(1991,p.200, withn. 22) makesthissamepoint. are Buton thesestelaithe foreigners describedas [oxaoxatDap5]apot, and oX^oxat, [X^]9\8°t,[D]apDapot [X] ,Bapa[po], respectively, Tox[ceoTax] perhapssuggestingthatSemachides' listingunderthe rubric'EArvOrpa0zv is distinctiveandmeantto separatehim outfrommerexorvox(althoughthere xorvotincludedon his areno "regular" casualtylist). 47.The AsoposRiveris mentioned at Hdt.6.108,whereit is calledthe betweenPlataiaandHysiai, boundary on the onehand,andThebes,on the other. 48. Camp1991,p.201.
HEROES
OF PHYLE?
389
ofline restoration is lessclearif weacceptRaubitscheks Thesituation included 69 as'E[Xrv0rtoa0rv] or'E[Xrv0rtorlq] on a decreethatoriginally would restoration Raubitschek's preserved. onlythelistnowfragmentarily bandand aboutthe sizeof Thrasyboulos's giveus muchlessinformation proposedby Merittbecause thanthe restoration its foreigncomposition inhabita strangemiddlegroundbetweenthecitizenandthe Eleutherians Thusit is possiblethattheywouldbehonoredona decreethat noncitizen. whoservedin Thrasyboulos's foreigners excludedany"real" nevertheless ona decreeofArchinosthatdelibof Eleutherians army.Butthepresence wouldbe an excitingnewpieceof the eratelyexcludedotherforeigners puzzleoverthestatusof Eleutherians. long-standing arenotknownto wasnotanAtticdeme,andEleutherians Eleutherai is listedunderthe havebeenAtheniancitizens.A certainSemachides at the endof a mid-5th-century rubric'EAzv0rtoa0rv, "fromEleutherai," ForThobytribe.44 (asregularly) listof Atheniandeadrecorded casualty werecitizensbut"outside masFigueirathisis evidencethatEleutherians casualty of 5th-century Butthereareotherexamples thetribalsystem."45 That forwhomno citizenstatusis assumed.46 liststhatincludeforeigners goodevidence is not listedunderanAttictribeis probably Semachides werenotAtheniancitizensat the timeof his thatmenfromEleutherai death. of anAthenian wasnot a demesman if anEleutherian Nevertheless, His villagelayin neitherwashe clearlya foreigner. demeof Eleutherai, between oncetheAsoposRiverwasfixedastheboundary Atticterritory in 506B.C., aftertheAtheniandoublevictory AtticaandBoiotia,probably On the otherhand,fromhis study overthe BoiotiansandChalcidians.47 of the towerson theborderbetweenAtticaandBoiotia,JohnCamphas statusmusthavechangedearlyin the nextcenarguedthatEleutherai's atleastuntillatein thefifthcenwas"Boiotian tury,andthatEleutherai hada traditionthatlinkedthem the Eleutherians Nevertheless, tury."48 (1.2.5;1.38.8),anEleutherian to Pausanias closelyto Athens.According to Athens. thexoanon of DionysosEleutherios namedPegasosintroduced usedto be thatalthoughEleutherai (1.38.8)reports Pausanias In addition, "joined theAthenians[zpootheborderof Attica,whentheEleutherians The becamethe border." Tovv], Kithairon xpr1cyavv 8r 'A0Y1vaiolg notbecausetheyweredefeated "joined continues, Pausanias Eleutherians, andbecauseof their inwarbutbecausetheywantedAtheniancitizenship anceswithEleutherian neighbor, An Eleutherian hatredoftheThebans." andprivyto Athenian, thatwasatleastsometimes tors,livingin a territory vis-a-visAtherelationship wasin a different tradition, thismythological It is niancitizensthan,say,a RhodianmeticorevenaThebandemocrat. wouldbe treateddifferently thatEleutherians therefore, not impossible, decree. onArchinos's thanotherforeigners wellintothe If Campis correctaboutBoiotiancontrolof Eleutherai on the of Eleutherians throughit),thepresence 5thcentury(andperhaps if therewasonlyonelist striking.Particularly decreewouldbe especially andArchinostherebyexcludedotherforeignpatriots,his decisionto inwouldbe a boldstatementaboutthe properborder cludeEleutherians stronglyindicatethat betweenAtticaandBoiotia.It would,furthermore, demes,wereperceived despitenotbelongingto Kleisthenic Eleutherians,
C. TAYLOR MARTHA 39o
dividebetweenAthenianand lieon the citizensideof a fundamental to if hemeantto makepointedcomments andthatArchinos, non-Athenian, withinit. includedEleutherians demosofAthenians, theindigenous about of appearon a decree Archinos If thisis the case,andEleutherians list,we wouldknowthatat leastfiftyincludedonlythe preserved that withstoodthesiegeof AtheniansandasmanyasthreeEleutherians three wereinvolvedatallat Thirty;butunlesswe believethatno foreigners the ortheratioof earlypoint,we cannotknowthesizeof thetotalforce, this to Atheniancitizens. foreigners REEVALUATIONOF THE DECREE decree:it clearaboutArchinos's whatseemsrelatively us summarize Let veryearlyin his campaignand menwhojoinedThrasyboulos honored not evidence withstoodthe earlysiegeof theThirtyat Phyle.It is who foreignat this wereoverwhelmingly the forcesunderThrasyboulos that if the decreeoriginallycontainedonlythe preOn the contrary, point. is correct,it of line 69 as £[ypa(poL list andMeritt'srestoration served native,with forceswereoverwhelmingly seemthatThrasyboulos's would Atheniansin a groupof fifty-fiveorso.Thisscenario, leastfifty-three at defeated wouldindicatethatalthoughArchinosopposedand however, not he proposalto grantallhis followerscitizenship, did Thrasyboulos's to the returnof of non-Athenians the contribution recognizing oppose equally rewarding Archinoshimselfproposed demos.On thisreading, the among"thosefromPhyle." thecitizensandforeigners to important answer" givea "conservative clearly Nordoesthedecree is correctand restoration evenif Raubitschek's aboutcitizenship, questions list, the preserved nometicswereincluded.If the decreecontainedonly would E[B£v0£p£C5 or of line69 as'E[B£v0£pa0£v restoration Raubitschek's rewards, Archinos's from foreigners "real" showthatthe decreeexcluded in villageof Eleutherai butpointedlyincludedresidentsof the marginal understanding our for implications ranks a movewithprofound citizen of theircitizenship. conceived and themselves viewed ofhowAthenians bandor the Thrasyboulos's of size Apartfromjudgmentsaboutthe therewas whether stand in it, theseconclusions of foreigners proportions as or, preserved) Raubionlyoneliston thedecree(theonenowpartially speech,two.As we Aischines' tschekproposedin orderto accommodate thattherewere haveseen,bothBuckandMunndiscountthepossibility thatthe statement by Aischines' over100 menon the decree(indicated offerings for gaveeachherolessthan10drachmas reward 1,000-drachma fortheirposiNeitherBucknorMunngivesanargument anda sacrifice). forit is not clearthatwe should tion,butthe pointdeservesdiscussion, rejectAischinesoutof hand. on thestoneallowedidenThe firstlettersof theepigrampreserved of it.Theyconfirmthat quotation tificationof thedecreefromAischines' andmightperof thedecreeis trustworthy, atleastpartof hisdiscussion On usto accepttherestof hisdescription. theotherhand, hapsencourage (mis)construes EdwardHarrishasshownthatin this speechAischines It is crownsto benefithis position.49 the termsof the lawsconcerning hero of moneyeach true,of course,thatAischinesmentionsthe amount
49. Harris 1995, pp. 143-144.
t
e
-
n
o
-
o
o
bfi
\
e
-
o
ONE HUNDRED HEROES OF PHYLE?
50. Sickinger(1999,p. 149)notes thatthe decreestoodin or nearthe Metroon.He arguesthatit wasthe the candiBoule'srolein scrutinizing datesforthe honorsof the decreethat the presenceof thisinscrip"explains tionin the Metroon." 51.0berl989,p.319. 52. Strauss(1986,p. 97),forexample,acceptsAischines'claimthat Archinos"limit[ed]theprizeto a smallsumof moneyandanolive accepts (andso,presumably, wreath" thatArchinoshonoredover100men), andusesthe politicalclimateof the timeto explainthe modestyof the the reward."Byde-emphasizing of the menof Phyle," achievement alsode-emphasized he notes,"Archinus thevillainyof theiropponents,thus reducingtensionbetweendemocrats andthe formerThreeThousand." Straussdoesnot revealwhetherhe thinksArchinoshonoredonlycitizens or not.His suggestion(p.96) that oppositionto ThrasybouArchinos's los'sbillgrantingcitizenshipto allhis maypossiblyhavestemmed supporters of meticsandslaves" from"abhorrence suggeststhathe thinksArchinos honoredcitizensonly.
39I
for honorsproposed receivedin orderto contrastit withtheostentatious share to reduceeachindividual's He hasa motive,therefore, Demosthenes. butit numberof meninvolved, theapparent of themoneyandso increase is notclearthatAischinesdidso or,indeed,wasfreeto do so.TheAtheniansarguedlongandhardoverwhomto honor(andinwhatway)forthe firstgenThrasyboulos's Archinosdefeated of thedemocracy. restoration fora carein favorof hisownmuchmoremodestrewards erousproposals fullyselectgroup.AischinesmighthaveexpectedAtheniansevenmany yearslaterto havea vaguesenseof whetherthe firstgroupof honored fifty-fiveornearlydoublethat,atmorethana hunnumbered individuals thedecreeis in theMetroon(intheexcavations says that dred.Aischines andso he mightwellhaveexpected of whichit was,in fact,discovered), someon thejuryto haveseenit.50Thispossibilitymayhavemadehim thenumberof honorands. hesitateto exaggerate commontoposofblame" JosiahOberrightlynotesthat"aparticularly betweentheAtheniansof thepresent in politicalspeechesis"thecontrast Obercitesthisspeechof Aischines ancestors.''5l dayandtheirillustrious deforthe pointthatearlierAthenians,in contrastto theirdegenerate "didnot giveexcessivehonorsto unworthymen."But thata scendents, Even pointis a commontoposdoesnotmeanthatit hasnobasisin fact.52 in thenumbers if we thinkAischineswaswillingandableto misrepresent point. thisparticular orderto makehiscase,he hadno needto exaggerate The amountof moneyArchinosgrantedto eachherofromPhyleis a to the more He movesimmediately minorpartof Aischines'argument. betweenthecrownof oliveArchinos contrast direct(andmoreimpressive) offeredto the men of Phyleandthe crownof gold thatDemosthenes stoodto win- a pointthatthedecreeprovesis notexaggerated. movedthattheheroesfromPhyle"each AischinesnotesthatArchinos be crownedwitha crownof olive- not of gold,forthenthe olivecrown washeldin honor,buttodayevena crownof goldis despised"£z£sa ,
-
-
oT£avUsaL
a
ov
oT£avX
,
avTXv
-
c,
£xasTov,
,
a
>,
,
-
,
\
A OV XOVOU T0T£ £V
\
yato
xaTas£
392
MARTHA
C. TAYLOR
andXenophonrespectively giveforthe numberof menwholeftThebes withThrasyboulos orparticipated in theinitialoccupation of Phyle.But thatis notthegroupthatAischinessaysArchinoshonored. According to Aischines,thecutoffpointforinclusionin thesehonorswasnotthecaptureof Phyle,butthelatersiege.Thus,anymenwhojoinedThrasyboulos afterhis firstbandhadtakenPhylebut beforethe siegeof the Thirty wouldhavebeenhonored.Archinos's decree,therefore, shouldreferto morementhanthesixtyorseventywhotookPhyle.It shouldhonormore thanthefifty-fiveorso menlistedin thepreserved fragments, justas his discussion of thesumeachheroreceived implies.53 AcceptingAischines' implication thatthereweremorethana hundredhonorands requires, asRaubitschek saw,thatwe posita second,lost, list of honorands.Othercontemporary Atheniandecreesinscribethe honorands in differentplaceson the stoneor clearlyseparate theminto morethanonelist.ThedecreeofTheozotides forAtheniansorphaned in thefightto restorethedemocracy listsnamesonitsleftsidein additionto thosethatwereinscribed belowthedecreeon theobverseof thestele,but whicharenowunreadable.54 A closerparallelcomesfromIG II210,the decreethatfinallygrantedcitizenship or isoteleia to Thrasyboulos's nonAtheniansupporters. Aswesawabove,thisdecree,inscribed onbothfront andback,separates the honorands intothreegroupsaccording to when theyjoinedThrasyboulos. At leastthesecondandthirdgroups,although listedseparately, receivedthe samehonors.55 It wouldnot be surprising, then,if Archinos's decreeseparated its honorands into two groups,inscribing thefirstlistabovethedecreeandepigram andthesecondbelowit (oronthesidesortheback),whilegivingthetwogroupsthesamehonors. The mostcompellingargument againstRaubitschekXs positionthat thisdecreeincludedover100honorands separated intotwolistsis thatwe havefoundfragments of onlythefirstlist.However, we haveclearlylosta greatdealof thestele.Wehavefragments of onlythefirstfourlinesof the decreeitself(Fig.1),forexample, andthepreserved fragments of thestele suggestthatmuchhasbeenlostbelowtheselines.Indeed,thereconstructionof thewholestelebaseduponRaubitschekXs andMeritt'sdrawings andpublisheddimensionsfor the fragments togetherwith conventions forproportions of stelaisuggeststhattherewasampleroomon the stele fortwolists(Fig.3).56 Theseconsiderations urgeus to recognizethe continuedplausibility of Raubitschek's positionthatthe decreeoriginally honoredovera hundredmendividedintotwolists,andto address thefurtherimplications of thisinterpretation. If the decreeoncecontainedtwolists,one mustexplainwhytheAtheniansseparated menwhowouldreceivethe samerewardsintotwogroups.Is thedivisionbasedon ethnicity withAthenian citizensin thefirstgroup,andforeigners in thesecond?Ordoesthedivisionshowwhena manjoinedThrasybouloswithmenwhocamewith Thrasyboulos fromThebesin the firstlist,andmenwhojoinedhimat Phylein thesecond? Meritt'srestoration of line69,whichmakesit a headingformetics,if correct, wouldshowunequivocally thatthefirstlisthonoredbothcitizens andforeigners. Thiswould,in turn,stronglysuggestthatthetwolistson the decreeservedto distinguish the honorands chronologically, withthe firstlistnamingthosefifty-fiveorso men bothAthenianandforeign-
53.Withoutaddressing the implicationsforhisunderstanding of the decree,Buck(1998,p. 74) arguesthat it wasonlyearlyreinforcements that allowedThrasyboulos to withstandthe firstassaultof theThirty.Anysuch reinforcements maywellhaveincreased the sizeof the originalcoregroupof sixtyor seventyto the forceof over100 thatAischinesimplieswithstoodthe Thirty'sassault. 54. See n.39, above.Seealso Stroud'scomparison (1971,p.295) of thisdecreewiththe decreein honorof the heroesof PhyleaswellasIG II210. 55. Osborne1982,p.33. 56. Raubitschek's andMeritt's restoreddrawingsof the stelesuggest thattheremayoriginallyhavebeen roomformorenamesbelowthe decree. It hasalwaysbeenclearthata portion of the lowerpartof the Phylestelewas missing,becausewe havepreserved onlythe firstfewlinesof the decree recordedon the stele.Butthe roughly squaredimensionsof Meritt'sand Raubitschek's drawingssuggestthat thelostlowerportionof the stelewas large.In contrast,the nearlycontemporarysteleof Theozotides(Stroud 1971,p.280) is 1.53m highby 0.67 m wide,morethantwiceashighasit is wide.Thereare,of course,no rigid rulesfortheproportions of stelai.Dow (1934,pp.142-144)proposedthata conventionprevailed wherebya stele's widthwas4.5 timesits thicknessand the height9 timesthe thickness,giving a heighttwicethewidth.If the decree forthe heroesof Phylehadthe same proportions asthatof Theozotides,it
ONE HUNDRED HEROES OF PHYLE?
