Sophilos' Vase Inscriptions and Cultural Literacy in Archaic Athens

Classical Association of Canada

Sophilos' Vase Inscriptions and Cultural Literacy in Archaic Athens Author(s): Martin F. Kilmer and Robert Develin Source: Phoenix, Vol. 55, No. 1/2 (Spring - Summer, 2001), pp. 9-43 Published by: Classical Association of Canada Stable URL: http://www.jstor.org/stable/1089022 . Accessed: 06/03/2011 14:55 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=cac. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. 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].

Classical Association of Canada is collaborating with JSTOR to digitize, preserve and extend access to Phoenix.

http://www.jstor.org

SOPHILOS' VASE INSCRIPTIONS AND CULTURAL LITERACY IN ARCHAIC ATHENS MARTIN F. KILMER AND ROBERT DEVELIN

I.

INTRODUCTION

IN THIS ARTICLEI EXAMINETHE INSCRIPTIONS on the extant vases and vase fragmentsof the earlyAttic black-figurepainterSophilos,whosefloruitis generally thought to be 590-570 B.C.(580-570, accordingto Boardman1974: 18). My purpose in undertakingthis study was to discoverwhether Sophilos ought to be considered literate, and if so, in what senses of that word. I find that he was a competent and consistent speller; that he knew how to constructsimple sentences; that he was well in tune with his society'smythology;in short, that he was literate both in the literal meaning of that word and in what we might call "culturalliteracy."Sophilos was one of the firstAttic paintersto paint letters on his vases, and he wrote at a time when the Greek alphabetwas probablyless than two hundredyearsold.l Under these circumstances,applyingthe standards of, for example, late twentieth-centurywestern Europe or North America for determinationof "literacy"is anachronistic. Judging early archaic inscriptions This paperis one of a serieswrittenin the contextof a long-termprojectundertakenby the two authors named. In each case, the first-namedauthor is the one most implicatedin the articlein question. Kilmer has consulted Develin at every stage of the preparationof this article, but has written this articlein the firstperson.Only the Conclusionsshouldbe consideredto be of truejoint authorship.I am particularlygratefulto Louis Kelly for his help in the linguisticportionsof this paper;to Eleanor Dickey for useful commentaryon linguistic questions and for putting me into contact with Rudolf Wachter, whom I also thank for comments. To David HarveyandJohn Was, my specialthanks for readingearlydraftsand for supportin manyguises. My thanksalso to the two anonymousreadersfor Phoenix,many of whose suggestionsI have incorporatedinto this finalversion.Any errorsremaining are my own. Epigraphicalnote: Sophilos' archaic letter forms (see Fig. 1) differed significantlyfrom those used in later Ionic script. We have preservedhis spellings in transcriptionto the lower-case system in common use for archaicvase inscriptions(see Immerwahr1990; Jeffery 1990). Sophilos' "heta" lambdaby X, (replacedeventuallyby the rough-breathingmark)is representedby h, his "check-mark" his lambda-shapedgammaby y, his sigma (almost alwaysthree-bar)by a and S, his "plus"-shapedkhi by X, and his pi with one short leg (easily confusedwith the classicalupper-caseform of gamma) by 7. Almost all forms used on the vases are reservedlater for upper-caseletters. We have thus chosen not to "correct"the spelling of words which in Classicaltimes were spelt with geminate consonants: this was not the usual archaicform. "Correction"in publicationsis one of the things which has led to the general-and quite erroneous-perception that the archaicAttic paintersof potteryhad as a rule a shakycontrolon the writing of Greek. For the most part, Sophilosuses letter forms similarto those used by Kleitias,Exekias,and the Amasis Painter. The cataloguenumbersgiven in parenthesesare based on Beazley 1956 and Beazley 1963 (B = ABV, R = ARV). For furtherdetail, see the Appendix,below, 37. 1For furtherdiscussion,see, e.g., Powell 1991; Robb 1994. This topic is controversial.

9 PHOENIX, VOL. 55 (2001) 1-2.

10

PHOENIX

on the basis of Attic spelling conventions of the Classical period, though less spectacularlyanachronistic,presents some of the same difficulties.The research of Leslie Threatte on phonology (Threatte 1980) has laid the foundations for establishingthe patternsof archaicAttic spelling;and Henry Immerwahr'sAttic Script(1990) presents a substantialbody of transcribedinscriptions, the great majority of them from pottery, including many new readings. Allen's Vox Graeca3(1987), though focused on classicalGreek literarylanguage,gives useful information on the development of the written language and its relations to the spoken language after the period of direct interest here. In many cases, extrapolationbridging to the evidence for the earlierperiod coveredby Threatte has led to unexpectedconclusions. In order to determine the "literacy"of someone in Sophilos' circumstances, we ought to look primarilyat two questions. First, when the painter writes the same word on two (or more) vases, is his spelling consistent? We might call this "internalconsistency."Second, when other vase painterscontemporary or nearlycontemporarywith Sophilos write these words, are their spellings the same as his, or at least close enough that we should most reasonablyassessthem as variants? I have cast my net ratherwide within the range of archaicAttic painted pottery. The degree of consistencyis significantenough overallthat it is clearthat many paintersof archaicAttic potteryhad little difficultyin writing whateverthey chose to write on a vase. Sophilos, as a pioneerin this new venture, might reasonablybe expected to have considerabledifficulty. In fact, as I hope this paper demonstrates,he exhibits a high degree of internal consistency, and his consistencywith other painters of the early Archaic period, and even with paintersdown into the earlyfifth century(at least for most words for which there are later matches), is remarkablyhigh. I emphasize that-with one exception, the "caption"identifying the (funeral)games of Patroklos-Sophilos has so far revealedonly two certain classes of inscriptions:"signatures"-allthose that are consisting of only two or complete or near complete are "painter"-signatures, threewords-and the names of heroes and divinities.This restrictionreducesthe probabilityof matches with other painters. We are fortunatethat some of the most crucialnames-names with spelling that has appearedidiosyncraticin the extremewhen comparedto their classicalversions-prove to have enough exact matchesthat we must acceptthem as normalspellingsfor Sophilos'time. Although we have remnants of many of Sophilos' vases, and several fully or almost fully reconstructible,there are relativelyfew inscribedvases extant in his oeuvre. He is early among black-figurepainters, belonging to the second generationof the style. His predecessorsin the developedmediumwerethe Nettos Painter,2the Gorgon Painter (whose careerbegan while the Nettos painterwas 2The name of the centaurwhom he shows in conflictwith Heraklesin his most famouswork, the Modern scholarshave regularizedthe spelling to amphorain Athens (B4/1), the painterspells N?TOS. conformto laterAttic usage, doublingthe tau. They havenot calledhim the "NessosPainter"-which

SOPHILOS' VASE INSCRIPTIONS

' - rt-, i .-?..*? -

l/ t:;^2a -nos I

11

Fig. 1: Fragmentfrom the ErskineDi(B40/16.1)illustrating Sophilos'normal letter-forms in his signatureandthe name Peleus.Photo:courtesyof BritishMuseum, London,afterWilliams1983,fig.25 (portionsof break"removed" by computerenhancement).

still active,andwho maywell havebeen Sophilos'teacheror at the leastthe senior memberof the workshopin which he startedhis career:see Bakir 1981: 1-3; Beazley 1986: 16), and a few other painterswhose careersare brieflynoted in Beazley'sABV and at greaterlength in his TheDevelopmentofAtticBlack-Figure (Beazley1986: 12-16 and pls. 10-15). BeforeSophilos,Attic paintinggiveslittle in the wayof inscriptionsapartfrom the names of a few divinities and some heroes and their opponents.3 Sophilos may be the first Attic potterypainterto give his name with the verb eypaaoev, meaning "drew"or "painted"(Fig. 1): later Attic orthographywould make this Eypavev; but the spellingwith cris standardfor the time in question(see below, wouldconformto laterspellingin mostof the Greekworld-perhapsto avoidconfusionwith the of theNewYorkNessos."I haveoftenwondered "Painter havenotadopted proto-Attic whyscholars thesensibleexpedient of callingthisearlyblack-figure artisttheNetosPainter,usingthespellingwe knowhimto haveused.Threattereports(1980:540),in discussing andvariants, thename'OSuoaace to the Prometheus occurrence of the spellingNEaooon a Tyrrhenian (B96/13)attributed amphora Painter.Tyrrhenian to be of Atticmanufacture, andthisspelling aregenerally considered amphorai remainsexceptional.The heroOdysseusis foundwith the spellingOXur8u; (retr.,Olyteus)on a vaseby Kleitias(B76/A)fromEgypt,as well as on the FrancoisVase(B76/1),in the fragmentary chariot-race a kantharos scene:Wachter1991:89, no.50. In anotherexample, bythe SoklesPainter, OXuTeu<;>is in attendancefor the armingof Akhilles (B173/1). 3See Immerwahr1990: 20-21 for examples.

PHOENIX

12

32-33). He is thus the first Attic potterypainterknown to have signed his work (Beazley1986: 16).4 He is much more generousin his use of labelledhuman and divine figuresthan was the norm with his predecessors;and he has left us many more inscriptionsthan have they, and he used the inscriptions,it would seem, primarilyin orderto identify these mythologicalfiguresfor his audience. II.

FRAGMENTARY VASES: SIGNATURES

(a) FragmentsfromtheAthensWeddingDinos (B39/15):MythicalWedding We may begin with the painter'ssigned vases, though they are not numbered among his earliest. His fragmentarydinos in Athens has preservedcomplete on cypacasv (Sophilos drew fragmentc the retrogradeverticalinscriptionEo4tiXoq5 of the in this Plate All 1). [this]; sigmas painter-inscriptionare in the form normal for orthogradesigma in later inscriptions(that is to say, their direction reversedfrom that of the standardretrogradesigma:thus their firstdown-strokes slope down towards the left).6 The alpha is of the closed type,7 the nu of the high-swungearlytype;and, in short, all of the lettersareof formsunexceptionable for their period. Sophilos has chosen a prominent position for the inscription. It is written vertically,retrograde,to our right side of the bronze-decorateddoor of a building. 4At least one of his inscribedvases, the Athens Wedding Dinos (B39/15) was dedicatedon the Akropolisat Athens. It was large and ornate(on the same scale as the ErskineDinos), and was most probablymade specificallyfor dedication. If so, it probablycarrieda painted dedicationinscription (Sophilos left no graffito inscriptions),which has not survived. It will certainlybe worth the effort to study the remainsof earlydedicationson pottery-and particularlythose of the sixth century-as a corpus. It can usuallybe determinedwith considerableprobabilitywhich graffitoinscriptionswere done before firing(thus belonging to the manufacturingprocess)and which were incised afterfiring. There is a subset of pots with dedicatoryinscriptions"painted"on them-some found far from their places of manufacture(for example, Ionian pottery found at Graviscaein Etruria: Various 1993), others, such as those from the Akropolisat Athens, verydose to home. Dedicatedmanufacture,often asalapXe "firstfruit,"denotes a specialrelationshipbetweenpotterand/orpainterand the divinityso honoured-or may reflect the intentions of a patron. Such dedicationsmay begin as earlyas the late eighth century.They certainlycontinue to the end of the sixth century.Sophilos'fine dinos was likely the potter/painter'sdedicationon the Akropolisto Athena, patronessof craftsmen(especiallypotters) and patronessof Athens. 5Vowel values in this name are not known; nor can we know whether later Attic usage would have doubled the lambda or not. Beazley (1956: 37) says it is "impossibleto say whether the name S is c6thtXoS, lo4iXoqS,or l64tXXo."These three forms of the name are attested;but they are not the Immerwahr (1990: 21) transcribesthe painter'sname as though it were E6jtkoS, only possibilities. without discussion. So also does Threatte (1980: Index). I retain the archaicspelling, and add no accentbecauseI am not convincedthat we know where the accentought to go. See Pape and Benseler 1959: nos. 1430-78; Kirchner1966: nos. 261, 299. 6Immerwahr(1990: 21) refers to these as "reversed."Either the "direction"of sigma had not as yet been fixed, or Sophilos was interestedin letters for their decorativequalitiesas well as for the additionalmeaning they could give his figures. 7Beazley 1956: 39, no. 15 gives (in a rareslip) the open alphaform.

SOPHILOS'VASEINSCRIPTIONS

13

This building is generallytaken to be the house of the bridegroom(Peleus?),the place towardswhich the procession (the vase's principal subject) moves. This suggestionis supportedby the similarbuilding on the ErskineDinos in London (B40/16.1; Fig. 1), which we shall discuss later (below, 14, 28-30); and, at a slightlylater date, by the building Kleitiasuses on the FrancoisVase (B76/1) for the starting-pointand ending-point of the same scene.8 The borderbelow the buildingis similarto that preservedin fragmentsb, f, andg of the samevase.9Also parallelto the door, and thus flankingthe inscription,area Doric column (painted white, and with what may be intended as a narrowbase-a feature not known in large-scaleDoric architecturein stone of the Archaicand Classicalperiods)to our right and an anta in black,with incision markingits base. The column capital is broad and shallow, set off by a convex necking-bandwhich somewhat recalls the shallowhawk's-beakmouldingsof some capitalsof the so-called "basilica"at Paestum.0lOn fragmentb is found the lower portionof the left anta of the house along with one foot of an individualwho must be facing awayfrom the building. On the analogyof the Erskine Dinos, this would be the bridegroom:the minor detail of horizontalincised lines decoratingthe ankleportionof the boots of both individualsgives some slight furthersupportto this hypothesis. On the Erskine Dinos, unfortunately,only two pairsof male feet arepreservedother than those of the bridegroom:those of Dionysos and the humanforefeetof the centaurKhiron, both of whom arebarefoot. Other fragmentsprovidesome of the guests in processionon their way to the celebration.On fragmentg, the names Acro (Leto) and Xapt9ko (Khariklo)are written orthogradeverticallybefore the two female figures, without interpunct, but with a space left between the words largerthan that between other pairs of letters.The names arewritten free-hand,with irregularlyplacedbases, not along the base-line (usuallyeither not drawn in, or meant to be invisible after firing) used often by laterpainters,and occasionallyused by Sophilos. Leto is the mother of Artemis and Apollo; Khariklo(much less well known to us) the wife of the centaurKhiron. Sophilosuses 9 (qoppa)in a numberof placeswhere laterAttic spellingwould use K (kappa). He uses it, in preservedinscriptions,only precedingk (lambda), as here, on the Athens Dinos fragment from Pharsalos (B39/16), and on the ErskineDinos (Fig. 3). On the other hand, Kleitias,workingmost likely no more than ten yearsafterthis, spells Khariklowith kappas(Wachter1991: 90, no. 82). Kleitiasdoes use qoppaonce on the FrancoisVase, for the dog 9opaX; (Qorakhs, "raven")in the KalydonianBoar scene (Wachter 1991: 87, no. 16). This is a "speakingname":gutturalqoppaperfectsthe imitationof the raven'scry. Kleitias' 8For the many inscriptions on the Erskine Dinos, see Bakir 1981: Al, pls. 1-2, figs. 1-4. Williams (1983: 13-34) gives additional fragmentsand transcriptionsof the inscriptions. For the building on the ErskineDinos, see Williams 1983: figs. 25-26. 9Numberingas in Bakir 1981: pi. 3. 1 See, for example,Lawrence1983: 156, fig. 121.