393
who helpedto capturePhyle,andthe secondlist honoringthosemenbeforethe siege. bothAthenianandforeign-whojoinedThrasyboulos coming The decreewould,then,not allowus to speakof Thrasyboulos forit wouldindicatethateven fromThebes"withseventyAthenians,"s7 coregroupin his marchon Phyleincludedcitizensand Thrasyboulos's The decreewouldreveal,however,thatthatcoregroupwas foreigners. Athenian,with onlyone to threemeticsin a groupof overwhelmingly fifty-fiveorso. wouldseemto confirmPausanias's Meritt'srestoration Furthermore, fromThebes figureof sixtyfor the groupthatcamewithThrasyboulos upfrom a rounding representing seventy(perhaps ratherthanXenophon's aboutfifty-fiveor indicatingthatmorenamesshouldbe restoredin the theethnicityof regarding list).Wewouldhaveno information preserved wouldoriginallyhavebeenmorethan andso couldnot in thesecondlist,however, thegroupof aboutforty-five twiceastallas the portiondepictedin armybeforethe siege joinedThrasyboulos's discusshowmanyforeigners Dow'smore the restoreddrawings. sinceit indicates speculation, on control as a 10 serves II2 IG Thirty. by the conventionwouldargue conservative helpedThrasyor in fought foreigners 116 than more no ultimately that thatthe stelewasoriginallytwiceastall boulos'sarmyat Phyle. as it waswide.Figure3 depictsthe stele withthesedimensions,andshowshow on the otherhand,is in severalwaysless restoration, Raubitschek's on muchspacemayhavebeenavailable a middle inhabited clearthanMeritt's,foraswe havenoted,Eleutherians the stone.The restof the decreeitself restoraboth Although the noncitizen. and citizen the between ground onlypor(ofwhichwe havepreserved tions indicatethat the decreedid not honoronly Atheniancitizens, tionsof the firstfourlines)wouldhave leavesopaque unlikeMeritt's, onatwo-listdecree, restoration RaubitscheWs filledsomeof the space,butperhaps not all,andit is clearthatthe amount werehonoredin a secondlist.The reason foreigners whetherornot"real" of spacerequiredforan additionalforty stronglyimpliesthatany for this differenceis thatMeritt'srestoration or so nameswouldnot havebeengreat. whereasRaubitschek's divisioninto two listswasbasedon chronology, FragmentB containsenoughspacefor oranethnicdivision. chronological a either with consistent is restoration namelines,andgivesus twenty-three honin Archinos's foreigners include obviously An ethnicdivisionwould anindicationof howmuchof the stele divisionwouldnotnecesors(albeitin a secondlist),buta chronological wouldhaveneededto be dedicatedto forty-sixnames(in two columns).If the the exclusionof sarilydo so. Indeed,if the divisionwas chronological, sameletterheightwereusedfora list of indicatethatforeigners fromthefirstlistwouldprobably foreigners "real" belowthe decree,forty-six foreigners fromthesecondlistaswell unlesswebelievethatno forwereexcluded namesin two columnswouldstillleave fromThebes.(In thatcasethe abwithThrasyboulos marched eigners muchspaceforthe completionof the notindicatetheirgeneralexcluneed first list the from of foreigners sence letter decreeitself.If a lessimpressive heightwereusedfora list of foreigners sionfromthedecree, butthisscenariois unlikely.) belowthe decree,even (orlatecomers) wouldargue restoration RaubitschekXs If thedivisionis chronological, andmore lessspacewouldbe required, camewith Atheniansandas manyas threeEleutherians fifty-three that forthe completionof spaceavailable leave would it restoration, Meritt's Unlike Thebes. from Thrasyboulos the recordof the decreeitself.These because originalband,however, the fullsizeof Thrasyboulos's uncertain canhardly,of course, crudecalculations in it. Simion possibleforeigners provethattherewasa namelistbelovv the decreewouldgiveno information the decree.Buttheydo suggestthat larly,althoughthe decreewouldindicatethatmorethanfortyAthenians enoughroomon therewasprobably atPhylebeforethesiegethere,it wouldleaveunclear joinedThrasyboulos a the originalsteleto accommodate thesizeof histotalforceatthatpoint,becauseof thelackof information secondlist. divisionimpliesthat a chronological A readingassuming aboutforeigners. 57. See,forexample,Munn2000, andcloseneighAthenians fifty-five about only lure could that a movement anda coreforce p. 235:"Thrasybulus of somesixtyto seventyAthenians in Thebeswasablein the dayor borsoutof Atticato joinThrasyboulos marchedto Phyle";Ostwald1986, two(?)beforethe firstattackandsiegenearlyto doublethe numberof p. 487:"Theseizureof Phyleby Atheniansinvolvedin therevoltthroughtheadditionof morethanfortyandhis bandof seventy Thrasybulus namedon the secondlist. fiveAthenians, andKrentz1982, Athenianexiles"; maysuggestthatnotallthe Theevidenceof thestoneitself,however, p. 70:"FromThebesa smallbandof fromThebes.In thatcase, AtheniansledbyThrasybulus...." menin the firstlist camewithThrasyboulos
OAP SY B °
A
o ,/S AT Yo kr o AE I TOE 4 I I JM A fH E S K rr koM, YA EA N NE I
T E A I
_4
I ho
° I NH I ao
E
/
E
-
k E
r A P,,
rA rANtI o N I LoE
rl
r H ;7fT I )/ /
A
/
I ozavo
E
E
T ANT
I 0>
B CR MEEtHN
8
Ag
k Nt+N / E YoE X yv
E
E k 1 ° I AH
S
1 toE
Q
s
<
IA
t¢
r
Ko tz kyeAT HET I mpk A E o E E I T EP
O
E Yx Op
s IE Y
EA PP I >* P oNot#P I o S o S SE Y S AE> 130 /L I °NYS / AMA k N T,LtOU A
| A
>
s
| r n oO NT I Lo E A P x I N EMY PrL 1\11 A° ° I N H I HE Ao E Av I ANT) A hIAP ot
°BA lEloS NIbor ME bl o E t YPPI oUloS S 1 0 A r Ft6EAH° ,<5 k YO H P) P I °E /
g
hENTI h0E
P
P Or
/ /
X / TTr °
Z
o
YS
/AXAPSE
A,>/A P S E Y > VSAX AP S E YS I ALUt r P°AAk rl I 0E XE I E P I ° ¢YhAt Y hA fi I o AbT I k P ATo E t A S I o E {f Y L | Xwo ° ¢ Y A sH M o m Y A A t I o t
A I rH I A O E
kE/\ H
t
kATAAA B oN T E E +N YA,4 / H oMN K A T H r A r o
oIA E ToN A
o D
4R,S
MARTHAC. TAYLOR
394
PE ExoH
S
I
HMAx
I SH
E
p
y
ATHd \ PAOE
N
/
P EPAAAI BOreN SEFEPAI /tlEAPETHEENEkASTEfANOIPOTETOSAtI FO I fi tHf\OtAOHNAIoNOI I kATAPAY EN (i)E9MO I SAP TANTAEPol\EnXrprtTo NAPA/V\ENOI I NLYNONE/V\ASI I/IANk g A^MATEYEEYkl\ElAHE Plox kO/XtYTEYtErP NTf«lBotH | tA IT<|8HM
// V}
HPx Xkv I xO+
|
AN $ EYx ErEtTATEApx
q <
oftheinscripFigure3. Restoration convention Dow's to tionaccording After forsteledimensions. Raubitschek 1941,p.289,fig.1
ONE HUNDRED
HEROES
OF PHYLE?
395
buton anydivisionbetweenthe two listswasmadenot on chronology, proposed. as Raubitschek ethnicity, of thedecreerecordin column2 thenames fragments Thepreserved of at leastfivemenfromthe demeof Phyle(lines43-47),aswellas the (lines39-41).They demeofAcharnai namesofthreemenfromthenearby Raubitscheks (following mayalsoincludeonetothreemenfromEleutherai MenfromPhyle,thatis, made of line69 as'E[X£V0£pa0£V]).58 restoration honoredon thefirstlist,a presence up 9%of thefifty-fourmencertainly Men fromthe nearbydemesand to Phyle'ssize.59 out of all proportion villagesof PhyleandAcharnaimadeup almost15%of the group.If we the figurerisesto over16%.Raubitschek includeevenone Eleutherian, thepresenceof so manymenfromthesedemesandvillagesby explained afterhe cameto Atticawhenhe thattheyjoinedThrasyboulos suggesting of somany Indeed,thepresence hadbroughthisrevoltintotheirbackyard. on the commemorated event the that proof wasforRaubitschek Phylians The largenumberof Phyliansand decreetookplaceat or nearPhyle.60 otherson the firstlist makesmostsenseif thatlist includesmenwho afterhe enteredAttica.Phyliansin sucha highprojoinedThrasyboulos if the firstlist honorsthe men portionwouldmakeless sense,however, in Thebesandcamewith him to capture who werewithThrasyboulos Phyle.ForwhywouldPhylianshavebeenthatmuchmorelikelyto join Chronology, in exilethanmenof otherAtheniandemes?61 Thrasyboulos intotwolists. thehonorands wasnotthefactorseparating then,probably factorandarguesthat(asRauThisleavesethnicityas the determining in anysecondlist(ifit existed)includedtheforeigners bitschekproposed) bandbythetimeof thesiegeof theThirty. Thrasyboulos's CONCLUSIONS consequences decreehasimportant ofArchinos's Theaboveinterpretation in their foreigners to debt Athenians' of the both forourunderstanding to acknowloftheThirtyandoftheirwillingness tyranny overthrowofthe band edgethatdebt.ThedecreewouldindicatethatwhenThrasyboulos's 1941,pp.293-294. 58. Raubitschek 59.The figureof fifty-fourassumes onlythe one metic(orEleutherian) mentionedin line 70. Phylewasa small deme.Traill(1975,p. 68) hascalculated Phyle'squotain the 500-manBoule astwo;thisis in contrast,forexample, quotaof to the twenty-two-man Acharnai. 1941,p. 287;see 60. Raubitschek alsoHarding(1985,p. 12, n. 1),who to thisrestoration" opinesthat"essential is "thepresenceof fivemenfromthe smalldemeof Phylein the list of names,whichsuggeststhatthe events recordedtookplacein thevicinityof Phyle." 61. One mightarguethatPhylians
wouldbe morelikely andAcharnaians in Thebesbeto joinThrasyboulos causeof theirproximityto Thebes. remained However,whileThrasyboulos in Thebes,his chancesforsuccesswere doubtfulto unclearandapparently mostin Attica.Onlycommitteddemorepulsedby crats,or menparticularly theThirty,wouldhavejoinedhim there.The smallnumbersthatThrasybouloshadwithhimin Thebes,according to oursources,showthatsuchmen hadhad werefew.BeforeThrasyboulos a successin Attica,personalideology, courage andindividual experience, wouldhavefiguredfarmorestrongly in a man'sdecithanmeregeography had sionto join.OnceThrasyboulos
provedhimselfandhis chancesby raisingevena smallband,andhad takenup a positionin Attica,however,the groupof menwho mightjoin himwouldwiden.Men of Phyleand Acharnaiwouldbe stronglydrawn (if at all inclinedto his cause)to join I a risingin theirownneighborhood. by Buck'sargument amnotpersuaded delib(1998,p. 72) thatThrasyboulos eratelychosemenfromPhyleand becausethey"hadintimate Eleutherai knowledgeof the area[of Phyle]and wouldhavebeenan assetin a force The small facingsuchanoperation." originalband sizeof Thrasyboulos's suggeststhathe hadlittleopportunity to pickandchoosehis men.
396
MARTHA
C. TAYLOR
numbered slightlymorethan100men,it includedatleastfifty-three Athenians,oneto threecloseneighbors, andoverfortyforeigners. Thus,over 40%of the menwhofirstfoughtagainsttheThirtywerenotAthenians butforeigners. Thisis not as impressive a figureasthe roughly85%that Krentz's interpretation indicated forThrasyboulos's troopsatthelatersurpriseattacknearAcharnai, butit is a significant number. Duringthecourse ofThrasyboulos's monthsat Phyle,however, theproportion of foreigners in his armygradually decreased to about10%,asindicatedon IG II210, beforethemoveto thePeiraieus.62 Onthisreading, theAthenians' dependenceon foreigners in the firststagesof the revoltwassignificant, but hardlyoverwhelming. This interpretation wouldalsohaveimportant implications for our understanding of Archinos's attitudeto foreigners. If Raubitschek's originalpositionis correct, it wouldshowthatArchinoswaswillingto honor fortyorso non-Athenians alongwithhisnative-born patriots(albeitin a separate list),andwaswillingto demonstrate graphically onhisdecreethe extentof thedebtowedto foreigners in theearlystagesof thereturnof the demos.
In theend,althoughthereis muchthatstillremainsuncertain about thisdecree,thisdiscussion hasestablished severalpoints,andclarified the possibilities thatremain.Archinos's decreecommemorated aneventvery earlyinThrasyboulos's campaign, anddoesnotdemonstrate thathisforces wereoverwhelmingly foreignatthattime.TheymayhavebeenoverwhelminglyAthenian(reading Meritt'soneto threemeticson a one-listdecree) orabout60%Athenianandnear-Athenian (reading Raubitschek's oneto threeEleutherians on a two-listdecree).If the decreeincludedtwolists, withAtheniansandforeigners in bothlists(reading Meritt'soneto three meticsonatwo-listdecree), Thrasyboulos's forceswereatleast53%Athenianandprobably more(sincewe wouldaddhowevermanyAthenians wereincludedin the secondlist whojoinedThrasyboulos betweenthe captureof PhyleandthesiegebytheThirty).63 Anyreadingof thedecreerequires usto temperrecentdiscussions of the churlishness of theAtheniansandthe conservatism of Archinos.As we sawabove,eventhe preserved fragments of thisdecreeshowthathe waswillingto honormeticsorEleutherians alongside Athenians. Raubitschek's proposal thatArchinoshonoredoverfortyforeigners in a second list,if confirmed, wouldbe anevenmoredramatic indication thatArchinosdidnot opposehonoringforeignheroes,butin eithercaseArchinos wasnotobsessedwiththeAthenianpurityof hishonorands. Thedecree's reference to wakaiX0xv 8nyog'A0nvaixv certainly soundspointed,butone cannotarguethatArchinosrefusedto honornon-Athenians at all.His decree,however, involvescrownsandmoneyforsacrifice, notcitizenship. The epigramon thisdecreemaybe marking outthatverydistinction: we Athenianswillacknowledge andhonorthe foreigners whohelpedus restoreAthensto democracy, butwe shallnotmakethempartof ourselves. The "indigenous demosof the Athenians" will not be sullied.Although ArchinosopposedanddefeatedThrasyboulos's proposalto grantall his followers citizenship, he didnotopposerecognizing thecontribution of at leastsomenon-Athenians to thereturnof thedemos.
62.The figureof 10%is basedon Xenophon's reportthatThrasyboulos had1,000troopsby the timeof his moveto the Peiraieus, butonly116 foreigners at mostarehonoredon the decreefortheirhelpup to thatpoint. 63. RaubitscheWs Eleutherians on a one-listdecreewouldleavethe total numberof Thrasyboulos's bandunclear (andtheproportion of foreigners in it) becausewe wouldnot knowhowmany foreigners (if any)wereamongThrasyboulos'sforces.
ONE HUNDRED HEROES OF PHYLE?
397
REFERENCES AgoraXVII = D. Bradeen, TheFuneral Monuments(AgoraXVII), Princeton 1974. and the Buck, R. J. 1998. Thrasybulus AthenianDemocracy(Historia Einzelschriften120), Stuttgart. Camp,J. McK. 1991. "Notes on the Towers and Bordersof Classical Boiotia,"AJA95, pp. 193202. Cloche, P. 1915. La restauration a Athenesen 403 avant democratique J.-C., Paris. Dow, S.1934. "The List of Athenian Archontes,"Hesperia3, pp. 140190.
Figueira,T. J. 1991. AthensandAigina in theAgeofImperialColonization, Baltimore. Funke,P. 1980. HomonoiaundArche (HistoriaEinzelschriften37), Wiesbaden. Harding, P.1985. FromtheEnd of the PeloponnesianWarto theBattleof Ipsus,Cambridge. Harris,E. 1995. AeschinesandAthenian Politics,NewYork. Hereward,D. 1952. "New Fragments of IG II2 10,"BSA 47, pp. 102117. and Hunt, P. 1998. Slaves,Warfare, Ideologyin the GreekHistorians, Cambridge.
againstthe Krentz,P.1980."Foreigners Thirty:IG II210 Again,"Phoenix 34, pp.298-306. .1982. TheThirtyatAthens, Ithaca. . 1986."TheRewardsfor ZPE 62, Supporters," Thrasyboulos' pp.201-204. . 1995.Xenophon,Hellenika II.3.11-IV.2.8, Warminster. Meiggs-Lewis= R. Meiggsand D. Lewis,A Selectionof GreekHistoricalInscriptions,rev.ed.,Oxford 1988. Meritt,B. D.1933."TheInscriptions," Hesperia2, pp.149-169. Munn,M. 2000. TheSchoolofHistory, Berkeley. Munn,M., andM. Zimmerman Munn.1989."TheStanfordSkourThe 1987and1988 ta PlainProject: Seasonsof Surveyon theAttic93, pp.274BoiotianFrontier,"AJA 275 (abstract). Ober,J. 1989.Massand Elite in DemocraticAthens,Princeton. Osborne,M.J. 1981.Naturalizationin Athens1, Brussels. . 1982.Naturalizationin Athens on theDecrees 2: Commentaries GrantingCitizenship,Brussels. . 1983.Naturalizationin Athens 3-4, Brussels.
MarthaC.Taylor LOYOLACOLLEGEIN MARYLAND OF CLASSICS DEPARTMENT 450I NORTHCHARLESSTREET 2I2I2 MARYLAND BALTIMORE, mtaylor@loyola . edu
Ostwald,M. 1986.FromPopular Sovereigntyto the Sovereigntyof Law,
Berkeley. A. E. 1941."TheHeroes Raubitschek, of Phyle,"Hesperia10,pp.284-295. Rhodes,P.J. 1993.A Commentaryon the AristotelianAthenaionPoliteia, Oxford. J.1999.PublicRecordsand Sickinger, Chapel Archivesin ClassicalAthens, Hill. Strauss,B. 1986.AthensafterthePeloponnesianWar,Ithaca. Stroud,R. S. 1971."GreekInscriptions: TheozotidesandtheAthenian Hesperia40, pp.280-301. Orphans," Tod,M. N. 1946.GreekHistorical Inscriptions1: Fromthe Sixth CenturyB.C. to theDeath ofAlexanderthe Greatin 323 B.C., Oxford. Traill,J.1975. ThePoliticalOrganization ofAttica,Princeton.
E. 1978."RoadsandForts Vanderpool, Attica,"CSCA11, in Northwestern pp.227-245. A Whitehead,D. 1978."ISOTEAEIA: Eirene16, Metaphorin Xenophon," pp.19-22. New . 1984."AThousand LCM 9.1, pp.8-10. Athenians," Defeat: Wolpert,A.2002. Remembering Civil Warand CivicMemoryin Athens,Baltimore.
HESPERIA
7I (2002)
Pages 399-4I3
A
N
FU
EW
N
I NSCRI
E RARY
F ROM
Al
BED
MONUMENT G
I NA
AB STRACT This articlepresentsan ancientmonumentdiscoveredon Aiginain 1999. The monumentis remarkable forits unusualshape:a rectangular slabwitha pyramidal top, a two-lineinscription,anda deepnichewith dowelholesin the floorandbackwalls.I arguethatthe monumentis funeraryin function, andthatits peculiarfeaturesarerelatedto its primaryuse.The inscription givesa malenameanda patronymic, AristoukhosAristomeneos,andcanbe datedto the 4th centuryB.C. It is possiblethatAristomenes,the fatherof Aristoukhos, is the heroof Pindar'sPythian8.