14

PHOENIX

use here confirms,for me, that he used qoppafor the gutturalstop corresponding to kappa. Sophilos' use of qoppa preceding lambda suggests a strongly gutturalised pronunciation. X following K forces K into the guttural. (For a concrete demonstrationof this, try speaking the words "capital"and "closet"aloud, and feel the positions of those consonantsin your mouth and upper throat.) Jeffery (1990: 33-34) describesqoppa as "thegutturalk before the vowels u and o."This descriptionworkswell with the earlyinscriptions(of the seventh and the earliest sixth centuries)as she reportsthem (1990: 70); and with the single preserveduse of qoppaby Kleitiason the FrancoisVase. Sophilos'preservedinscriptionsnever show 9 immediatelyprecedinga vowel, nor do we have any case, in his oeuvre, of K immediately preceding o or u. We do not know what his choice would have been for this combination.Threatte comments (1980: 21) that "(Qoppa)is regularlyused before o-vowels, even when 1 intervenes."In this context he may see qoppaas a spellingvariant,ratherthan as a letter indicatinga differentsound. It is certainlyalso true that Sophilos' and Kleitias'knowledge of letters is not based solely on the Attic tradition (see Immerwahr1990: 155; Threatte 1980: 21-23). The animalfriezes of both paintershave strong Corinthianties; and the Corinthianpainterscontemporarywith these two Attic painterswere becoming adept at supplyingnames for their mythologicalcharacters.Like Sophilos, these painters regularlyuse qoppa before lambda: see, for example, neptQ9uuevo; (Periklymenos)on the amphorafrom Cerveteri(Paris E640: Amyx 1989: 270, no. 6; Carpenter1991: 269). Sophilos'retrogradedipintoInarpoQ9uSi; atka (see Plate 2) providesanotherparallel.1l Khironmay also have been shown on this dinos, as he is on both the Erskine Dinos and the FranpoisVase. As in both of those, he would apparentlyhavebeen shown separatefrom his wife, not beside her-otherwise we ought to see at least tracesof his figureor of his name on this fragment.Both Leto and Kharildowear complex dresseswith polychromeanimalsreminiscentof the Corinthianpainted creatureswho serve as models for so much in Sophilos' animal parades.12To our right, walking in front of these two figures,is anotherpair of female figures: Ae[!]ETcp (Demeter) is written retrogradetop-to-bottom along the line of her back, letter-basestowardsher (the first two epsilons and the tau partial),haccta (Hestia) orthogradecurvingdownwardbefore the face of the woman in the lead (with closed heta, c in the form later normal for retrogradeinscriptions): the name Hestia clearlygoes with the figurefartherfrom the viewer.The placement of Demeter'sname, written as it is along the back of the nearerfigure,reinforces that. Next to our right is a figure only partiallypreserved,probablyIris. No inscriptionremains;but comparisonof her apparentposition at the head of the 11Threatte (1980: 22) transcribesthis fIaTpoQxu;with Kin placeof 9, and cites it as an exception to Sophilos'usage in Xaptl9o. Immerwahr(1990: 21, no. 62) transcribesthe hero'sname correctly, but does not comment on the qoppa. See below, 20. 12On such dresseswith "story-friezes"and the techniquesneeded to makethem in the absenceof true tapestry,see Barber1992: esp. 112-117.

SOPHILOS'VASEINSCRIPTIONS

15

procession,her costume, and her attributesto those of the figurelabelled"Iris"on the Erskine Dinos can leave little doubt. There is the same white flesh, similar short-sleevedand short-skirtedkhitoniskos,13 and the lower end of the herald's staff (complete on the ErskineDinos) which shows the goddess'soffice. For this occasion,she maybe imaginedto act as a sort of travellingbutler,introducingthe divinitiesto the bridegroomas they arrive,as well as announcingthe arrivalof the processionas a whole. On the analogy of both the Erskine Dinos and the parallel scene on the Francois Vase, this fragment ought to represent the leading figures of the wedding procession (those closest to the bridegroom'shouse), although on the Francois Vase Iris walks beside Khiron the centaur, whose human foreparts actuallyslightly precedeher. Kleitiashas not felt bound to follow everydetail of any earlierversion known to us:14for instance, his Iris' short chiton is decorated in a form of plaid ratherthan the solid colour of both Sophilan examples;and KhirontakesPeleus'hand abovean altarwhich Sophiloscertainlydid not include on the ErskineDinos, and cannot be shown to have used here. Fragmentjgives us the head of noac[i5ov] (Poseidon),his name slopingdown retrogradebehind his head, the c partial. Partly obscuredby him and a touch ahead is his wife Av[ItTptpTc] (Amphitrite),written orthogradeto the right from her forehead(v partial).15Though her name is poorly preserved,the reading is highly probable:the couple appeartogether on the ErskineDinos, the goddess's name spelt with initial Av as here, and also on the FrancoisVase, with the same spelling (Wachter 1991: 90, no. 68; 99). On the latter two vases, the divine couple are mounted on a chariot. Here, although it would be consonant with their riding in a chariot that their heads almost touch the upper border of the scene (we can comparethe figures of fragmentc, for whom the reins and whip or goad are partiallypreserved),the heads of others who are not chariot-borne come almost into contactwith that border;and the walkingIris'head must have touched. Fragment d shows a serpentine creaturewho (again on the analogy of the ErskineDinos and-though less completely-the FrancoisVase [Wachter 1991: 89, no. 91; 99]) could be Okeanos.16The drapedmale figurewalking on the far side of the sea-serpentbody, however, has no equivalenton the Erskine Dinos, where the space behind Okeanos'human torso is taken up by two female figures (see below, 30).17 Beazley had alreadysuggested (1986: 27) for the equivalent 13Thoughherethe hem/selvageis markedby shortverticalincisions,ratherthanthe "painted" maeander of thefigureon the ErskineDinos. 14Whether that"model" wasvisual,poetic,or basedmoreon his ownimagination, whateverits originalinspiration.

15Forv in p beforeplosives,seeThreatte1980:595-596.AvIvrptpIT is placeof thelaterstandard thestandard collectionwhich spellingon the manyCorinthian pinakesin the BerlinCharlottenburg presentthesegods(vidi). 16Butsee Beazley1986:27 andbelow,n. 18. 17Theremaybe twowalkingfiguresratherthanone. The darksegmentof clothingon the right, however,looks-even in enlargedcolourphotographs-justlikethe insideof the sleeveto ourleft.

16

PHOENIX

figures (poorly preserved)on the Franfois Vase that the human-bodiedfigure must be Okeanos, and the serpent/compositecompanion perhapsone of the Tritons.18 This seems also the best interpretationhere, with Okeanos (human in body; nothing remains of his head) in charge of the sea-creature.The fact that the human figurewalks aft of the forepartsof the sea-creaturemakesgood sense. After all, people on Greek vases (and in reallife) herd cattle like this; and literarymetaphorsfor Okeanos and other marinedivinitiesmakethem herdsmen of various underwaterbeasts.19 No part of either name is preserved,leaving speculationlittle to grasp. Stewart'ssuggestion(1983: 59) that this is Okeanos with Hephaistos walking beside, is problematic. Early images of Hephaistos, which generallygo out of theirway to show the deformityof his leg(s), makethis improbable.For Sophilos on the ErskineDinos, as laterfor Kleitias,Hephaistos is too badly crippledto walk or to stand in a chariot,and must ride his mule or donkeyside-saddle.

Fig. 2: Fragmenti of AthensWeddingDinos (B39/15)with inscription namingtheNyssai.Photo:M. Kilmer. Fragmenti preservesone frontalfemalefaceandtwo morefemalefacesat right, turned to our left. The frontal female plays the syrinx. The inscriptionsloping down to the right of the two profileheads (orthograde,though the a is writtenas laterretrogradesigmaswill be) is the wordNuaat (Nyssai;Fig. 2). Stewart(1983: 59) and Beazley (1986: 17) take the two profilewomen as Nyssai, "Dionysos' companions,"so also does Immerwahr(1990: 186). These nymphs of Mount Nyssa, though minor charactersfor us moderns, are also known from myth as nursesof Dionysos.20Despite this simpleinterpretation,the inscriptionhas given (and continues to give) rise to an inordinateamountof difficulty,causedat least The incisedlinewhichmeetsthecurvedbackof theseamonstermakessenseastheunderside of the leftforearm of thesamefigure:thearmis stretched outbeforehim,andhiscloakmusttravelwithit. The cosed handto ourleftis besttakenasa dosedrighthand:theshortstraightlinebelowthebase of thethumbis notpartof thepainter's intention. on the FrancoisVasehas"ahumanbodybut the headandneckof a bull"(Beazley 18Okeanos 1986:27). We mayalsocomparethe figureof Nereus(perhaps rather"aTriton")shownperhaps in combatwith Herakles(?) on Sophilos'column-krater in Athens(B40/24).The bellypatternis somewhat butthissea-monster hasa beardedserpentgrowingfromhisback.Seealsobelow, simpler; 25-26. 19Foran extendedscene enlargingthe metaphor,of Proteus,see Hom. Od.4.400-570. 20See Threatte 1980: 260-261; Immerwahr1990: 21, no. 61. Henrichs(1987: 116, n. 41) follows West (1978: 373-374) in identifying the Nyssai as tree-nymphs. Immerwahr(1990: 186 Add. et corr.)supportsthis. For the Nysai (or Nyssai) as nursesof Dionysos, see Graves1960: 104; for this

SOPHILOS'VASEINSCRIPTIONS

17

in partby scholars'assumptionthat what is difficultto understandin inscriptions is likely to be ancient error. Vanderpool (1953: 322) offered the suggestion that the word should be read as <M>uacL (i.e., Moiotai, Muses) "althoughit involves a rare case of v for p." This reading also requiresan unusual use of upsilon, ratherthan omikron,for the ou of laterAttic.21Sophilosmayuse upsilon to representthe spurious diphthong later written ou in the name FIarpo9Xu; (probablyfor the latergenitive form l-aTpoKoo6;)on the Athens Dinos fragment from Pharsalos(B39/16; see Plate 2). Considerablylater(ca 540), a parallelwhose implicationsare much clearerthan in our text, is the inscriptionon an amphora of Panathenaicshapein London (B307/59): AUVEIKECU . htno; . vtKat ("thehorse of Dy(s)niketos is the winner"),where the final u of the owner'sname is clearly intended to be the Attic ou genitive ending. The preservedtext of Sophilos' inscription, however, makes good sense as it stands. If we were dealing here with a manuscript,ratherthan with an inscriptionon a pot, few scholarswould argue in favour of an emendation which does not improve on the manuscript reading for sense or for context. Most scholarswould be doubly cautiousabout an emendationwhich additionallyrequireschangeswhich go againstthe normal usage of the author(and of the "manuscript"-thatis, the preservedinscriptions) in question. The emendation requiresus to accept that Sophilos here wrote v where he ought to have written lp (here and nowhere else). It requiresus to acceptthat Sophiloshere made a hash of writinga word (Mooaat) which in other places gives him no difficultyat all. This is not reasonable.These women are Nyssai. The syrinx-playerStewarttakes to be the Muse Kalliope,though this requires a somewhat difficult logic. Kleitias names as Kalliope, among the Muses in the wedding procession, the frontal syrinx-player. Because of Kleitias'special treatmentof Kalliope, Stewart(1983: 57-58) takes her to be, for Kleitias,chief of the Muses.22On this basis, the syrinx-playingfemalewhom Sophilosgrouped with the two women clearlynamed as Nysai, Stewartbelieves to be the Muse Kalliope. On the Erskine Dinos, in the first group of Motsa (Muses;Williams 1983: fig. 31) Stewart (1983: 61) once again identifiesthe frontal syrinx-player as Kalliope (with rather more probability, as the painter calls these women Muses). On the Athens fragment, she could be simply a Nyssa playing the syrinx-nymphs, after all, are expected to be musical. Kleitias,however, names of the Hyades.The Nyssaiareto be treatedin theywerelaterto be honouredas the constellation a supplement to LIMC,alongwith "Kentauroi" andassortedothers.On thismatterI alignmyself with Beazley,Stewart,andImmerwahr ratherthanwith Vanderpool and(thoughmoretentatively) Threatte.Eithergroupmakesa formidable ally. 21On u forou, seeThreatte1980: 1990:162.Forthestandard 260-261;Immerwahr orthography of thisperiod,omikronrepresenting whatlaterAtticGreekwritesasthespurious diphthongou, see, forexample,Buck1955[1933]:30. 22So alsoWachter1991: 107-108, of mostof Kleitias'Muses'namesto stressingthe similarity thosein Hesiod.