1. I wouldliketo thankmany friendsandcolleaguesat theAmericanSchoolof ClassicalStudiesat Athenswho havelenttheirtimeand expertiseto the discussionof this monument: A. Ajootian,N. Bookidis, W.Closterman, C. Gray,J.Grossman, M.Langdon,C. Lawton,J.Morgan, M.Richardson, D. Romano, R.Stroud,M. Sturgeon,andC. Williams.I owespecialthanksto M. H. Jameson andH. Kritzas,whosehelpful commentsandsuggestionssavedme frommanyerrors.I wouldalsoliketo thankthe 2nd Ephorateof Prehistoric andClassicalAntiquities,andespecially itsdirector, GeorgeSteinhauer, for granting me permission to studyand publish thismonument.Finally,I am grateful to the anonymous reviewers of Hesperia fortheirrecommendations, whichI havetakento heart. 2. Thereis no sourceof marbleon Aigina,so the stonewouldnecessarily havebeenpurchased andbroughtfrom theoutside.
An ancientmonument discovered in 1999 on theislandof Aiginamerits attentionfor two reasons.lFirst,it displaysan unusualcombination of architectonic features: a pyramidal topcrowning a rectangular monolithic slab,anda deepnarrow nichecutinthefaceof theslab.Second,themonumentbearsaninscription, whichcouldbethefirstepigraphic evidenceon Aiginafortheclientsof Pindar's Aiginetanodes. THE MONUMENT Themonument wasfoundin thechurchof AgiosNikolaosin theareaof Kavouropetra, on thenorthern coastof Aigina,about2.5 km northeast of thetownof Aigina.It is setflushwiththeedgeof thethreshold insidethe churchandis 1.5 mhigh,0.48 mwide,and0.24 mthick,includingapyramidaltop thatbegins1.29 m fromthe bottomof the block(Figs.1-2). The stoneof the monumentis a medium-grained marble,grayishin color.2 A two-lineinscription on the smoothfrontfaceextendsfromleft to rightbelowthe top (Fig.3). A rectangular cutting0.08 m belowthe baseof thepyramidal topandroughlycenteredbetweenthelongsidesof themonument(0.115 m fromthe left edgeand0.12 m fromthe right) measures 0.66m in length,0.245 m inwidth,and0.225 m deep.Thebottomedgeof themonument is brokenofffrontto backattheleftandright. Thetipof thepyramidal topis cutoff,leavinga flatrectangle (L.0.08m, W.0.065 m)witha circular holein thecenter(Diam.0.015 m).Thehole
4oo
I REN E PO LI N S KAYA
Figure1. Funerarymonumentof Aristoukhosinsidethe churchof Agios Nikolaos,Kavouropetra, Aigina.Photoauthor
Figure2. Funerarymonumentwith cleanedniche.Photoauthor
Figure3. Inscriptionon the funerary monument.Photoauthor
INSCRIBED FUNERARY MONUMENT FROM AIGINA
40I
its to determine sothatit is impossible is filledwithdebrisandwhitewash, originaldepth. The left andrightwallsof the cuttingareworkedsmoothsurfaces, In thebackwallof thenicheis a circurougher. buttheflooris somewhat goesthroughthe stone. edgesthatpresumably larholewithchipped-off vis-a-visthesides,andis located0.2 Thehole(Diam.0.02m)is centered in shape(L.0.065 mbelowtheceilingoftheniche.Anotherhole,elliptical m,W.0.035 m,D. 0.06 m),is cutintothefloorofthecutting,0.05 mfrom andslightlyoff center,0.093 m from the frontsurfaceof the monument theleft,and0.117m fromtherightwallof thecutting.The edgesof this hole slopeinwardandlookworn.In addition,the remainsof two iron nailscanbe seenin thewallsof theniche,closeto thebottom. is enoughto recognizeits unusual A singleglanceat thismonument form.To distinguishits primarydesignandpossiblesecondaryalterationswe needto considerthe five majorelementsof the monument's top,thecutting,theholesinshaft,thepyramidal design:therectangular Leavingasidethe holesforthe mosidethecutting,andtheinscription. of andcontemporaneity forthe originality ment,the strongestargument to eachother. relationship the otherfourelementsis theirarchitectonic cutting,whichI identifyas a niche(see The positionof the rectangular topandthesidesof bythebaselineofthepyramidal below),is determined to thebaselineof shaft:thetopedgeofthenicheis parallel therectangular top, andthe nicheis centeredbetweenthe sidesof the the pyramidal monument. at an equaldistancebetweenthe The placementof the inscription top andthe top edgeof the nicheis another baselineof the pyramidal top,the niche,andthe indicationthatallthreeelements(thepyramidal floorof thenichecan the in hole The together. designed were inscription) of a three-dimensional as a dowelholeforthe attachment be explained object.It is possiblethattheholein thebackof the nicheis a lateraddiholesareoftenfoundon reusedgravestelaiandareusually tion.Circular identifiedasconduitsforawaterpipeof a fountainhead.3If thiswerethe onewouldexpectto find however, casewiththe Aeginetanmonument, darkstainsormarksofwaterwear,noneofwhicharepresentontheinner as an addiwallof the niche.If thisholeis original,it canbe explained of anobjectin theniche.Thus,thereare tionaldowelholeforthesecuring aspartsof stronggroundsforviewingallthe elementsof the monument theoriginaldesign. INSCRIPTION asfollows: on thestone(Figs.4-5) canbe described The inscription
3. E.g., Clairmont,CAT 2.382c [= Conze1893-1922,II, cat.no.658, pl. CXX];Conze1893-1922,IV,cat. no. 1871,p. 37.
L.H.0.015-0.018 m, omicronDiam.0.013-0.014 m. Spacing betweenthelettersof thefirstlineis 0.02-0.025 m;betweentheletters of thesecondline,0.01-0.015 m, exceptbetweenthefirstalphaandrho, whichis 0.02 m.The lengthof thefirstlineis 0.31 m,of thesecond, 0.30 m.The distancebetweenthetwolinesis 0.015-0.02 m.
IRENE
402
POLINSKAYA
Figure4. Squeezeofthe inscription. Photo author
saec.IVa. 'Apt.axovx°S 'ApaxoyeveoS
The twowordstogether,A*oaxouXoS A*oaxo,ueveoS, giveus a personalmalenameandapatronymic, whichin theabsenceof anyadditional words(e.g.,thosethatcouldsuggesta votivestele)mostlikelyindicate thatourmonument is funerary, honoringandpreserving thememoryof a deceased person. In general,theletteringof the inscription is verycarefully executed, with a slightwideningof the endsof somehastae.All but one of the elevenlettersthatappearin theinscription arefamiliar fromArchaicand Classical examples onAigina.4 Theformof nu (straight verticalstrokesof equallength)is unattested on Aiginain theArchaicandEarlyClassical periods.DuringthePeloponnesian War,theAiginetanpopulation wasin exile,andAtheniansettlersoccupiedtheisland(Thuc.4.56.2).Thus,we haveno evidenceofAiginetan writingfromthisperiod,andtheexamples ofpost-war inscriptions areextremely meager.5 Underthesecircumstances, wecannotdetermine whetherthenuontheAiginetan blockwasa regular formusedin thatperiod.Althoughit appears to be a hapaxin theknown Aiginetan usage,therecanbelittledoubtthattheinscription isAiginetan. The dialectof the inscription is Doric,as the uncontracted genitiveof A*oaxo,uevYIS (A*oaxo,ueveoS insteadof A*oaxo,uevovS) demonstrates.6 Withoutindications of a foreignorigin,dueweightmustbe givento the 4. LSAG2,Aigina,a4, £4,x,ll2, o, p1,o2, , X The lettersof ourinscriptiondisplaysomestylisticvariations of the knownforms:alpha-the right leg is slightlyshorterthantheleft and endsabovebaseline;epsilon the middlestrokeis shorterthanthe upperandlower,andnot touchingthe vertical;mperceptible, butvery slightinwardcurvingof outerhastae; sigma-shortercentralstrokesof the 4-barsigma.The upsilonis foundin the inventoryof the Sanctuary of
Aphaia(IG IV 39, secondhalfof the 5th centuryB.C.). The unattested letterformis nu. 5. Onlyonepublishedinscription on a sarcophagus datesto the 4th centuryB.C.; seeArchDelt32 (1977) B'1[1984],p. 43 = SEGXXXIV 270. Alreadyin the secondhalfof the 5th century,someof the known Aiginetaninscriptions maynotbe reflectingpurelyAiginetanletter forms.In 457 B.C., Aiginawas defeatedbyAthensandforcedinto
theDelianLeague(Thuc.1.105.2-4, 108),andtheislandwasoccupiedby AthensduringthePeloponnesian War. A numberof the horosstonesand inventories of the sanctuaries dateto the secondhalfof the 5th centuryon Aigina,andcouldbe seenasthe productof Athenianpresenceon the island.Theyposemanyproblemsin identification of scripts(Barron1983; Figueira1991,pp.115-120). 6. Buck1955,p. 40.
p INSCRIBED
T
O
M
E
N
FROM AIGINA
4o3
o
x
y
o
I I t
MONUMENT
FUNERARY
Es
Figure5. Drawingof theinscription. andG. Lavoie I. Polinskaya
7. LSAG2,Aigina,a2, vl, v2,v3,v4, p.109.
8. Woodhead1959,p.64. inscriptions 9. Classicalfunerary andallbut areespeciallyinfrequent, one dateto the 5th centuryB.C.: IG IV of Antistatesthe Athe50, gravestone nian,ca.450 (SEGXXV 332,XXIX of Gleu295);IG IV 49, gravestone century(?) kitasthe Salaminian,5th (SEG XXV 323,1X 296);IG IV of Hermaios,ca.47547, gravestone 450 (SEG XI 30,XV 187,XXIX297); ArchDelt34 (1979)B 1 [1987],p.69, graffition a potsherdandon a completepot froma grave,5thcentury
onAigina.In sum,thedialectof theinscription, stone'splaceof discovery andtheprovof mostof thelettersin theAiginetanalphabet, thepresence stronglysuggestthatit is anAiginetanproduct. enanceof themonument The abis difficultto determine. The precisedateof theinscription senceof the diagnosticArchaicAiginetanletterformsrulesout an Archaracteristics chaicdate.7At thesametime,theabsenceof theprominent upper of theHellenisticandRomanletterforms,suchasstraightparallel militates serifs, andlowerbarsforsigma,lunatecurves,andpronounced inscripagainsta Hellenisticorlaterdate.8Sincethenumberof Classical in the local tionson Aiginais small,9the chronologyof developments We thusmaybejustifiedin using scriptin thisperiodis notestablished. dating.Comparison of approximate forthepurposes outsidecomparanda of theletter suggeststhatsomespecialcharacteristics withAtticexamples of a datefollowing mightbeindicative formsin theAiginetaninscription War.The shortermiddlebarforepsilon andtheslight thePeloponnesian arefamiliarfrom curvingof theverticalstrokesof muin thisinscription of the4thcenturys.c.l°Thewidening of thefirstquarter Atticexamples of thefreeendsof theletterstrokes(asin thetauandepsilon here)canbe datingfromca.350 to 325 s.c.ll inscriptions seenin Atticfunerary suggesta datesometimein the 4th Theseandearlierobservations centuryB.C. A moreprecisedatewouldrequirea largersampleof local to drawon,whichwe lackatthepresenttime. inscriptions (SEG XXXVII254 and262);Welter
1938a,p.58, fig.49, and1938b, pp.521-523,namesof the deceased paintedor carvedin the chamber tombs. steleof Chrysallis, 10.Funerary andMyrta,IG II25649, Phaidrias, 390-365 B.C., Athens,Nat.Mus.750 (Conze1893-1922,I, cat.no.392, steleof p. 91, pl.XCVI);funerary IG II212086,390Menekrateia, 365 B.C. (Conze1893-1922,I, cat.no.161, p. 40, pl. L);funerary steleof Tito,IG II210231,firsthalf steleof of the 4th century;funerary Artemisia,IG II210840,beginningof
the 4th century.See alsothe laterfunerarysteleof Epithales,IG II29157, II, 350-340 B.C., Ker.4857(Kerameikos pp.40-41, pl. ll). monumentof Dama11. Funerary sistrate,IG II211037,350-330B.C., Athens,Nat.Mus.743 (Conze 1893-1922,I, cat.no.410,pp.94-95, steleof Sympl.XCVII);funerary machia,IG II29337,slightlybefore ca.317/316B.C., Athens,Nat.Mus. 1728.SeealsoIG II26942,firsthalf of the 4th century,Athens,Epigr.Mus. 2674 (Clairmont,CAT suppl.,no.PE or RSE33, fig.p.130).
4o4
IRENE
POLINSKAYA
PRO S OPO GRAPHY The name'AtotoTo,uevYIS is well attestedthroughout the Greekworld.l2 Thename'AptoTouXog is muchmorerare.An'AptotouXog is knownfrom Epidauros (IG IV2187,4th centuryB.C.; notethevarianta°X )- Seven personsof this nameareknownfromthe Aegeanislands,of whichthe earliesttwo aredatedto the 4th centuryB.C. No 'AptoTouXog is known fromcentralGreece,andtheonly'AptoTouXog in Atticawasapparently a meticfromKythnos.13 Nor is anyother'AtoaTouXog knownfromAigina.The only'ApoTo,urvYls attestedon Aiginais the heroof Pindar's Pythian8. This epinikionis considered to bethelastwrittenbyPindarandis usuallydatedto 446 B.C., "theonlypoem,so faraswe know,thatPindarwroteforAegina aftershehadlostherindependence as a resultof theAthenianvictoryat Oenophyta (Thuc.1.108).''14The Aristomenes of Pythian8 wasthesonof Xenarkes, fromtheclanof Meidylidai.15 Otherrelatives mentioned in the epinikionarehis maternal uncles,TheognetosandKleitomakhos (Pind. Pyth.8, lines36-38).Aristomenes is honoredwithanepinikionfora victoryinwrestling atthePythianGames.He wonothervictoriesathomein thepentathlon of Apollo,l6andabroad, atMegara,Marathon, andArgos (lines78-80).17 In 446 B.C., Aristomenes couldbe described asbeing"on thevergeof manhood,''18 a description reinforced bytheprominent place givento thefigureof hisfatherin thepoem(lines71-75),aswellasbythe directaddressing of theathleteas"achild,"Ct) woct (line33).If Aristomenes wasfifteenyearsoldin 446,he wouldhavebeenthirtyin 431,whenhis familywasforcedintoexile.A hypothetical sonof thisAristomenes could stillhavebeenalivein thethirdquarter ofthe4thcenturyB.C. Althoughit wouldbedangerous to insistontheidentityof theMeidylidAristomenes, thesonofXenarkes, withAristomenes, thefatherof Aristoukhos, atleast the datesfortheyZoruit of the formerAristomenes' hypothetical sonand thedatingof theAiginetanfunerary inscription arefilllycompatible. 12.In the Peloponnese, western Greece,andMagnaGraecia,the name Aristomenes is attested65 times.On theAegeanislandsandCyrenaica, thereare108 attestations, of which27 are4th centuryorearlier.In Attica, thereare58 attestations; in central Greece,43, of whichallbuttwo are laterthanthe 4th century(LGPNI, II, IIIA,IIIB,S.v.AoxoToXurwrlg) 13. LGPNI, II, IIIA,s.v.AoxoTouX°5Forthemetic,seeMeritt1954, p. 271, no. 110 (SEGXIV 207),a 4th-centurygravestone; I wouldlike to thankJohnTraillforbringingthis nscrlptlonto myattentlon. 14. Burton1962,p. 174. 15. Pyth. 8, line38, waoav Mrx8U?W8av,Pind.fr.190,a MeduBou 8' au .
.
.
.
yzvza. Burton1962,pp.181-182; Figueira1981,pp.311-313. 16. Scholiato theselinesof Pyth. 8 indicatethatthe pentathlon waspart of theDelphiniafestivalin honorof Apolloon Aigina(Drachmann 1910, p. 215).A templeof Apolloin the townof Aiginaanda stadiumnearby arementionedby Pausanias (2.29.112.30.1). 17. Contraryto mostcurrent opinions(e.g.,Simon1980,p. 44), I identifyHoasayxv rstxotov as theArgiveHeraia,not as a localAiginetanfestivalin honorof Hera Moderninterpreters relyon the lonescholiastwho saysthattherewasa local Heraiaon Aigina(Drachmann 1910, p. 217),yet it is likelythatthe scholiast
himselfmisunderstood the reference. The "localagonof Hera"appearsin the list of Aristomenes' foreignvictories,andservesas a paraphrase forthe festivalandplacename(Heraiaat Argos,the mostwellknown"local" agonof Hera),whichis a common practicein Pindar(cf.01. 9, lines9899).Whenpromptedaboutathletic contestson Aigina,the scholiaprovide two namesof festivals: Delphiniaand Aiakeia.If the Heraiawereanother competitionon Aigina,one mightexpectto findthatotherathletescelebratedin Pindar'stwelveAiginetan odeswouldbe mentionedasvictors in theseallegedlylocalgames,but Pyth. 8 remainsthe solereference. 18. Burton1962,p. 174.