18

PHOENIX

Kalliopein a scene in which he writes the names of all nine Muses.23 Sophilos on the ErskineDinos shows two groupsof Muses: one Muse playingthe syrinx, flankedby two others to each side, and a groupof threeMuses walkingbeside the horses that drawHermes and Apollo-eight Muses in all. Each group bearsthe group label Moaat. No individualMuse is named. Since the numberof Muses is not the (later) canonicalnine, and none of them bears an inscribedpersonal name, there seems little reason to transferKleitias'iconographyof Kalliope to the Muses on the Erskine Dinos; and still less to assume that the syrinx-player shown among the Nyssai here is anything other than one of their number. On the analogyof the group of five Muses on the ErskineDinos, we might suggest completionof the fragmentaryvignette herewith two profilefemalesto our left of the syrinxplayer. Sophilos'reuse of the iconographyfor a separatemythological groupwould be well in characterfor him. His placementof the Mooat inscription over the right-side flanking pair on the Erskine Dinos is like the placement of the Nuaal inscriptionhere. Sophilos likes paratacticalarrangements-and he is particularlyfond of "ringingthe changes"on what often seem on first exposure like mere repetitions. One more fragment,c, has a partialinscription.The nearerfigureis a bearded male, the farthera female. Startingjust before her forehead, the letters he are complete (closed heta, as is usual for Sophilos; the horizontalsof the E slant down to the right, another preferredform). Following the E there is only the short lower end of the hasta of a letter, spaced slightly fartherfrom the ends of the bars of the e than the vertical hasta of the e is from the nearervertical of the heta. In a stoichedon inscription,particularlyone on stone, the spacing would be important.Here, with the painterdoing his letteringentirelyfreehand, spacingcannot weigh heavilyin the argument.Immerwahr(1990: 21) takes the remnant as the hasta of a rho,24as does Bakir (1981: 65); and, on the basis of the perceivedparallelwith the Erskine Dinos, makes this hep[a] (Hera) and the male figure Zeus. The pairing is natural;and on the Erskine Dinos Zeus' name is written between goad and reins, which could also be done here, starting just to the right of the break. It should be borne in mind, however, that even should the partial letter be a rho, Hera's is not the only name that could fit. h?pgLe;(Hermes) works as well for the letters, though it is difficult to see how the painter could then follow his common patternfor writing the name of the female figure. Maia, Hermes' mother, would be suitable, and has a nice short name-and that is the pairing on the Francois Vase (Wachter 1991: 89, nos. 92-93). Even with Sophilos' exuberantlettering, it is difficult to make this a P (for example,for Hebe, whom we find on the ErskineDinos though not on the FrancoisVase). We ought not, however,to excludeapriori such relativelyminor 23Wachter1991: 90, nos. 66-67, 70-73, 76-77. On Kleitias'non-canonicalMuse nameSTeostope (Stesikhore),see Stewart1983; Wachter 1991: 107-108. 24This would be the long-stemmed p (rho) used several times in other words on this vase rather than the stemless (or nearly stemless) D-shaped p of Khiron and Hera on the Erskine Dinos.

SOPHILOS'VASEINSCRIPTIONS

19

figuresas hepas (Herse);25and even hcpaKcK;is possible,though one would like to see an attribute. The hasta could work (its position, ratherfar from the e, gives some slight supportfor this) as that of a ), with Hephaistos the most probablesupplement. It would be surprisingto have the lame Hephaistos in a chariot,26and still more curious to have his lawful wife, Aphrodite, beside him: on both Erskine Dinos and FrancoisVase Aphrodite rides with her lover Ares, not with her husband, though on the FrancoisVase we see only the inscribednames,not the couple:the handle is attachedas though on top of their figures.On the whole, however,the positioning of the fragmentaryname, when comparedwith the rest of Sophilos' name-placements,connects it most probablyto the female figureratherthan the male. Suspensionof judgment might be an appropriateresponseat this stage of the investigation. (b) DinosfromPharsalos(B39/16) Of the second signed vase, the dinos from Pharsalos(B39/16), there survives only a single inscribed fragment which shows an animal frieze on the lip, a patternof tongues separatinglip from body, a small but telling segment of the main pictorialscene, and below that a magnificentlion to one side of a complex floral.27Of the principalscene (Plate 2) there arepreserveda pairof horsesin the foreground;to the far side of them, the heads and forehoovesof a second pair of chariot-horses.In the space before the horsesis written (retrograde,and sloping down towardsthe horses'forelegs),the "painter" inscriptionXo).t0o; l'eypacsgev. The initial a is written in the form which will become standardfor retrograde inscriptions;and what remainsof the cain eypac)a?v,though it is badlydamaged, looks to belong to the same form. Curiously,the initiala, though surelyintended as a three-barsigma,has enough of a bend in its lowerlimb to suggestthe four-bar formwhich becomes standardin Attic inscriptionsonly substantiallylater,within the fifth century.28This peculiarityoccursin others of Sophilos'sigmas, making them look sometimes more like the Corinthiancrookediota than like a sigma. Uncharacteristicof Sophilos is the initial e of eypaaocv which is inserted into the narrowspace between the p and the y as though it were an afterthought;so much so that he has had to fit it above the centre-line of the rest of the letters, taking advantageof the divergenceof the lines of alpha and gamma. Different also from the signatureon our first fragmentarydinos is Sophilos'inclusionhere of the pronoun etc. Inclusion of pc in this phrasewas clearlyoptional; and we 25What scholarwould havebeen so foolhardyas to conjectureNualt with only the firsttwo letters complete, or only the last three? 26See above, 16. 27Bakir1981: A.3, pl. 6. Two furtherfragmentsareidentified(Bakir1981: pi. 7), neitherof them inscribedand neitherjoining. 28Immerwahr1990: xxiii S 4 and 5. On the forms of sigma, see Immerwahr1990: 157-160. Four-barsigma is fairly common in sixth-centuryvase inscriptions,particularlyin early red-figure (Immerwahr1990: 158).