INSCRIBED
FUNERARY
FROM AIGINA
MONUMENT
4o5
ofa noble weremembers andAristomenes It islikelythatAristoukhos Xenarkes. While andhisfather, family,19 asweretheMeidylidAristomenes wasnotthe on thepresentmonument it is possiblethattheAristomenes in Pythian8, he couldhavebeena memberof the same one celebrated extendedfamily.Whethera relativeof the Pythianvictoror not,Arislivedin theperiodof lostgloryin thehistory toukhossonof Aristomenes in 404 B.C., theAiginetans to theirhomeland of Aigina.Thoughrestored arenathat in theinternational wouldneveragainachievetheprominence Bothtextual theyhadheldbefore457,theperiodof theirindependence. evidencefromthe4th centuryareverypooron Aigina.Perandmaterial population of exile:the returning hapsonereasonliesin thevicissitudes astheonethathadlefttwoanda halfdecadesearwasnot asnumerous wasoneof sonof Aristomenes lier.20 Ifweareto imaginethatAristoukhos to a lootedhouse exiles,he maywellhavebeenreturning the returning andanemptycoffer. on Aiginawiththelargemonumental the monument If we compare tombsof the 4th centuryin Attica(e.g.,the Dexileosmonument,the we mightasmonument),21 andtheKallithea monument of Aristonautes, and sessits valueasmodest.Yetit mustbe concededthatthe acquisition of of marblefromoutsidetheisland,andthe commission transportation somefinancialexrequired the designandexecutionof the monument, modestmonumentmaybe penditure. It is possiblethatsucha relatively in the a reflectionof the reducedfortunesof the Aiginetanaristocracy 4thcenturyB.C. At thesametime,weshouldnotethatevenin thetimeof B.C., it wasnotanAiginetan prosperity, in the6thandearly5thcenturies gravemarkers. customto setupostentatious Typicalburialson Aiginathroughoutantiquitywereunderground Rarelyweregravemarkchamber tombsorshaftgraveswithstairways.22 gravestones of the Twosurviving erserectedabovegroundovertombs.23 pillarsmadeof trachyte. 6th centuryB.C. (IGIV 47, 48) areundecorated arecommonreliefstelai.24 gravemarkers TheonlythreeknownSth-century duringtheAthenianoccuof localburialpractices Aftertheinterruption tombsareagainused; pationof theislandfrom431 to 404 B.C., chamber 19. It must be grantedthat while names in AoesTo-are appropriateto aristocrats,they hardlyguaranteethat status. 20. Xen. Hell.2.29.9: Ausavrx5Atyevav 8oos 8r afpexo,urvos osoug asrdxxz rrv woRevAxytvrlTaxs, rduvaTowArestousau9v a0ootoas. See also Figueira1993, p. 323; Plut. Lys.14.4. 21. Dexileos monument,394/3, Ker.Mus. P 1130 (Clairmont,CAT 2.209; Stewart1990, p. 172, fig. 482); Aristonautesmonument,IG II25462, second half of the 4th century,Athens, Nat. Mus. 738 (Clairmont,CAT 1.460; Rolley 1994, p. 378, fig. 397);
Kallitheamonument(tombof NikeratosandPolyxenosof Istria),330-320, Peiraieus Mus.,no inv.no.(Steinhauer 2001,pls.458-459;Ridgway1990, pp.31-32, 64, n. l5). Welter 22. On Aiginetanburials: 1938a,pp.55-62; 1938b,pp.495-524; Papastaurou 1990,pls.7-14. Tombs ArchDelt18 of the 6th-Sthcenturies: 21 (1963)B 1 [1965],p.52;ArchDelt (1966)B 1 [1968],pp.l00-102; ArchDelt 33 (1978)B 1 [1985], 34 (1979)B 1 [1987], p.53;ArchDelt pp.68-71. found 23.The onlygravemarkers 36 in situ:IG IV 47,48;ArchDelt (1981)B'1 [1988],p.67, at Leophoros
174in ChalikakiAgiasParaskevis 289); Meristos(SEGXXXVIII 1986[1990],p. 58, n. 61 ArchEph (SEGXXXJX331). 24. AiginaMus.733,ca.450 B.C. (Alt-AginaII.2,no.55,p.82,pl.43).
AiginaMus.729,early5th centuryB.C. II.2,no.54, pp.80, 82, (Alt-Agina pl.42),is sometimesidentifiedas Hiller1975,p. 72, n. 20. A funerary: steleof a youngman,Athens, funerary Nat.Mus.715,430-420 B.C. (Fuchs 1993,pp.487-488,fig.571),datesto the periodof theAthenianoccupation of Aigina,andwasprobablyanAthe.
nlan
.
commlsslon.
.
406
IRENE
POLINSKAYA
surviving examples datefromthe secondhalfof the 4th century.25 Aside fromthe presentmonument,no gravemarkersareknownfrom4thcenturyAigina.The inscribedmarblemonumentwith a niche thus appears to havebeenexceptional in thecontextof Aiginetanburialpractices,andalthough modestin comparison to contemporary Athenian grave markers, it represents an effortto producea distinctiveandrespectable memorial forthedead. FORM AND FUNCTION Threefeatures of themonument defineits peculiarity: thepyramidal top, niche,anddowelholes,presumably indicating thepointsatwhichanobjectwassecuredin theniche(Fig.6). P Y RA M I D A L
To P
The pyramidal top of the monumentwasoriginally fittedwith a finial, whichis indicatedby a smallroundholeon the flatareacreatedby the levelingof thetip of thepyramid. Parallels fortheuseof attachments on topof funerary monuments, however, arehardto find.26 No othermonumentswithpyramidal topsareknownfromAigina.A pyramidal top in generalis a rareelementin the designof ancientGreekfunerary monuments.27 Oneunambiguous Classical exampleof theuseof a pyramid asa gravemarker is theSth-century B.C. monument of Parthenia, thedaughter of NadystheCarian,fromSinope.Thethree-sided pyramid probably representsonlythetoppartof themonument, therestof whichhasnotsurvived.28 A gravemarker of unknowndatewitha four-sided steppedpyramidaltop is knownfromthe cemeteryof Canalicchio in Syracuse, and representations offour-sided stepped pyramidal roofscrowning whatmight be heroaarefoundon Apulianvasesof the4th centurys.c.29Apartfrom thevisualresemblance, no stylisticlinkscanbe established betweenthese monuments andtheAiginetanexample. Nor,outsidea funerary context,cantwoothertypesof Greekmonumentsincorporating pyramidal shapesinto theirdesignbe considered 25.Welter1938b,p.498. 26. Althoughnot numerous, examplesareknownof roundholes,aswell asrectangular dowels,forarchitectural attachments on top of the so-called Totenmahlrelief stelai,at leastfromthe Hellenisticperiod(Fabricius 1999, pp.138-143,fig. 19:c-k). 27.Twoexamplesof three-sided pyramidlike funerary monumentsof the Romanperiodareknownfromthe Kerameikos: P 673 (Kerameikos II, p.50, no.47, pl. 14) andP 210 (KerameikosII, pp.S0-51). 28. Istanbul,Arch.Mus.3868,
localAsiaMinormarble;H. 0.33 m, secondquarterof the 5th centuryB.C. (PfuhlandMobius1977-1979,I, p. 16,no.22, pl. 6; Clairmont1970, no. 10,pl. 36).All threesidesof the pyramidwereinscribed,andthese epigramstwicereferto a stele,indicating,as Clairmont(1970,p. 34) noted,thatthepyramidsurmounted a stele.Clairmont(1970,p. 34, n. 119) identifiesthispyramidal monument asunique,butrefersto additional evidenceforthe useof pyramidsas gravemarkers in classicalantiquity. In addition,Nakayama (1982,pp.43-
45, fig.4:a),on the basisof vasepaintings,identifiesa typeof gravemarker (GB-I)in the shapeof a "blockformigeMonumentemitpyramidalem Oberbau," the pyramidof whichis of steppedconstruction, andtopped witha stele. 29. GravemarkerfromSyracuse: Mus.Naz.40089;Apulianamphora, CVAKarlsruhe 2 [Germany8], pl. 60. Anotherpotsherdis in the personal collectionof N. Neuerburg, who publishedanillustration of thissherd withoutassigningit a number (Neuerburg 1969,pp.111-112).
INSCRIBED
drawing Figure6. Reconstruction of monument of thefunerary withoutfinialand Aristoukhos, DrawingM. Kapgan sculpture.
FUNERARY
MONUMENT
FROM AIGINA
4o7
fromthe astheydisplaymanydifferences orinspirations, stylisticparallels tops(withconcave here.Forexample,pyramidal monumentconsidered by singleor sides)thatcrownvotivepillarsfromArcadiaareseparated inplan.30 doublebandsof moldingfromtheirpillarshafts,whicharesquare planarepartof thedesign northesquare normoldings, Neitherconcavity, stele-likeobjectsof fromAigina.Kyrbeis,"freestanding, of themonument bronzeorstone,havingeitherthreeorfoursides,andcrownedatthetop onwhichofficap,''3l wereusedin Athensasmonuments bya pyramidal evidencethat The onlyarchaeological wereinscribed.32 cialregulations consistsofbaseswithtriangusupports thisdefinitionof kyrbeis,however, larcuttingsthatindicatethatthe shaftsof the insertedpillarswereof fromthe designof the triangular shape.Thisfeaturedifferssignificantly elements, Thus,in spiteof the commonpyramidal presentmonument. neitherthe kyrbeisnorArcadianvotivepillarsarecloselyrelatedto the Aiginetanblock. NICH E shouldbeidencuttingonthefaceofthemonument Thedeeprectangular of anobject.Theproporfortheplacement tifiedasa niche,anenclosure tionsof the niche(H. 0.66m,W. 0.245m, D. 0.225m) suggestthatit is to havebeenusedfora relief.Thealmostsquarefloor toodeepandnarrow of thecutting(L.0.245m,W. 0.225m)witha large,deepholesomewhat and off centersuggeststhatthe nichewasmadefora three-dimensional
30.Thesevotivepillarscomefrom the Tegea,Mantinea,andPallantion, earliestdatingfromthe firsthalfof the 5th century,andbearvotive
inscriptions on one of the foursides 1906, of the shaft(Arvanitopoullos pp.63-66, nos.16-17;Rhomaios 1967,p. 408).These 1911;Papachatzes
arecollectedin Stroud1979, references p.47,n. 148. 31. Stroud1979,p. 47. 32. Stroud1979,pp.1-40, 49-50.
408
IRENE
POLINSKAYA
nota flatobject.A rectangular nicheis theonlytypeknownfromArchaic andClassicalGreece,andit is mostlyusedas an architectural element, awallniche.33 Nichesin funerary contextsarealsofrequently wallniches, usedfor the placementof funerary urns.34 The mainpeculiarity of the Aiginetanexample,however,is thatthe nicheis locatedwithina freestandingmonolithicmonument. A surveyof the Archaic,Classical, and Hellenisticperiodson the Greekmainland, AsiaMinor,thePontusarea, andsomeof the Aegeanislandshasthusfarnot revealedanyotherexamplesof freestanding monolithic funerary monuments witha deepniche fora three-dimensional object.35 DOWELHOLESANDTHE OBJECTIN THENICHE If theholesin thefloorandbackwallof thenichewereindeedassociated withtheoriginaluseof themonument, it seemsmostlikelythattheyheld dowelsusedto anchoranobject.The shapeof the niche,withtheheight roughlythreetimesthewidth,suggeststhatit wasdesignedto holda tall andnarrowobject.Suitableobjectsforthisshapeandtheseproportions includeavase,suchasaloutrophoros orlekythos,36 stoneversions ofwhich wereoftenusedas gravemarkers in the Classicalperiod(Fig.7),37or a statuette. A tall,narrow stonevesselwouldbe anespecially attractive candidate fortheinhabitant of theniche.38 Lekythoiandloutrophoroi areoftenrepresentedon ClassicalAtticgravereliefsas standingon a shelfor ledge, illustrating thepracticeof settingup ceramicshapesof the samekindon tombs.39 Sometimesthesevesselsareshownon gravemonuments in a 33. Someof the earliestexamplesin the historicalperiodarethe so-called Wandnischen, wallnichesfromThasos thatwereconstructed in thewallsof the citygates.Theyareusuallyconsideredcultic.Otherexamplesof wall nichescomefromMagnesia,Ephesos, Priene,Messene,andDelos(Hornbostel-Huttner 1979,pp.33-48).It is quiteclearfromthe surviving basesin the nichesof Delianhousesthatstatuettesandvotivereliefswereplaced in them;see Kreeb1988,pp.43-46, pl. 8.1 (aninscribedbasein the niche in HouseE on the peribolos-street); pl. 8.2 (a basein the nichein House Th Vl D); andalsopl. 10 (reconstruction:statuettein a wallniche).Fora nicheintendedfora lampin the Erechtheion, seePalagia1984. 34. In the Hellenisticperiod,niches werebuiltinsidetombsandusedfor theplacementof funerary urns,e.g.,in theTaurianChersonesos, in the abovegroundtombsbuiltagainstthe fortifi-
cationwallof the city(Koshelenko, Kruglikova, andDolgorukov1984, p.52, pl.xix.8). 35. Nonearedocumented in Conze 1893-1922;Clairmont,CAT,and 1970;Diesantz1965;FraserandRonne 1957;Papapostolou 1993;Pfuhland Mobius1977-1979;Cremer1991;Firatli1964;Kieseritzky andWatchinger 1909;Fraser1977;or Schmidt1991. 36.Theseceramicshapes,imitated in stone,rangein ratioof heightto widthbetweenroughly1.7 and3.5;the mostcommonratiosareabout3:1 (Caskey1922,p.19, diagram XXXIX, pp.209-225). 37. Grossman1995,p. 228:"The practiceof usingsculptedmarblelekythoiasgravemarkers beginsat the end of the fifthcenturyB.C." SeealsoVedder1985,pp.43-48;KurtzandBoardman1985,pp.148-151.Although othervaseshapes,e.g.,amphoras, especiallythoseof Panathenaic shape, servedfunerary functions(Neils2000,
p. 130;Valavanis 2000),theyareless likelycandidates fortheAiginetan nichebecauseof theirunsuitable proportions;the ratiosof heightto width foramphoras rangebetween1.2 and 1.6 (Caskey1922,p.36). 38. A numberof marbleloutrophoroiandlekythoiof an appropriate sizeforthe nichesurvive: e.g.,the loutrophoros thatsurmounted the trapezaof Parthenios theMessenianin the Kerameikos (Fig.7), MG 47, Ker. 11174,H.0.52m,Diam.0.18m, Diam.(base)0.085m (Kerameikos X;]V, pls.14.2,23.2);anda lekythosfrom the Dipylonarea,Athens,Nat.Mus. 851,H.0.58 m (Clairmont, CHT 2.211;IG II213033). 39. Kokula1984,p. 15:"Dieerste UmsetzungderLutrophore in Stein zeigtnachderMittedes5.Jhs.v. Chr. ein doppelseitiges Grabrelief ausBrauron(Athens,Nat.Mus.4468),auf dessenVorderseite eineLutrophore und aufdessenRuckseiteeineLekythos
INSCRIBED
FUNERARY
MONUMENT
FROM AIGINA
4o9
Figure7. Trapezaof Partheniosthe Messenian,surmountedby a loutrophoros.Kerameikos,Athens. Photo author
recessed frame(Fig.8),asif intendedto depicta nichewithavesselin it.40 YetI knowof no examples orrepresentations on vasesof slabstelaiwith nicheshousingvessels.Stoneloutrophoroi andlekythoiwereusuallyset up as freestanding sculpture on stonebases.In mostcases,a roughhole, moreorlesscircular, wouldbe madein thepedestal; the footof thevase wouldbe set into the prepared holeandthe remaining gapsfilledwith lead.4l Accordingly, onemightexpectto findonthefloorof theAiginetan monument a roughlyroundholesomewhat largerthanthecircumference of a vessel'sfoot,centeredbetweenthe sidesof the niche.Althoughthe holewe findis ellipticalandsomewhat off center,we shouldnotruleout thepossibility thatthenichewasnevertheless designedto accommodate a stonevessel.Thecuttingmayhavebeenfora dowelhole,whichwouldnot dargestellt ist.BeideVasensind deutlichalsTongefasse aufdie Stele gesetzt,so dassim Gebrauch befindlichegefassederGrabkults im Bildauf derStelegezeigtwerden." Otherloutrophoroion a ledgeincludeAthens, Nat.Mus.2319, Clairmont,CAT 2.267; Athens,Nat.Mus.884, Clairmont,CAT2.710. 40. Loutrophoroi: e.g.,Athens,Nat. Mus.985 (Clairmont, CAT 1.947); Paris,LouvreMa 3119 (MNC2279) (Clairmont, CAT2.336); Athens,Nat.
Mus. 984 (Clairmont, CAT2.23); Athens,Nat.Mus.879 (Clairmont, CAT2.887); Athens,Nat.Mus.899 (Clairmont, CAT2.889); Athens,Nat. Mus.2553 (Clairmont, CAT3.406a). Kantharoi: e.g.,fromCyzicus,Istanbul, Arch.Mus.2215; fromMesambria on Pontos,Sofia,Nat.Mus.4739; from Dionysopolis, VarnaMus.II1595; fromMesambria, NessebarMus., no inv.no.(PfuhlandMobius19771979, II, nos.2259-2262, pl. 319). Kraters: e.g.,fromCos(?),CosMus.,
no inv.no.;fromLinopoti,on Cos, CosMus.,no inv.no.(Pfuhland Mobius1977-1979,II, nos.2263, 2264,pl.320). 41. E.g.,the funerary trapezaiof Philoxenos,Dion,andParthenios, theMessenians, in the Kerameikos, ca.340 B.C. (Ker.I 367,I 368,I 369; Kerameikos XlV, pp.105-110,figs.SS56, pls.22.1-3;23.2;14.2)andthe loutrophoros of Hegetorin the Kerameikos,350-340 B.C. (MG 32; Kerameikos XlV, p. 81, pl.19.2-3).