20

PHOENIX

have no reliableway of recoveringthe motivationfor a painter'schoice between speakingin his own voice (albeitin the third person)or havingthe vase speakfor him, as here:"Sophilosdrew me."29 Below this signature,and roughlyparallelwith it, is the retrogradeinscription lnarpo9Xu; : acka (for &O0a)30("the games [or prizes] of Patroklos"; Plate 2).31 Spellingof the hero'sname with qoppa,fIacpo9Xu;, also occurs,ratherlater, on the SoklesPainter'skantharosB173/1 (vidi).32The shapeof this Berlinkantharos strikes me as quite close to kantharoifrom the workshop of Nikosthenes (for example,one signed by Nikosthenes aspoietes,R132/a, whose date is not likelyto be much earlierthan 520-510). 510 is too late for the Sokles Painter'skantharos; but a date as late as 530-520 seems reasonable. In discussionsof his version of the same funeralgames on the FrancoisVase, Kleitias is sometimes said not to know the Iliadic version of the two-horse chariotrace, which Diomedes won (Wachter 1991: 96-97; for refinements,see Snodgrass1998: 119-121). Given the fact that the remainsof earlyepic other than the canonicalIliad and Odysseyare scant, it may be foolhardyto make such accusations. Major contests, military or sporting, make excellent materialfor poetry with an epic flavour. Why should there not have been other (poetic?) versions of these games extant and known to Kleitias, and to Sophilos before him? Sophilos'inscriptionestablishesthe contextof his chariotrace.He has provided stands for spectators,33who look out in two directions,one grouplooking at the chariot race, the others now looking out across the break to the right of our fragment.One spectatorin the upperseats turnshis head right aroundto look at what is going on behind him-and thus strengthensthe link between the events to either side of the stands.Over the spectatorson the right side the name AXtX,; 29Svenbro (1993 [1988]: 26-63) discusses this phenomenon. I do not find his presentation convincing. 30Threatte1980: 453. The reverseof this spellingvariant,0 written for t, occursfor exampleon the group of ostraka(manywritten by a single hand) nominatingOeCltaoKkeS(Themisthokles)for ostracism:see, for example,Harris1989: fig. 1; Immerwahr1990: no. 579, fig. 155; no. 581, fig. 152. Immerwahr(1990: 21, n. 4) suggests "Tau for theta in &Tkamay be due to the following lambda" (citing Threatte 1980: 453). 31On this inscription,see also n. 11. On the scene and the findspotof the fragmentaryvessel, see Snodgrass1998: 116-118. 32Although Beazley classifiesthe Sokles Painter'swork with the Little Master cups (and Droop cups), he notes that the name Sokles (aspoietes,and spelt SoKXeeq-not with qoppa,but with double epsilon)occurson a red-figureplate (R164/a; ParisCA 2181) from Chiusi "whichrecallsthe Cerberus Painter[Paseas]"(Beazley 1956: 173). Paseasis roughlycontemporarywith Oltos. Beazleynotes that this Sokles need not be the same potter for whom the Sokles Painterworked, but certainlydoes not excludethe possibility,he also notes that the "signature" may not be genuine. 33Thesemayhavebeen referredto as cKpta:cf. Eustathiuson Od.1.132; Sudas.v. TKpcta.Sophilos, unfortunately,does not name them-something Kleitiasmight well havedone had he includedstands on the Franfoisvase.

SOPHILOS'VASEINSCRIPTIONS

21

and that hero may be present(though (Akhilleus;Plate 2) is writtenretrograde;34 his figure is completelylost) either as an umpire as on the FrancoisVase, or as a participantin whatevercontest was taking place in the missing portion of the dinos next to right of our fragment.The sigma in this word is of the form which was to become standardfor retrograde. I suggest that the stands on this vase fragmentservedmuch the same purposeas the house on the ErskineDinos: that they mark the key place in the racing oval where the race began-and where it ended afterrunningright aroundthe dinos. There is preserved,above the horses at the left side of the fragment, a pair of letters: os. There is also a fragment of a third letter, the top of a vertical hasta which could belong to t, 4, or x (in its then-standardform +). Beazley took it as ), and the group of three letters as the start of a retrogradepoietes signature,for example, Eo4)[tXo;: cEXOIEccv](Sophilos made <me>). This is one of only two places in Sophilos' preservedwork where a poietes-signature is possible. The word 7toirlacv in such contexts is taken by some scholars to mean something more in the line of "fromthe workshop of," or "designed by."35In examples as early as Sophilos, it seems to me probablethat the most straightforwardtranslationof the verb, as "made"-i.e., "potted"-is the correct one. Martin Robertson's"poietes-signatures" (1992a, 1992b) providesan elegant solutionto the dilemma.The "doublesignatures"aspoietesand as "painter" which arecertainfor Exekias(and all but certainfor Nearkhos)aroundthe middle of the sixth century(they may begin as earlyas Sophilos, though apparentlynot in the same format) make it abundantlyclear that the functions of poietesand painter were seen as separate,or at least separable. Poietes-"signatures" started as early as the mid-seventh century,when we find the "signature"of Aristonothos on a krater(now in the Palazzo dei Conservatoriof the Capitoline Museum, Rome) made apparentlyat Cerveteriin Etruria. "Made"in this case may include both potting and painting. The introductionof the eypa4)ac-v-"signatures" suggests that cEioteoevwas tied stronglyto the making of the vessel. Thus a personwho wished his work as painterto get the same recognitionas the work of the potter signed as painter. In Attica, Sophilos appearsto have been the first painter to do this. Use of eiroiflaTvto mean the "maker"seems also incontrovertible in the case of the bronze diskos (findspot unknown, thought to belong to the late sixth century, and characterizedas Boiotian by its lettering) in the Ortiz collection (Ortiz 1994: cat. no. 128; see also Neils 1992: 166, cat. no. 33). This discus bears the inscription ZtPo; 'cEnOtF1tEc? (all sigmas three-bar, the certain and omitted in Neils' intentional, digamma transliteration). I find it 34Here, as usual, Sophilos uses the single consonant (A) where later Attic writers will use the rather geminate. Immerwahr(1990: 21, n. 4) suggests the spellingwith -Eqas a variantfor AXlXXcuq than an error. 35For recent discussion of the matter, see M. Robertson 1992a: 4-5, 44-45, 186, 294, and commentspassim,with substantialbibliography,1992b.

22

PHOENIX

improbablethat this meant that Simos was the designerof this particulardiskos, or the owner of the bronze-workswhere the diskos was made. This, of course, representsanother medium at a later date. The diskos belongs even more to the mundane world than does painted pottery, but the parallelis not without significance. Immerwahr(1990: 21, no. 62) takes the letters as --]toS (t partial),a wordending with the a reversed. This would best be taken as a propername, most likely that of a charioteer(as on the games scene on the FrancoisVase). Sophilos, unfortunatelyfor any attempted solution, shows flexibilityin the forms of his forms moreoften than to the "standard" sigmas,tending perhapsto the "reversed" in both his orthogradeand his retrogradewriting. Even on this single fragment, we see "correct"a in retrogradein the word AXtXcqand in the initial a of the painter-signature. The ; in frlapoo9uo (retrograde)is "reversed";and the a in cypaocev is badly damaged, though it looks as though it may have been a "correct"retrogradesimilar to the initial letter of the painter'sname on the same line. Thus on this one fragment we have two certain "correct"sigmas, one certainly "reversed,"one damaged-and the a in the fragmentaryword, which works as a "correct"sigma if the word is retrograde,and a "reversed" one if it is orthograde. Statistical analysis is not going to help with this problem. It may in fact be time to question how firmlya directionshad become establishedby this time. The remainderof Sophilos' inscriptionswill, I think, strengthen the sense of vacillation or, perhaps better, variation on Sophilos' part. (c)Excursus:TheSignatureon theErskineDinos (B40/16.1) There is one further complete painter signature,that on the Erskine Dinos (B40/16.1; Fig. 1). As on the Athens Wedding Dinos discussedabove (B39/15; see Plate 1), the signatureis placed along the side of a building,this time between column and anta, and to the left of the doorway.This time the painterhas had no problemsof spacing(or of afterthought)as on the dinos fragmentfrom Pharsalos. o4tXSoq;: i'cypa4aev is written retrograde36in a remarkablyneat hand-the principalanomalybeing that the letters get broaderas the painterwrites. Once the again Sophilos has left us a problemwith sigmas:the initial one is "correct"; other two are "reversed."The phi of cypaokev has an extra short line at upper right, which makes it look as though the painter may have written-or started to write-his ("reversed") last sigma one space too soon, but made the correction more or less immediately. 36The recentimpressiverestorationhas inserteda broadspacebetweenthe finalnu and the epsilon precedingit. An earlierrestorationusing inappropriatematerialsis the principalculprit:see Williams 1983: 9. The letter spacing would surelylook better if the original arrangementcould be restored. Even without the modern problem, the nine letters of a'cypa4a?vtake close to twice the space used by the seven letters of the name plus the interpunct.