4IO
IRENE
POLINSKAYA
Figure8. Funerary stele,National Archaeological Museuminv.2553, Athens.CourtesyMuseum
haveneededto be roundorcentered if it wasusedto affixa stonebaseor plinth.The surfaceof thefloorof the nicheis notaspolished(especially towardthebackof theniche)asitswalls,whichsuggeststhatthesurface maynothavebeenintendedtobeseen,whichwouldbethecaseif a plinth wereadded. Anothercandidate fortheobjectin thenicheis a statuein theround. On a muchlargerscale,funerary sculptures in highrelieforin theround withinarchitectural framesarewellknownfromthe LateClassicaland EarlyHellenistic periods.42 Theproportions of thenichecouldaccommodatea medium-sized statuette(H.0.5-0.6m).Thecuttingonthefloorof thenicheis located0.05m fromthefrontof themonument, leaving0.11 m betweenthe backof the cuttingandthebackwallof the niche.One dowelwouldhavebeensufficientto holda smallstatuettein place.The holein the backwallof the niche,if it is original,couldhaveservedto 42. Fuchs1993,pp.496-498; Clairmont1970,pp.46-50, pls.11-89 (Attic,Thessalian, Macedonian, Ionian, andCretanmonuments); Kurtzand Boardman 1985,p. 156.Classical vasesthatdepicta gravemonument
with a human figureor a group of figureson top (e.g., white-ground lekythos,Bonn, Akad. Kunstmus.66, ARV2 1229, no. 15; CVABonn 1 [Germany1], pls. 43.2, 4; 44.2, 4) do not find supportin the con-
temporaryarchaeologicalrecord (Lohmann 1979, p. 40), and are not strictlycomparableto the figural sculptureproposedto have been in the Aiginetan niche.
INSCRIBED FUNERARY MONUMENT FROM AIGINA
4II
The bottomholebyitselfdoesnotallowus to destabilizethe statuette. or the materialof the techniqueof the attachment, terminedefinitively holessimilarto thoseseenontheAiginetan Smallelliptical thesculpture.43 wereusedwithdowelsto secureArchaicandEarlyClassical monument Bronze,howbyhollowcasting.44 thoseproduced especially bronzestatues, sculpof filnerary ever,asfaraswe know,wasnotusedfortheproduction mention turein the Classicalperiod.DonnaKurtzandJohnBoardman madeof bronze.45 sculpture Hellenisticepitaphsthatreferto funerary While a stonevesselor a stoneor bronzestatuettearenot unlikely fortheobjectin theniche,we haveto allowforthepossibility candidates objectwas placedthereas a that someother,perhapsunconventional, The objectin the nichemaynot havebeen memorialfor the deceased. piece,butwasusedinsteadduringthelifeasa funerary madespecifically It maybe thatthedesireto timeof thepersonhonoredbythemonument. withsuchanobjectcalledforthecreationof thedeceased commemorate the deepniche,if not in particular an unusualshapeforthe monument, top. thepyramidal CON CLUSION
43. Marblewasthe materialof sculptureandreliefs choiceforfilnerary of the ClassicalandEarlyHellenistic in the round periods.When sculpture wasused,it wastypicallycarvedon a plinththatwasthenset intoa cutting in the bottomof the naiskosframeof Thesecuttingswere the monument. mostlywideandflat,andthe plinths couldbe solderedontothebaseswith lead. 44. See,e.g.,Keesling1995, 1938, pp.146-147; Raubitschek p.133; 1949, p.61; Haynes1992, pp.100-105, fig.8. 1985, 45. KurtzandBoardman pp.295,315.
fromthechurch on themonument Whiletheletteringof theinscription ofAgiosNikolaosonAiginapointsto a datebetweenca.400and300B.C., in eitherthevodoesnotfindcloseparallels the shapeof themonument concontextsof thisperiod.The formof theinscription, tiveorfilnerary function,and suggestsa funerary sistingof a malenameandpatronymic, honorsthe sonof theMeidylidAristoit is possiblethatthe inscription menes,who wasthe heroof Pindar'sPythian8. At the sametime,the as monumentstandsas anoddityin the contextof localburialpractices, on Aiginain the Archaic,Classical,or Hellenistic it wasnot customary it In spiteof thelackof parallels, gravemarkers. periodsto erectelaborate of the monumentis dueto is veryunlikelythatthe presentappearance designandwellwithitspeculiar The monument, remodeling. secondary datafor newpieceof archaeological is a significant inscription, preserved hope that my It is evidence. material theperiodonAiginamostlackingin fromAiginato theattentionof other bybringingthisunusualmonument bebetterdefined. its functionandplacewillultimately scholars,
IRENE
4I2
POLINSKAYA
REFEREN C ES .
.
Grossman,J.1995."The Sculptured FuneraryMonuments of the ClassicalPeriod in the Athenian Agora"(diss. New York University). Haynes, D.1992. TheTechnique of AoTot £ztypafpal xal 7rkaxa GreekBronzeStatuary,Mainz. Zvu£a T£y£a$,"ArchEph 1906, Hiller, H. 1975. IonischeGrabreliefs pp.23-66. dererstenHdlftedes5. Jhs. (IstMittBarron,J.1983."The Fifth-Century BH 12), Tubingen. Horoi of Aegina,"JHS 103, pp. 1Hornbostel-Huttner,G. 1979. Studien 12. zur romischen Nischen-architektur Buck, C. D.1955. TheGreekDialects, (Studies of the Dutch ArchaeologiChicago. cal and Historical Society 9), Burton, R. W. B. 1962. Pindar'sPythian Leiden. Odes,Oxford. Keesling,C. 1995."Monumental Caskey,L. D. 1922. Geometryof Greek PrivateDedications on the AtheVases,Boston. nian Acropolis,ca.600-400 B.C." Clairmont,CAT= C. W. Clairmont, (diss. Universityof Michigan). ClassicalAtticTombstones, 6 vols., Kerameikos II = H. Riemann,Die Kilchberg1993. Skulpturen vom5. Jahrhundertbis Clairmont,C. W. 1970. Gravestoneand in romischeZeit (Kerameikos II), Epigram:GreekMemorialsprom the Berlin 1940. Archaicand ClassicalPeriod,Mainz. Kerameikos XIV = W. K. Kovacsovics, Conze, A. 1893-1922. Die attischen Die Eckterrasse an derGrdbersterrasse Grabreliefs I-IV, Berlin. desKerameikos (Kerameikos XIV), Cremer,M. 1991. Hellenistisch-romische Berlin 1990. Grabstelen in nordwestlichen KleinKieseritzky,G., and C. Watchinger. asien1-2, Bonn. 1909. Griechische Grabreliefs aus Diesantz, H.1965. Die thessalischen Suirussland,Berlin. Grabreliefs. Studienzur nordKokula,G.1984. Marmorlutrophoren griechischen Kunst,Mainz. (AM-BH 10), Berlin. Drachmann,A. B. 1910. Scholiavetera Koshelenko,G., I. Kruglikova,and in PindaricarminaII, Leipzig. V. Dolgorukov,eds. 1984. AntiFabricius,J. 1999. Die hellenistischen chryegosudarstvaSevernogoPriTotenmahlreliefs. Grabreprasentachernomoria [Ancient states of tion und Wertsorstellungen in the northernBlack Sea coast], ostgriechischen Stadten,Munich. Moscow. Figueira,T. 1981. Aegina.Societyand Kreeb,M. 1988. Untersuchungen zur Politics,New York. f gurlichenAusstatungdelischer . 1991. AthensandAeginain Privathduser,Chicago. theAgeofImperialColonization, Kurtz,D. C., andJ. Boardman.1985. Baltimore. Thanatos.TodundJenseitsbeiden . 1993. "FourNotes on the Griechen,Mainz. Aiginetans in Exile,"in Excursions LGPN= P. M. Fraserand E. Matin EpichoricHistory:Aiginetan thews, eds., ALexiconof Greek Essays,Lanham,Md., pp.293-324. PersonalNames,4 vols., Oxford Firatli,N. 1964. Lesstelesfune'raires de 1987-2000. Byzancegre'co-romaine, Paris. Lohmann, H. 1979. Grahmdlerauf Fraser,P. M. 1977. RhodianFunerary unteritalischen Vasen,Berlin. Monuments,Oxford. LSAG2= L. H. Jeffery,LocalScriptsof Fraser,P. M., and T. Ronne. 1957. ArchaicGreece,2nd ed., Oxford Boeotianand WestGreekTombstones, 1990. Lund. Meritt, B. D. 1954. "GreekInscripFuchs,W. 1993. Die Skulpturder tions,"Hesperia23, pp.233-283. Griechen,4th ed., Munich.
Alt-Agina II.2 = E. Walter-Karydi,Die aginetischeBildhauerschule: Werkeund schriftliche Quellen(Alt-AginaII.2), Mainz 1987. Arvanitopoullos,A. S. 1906. "'Av£x-
INSCRIBED
FUNERARY
MONUMENT
I-II, Grabreliefs Die Ostgriechischen
FROM AIGINA
4I3
Stroud,R. 1979.Theaxonesand IyrbeisofDrakonandSolon (University of CaliforniaPublications,ClassicalStudies19), Bulletinde l'Institutarcheologique Berkeley. bulgare12.1,pp. 132-181. P.2000."Panathenaische Valavanis, AmphorenaufMonumenten the . 1949.Dedicationsprom reisamphoren zu denPanathenaischen Cambridge, AthenianAkropolis, hellenistischer 25.11.-9.11.1998, spatklassischer, Rauischholzhausen Mass. M. Bentz and N. Eschbach,eds., undromischerZeit,"in PanatheRhomaios,K.A.1911. "'ApxadLxoL zu denPanathenaika:Symposion Mainz, pp. 125-130. EpllaL,'' ArchEph1911,pp.149-159. Rauischnaischen Preisamphoren Neuerburg,N. 1969. "Greekand Ro22:2, Archaeology man Pyramids," Ridgway, B. S.1990.HellenisticSculpture holzhausen 25.11.-29.11.1998, pp.106-115. 1, Madison. M. BentzandN. Eschbach,eds., Rolley,C.1994. La sculpturegresque, Palagia,O. 1984."A Niche for KallimaMainz,pp.161-173. Paris. chos' Lamp,"AJA88, pp. 515-521. zur Vedder,U. 1985.Untersuchungen Papachatzes,N. D.1967. IIowaorvc'ov Schmidt,S. 1991.HellenistischeGrabattischer plastischen Ausstattung II£,0CU5CN: >CX=, undchronologische 'ERAordof reliefs:Typologische Grabanlagen des4.Jhs.v. Chr. Cologne. Beobachtungen, Athens. (Archaologie, ser.38,vol.7), Papapostolou,J. A. 1993. Achaean Simon,E. 1980.Die GotterderGriechen, Frankfilrt. GraveStelai,Athens. Munich. Berlin. Welter,G. 1938a.Aegina, Papastaurou,E. 1990."EvvoRoopaSteinhauer, G.2001. To A,oxorcoRoycxo MII. 1938b."Aeginetica 8CzxvTazv omvALyLva,"ArchEph MOVC7£C'OII£C,OO(CCR)N, Athens. XX1V,"AA 1938,pp.480-540. 1986 [1990], pp. 49-59. Stewart,A. 1990.GreekSculpture:An Woodhead,A. G. 1959.TheStudyof Pfuhl, E., and H. Mobius. 1977-1979. Exploration,New Haven. Cambridge. GreekInscriptions,
Nakayama,N. 1982. "Untersuchung der aufweissgrundigenLelythen dargestelltenGrabmaler"(diss. UniversitatFreiburg). Neils, J. 2000. "Panathenaicsin the Symposion West,"in Panathenaika:
IrenePolinskaya BOWDOINCOLLEGE OF CLASSICS DEPARTMENT STATION 7600 COLLEGE BRUNSWICK, MAINE O4 OII - 84 76 ipolinsk@bowdoin. edu
Mainz. 1938."ZurTechnikund Raubitschek,A. Statuenbasen," Formderaltattischen
HESPERIA
71 (2002)
Pages 415-433
NOTES RESEARCH
FROM
THE
IN THE STOA
TINS
2
OF
ATTALOS
1. The authorswould like to thank John McK. Camp II, Director of the Agora Excavations,for his supportof this project.We are also gratefulto the anonymousHesperiareviewers and to the membersof the Publications Committee of the American School of ClassicalStudies at Athens for their helpful suggestions.
This series of researchnotes was conceived as a way to bring to light various details of the archaeological record at the Athenian Agora Excavations that might otherwise be lost within broaderpublications.' As noted in the first installment, the title "Notes from the Tins" refers to the feta cheese and oil tins used by the excavationsfor the storage ofuninventoried pottery and other finds from specific deposits or stratified levels. Study of finds in these tins led to the discovery of the details presented in these notes. Some of the pieces discussed had alreadybeen inventoried and so do not, sensustricto,come from the tins. The tins, however, serve equally as a metaphor for the study of items long ago inventoried, whose importance to archaeology might rise and fall with changing scholarly interests, and whose interpretation might change with further considerations of contexts or more recent discoveries. This is not to say that the tins do not figure strongly in the notes presented here. My study of amphoras as paint pots combines material found in the tins with jottings in the excavation notebooks and one previously inventoried example. No less important is the absence of similar fragments from hundreds of other tins, reminding us that negative evidence from the tins is sometimes useful as well. Kathleen Lynch'sstudy of early black-glazed mastoi clarifies the importance of rare, inventoried, but largely overlooked Attic examples within the context of known blackfigure examples and representations of the form by Attic vase painters. She also examines the contexts within the Agora (including the contents of the tins) in which the black-glaze examples were found. John Papadopoulos's commentary on crudely reworked clay disks is based on the hundreds of such disks found in the excavations, the many examples kept in the tins, and the few he has selected for inventory.These disks have piqued the curiosity of scholars over the years, but they have never received as broad a consideration as they do here. Finally, Susan Rotroff's study of the context pottery tins brought to light a raresignature on a moldmade bowl. Presumably this signature was overlooked and relegated to the tins instead of the glamour of inventoried status when the deposit was originally excavated. M.L.L.
MARK
416
L. LAWALL
ET AL.
AMPHORAS AS PAINT POTS? MARK L. LAWALL WITH
A CONTRIBUTION
BY AUDREY
JAWANDO
The secondaryuse in antiquityof transportamphorasas smallcoffinsor layersfor drainageis well attested.2Othercontextsof reusearerarelyattestedwith certaintyin archaeological finds.Three examplesof imported as reused amphoras paintpots appearedin excavationsof two depositsat the Agorarelatedto the Persiansackof Athens in 480 B.C.These paint pots arenot uniqueamongthe Agorafinds,but the identificationof the jars,preliminarystudyof the pigment,and considerationof the archaeologicalcontextsof thejarsprovidenew evidenceforthe studyof Athenian history,commerce,andtopography. 1
P 1334
Toe and lower part of amphora
Fig. 1:A
G 6:3 (Rectangular Rock-Cut Shaft, upper fill at 10.9 m) P.H. 19.6 cm; Diam. (toe) 6.3 cm Traces of white slip on exterior with wide horizontal brown band around top of preserved fragment. Red-brown deposit on interior with some thicker areasof preserved pigment; pigment spills over ancient breaks. Dusky pale reddish and gray-brown fabric.Wide scatter of gray glassy inclusions, dark gray/blackish stony bits, and yellowish lime chunks. Fabric color: 5YR 6/6 and grayer.Pigment color: 10R 5/8. Comments: Brief mention of this piece as "anunpublished amphora fragment"is found in Vanderpool 1946, p. 266, n. 6. This amphora type is attributed to the area of Klazomenai; see Doger 1986; Lawall 1995, pp. 48-53; Dupont 1998, pp. 151-156. Deposit G 6:3 is a deep rectangular pit on the Kolonos Agoraios; see Shear 1993, pp. 445-449; Vanderpool 1938 and 1946. The upper fill of the shaft is reported to have included miltos (notebook for section A, p. 1002; Vanderpool 1946, p. 266). The upper fill also included a lekythos, broken and reused as a much smaller paint pot (Vanderpool 1946, p. 266, no. 146). Date: late 6th century B.C.
2
AS-P 1049
Series of joining lower body fragments
R 12:1, tin EA 150 Smooth orange buff exterior;resinated interior with red pigment thickly preserved over the resin. Fabric with some mica; grainy break, moderate scatter of gray glassy and bright white opaque bits, all fairly small. Fabric color: 5YR 5/8. Pigment color: 10R 4/8. Comments: R 12:1 is a well deposit at the southeast corner of the Agora excavations,just east of the Stoa of Attalos (Shear 1993, pp. 469471). A nearby deposit of the same period is Q 12:3, the Stoa Gutter Well (Roberts 1986). The attribution and date of these fragments
2. Forusein drainageconstructions,seeMattioli1998.Forthe use of amphoras ascoffinsorurns,see Kerameikos IX, passim.On the reuseof see, amphoras, in general,Grace 1979, text with fig. 10, and for the rarityof reexportation,see van Doorninck 1989, esp. pp. 247, 256. I am gratefulto AudreyJawandofor carryingout the analysisof the pigment and providing the reportpresentedhere. I also thank Julie Unruh for her generousassistance and advicein preparingthis note.
NOTES
FROM
THE
TINS
2
417
A
m=/
r^ D
Figure1. A) Klazomenianamphora (P 1334);B) NorthAegeanamphora (tin EA 148). DrawingM. L. Lawall
_
-i
cannot be determined on the basis of the remains;neither the fabric nor the form is diagnostic. Date: context date, before 480 B.C. 3
Toe and lower part of amphora
Fig. 1:B
R 12:1, tin SA 148 P.H. 22.5 cm; Diam. (toe) 5.3 cm Resinated with reddish brown discoloration and a few thicker deposits of red pigment. Smooth, hard, micaceous orange-brown surface,fine-grained break;moderate sized, poorly sorted mix of dark gray opaque, very small white, and a few larger red-brown inclusions. Fabric color: 5YR 5/8. Pigment color: 10R to 2.5YR 4/6. Comments: The amphora is from the North Aegean; see Lawall 1997, especially pp. 116-117; Lawall 1995, pp. 116-175. Date: context date, before 480 B.C.