SOPHILOS'VASEINSCRIPTIONS

23

(d) MenidiLouterion(B40/21) A furthersignatureis found on the fragmentarylouterionfrom Menidi, now in Athens, which Bakir takes to provideour earliestsignature.37On one of the smallerfragments,whose surfaceis ratherbadly worn, we can make out to our left of the attachment point of a spout a horse's head with chariot-harness,a bearded serpent, the sandaledfoot of a woman facing to our left (lower right of fragment), and the inscription [Zo4]ItoS;: t'C.O9EgE.Unfortunatelyfor us, the diagnosticletters of the verb are missing apartfrom extremelysmall traces. E following It, though damaged, is virtuallycertain. The scrap of the second preservedsigma shows that it-like the final sigma of the name-was "reversed." There is comfortablespace for three letters done to the scale of the surviving ones. We have seen, however, that Sophilos was no purist about letter widths. Given the exampleof the painter-signatureon the Pharsalosdinos fragmentwith its wedged-in c, we should be ill-advised to be dogmatic about this. The four letters ypca could be made to fit the space with no more than Sophilos' usual gerrymandering;and thus the hoped-for makersignatureis not to be found here with significantlygreatercertaintythan on the Pharsalosdinos.38 There is anotherline of writing "above"the signatureline; and in this can be made out with some securityan interpunct(the last element of the line, at farleft: this suggeststhat whateverwas written here was meant to be read as continuous with the better-preservedline) precededby what might be a "standard" retrograde sigma, a long vertical (possibly the hasta of a tau) at about the middle of the length of the inscription,and a few more scrapsincluding, at far right, remains of a gamma or (less likely) an orthogrademu or nu. Bakir (1981: 68) makes no attemptto reconstructthis line, callingit indecipherable. Both lines are written as though along a vertical baseline, invisible to us. "Below"them-spaced far enough away that they are clearlymeant to be taken as quite distinct-are three letters: ]cp9[ . The e and the p are both in their retrogradeforms;39the o is partialbut virtuallycertain.There appearto be scraps of at least two lettersfollowing the incompleteomicron;but to my knowledge,no convincing supplementhas been offered. This may well be part of the name of the woman whose sandal-shodfoot is preserved:the name could begin near her face;and the letter-basesare turnedtowardsher. The second and substantiallylargerfragmentof this louterionhas as its lower element a Corinthianizingrow of animals:to our left, a lion, then a panther,and finallya siren. Above the animalsis the principaldecorationof the piece, which illustratesa battle between a human and some centaurs.There are three letters, 37Bakir1981: 68-69, cat. no. A20 and fig. 7 (drawing)and pls. 8-10; for its date, see ibid.25. 38For a useful and realisticdiscussionof this, see Bakir1981: 6-7. 39Since Sophilos has used the stemmed rho here,we can be certainof its orientation.The stemless form of the letter is often difficultto distinguishfrom delta, and is commonlyimpossibleto "read"for orientation. Since the c has the same orientationas the rho, we can be reasonablysure that this was an intentionalretrogradeinscription.

24

PHOENIX

and a part of a fourth, preservedon the right upper corner of this fragment: .....].40 The human figure is the one closest to the start of the inscription. KeTCa[ He has dropped his bow (it is still strung, and obviouslya powerful composite recurvedbow like the one Odysseus knew how to handle) and holds a centaur by the hair, with his (lost) weapon drawnback for a mortalblow. The centaur holds the hero'sleft forearmor perhapsreachestowardshis chin in an appealfor mercy.41His equine forelegs are alreadygiving way in a mannerthat recallsthe Nettos Painter'samphora(B4/1). He has alreadylost whateverweapon he once held in his right hand. It is clearhe has not much longer to live. Another centaur is alreadydown:we can see his bodyjust to the left of our humansubject.A third, this one with his equine hairincised (recallingthe shaggypelts of satyrs),is about to tryto makeoff to ourleft or is confrontinganotherof the opposingforce.At all events, he is alreadyseriouslywounded. Blood poursfrom a wound in his equine chest, and we can see his right hand reachingdown towardsthe wound, perhaps hoping to staunchthe flow, perhapssimplybecausehis humantorso is collapsing. The scene has been identified as Herakles and the centaurs.42This would have to be the clash after the most unfortunateopening of the wine at the house of Pholos, not the fight with Nessos. That combat is alwaysshown as a duel, not part of a general brawl. The attributeswhich suggest Herakles (his iconographywas not yet standardized,though some elements do appearto be alreadyin place by the late seventhcentury)arethe bow (dropped)and the quiver full of arrowsat his waist, and the fight in which no other human is seen. The assumptionis plausible, but perhaps should not be taken as proven. We have Herakles certain (becausehe is named) on the Nettos Painter'sname amphora. There his weapon is the sword, as it probablyis here.43 We have a least one other possible Herakles from the hand of Sophilos, perhaps in combat with a ratherco-operative-lookingTriton (B40/24). The serpent "Triton"carriesand the serpentgrowing out of his back suggest that, like Nereus and Proteus-and 40See Threatte 1980: 487 on the omission ofv beforeT in KEvtaupot:could this ratherbe a proper name for the centauror for the individualhumannext to it? Position and directionof writing"attach" it more naturallyto the humanthan to the centaur.IfKeta is the correctsupplement,therecould be a name connectedwith the verb KCVT(CO(to prick,goad, spur). Another connectedword could be KSVTpov and relatedwords (goad, sharp point, etc.). If the transcriptionshould ratherbe KET,there are KrTEtot,an unknown race in Mysia (Hom. Od. 11.521; cf. Strabo 13.1.70). For Pindar (Pyth. 2.78-90) the Kentauroiare childrenof Kentaurosand the wild maresof Mount Pelion in Magnesia. Kentauroswas the son of Ixion and the cloud-HeraZeus had made to preservethe realHera'shonour (and his own). 41In photographs,the hand looks dosed. 42So Beazley 1956: 40, no. 21; Bakir 1981: 68-69, cat. no. A20 fig. 7 (drawing),and pls. 8-10; Immerwahr1990: 21, cat. no. 63-64; Threatte 1980: 487. 43B-W 1, a proto-Attic standedovoid kraterin Berlin (SMPK A 21; vidi) by the Checkerboard Paintershows a hero holding a centaurby the forelock,and runninghim throughwith a sword. The Nessos-if Nessos he is: Morris (1984: 74) prudentlycalls the human "a man,"and does not name the centaur-appears to drop the roughbranchhe has used as a weapon, and reachesout his left hand to touch the hero'schin.