3. Koehler1986,pp.50-52.
The amphoras described above were not initially intended as jars of red paint. Fragments2 and 3 preserveresinous lining, which is most often associated with sealing the interiors of amphoras for shipments of wine.3 The red pigment covers these linings, so it must have been added after the
MARK
418
L. LAWALL
ET AL.
jars were emptied of wine. Furthermore,the red pigment on 1 covers the ancient break around the upper edge of the fragment. Only the bottom part of the amphora was used for the paint pot. Audrey Jawando studied the pigments found on the amphora fragments and supplied the following report:"The test for an iron base in the red pigment was based on the fact that an iron (III)-containing compound combined with hydrochloric acid produces iron ions (Fe3+). Pigment + HC1 (3M) -* Fe3+(aq) "Ironions combined with potassiumferrocyanide(K4Fe(CN)6)createssolid ferric ferrocyanide(Fe4(Fe(CN)6)3),or Prussian blue. 4Fe3+ (aq) + 3K4Fe(CN)6 (aq)
Fe4(Fe(CN)6)3 (s) + 12K+ (aq)
"If a blue precipitate forms when an unknown compound is acidified and combined with potassium ferrocyanide,then the test for iron is positive.4The reaction of the pigment with HC1 and K4Fe(CN)6was tested against two control samples:yellow ocher,which is known to contain Fe3+; and white lead pigment, which contains no Fe3+.Both the yellow ocher sample and the red pigment turned blue-green, while the color of the white lead pigment did not change. This indicates that the red pigment contains iron. The ancient term for red, iron-based pigment, t-iroS,5 can refer to pigments from a wide range of sources, including perhaps Attica,6 most 4. Odegaard,Carroll,and Zimmt commonly Kea and Lemnos, and also Cappadocia, Carthage, and Egypt. 2000, pp. 62-63. 5. For an overview,with discussion Indeed, iron oxides suitable for red pigment are quite common.7 Relatively of ancient references,see RE XVa, recent efforts to characterize miltos from different sources preclude as1932, cols. 1851-1854, s.v.Minium suming a particularsource for the red pigment found in these amphoras (W. Kroll). from the Agora.8 6. Caley (1945, p. 155) notes the Nevertheless, the examples of miltos found in Athens raise important presenceof ochers in Attica, but historical questions. If the miltos could be connected to Kea, then these without furtherreferences. 7. Photos-Joneset al. 1997, amphoraswould document Athenian use of Kean resourceswell before a Kean inscription attests to the resumption of Athenian control of that ispp.359-360. 8. Photos-Joneset al. 1997, passim. land'smiltos exports sometime before 350 B.C.9 A Cappadocianprovenance 9. IGII2 1128;Tod 1948, no. 162, for the miltos in these amphoraswould provide early and rarearchaeologipp. 181-185; and Bockh 1886, pp. 312cal evidence of imports from the Black Sea; accordingto Strabo,this miltos 317. was exported through Sinope.10Evidence for late-6th-century Athenian 10. On Sinope'srole in Cappadocian exports,see Strab.12.2.10. imports from either Egypt or Carthage is also quite rare.1 For securelyidentifiableBlack Sea The three paint pots from pre-Persian contexts in the Agora excavaamphorasin 4th-centurycontexts in tions also reflect changing activities in the area. No such residues appear the Agora, see Grace 1985, p. 21, among roughly 150 other Agora deposits closed between ca. 525 and n.52. 86 B.c. Therefore, sometime before 480 B.c., different activities seem to 11. Habermann1986; Lawall2001. have taken place in the Agora, with the result that miltos pots were no 12. Papadopoulos1996, p. 112. 13. For uses of miltos, see Photosthe that has left behind. John Papadopoulos developsuggested longer et al. 1997, pp. 359, 369. For use Jones late as ca. ment of the Agora area as a civic center may have begun as 490, of miltos in Athenian potteryproducwith a concurrent decline in the use of the area as a potter's quarter.12 tion, see Richter 1923, pp. 53-59, 96Although decoration of pottery is only one of many uses for miltos,13the 98; Vanderpool1946, p. 266; Noble apparent disappearance of miltos pots after 480 may be related to the 1988, pp. 125-127; and Schreiber decline in the use of the area for craft production. A full survey of the 1999,pp.48-52.
NOTES
FROM
THE
TINS
2
419
Agora excavation notebooks for references to miltos has not been undertaken; nevertheless, scatteredfinds from the late 5th century B.C.appearto involve either very small vessels or the use of miltos with other materials (such as tiles).14The reuse of amphoras as paint pots may have been an element of late-6th-century pottery production in the area of the Agora, while later traces of miltos may be associated with different activities. Most directly,these remains of paint pots illustrate the reuse of emptied imported amphoras.The fragments also highlight problems encounteredin studyingpoorly preservedimported goods. Futureprovenancestudies of miltos residues might clarifywhich, if any,of the sources noted above is the correctone. The fact that these samples come from fully documented archaeologicalcontexts will then allow any evidence for provenance to be considered in an appropriatesocial and historical setting. ADDENDUM
While this note was being revised for publication, Julie Unruh, a conservator at the Agora Excavations,informed me that a Corinthian B amphora bottom with toe, found in 2000 in well J 2:14, contains a large mass of red and yellow pigment. Analyses of this very well preservedfragment are expected to shed furtherlight on the reuseof amphorasas paint pots around the Agora.
THREE MASTOI FROM THE ATHENIAN AGORA KATHLEEN
M. LYNCH
Mastoi, as their name implies, are cups shaped like a female breast.15The shape is relatively rare16and is decorated in either black-figure or blackglaze with black-figure subsidiarydecoration. Although no mastoi in figural black-figure are known from the excavations of the Athenian Agora, three largely black-glazed Attic examples have come to light (Figs. 2-4), two of which have not been previously published.17The three examples from the Agora are representative of developments in the form of the mastos. The purpose of this note is to introduce these vessels and place them in a chronological and typological context. The form of the mastos is conical or slightly bulging, and ends with an articulatednipple.18The Attic examples may originally have had either 14. Apart from G 6:3 and R 12:1, two other deposits included remainsof miltos. An unused section of the foundation trenchfor the New Bouleuterion, filled late in the 5th century, included a small bowl with tracesof miltos (section B notebook,pp. 18821883), and a tile fragmentwith miltos remainswas found in the late-5thcenturydeposit F 6:2 (section KK
notebook,p. 769). Caley (1945, p. 153) lists three fragmentsof a black-glaze skyphos(P 3448) with miltos residue found in mid-5th-centuryfill in section K, nearthe RectangularPeribolos and the SouthwestFountainHouse. 15. On the name of the shape, including ancient references,see Richter and Milne 1935, p. 30; Kanowski 1983, pp. 105-106; REXIVb, 1930,
col. 2175, s.v.Mastos (H. Nachod). There is also a mastoid cup that has a flat bottom. 16. There are approximatelytwenty true mastoi listed in the Beazley Archive Database. 17. All three are mentioned in Mertens 1979, p. 23, n. 16. 18. See Schreiber1999, pp. 194-195 on potting technique.
420
MARK
L. LAWALL
ET
AL.
two horizontalcup handles or one horizontaland one verticalstraphandle.l9 The straphandle allowedthe drinkerto hold the cup.The horizontalhandle presumably allowed the cup to be hung on a wall so that the view was "anatomicallycorrect." The form is thought to originate at Corinth, for there are Corinthian mastoi dating to the first half of the 6th century B.c.20Although there are no surviving Attic mastoi from this date, mastoi appearin scenes of symposia on cups by the Attic KX Painter dated to ca. 580-570 B.c.21 Attic mastoi with black-figuredecorationappearafter550 and continue in blackfigure until ca. 500
B.C.22
The shape of the Attic mastos shows some variation, and parallels between the black-glaze examples and the black-figureforms provide some indication of chronology.23Examples earlier in the series have delicate thin walls, straightersides, and smaller nipples, resulting in an overallconical appearance.The fine potting and delicacy of subsidiarydecoration of 1 (Fig. 2) are comparableto features of an unattributedblack-figure mastos in the British Museum, dated to ca. 525 B.C.on the basis of its figural style.24Later versions have thicker walls, a more bulging, hemispherical profile, and larger, acorn-shaped nipples.25Mastos 2 (Fig. 3) is a transitional form with heavier walls, but a conical nipple. Mastos 3 (Fig. 4) is a fragment of a late mastos with thicker walls and less careful application of ornament. Its profile is similar to that of Munich 2003, which is dated 510-500
B.C.
19. Beazley 1928, p. 4, n. 2, on handles.In the Corinthianexamples the handles are horizontalskyphos handles. 20. Payne [1931] 1971, p. 312, n. 2, nos. 999, 1000, cataloguedas Middle Corinthian(600-575 B.C.)but possibly later;Amyx (1988, p. 503) suggests a Late CorinthianI date (570-550 B.C.). Bothmer (1975, p. 123) states incorrectlythat the earliest-knownmastos is a black-glazedexamplewith blackfigure subsidiarydecorationat Corinth. Nancy Bookidis and Ann Brownlee (pers.comm., Oct. 2000) concurthat no earlyblack-glazedmastoi exist in the collection in Corinth. 21. Greifenhagen1977, p. 135. Three mastoi appearhanging behind symposiastson a cup by the KX Painter,Samos 1280 (575-570 B.C.), ABV26, no. 27; SamosXXII, no. 200, pls. 37-38 (= PapaspyridiKarusu1937, pl. 57.1). Also on a skyphos by the KX Painterin Athens, NM 640 (585-580 B.C.), a figure
carrieswhat looks like a footed mastos; see CVAAthens 4 [Greece 4], pl. 3.2 and commentary,pp. 15-16 (= Papaspyridi-Karusu1937, pl. 58.1). Both of these cups have previouslybeen attributedto Sophilos. 22. Greifenhagen(1977, pp. 135137) lists twelve Attic black-figure examplesin rough chronologicalorder; Mertens (1979, p. 23, n. 16) adds several others.The latest black-figured mastos is probablythe white-ground Munich 2003, dated by Mertens (1977, p. 87, pl. 12.3) to 510-500 B.C.on the basis of its black-figurestyle. Flatbottomed mastoid cups continue the form to ca. 475 B.c.The Sotades workshoprevivesthe form in the mid-5th centuryB.C.(BritishMuseum D9, D10). 23. Greifenhagen1977, pp. 134135; Mertens 1979, p. 23. It is likely that the same potters made mastoi for decorationin black-figureand blackglaze. Unfortunately,most of the mastoi arepublishedwithout profile
drawings,making studies of the potters difficult.For other publishedblackglaze mastoi, see Dohan 1934, p. 530 (Philadelphia,UniversityMuseum, MS 4869); CVAAdria 2 [Italy65], pl. 28 [2941]:9, pl. 29 [2942]:1 (Adria, IG 2291, inv. Bocchi A32); CVAAdria 2 [Italy65], pl. 29 [2942]:2 (Adria, inv. Civico A256). Two fragments preservingonly nippleswith surrounding decorativebands may or may not be figural:Samos K 6891 (SamosXXII, no. 181, pl. 35); and Heidelberg S23, CVAHeidelberg4 [Germany31], pl. 165 [1504]:4. 24. BritishMuseum B377, Greifenhagen1977, pl. 38.1-2. 25. The three black-figuremastoi by Psiax have hollow nippleswith beadswithin to createa rattlingeffect when lifted; see Mertens 1979, p. 23. Mertens does not give a firm date for these, but links them with Psiax's associationwith Nikosthenes (ca. 525500 B.C.).
NOTES
FROM
THE
TINS
2
42I
The contexts of the three mastoi from the Agora agree with this proposed typological development. Mastos 1 is from the use fill of a well deposit, G 15:2, closed around 525 B.C.26Mastos 2 comes from deposit Q12:3, known as the Stoa Gutter Well, closed in association with the Persian destruction of Athens in 480.27Material from this deposit ranges from 525 to 480 B.c.; thus, 2 could comfortably date after 1. Mastos 3 comes from a deposit, R 12:3, closed ca. 500.28 Each of these deposits contained examples of domestic pottery, and it is likely that the mastoi represented an exotic element of a private sympotic set.
Figure 2. Mastos 1 (P 1217). Scale2:5. Courtesy AgoraExcavations, drawingK.M. Lynch.
1
P 1217
Mastos
Fig. 2
G 15:2 H. 8.4 cm; est. Diam. 5.2 cm
26. Unpublished.See Shear 1933, p. 465;AgoraXII, p. 391;AgoraXXI, p. 98. 27. Roberts1986, p. 30. On Persian
destruction deposits,see Shear1993. 28. Thompson 1956, p. 61.
Complete profile preserved.Missing much of wall, one vertical strap handle preserved.Very thin walled and light, with lustrous black glaze. Slightly convex profile; biconical nipple separatedfrom body by a raised fillet. Concave strap handle attached below rim to mid-body. Nipple reservedwith three lines of black glaze: below tip, at carination, and below fillet. On lower wall, a reservedband with a frieze of alternating black and red tongues. Above and below the tongues, three very fine, equally spaced lines. Tongues separatedby thin relief lines. Tiny added red dots at tips of scalloped upper edge of tongue frieze. Added red lines: single line in the black-glazed band above the nipple; single below reservedband; double above the reserved band; double at base of strap handle; single at the top attachment of strap handle (fugitive). Added red on sides of handle. Interior black glaze. Date: period of use, 550-525 B.C.
MARK
422
L. LAWALL
ET AL.
\V I^^^^/^WtWXE~ ^^l^ ^Figure
2
P 24556
Mastos
3. Mastos2 (P 24556). Scale2:5. CourtesyAgora Excavations, drawingK. M. Lynch.
Fig. 3
Q12:3 (Stoa Gutter Well) P.H. 5.0 cm; P. Diam. 7.25 cm Missing handles and much of upper wall. Dull and streakyblack glaze. Bulging conical profile; flat, conical nipple. Heavy wall, especially at nipple. Nipple with black-glazed top, reservedbelow. Bottom of wall reservedwith bands of decoration:frieze of irregularblack-glazed tongues separatedby vertical relief lines between two dilute horizontal lines; row of dicing dots between two thin, dilute lines; darkerdilute line; wide streakyblack-glazed band; uneven dilute line. No trace of added red lines. Interior black glaze. Date: before 480
B.C.
Published: Roberts 1986, p. 30, no. 63, fig. 20 (profile).
Figure4. Mastos3 (P 25277). Scale 2:5. CourtesyAgora Excavations, drawingK. M. Lynch.
3
P 25277
Mastos
Fig. 4
R 12:3 P.H. 6.5 cm; P. Diam. 10.8 cm Missing handles and much of upper wall. Good black glaze. Trace of reserved handle panel indicates at least one horizontal cup handle. Bulging conical profile; acorn-shaped nipple. Nipple and very bottom of wall reserved.No trace of added red lines. Interior black glaze. Date: last quarterof the 6th century B.C. Well in Stoa Shop II; see discussion in Thompson 1956, p. 61, pl. 19:c-g.
NOTES
HICXL&)
FROM
THE
TINS
2
423
N x?Ci?
A CONTEXTUAL APPROACH TO PESSOI (GAMING PIECES, COUNTERS, OR CONVENIENT WIPES?) JOHN K. PAPADOPOULOS
Much has been written about small disks or roundelsof clay fashioned from sherdsof potteryby chippingand smoothingaroundthe edges.29 They arecommonoccurrencesin varioustypes of depositsin the areaof the ClassicalAgora,andespeciallycommonin manyof the LateGeometric and EarlyArchaicwells. Among numeroussuch contexts,I illustrate hereonly a selectednumberof disksfromwell I 13:4 (Fig. 5), a well deep below the Middle Stoa excavatedin 1996 that was abandonedca. 700 B.C.30 In
discussing the possible function of the disks, Eva Brann writes:
What theywereusedfor,whetherfor counters,pucks,coversor plugs,is uncertain.Those with holes ... mayhavehad a stringto serveas the handleof a lid. Remainsof plugshaveactuallybeen foundin Mycenaeanstirrupjars... but the 7th centuryhas no commonround-mouthed,narrow-necked shape.Some,most likely,weregame-counters.Severalgamesrequiringsherddisks wereplayedin the streetsandpublicplacesof Athens (cf. PaulyWissowa,R.E., "Spiele")andthe Agoraeven possessesa die of the period(AgoraMC 84).31
29. See, among others, Burr1933, pp. 546, 603-604; Young 1939, pp. 86, 191-192, figs. 57, 142, nos. XVII 23 and C 163-173; Brann 1961, p. 342, under no. F 62; Lalonde 1968, p. 131. 30. Camp 1999, pp. 260-262. 31. Brann 1961, p. 342, no. F 62; with furtherreferenceto CorinthXII, pp.217-222. 32. Kurke1999a; 1999b, esp. pp. 254-274. The fullest overview of boardgames (lusoriatabula)remains that in REXIII.2, 1927, cols. 1900-2029 (H. Lamer). 33. Burr 1933; Kurke1999b, p. 274, fig. 9. 34. Kurke1999b, p. 273. 35. Burr 1933, p. 603, fig. 71, nos. 275-276. 36. LSJ s.v. 7oeaso6.
Amongthesevariousfunctions,thatof gamingpieces(pessoi)hasbeen singledoutby philologists,mostrecentlyby LeslieKurke,who haswritten muchon Greekboardgamesandhow to play-or not play-them.32The examplesmost commonlyillustratedfromthe areaof the ClassicalAgora arethosefromthe so-calledProtoattic"votivedeposit"publishedby DorNot all of these,however,were"ornamented bits of old potothy Burr.33 as Kurke In maintains.34 addition to terracotta tery," disks-including one of Corinthianfabricand severalcut from coarseundecoratedpotterythe deposityielded a numberof relatedstone disks,some markedwith crosses.35 It is exactlysmall stones such as these and smoothedcircular disksof claythatwereusedforvariousboardgames,includingthatin the celebratedsceneof AjaxandAchilleson the Vaticanamphora(Fig. 6). The purposeof this note is to returnto a multifunctionalinterpretation for the disks,to suggesta possiblefurtherinterpretation not notedin previousscholarship,to emphasizethe evidenceof context-which is often overlooked-and to drawattentionto the factthat a ersaoo6in Greek does not referonly to an oval-shapedstone for playingboardgames.36 The greatvarietyin the size of the disksand the type of vesselfrom whichtheywerefashionedhighlightsthe factthat not all servedthe same function.Of the examplesassembledin Figure5, the smallesthavea diameterof 0.030 m anda thicknessof 0.006 m,whereasthe largestmeasure 0.098-0.105 m in diameter,with a thicknessof 0.017-0.022 m. Manyare cut from earlierdecoratedpots ranging in date from Protogeometric throughLate Geometric.Othersarefragmentsfromvariouscoarseware
MARK
424
P 32704
P 32692
P 32707
P 32706
P 32713
P 32712
P 32720
L. LAWALL
P 32709
P 32710
P 32714
ET AL.
P 32708
P 32711
P 32705
P 32715
P 32701
P 32698
P 32697
P 32718
vessels, including massive pithoi. Although many of the disks illustrated in Figure 5 may have been used as gaming pieces, others, such as the three examples shown on the bottom row, are far less likely to have served such a function. In any discussion of the uses for such disks, the evidence of context is paramount.A few of the disks, such as those first published by Burr,clearly derive from a context with cultic overtones. In discussing the twelve clay disks from the 5th-century triangularhieron, Gerald Lalonde noted that although their provenience may tempt some scholars to identify them as "cult objects, their appearance suggests nothing so much as simple jarstoppers."37Occasionally,disks, specifically identified as pessoiby their excavators,are found in tombs, such as the group of Archaic disks recovered from tomb 11 at Eretria.38Such contexts, however, are the exceptions, not the rule.
P 32716
P 32717
P 32703 Figure5. Claydisksfashionedfrom Protogeometricand Geometric potsherdsfromwell I 13:4. CourtesyAgoraExcavations
37. Lalonde 1968, p. 131. 38. EretriaMuseum, inv. 1658816595. The majorityarefashionedfrom coarsewareor cooking vessels;only one is actuallyfrom a decoratedpot.
NOTES
FROM
THE
TINS
2
425
Figure6. Detail,Athenianblackfigurebellyamphoraby Exekias, ca. 540-530 B.C.Vatican Museum, inv. 344. CourtesyHirmerVerlagGmbH
39. See AgoraVIII, pp. 107-108; Papadopoulos1996; Papadopoulos, forthcoming. 40. AgoraVIII, p. 108. 41. AgoraVIII, p. 108. 42. LSJ s.v. :caoc6.
In Athens, the vast majority of disks were found in abandoned wells, filled with either domestic refuse or the debris from potters'workshops.39 The fill of the average well consists of the period-of-use fill at the bottom-often comprising plain or banded pots inadvertently dropped by their owners in the processof extractingwater-below a dumped fill. Apart from the sometimes copious quantities of potters'debris found in some of the wells, the dumped fill, which usually contained the most interesting pottery and other small finds, was normally deposited all at the same time when the well was abandoned in order to prevent people from falling in.40 Brann wrote that the dumped fill "wasvery probablycarted from near-by rubbish heaps or swept together from local debris. It contains, therefore, the disiecta membra of the furnishings of the neighborhood pantries, kitchens, courtyards,burial plots, and sanctuaries,but from the latter two there are only the most fragmentary remains."41Some wells contained only a handful of clay disks, while others produced numerous examplesoccasionally twenty or more disks-among a fill that yielded, on average, anywhere between a few hundred sherds to several thousand. The quantity and varying sizes of clay disks in such contexts not only underscore a multifunctional purpose, but also suggest that they were used as common, everydayitems, as was stated by Brann. A further function is plausiblysuggested both by the evidence of context as well as by the meaning ofpessos.The word IeaCoo6in Greek has a variety of meanings. In addition to referringto a stone for playing board games, it can also mean a "medicatedplug of wool or lint to be introduced It is frompessosthat the English word into the vagina, anus, etc.,pessary."42 pessary derives, and it is this meaning, in both Greek and English, that complicates the issue. In the same way that we cannot be surewhether the gaming pieces on Exekias's amphora were of stone or clay, we cannot be sure of the material that is used by the man on the cup interior, now in Boston (Fig. 7). In attributingthe cup fragment to the Ambrosios Painter, John Beazley laconically described the scene as "man o&cotcOjICvoc"
MARK
426
L. LAWALL
ET AL.
Figure7. Athenianred-figurekylix tondo fragment,with edges trimmed,by the AmbrosiosPainter, ca. 510-500 B.C.Boston,Museumof Fine Arts, Res. 08.31 b (from Orvieto). CourtesyMuseum
(oCrOdtccomeaning "wipe off,""wipe oneself").43It is ironic that the edges of the fragment itself are trimmed to define a disk resembling a pessos.A bearded man wearing a cloak loosely over his shoulders squats, leaning against a staff. According to Emily Vermeule, following Edward Perry Warren, he is using one of the proverbialthree stones, T-rEpSelov ixcvoL 7pcoxTov
Vermeule goes on to cite Paul Hartwig's sug7TioVd6c,LXtOo0."44
gestion that this representationillustratesthe passagein Aristophanes'Peace in which Trygaios, taking a corselet from the arms dealer,notes: evacTrorcwczTvyaop Sat'
; T7avOmTCMjS.eoS
Look, this will make a handy crapperconcluding: LThepec; Xs0oue.o06 NeaCOe; ()i8, rapc,p?vTz this way,if I prop it up with three stones. Neat, eh?45 In Acharnians,Aristophanes clearly states that a stone was one meansevidently the most common-for wiping oneself: .. 6. 8? Xov Xocpelv OvXO6Trc 3ouX6[?VO; V Xa3pOL XitXL?P
CtXs00ov
&pTL(o; X?xs?aClVOV'
?1Tc cL?v 8' ?XCto
T'v [doc&ptapov, xst?0l' &[ltoprcov pa3Xo[ KpocTzvov.
43. ARV2 174, no. 22. For the iconographyof defecationin Athenian vase painting,see, most recently, Cohen and Shapiro2002, pp. 88-89, pl. 22:d-e, with references. 44. Vermeule1969, p. 14, no. 14, pl. 11.2. 45. Ar. Pax 1228, 1230, J. Henderson, trans.,Cambridge,Mass., 1998.
NOTES
P 32544 Figure8. Selectedostrakafromthe areaof the ClassicalAgora.Left to right:Themistokles,Aristeides, Xanthippos,Xanthippos.Scale1:2. drawing CourtesyAgoraExcavations, A. Hooton.
46. Ar.Ach.1168-1173,J. Henderson,trans.,Cambridge, Mass.,1998. 47. QuotedbyVermeule1969, p. 14.The rhymeworksaswellin EnglishandAmericanpronunciations usingarseor ass.Fora perceptive treatmentof the fateof humanwaste, see Laporte2000. 48. AgoraXIV, p. 197.
49. Camp1999,p. 268. Forthe contextsof otherostraka,seeAgora XXV. 50. Althoughthe ekklesia regularly metat the Pnyx,the one occasionwhen the Demosconvergedon the Agora waswhenan ostracismtookplace;see AgoraXIV, pp. 50-51.
FROM
P 32573
THE
TINS
2
427
P 32903
P 32525
... and when he wants to grab a stone I hope in the darkness he grabs in his hand a fresh-shat turd, and holding that glittering missile let him charge at his foe, then miss him and hit Kratinos!46 The same was true in 19th-century England, if the statement attributedto Alfred Lord Tennyson is correct:"Iwipes me ass [arse] on a piece of grass or sometimes on a stone."47 Such a meaning is usually limited to stones. But in the same way that Kurkesuggests that clay diskswere used as gaming pieces, I wonder whether they also served for wiping oneself? Their context-common neighborhood rubbish or debris from potters' establishments dumped into abandoned wells-in Late Geometric and Archaic Athens is certainly suggestive. Many of the clay disks were found in wells that also contained stone disks or conveniently shaped pebbles. And it is useful to remember that public latrines, with running water, were a luxury not known in Athens until Roman Imperial times.48In what context other than common household debris would we expect to find discardedwipes in the Late Geometric or Archaic period? If some of the numerous clay disks recovered from early Athenian wells were used in the manner illustratedin Figure 7, then their contextif not their quantity-is remarkablysimilar to another class of Athenian sherds: ostraka (Fig. 8). At about the same time when the Agora excavators uncovered the well with the pessoi (Fig. 5), they also came across a "collection of about 144 ostraka found scattered throughout a layer of fill in the areabehind a Classical commercial building."49Although it is clear that many ostraka are simply too large and cumbersome to have served as convenient wipes, others seem ideally shaped. Again, it is their context that is suggestive, particularlythat of the numerous examples thus far recovered from the Agora and Kerameikos excavations. Indeed, it is clear that many of the latter,especiallythose from the Kerameikos,were dumped a long way from their intended place of use in any ostracism.50Epigraphers and philologists have been reluctant to consider the context and, more particularly,the afterlife of ostraka.And it is to them that I pose the following question:is it possible that some Athenians used some of the ostraka, appropriatelyinscribed with the name of a worthy political figure-and pessoi-in a manner not unlike that illustrated in Figure 7?
MARK
428
ET
AL.
BOWL OF 'ApyzeoS
A MOLDMADE SUSAN
L. LAWALL
I. ROTROFF
Examinationof the contextpotteryfrom the buildingfill of the Middle Stoa broughtto light a rareinstanceof an Attic moldmadebowl with a signature.51 E
-
~~~ ~
~
.dl
~~~~~~
3i$\
r\\2
5!,1
I~ IIJ
Figure9. Moldmadebowlof 'Apyeios(P 31702). Scale1:2.Courtesy drawingA. Hooton. AgoraExcavations,
I,
P 31702
Moldmade bowl
Fig. 9
H-K 12-14 (Middle Stoa Building Fill) P.H. 6.6 cm; est. Diam. 14.5 cm Single fragment preserving part of wall from tip of calyx to rim. Calyx: diamond-shaped tip of fern preserved,with tips of two petals of rosette to left. Wall (left to right): large bird flying left (tail and right wing preserved);Eros flying right, his arms upraised in front of him; signature APFEIOT,retrogradeand vertical, reading from bottom to top; frontal Nike (right wing, right lower arm, top of head, trace of top of left wing [not shown in drawing] preserved), small bird flying left above her. Rim: simplified guilloche, beading, double spirals crowned by leaves. Scraped groove with miltos below slightly outturned rim. Metallic black gloss, partly missing outside; fine reddish yellow fabric (5YR 6/6) with tiny sparkling inclusions. The date of the fragment can be established with some assurance.Moldmade bowls were first made in Athens in the second half of the 3rd century,probablybeginning in 224.52A lower terminus of ca. 175 for this new fragment is indicated by both stratigraphicand stylistic criteria. Its context, the building fill of the Middle Stoa, contained about 1,500 stamped amphora handles. With five exceptions that may be dismissed as Late Hellenistic intrusions, the latest date to 183 or 182 B.c., or perhaps fifteen years later,53providing a date in the first quarterof the 2nd century or earlier for the fragment in question. The motifs on the fragment point to the same conclusion, for they are amply paralleled on bowls of the Workshop of Bion, which was active between ca. 224 and 175 B.C.54Good examples of the relevant motifs, at precisely the same size, can be found on the following published Athenian bowls: Calyx
P 7001, P 20190: AgoraXXII, pp. 62, 64, nos. 154, 168, pls. 28, 31, 78, 95 (both signed Biovo;)
51. Including this new piece, only seventeensigned bowls are known from the Agora, six of which date before the middle of the 2nd century. For signatureson bowls from the Agora, see AgoraXXII, pp. 40-41. 52. For a detailedexplanationof this surprisinglyprecisedate, see AgoraXXII, pp. 6-13. 53. VirginiaGrace dates the latest Rhodian and Knidianamphorasto 183 or 182 (Grace 1985). On the basis of his reappraisalof the Rhodian chronology, GeraldFinkielsztejnputs the closing date of the fill at ca. 169/167 (Finkielsztejn2001, p. 177). 54. For the Workshopof Bion, see AgoraXXII, pp. 26-27.
NOTES
FROM
THE
TINS
Frontal Nike Small bird flying left Rim pattern
2
429
P 401: AgoraXXII, p. 62, no. 152, pls. 28, 78 P 401, P 7001: AgoraXXII, p. 62, nos. 152, 154, pls. 28, 78 (the latter signed Bicovoq) P 18654, P 18646: AgoraXXII, p. 59, nos. 125, 130, pls. 24, 25
The flying Eros on the new fragmentis not paralleledwithin the Workshop of Bion, but precisely the same figure, at a smaller scale, appears on bowls of Class 1.55Contexts suggest that bowls of Class 1 are later than those of the Workshop of Bion, dating in the second quarter of the 2nd century. They share several motifs with the earlier workshop, but in all cases at a smaller size, showing that Class 1 is derivative from the Workshop of Bion.56The discovery of the Eros motif at larger size in the oeuvre of the Workshop of Bion provides another example of the relationship. More importantly, however, the new bowl is signed. The maker inscribed his name lightly in the mold, so that it appears in very low and delicate relief on the bowl, barely visible except in raking light. This explains how the fragment escaped inventory at the time of discovery.As is normal on moldmade bowls, the name is given in the genitive: 'Apyeioo. The hand appears to be different from the one that signed two bowls of this workshop with the name Bion,57demonstratingthat at least two craftsmen worked there, probably at about the same time. Several individuals named 'Apysio are known from Hellenistic Athens.58The name could also, however,be an ethnic, and in this connection it is worth noting that there is an Argive workshop that produced bowls with marked similarities to those produced by the Attic Workshop of Bion. Gerard Siebert has described the output of an Argive shop that signs its work with a complex monogram (henceforth, the Argive Monogram Workshop).59By his estimation, it is the earliest of the Argive workshops, commencing production around 220, and therefore about contemporary with the Workshop of Bion.60It employs some motifs that are so closely similar to those used by the Athenian shop that they can only be explained by mechanical copying or shared stamping devices. The bird 55. P 589, P 9849, P 25444: Thompson 1934, pp. 378-379, D 34, fig. 65; AgoraXXII, p. 61, no. 143, pl. 26 (at far left and right in the photograph),and no. 145, pl. 27 (just right of center in the photograph).The Eros on bowls of Class 1 measures2.5 cm from toe to head, as opposed to 2.8 cm on the fragmentunder consideration here. 56. For Class 1, see AgoraXXII, 30. The other matching motifs of p. differentsizes that supportthe relationship arethe gorgoneion medallion (cf. P 11426, P 401 [AgoraXXII,
pp. 59, 62, nos. 124, 152, pls. 24,28, 78] from the Workshopof Bion with P 13684 [p. 61, no. 144, pl. 27] of Class 1); Athena Parthenosmedallion (cf. P 18662 [AgoraXXII, p. 56, no. 104, pls. 18, 98] of the Workshop of Bion with P 12068 [p. 68, no. 208, pl. 41] of Class 1); frontalNike (cf. P 401 [AgoraXXII, p. 62, no. 152, pls. 28, 78] and the fragmentpublished here,both of the Workshopof Bion, with P 9849 [p. 61, no. 143, pl. 26] of Class 1); Eros on a panther (cf. P 16208 [AgoraXXII, p. 63, no. 163, pl. 30] of the Workshopof
Bion with P 23606 [p. 63, no. 157, pls. 29, 79] of Class 1); Eros on a goat (cf. P 8101 [AgoraXXII, p. 64, no. 169, pl. 31] of the Workshopof Bion with P 12062, P 23606 [pp. 62-63, nos. 156, 157, pls. 29, 78] of Class 1). 57. P 7001, P 20190: AgoraXXII, pp. 62, 64, nos. 154, 168, pls. 28, 31, 78. 58. Osborne and Byrne 1994, s.v. 'Apy?io;. 59. Siebert 1978, pp. 50-63. The monogramcannot be readwith certainty, but it cannot representeither of the names in questionhere (Argeios,Bion). 60. Siebert 1978, p. 170.
430
MARK
L. LAWALL
ET
AL.
flying left that occurs on the fragment published here and on many other bowls from the same shop is very close to a bird found on bowls of the Argive Monogram Workshop (but not on other Argive bowls).61The rampant goats common within the Workshop of Bion are also closely similar to those on Argive bowls.62Charles Edwards, who examined and measured Attic and Argive fragments with the bird motif, reported that the Argive motif was smaller than, and thus derivative from, the Attic one.63 There are also compositional similarities between the Argive and the Attic bowls: unlike other Argive workshops, but similarly to the Workshop of Bion, bowls of the Argive Monogram Workshop display a single row of figures above a leafy calyx and, also like bowls of Bion's workshop, the points of the guilloche of the rim pattern are almost always oriented to the right. It is clear, then, that there was a close relationship of some sort between the Argive Monogram Workshop and the Attic Workshop of Bion. The reduced size of the bird motif in the Argive repertoireshows that the Argive shop borrowed motifs from the Attic one. The 'Apydo; signature, however, invites further speculation. What lies behind the name is difficult to say-an Argive metic, perhaps, or a family with ties in the Peloponnesian city. Such a person, having established his business with some success in Athens, may then have used familial connections in Argos to set up another there. Whoever owned the Workshop of Bion, he was remarkably entrepreneurial.A workshop on the island of Lesbos also manufactured bowls closely similar to those of the Attic shop, of which it is likely to have been a branch.64This degree of enterprise,unusual for the middle years of the Hellenistic period, anticipates by over a century and a half the international manufacturing networks that were to be established by Roman potters in the early years of the Roman empire.65
61. Siebert 1978, pp. 53, 354-356, 359, nos. M 45, M 50, M 55, M 87, pls. 27,28, 30. 62. Cf. P 18666 (AgoraXXII, p. 56, no. 105, pls. 18, 75, Athens, Workshop of Bion) and Siebert 1978, pp. 57, 357, nos. M 67, M 68, pl. 29 (Argos,MonogramWorkshop). 63. Edwards1986, p. 397, fig. 1. 64. Massa 1992, passim,with comments on pp. 243-244; Rotroff 1994, pp. 578-579. 65. For example,the establishment of subsidiaryworkshopsat Pisa, Lyon, and Ephesos by manufacturersof Arretinepottery (ZabehlickyScheffenegger1995).
NOTES
FROM
THE
TINS
2
43I
REFERENCES AgoraVIII = E. T. H. Brann,Late Geometric andProtoatticPottery, Mid 8th to Late 7th CenturyB.C. (AgoraVIII), Princeton 1962. AgoraXII = B. Sparkesand L. Talcott, Blackand Plain Potteryof the 6th, 5th, and 4th Centuries B.C.(Agora
XII), Princeton 1990. AgoraXIV = H. A. Thompson and R. E. Wycherley,TheAgoraof Athens:TheHistory,Shape,and Uses of an AncientCity Center(Agora XIV), Princeton 1972. AgoraXXI = M. Lang, Graffitiand Dipinti (AgoraXXI), Princeton 1976. AgoraXXII = S. I. Rotroff,Hellenistic Pottery:AthenianandImported MoldmadeBowls(AgoraXXII), Princeton 1982. AgoraXXV = M. L. Lang, Ostraka (AgoraXXV), Princeton 1990. Amyx, D. A. 1988. CorinthianVasePaintingoftheArchaicPeriod, Berkeley. Beazley,J. D. 1928. GreekVasesin Poland,Oxford. Bockh, A. 1886. Die Staatshaushaltung derAthenerII, 2nd ed., Berlin. Bothmer,D. von. 1975. Rev. of J. Boardman,AthenianBlack FigureVases:AHandbook,in ArtB 57, pp. 120-123. Brann,E. T. H. 1961. "ProtoatticWell Groups from the Athenian Agora," Hesperia30, pp. 305-379. Burr,D. 1933. "AGeometric House and a Proto-Attic VotiveDeposit," Hesperia2, pp. 542-640. Caley,E. R. 1945. "AncientGreek Pigments from the Agora,"Hesperia 14, pp. 152-156. Camp,J. McK., II. 1999. "Excavations in the Athenian Agora, 1996 and 1997,"Hesperia68, pp. 255-283. Cohen, B., and H. A. Shapiro.2002. "The Use and Abuse of Athenian Vases,"in Essaysin Honourof Dietrichvon Bothmer,A. J. Clark andJ. Gaunt, eds., Amsterdam, pp. 83-90. CorinthXII = G. R. Davidson, The Minor Objects(CorinthXII), Princeton 1952. Doger, E. 1986. "Premieresremarques
sur les amphoresde Clazomenes," in Recherches surlesamphores grecques (BCH Suppl. 13), J.-Y. Empereur and Y. Garlan,eds., Paris,pp. 461471. Dohan, E. H. 1934. "SomeUnpublished Vasesin the University Museum, Philadelphia,"AJA38, pp.523-532. Dupont, P. 1998. "ArchaicEast Greek TradeAmphoras,"in East Greek Pottery,R. M. Cook and P. Dupont, London, pp. 142-191. Edwards,C. M. 1986. "Corinthian Moldmade Bowls:The 1926 Reservoir,"Hesperia55, pp. 389419. Finkielsztejn,G. 2001. Chronologie detailleeet reviseedeseponymes rhodiens,de270 a amphoriques 108 av.J.-C. environ:Premierbilan (BAR-IS 990), Oxford. Grace,V. R. 1979. Amphorasand the AncientWineTrade(AgoraPicBk 6), 2nd ed., Princeton. .1985. "The Middle Stoa Dated by Amphora Stamps," Hesperia54, pp. 1-54. Greifenhagen,A. 1977. "Mastoi," in Festschriftfur Franz Brommer, U. Hbckmann and A. Krug,eds., Mainz, pp. 133-137. Habermann,W. 1986. "Die athenische Handelsbeziehungenmit Agypten, Karthago,und Kyrenewahrenddes 5. Jahrhundertsv. Chr.,"Miinstersche Beitrdgezur antikenHandelsgeschichte 5:2, pp. 96-105. Kanowski,M. G. 1983. Containers of ClassicalGreece,St. Lucia, Queensland. Kerameikos IX = U. Knigge,Der IX), Berlin Siidhiigel(Kerameikos 1976. Koehler,C. G. 1986. "HandlingGreek TransportAmphoras,"in Recherches surlesamphores grecques(BCH Suppl. 13),J.-Y. Empereurand Y. Garlan,eds., Paris,pp. 49-67. Kurke,L. 1999a. "AncientGreek Board Games and How to PlayThem," CP 94, pp. 247-267. .1999b. Coins,Bodies,Games, and Gold:ThePoliticsof Meaning in ArchaicGreece,Princeton.
432
Lalonde, G. V. 1968. "AFifth-Century Hieron Southwestof the Athenian Agora,"Hesperia37, pp. 123-133. Laporte,D. 2000. Historyof Shit, N. Benabidand R. el-Khoury, trans.,Cambridge,Mass. (originally publishedas Histoirede la merde [Prologue],Paris 1978). Lawall,M. L. 1995. "TransportAmphorasandTrademarks:Imports to Athens and Economic Diversity in the Fifth Century B.C."(diss. Universityof Michigan). .1997. "Shapeand Symbol: Regionalismin 5th-CenturyTransport Amphora Productionin NortheasternGreece,"in Tradeand Productionin PremonetaryGreece: Productionand the Craftsman (SIMA-PB 143), C. Gillis, C. Risberg, and B. Sjoberg,eds.,Jonsered, pp.113-130. .2001. "ALate Archaic Punic Amphora,"Hesperia70, pp. 169171. Massa,M. 1992. La ceramicaellenistica condecorazione a relievodellaBottega di Efestia(Monografiedella Scuola archeologicadi Atene e delle Missioni italianein Oriente 5), Rome. Mattioli, S. P., ed. 1998. Bonifiche e drenaggiconanforein epocaromana: Atti del Aspettitecnicie topografici. Seminariodi studi,Padova,19-20 ottobre1995, Padua. Mertens,J. R. 1977. Attic WhiteGround:Its Developmenton Shapes OtherthanLekythoi,New York. 1979. "SomeNew Vasesby Psiax,"AntK 22, pp. 22-37. Noble,J. V. 1988. TheTechniques of PaintedAttic Pottery, rev. ed.,
London. Odegaard,N., S. Carroll,andW. S. Zimmt. 2000. Material Characterization Testsfor ObjectsofArt and Archaeology, London.
Osborne,M. J., and S. G. Byrne. 1994. A Lexicon of GreekPersonal Names 2: Attica, Oxford.
Papadopoulos,J.K. 1996. "The Original Kerameikosof Athens and the Siting of the ClassicalAgora," GRBS37, pp. 107-128. .Forthcoming. CeramicusRedivivus: The Early Iron Age Potters
MARK
L. LAWALL
ET AL.
Field in the Area of the Classical Athenian Agora (Hesperia Suppl. 31),
Princeton. S. 1937. "SophiPapaspyridi-Karusu, los,"AM 62, pp. 111-135.
Payne,H. [1931] 1971. Necrocorinthia, repr.College Park. Photos-Jones,E., A. Cottier,A. J. Hall, and L. G. Mendoni. 1997. "Kean Miltos:The Well-Known Iron Oxides of Antiquity,"BSA 92, pp. 359-371. Richter,G. M. A. 1923. The Craftof Athenian Pottery:An Investigation of the Techniqueof Black-Figured and Red-FiguredAthenian Vases,New
Haven. Richter,G. M. A., and M. Milne. 1935. Shapes and Names ofAthenian Vases,
New York. Roberts,S. R. 1986. "The Stoa Gutter Well, a Late ArchaicDeposit in the Athenian Agora," Hesperia 55,
pp. 1-74. Rotroff,S. I. 1994. Rev.of Massa 1992, in BJb194, pp. 578-580. Samos XXII = B. Kreuzer, Die attisch schwarzfigurige Keramik aus dem Heraion von Samos (Samos XXII),
Bonn 1998. Schreiber,T. 1999. Athenian Vase Construction.A Potters Analysis,
Malibu. Shear,T. L. 1933. "The Campaignof 1932,"Hesperia2, pp. 451-474. Shear,T. L., Jr. 1993. "The Persian Destruction of Athens: Evidence from the Agora Deposits,"Hesperia 62, pp. 383-482. Siebert, G. 1978. Recherchessur les ateliers de bols a reliefs du Peloponnese a l'epoquehellenistique (Bibliotheque
des Ecoles frangaisesd'Atheneset de Rome 233), Paris. Thompson, H. A. 1934. "TwoCenturiesof Hellenistic Pottery," Hesperia3, pp. 311-480. .1956. "Activitiesin the Athenian Agora: 1955," Hesperia 25,
pp. 46-68. Tod, M. N. 1948. A Selection of Greek Historical Inscriptions 2, Oxford.
Vanderpool,E. 1938. "The Rectangular Rock-Cut Shaft:The Shaft and Its Lower Fill,"Hesperia7, pp. 363411.
NOTES
.1946. "The RectangularRockCut Shaft:The Upper Fill," Hesperia15, pp. 265-336. van Doorninck, F. H. 1989. "The Cargo Amphorason the 7th-Century YassiAda and llth-Century Serge Limani Shipwrecks:Two Examples of a Reuse of ByzantineAmphoras as TransportJars,"in Recherches sur
FROM
THE
TINS
la ceramique byzantine(BCH Suppl. 18), V. Deroche and N.-M. Spieser, eds., Paris,pp. 247-257. Vermeule,E. 1969. "Some Erotica in Boston,"AntK 12, pp. 9-15. Young,R. S. 1939. Late Geometric Gravesand a SeventhCenturyWell in theAgora(HesperiaSuppl.2), Princeton.
MarkL. Lawall UNIVERSITY OF MANITOBA DEPARTMENT
OF CLASSICS MANITOBA
WINNIPEG,
R3T
2M8
CANADA
lawall@cc.
umanitoba.
ca
AudreyJawando FREDERICK 99
LAW
WARREN
OLMSTED
NATIONAL
HISTORIC
STREET MASSACHUSETTS
BROOKLINE,
02445
audrey_j [email protected]
KathleenM. Lynch UNIVERSITY
OF CINCINNATI
DEPARTMENT P.O.
BOX
OF CLASSICS
210226 OHIO
CINCINNATI,
kathleen.lynch@uc.
45221
edu
JohnK. Papadopoulos UNIVERSITY
OF CALIFORNIA,
DEPARTMENT THE A2IO LOS
OF CLASSICS
COTSEN
INSTITUTE
ANGELES
AND OF ARCHAEOLOGY
CALIFORNIA
[email protected]
SusanI. Rotroff WASHINGTON
UNIVERSITY
DEPARTMENT
OF CLASSICS
CAMPUS ST.
LOS
FOWLER ANGELES,
LOUIS,
BOX
1050
MISSOURI
63130
[email protected]
2
90095-1510
SITE
433 Zabehlicky-Scheffenegger,S. 1995. "SubsidiaryFactoriesof Italian SigillataPotters:The Ephesian Evidence,"in Ephesos: MetropolisofAsia, H. Koester, ed. (HarvardTheological Studies 41), ValleyForge, pp.217-228.
HESPERIA
7I (2002)
GREES
EPIGRAPHICALINDEX
PERSONS 'A¢oyetog, potteron moldmade bowl,224-ca.175a.,428 'AtoesTouxog /'AtoesTotueveog, on filnerary monument fromAigina, s. IV a., 4021 'Atoesto,uevg, s. IV a.,fatherof 'A¢ototouXog, 4022 'Otosotoc8[ocg], on ceramic beehivefromIsthmia,fin.III a.,349,355 (5) INSCRIPTIONS
STUDIED OR EMENDED
AgoraInventory Number I 16,16b 378-395 I 17 378-395 I 18 378-395 I 93 378-395 P 31702 428 CorpusCultusEquitisThrx aczz I33 216 I 34 211 I40 212,214 I 54 214 I 85 214 I 162 212 II 200 216 II 445 216 IV29 212 V23 218 .
.
Etudesthasiennes III 21 III 22
284 284
FouillesdeDelphes III.3 212
268
436
Hesperia
2,1933,pp.149-169 10,1941,pp.284-295
378-395 378-395
Die Inschriften vonEphesos
1518
265
Die Inschriften von Olympia
651
284
Inscriptiones Graecae I3
590 596 605 948 1357
266 266 267 267 188-189
Inscrittiones Graecae II2
10 3040 3042
385-386,392-393 284 285
Inscriptiones Graecae X1V
1
263-264
Inscrittiones Graecae in Bulgariarepertae
I 345 II 512 II 569 II 796 IV2134
226 217 216-217 225-226 213
IsthmiaInventory Number IP 586
355 (5)
Labraunda
III.2,p.8 (no.13) III.2,pp.9-10 (no.14) III.2,p.13 (no.16) III.2,pp.14-15 (no.17)
275 274 275 275
Supplementum Epigraphicum Graecum
XXII79 I 841
188-189 263-264
SyllogeInscriptionum Graecarum3
6 l9a 29
265 268 269-270
ForIG references andHesperia,SEG,andAgoraXVI references to textswithnoIG reference in A. S. Henry, "TheAthenianStateSecretariat andProvisions forPublishing andErectingDecrees," pp.91-118,seeappendix,pp.115-117.Notealsothecorrigenda oppositep.324.
Published Recently
JOSEPH W. SHAW, ALEYDISVAN DE MOORTEL, PETERM. DAY, VASSILIS KlLlKOGLOU A
and
175 pages,66 figures,16 tables 30. HesperiaSupplement ISBN 0-87661-530-2. July2001.Paper$35.00
Function
Crete:
Central
South-
in
Kiln
Ceramic
IA
LM
Production
Pottery
kilnfoundin excaAn in-depthstudyof theLateMinoanIA cross-draft vationat Kommos,Crete.The kilnis of a typepopularduringthe Neohasallowedtheauthors palatialperiod,andits goodstateof preservation useaswellasaboutthe and layout internal original its about to speculate locallyproroofthatcoveredit. Muchof thelargequantityof obviously in detail,allowing withthekilnis analyzed ducedpotteryfoundassociated characandtechnical forthefirsttimethestudyof theshapes,decoration, The kiln. IA LM specific a in fired been teristicsof vasesknownto have usedto iltechniques of analytical program bookpresentsan integrated similarities the compositional lustratethe rangeof firingtemperatures, in theclayused,andaspectsof thefiringprocessandthe anddifferences to ourknowlcontribution Offeredhereis avaluable upperkilnstructure. atthebeof ceramicproduction andorganization edgeof thetechnology ginningof theLateMinoanperiod,whichwillforma basisforstudiesof andexchange. potteryprovenience
,
.
.
-
-
-
w
{
tEr
S
'
t -
@s
a
% t
.i z
) -
d . ''<S,',,
,Q,:3
(HARLES K. WILLIAMS 11AND NANCY BOOKIDl 5, EDITORS Corinth,
the
Centenary:
I896
Forthcoming
I996
Thisbookpublishes twenty-five of thepaperspresented attheDecember 1996symposium heldin Athensto celebrate theone-hundredth anniversaryof theAmericanSchoolof ClassicalStudiesexcavations at Ancient Corinth.Thepapersareintendedto illustrate therangein subjectmatter of research currently beingundertaken by scholarsof AncientCorinth, andtheirinclusionin onevolumewillserveasa usefulreference workfor nonspecialists. Eachof the topics whichvarywidelyfromCorinthian geologyto religiouspractices to Byzantinepottery is presented by the acknowledged expertin thatarea. Thebookincludesa fullgeneralbibliography of articlesandvolumes concerning material excavated at Corinth.As a summary of onehundred yearsof research, it willbe usefulto generations of scholars to come.
Ca.500 pages,400 illustrations Corinth XX ISBN0-87661-020-8 January 2003. Clothca.$75/ £57.50
HESPERIA SUPPLEMENTS 13 MarcusAurelius:Aspects of Civicand CulturalPolicyin theEast,
byJamesH. Oliver(1970).$15.00 14 ThePoliticalOrganizationofA2ica, byJohnS.Traill(1975).$15.00
16 A Sanctuaryof Zeuson MountHymettos,byMerleK.Langdon(1976). $15.00 17 Kalliasof Sphettosand theRevoltofAthensin 286 B.C., byT. LeslieShearJr.(1978).$15.00 19 Studiesin A2ic Epigraphy,History,and Topography Presentedto Eugene Vanderpool (1982).$15.00
20 Studiesin AthenianArchitecture, Sculpture,and Topography Presentedto HomerA. Thompson(1982).$15.00 21 Excavationsat Pylosin Elis, byJohnE. Coleman(1986).$25.00 22 A2ic GraveReliefsThatRepresentWomenin theDressofIsis, by ElizabethJ.Walters(1988).$40.00 23 HellenisticReliefMoldsfrom theAthenianAgora,by ClaireveGrandjouan (1989).$25.00 24 ThePrepalatialCemeteries at Mochlosand Gourniaand theHouseTombsof BronzeAgeCrete,byJeffreyS. Soles(1992).$35.00 25 Debrisproma PublicDiningPlacein theAthenianAgora, by SusanI. RotroffandJohnH. Oakley(1992).$35.00 26 TheSanctuaryofAthenaNike inAthens:Architectural Stagesand Chronology,by IraS. Mark(1993).$50.00 27 Proceedingsof theInternationalConference on GreekArchitectural Terracottas of the Classicaland HellenisticPeriods,Decemberl2-15, 1991,
editedby NancyA. Winter(1994).$120.00 28 Studiesin ArchaicCorinthianVasePainting,by D. A. AmyxandPatricia Lawrence(1996).$65.00 29 TheAthenianGrain-TaxLaw of 374/3 B. C., by RonaldS. Stroud(1998). $35.00 30 A LMIA CeramicKiln in South-CentralCrete:Functionand Pottery Production,byJosephW. Shaw,AleydisVande Moortel,PeterM. Day, andVassilisKilikoglou(2001).$35.00