SOPHILOS'VASEINSCRIPTIONS

25

like the sea itself-he is a shape-changer.Here "Herakles" wearsa sheathedsword slung on a baldricand has a quiveron his back.Hermes-at all events, a man with a kerykeion-appearsto greet him, and two more adult males (one at least armed with a long slender spear) accompanyhim: possibly Zeus and Poseidon. The difficultyin accepting this Triton-rider as Herakles stems in part from the fact that heroesother than Heraklesareknown to have done battlewith sea-monsters: Menelaos with Proteus, Peleus with Thetis. The ErskineDinos providesample evidence that the story of Peleus and Thetis was one which interestedSophilos and presumablyalso his clients. For Herakles,we might also cite the fragmentary dinos in Athens discussedearlier(B39/15): fragmentc with its inscriptionhsp[ could possiblypresentHeraklesagain-in which case the female figurewith him could be anticipatedto be Athena. Unfortunately,this is no help in the dilemma of the centauromachy. The wordwrittenon the centauromachyfragmentis takenby all commentators I have encounteredas KeTa[upot] (centaurs).44Like Sophilos' Moactafor the Muses and Xaptrs; for the Graces(both on the ErskineDinos), and like his Nucat on the Athens Wedding Dinos (B39/15; Fig. 2), this would then be a collective name-much as Kleitias, a few yearslater, will write EItcvot over his silens and Nu4(a over his nymphs on the Francois Vase ratherthan giving an individualname to each satyrand each nymph.45Little notice seems to havebeen given to the possibilitythat the name goes with the human, not with the centaur at all. The alpha'suncertaintycalls for at least a little caution.We have seen that Sophilos has a slight preferencefor beginning names quite close to the faces of those they identify. This is borne out more forcefullyby the inscriptionson the ErskineDinos (B40/16.1), the best-preservedof all of Sophilos'inscribedpieces (see below, Section iii). (e)AthensFragment(B43/4) A fragmentfrom the Agora of Athens (B43/4) Bakir(1981: 75, cat. no. B19 and pl. 83 [photo]) places in the circle of Sophilos, ratherthan ascribingit to Sophilos himself. There is part of a chariot-horse,an old man with his hand at his head as though grieving, and the middle portion of another figure at right. The preservedinscription,above the old man, is just three letters long: --]Xo?, surelyto be taken as the last letters of a name. If we supplementAntilochos (or Antimachos,both good Homeric warriors'names;other supplementsare equally possible),it would go with the chariot-riderratherthan with the old man, whose pose is remarkablyreminiscentof the old seer on the pediment of the Temple of Zeus at Olympia. The lettering is crisperthan Sophilos' norm, and appearsto have higher relief, suggestinguse of a thickerslip for this inscription. 44See above, n. 40. 45Wachter1991: 91, nos. 124-125.

26

PHOENIX

0) Izmir Fragment(B40/20) This fragment(B40/20), with rathermorepreserved,is thoughtto show Kastor and Polydeukesat the wedding of Menelaos to their sister Helen.46 (There was also a variantof the legend in which Theseus marriedHelen afterabductingher. It would be interestingif Sophilos had representedit here;but this cannotgo beyond conjecture.)47This is a fragmentarystanded nuptial lebes (lebesgamikos)from Old Smyrna. Between draped figures and their four horses is written K]caoTop (horizontal above the reins); above that, also horizontal,lnoX6[s?]K[E;]. The twins must be the figuresin the chariot(s).Their namesarewritten startingfairly close to their faces. The next figures could representthe bride and either her mortal step-fatherTyndareos or-perhaps more likely-the groom.48The man drivesthe chariot. Only one name is preserved,written this time abovethe level of the characters'heads:hEcXy[e] (Helen). The man's name would have been written either just below this (as with Zeus and Hera on the ErskineDinos) or immediatelybehind him. Both of these areasof the vessel aremissing. Although the photographin Bakir(1981: pl. 45) is not of excellentquality,the initial letter evidentlyhas the squaredcornersof closed heta. The letteringis verylargeindeed, and with its placement such that it virtuallyfills the spaces between passengers and horses,we can see that the letters, in additionto informingus of the names of the participantsin the mythologicalscene, are a very importantpartof the purely decorativeaspect of the piece. (B40/17-18) (g)AthensKotyle-Krater(s) of a Fragments kotyle-kraterin Athens (B40/17-18)49 give partsof a mythoscene, logical though what survives does not allow us to establish the myth illustrated. On one fragment there are two draped female figures moving to the left, one of them holding out a corner of her himation in her right hand. Before the women is the name Ilavbposo; (Pandrosos), retrograde,vertical, 46Bakir(1981: 69, cat. no. A21; pls. 39-45 [photos]; figs. 32-38 [drawings];33-38 [discussion]) places this in Sophilos'middle period, along with the ErskineDinos. In the BeazleyArchive there is an excellentreconstructiondrawingby SirJohn Boardmanof the section of the vase which bearsthe names of the Dioskouroi. His transcriptionof the name heXevgives its firstletter as theta. The sigma in the name Kastorhe transcribeswith four bars ratherthan Sophilos'more usual three. Although I have not seen this fragmentother than in photographs,I would expect the initial letter of Helen's name to be heta, perhapswith the cornersof the letter more roundedthan usual. 47Simon 1996: 13, referringto Shapiro1992. 48This would put the iconographyinto dose relationshipwith the Amasis Painter'sNew York lekythos (B155), and would show that this second mode of presentingthe wedding (mythologicalor human)existedalreadyin Sophilos'lifetime. The Amasis Painterdepictsa muchless grandprocession. The bride is surely illustratedin the cart beside her groom, who controls the donkeyswith a long flexiblerod. The bridecarriesa wreath, and holds her mantleout in much the way that Hera laterwill as a symbolof her patronageof marriageand particularlyof brides.Here also only one house is shown. 49Beazley (1956: 40, nos. 17-18) listed these two fragments separatelyas from lipless dinoid vessels, making their separatenessspecific.Bakir (1981: 26) assignsthem to a single vessel. I have not seen the firstof these two fragments,and so cannot add usefullyto the discussion.

SOPHILOS'VASEINSCRIPTIONS

27

letter-basestowardsthem. Both sigmas are "reversed,"and there is considerable spalling as well as surfacewear. None of the letters is so badly damaged as to make the readinguncertain. Behind the two women comes a beardedman with a scepter-probably Kekrops,Pandrosos'father. Bakir (1981: 26) suggests the second woman should be either one of Pandrosos'sisters (Aglaurosand Herse) or her mother. If this woman'sname was included,it most probablywas written in line with the name Pandrosos,much like the names of Leto and Kharikloon fragmentg of the Athens Wedding Dinos (B39/15). Immediatelyin front of the women are the remains of a kerykeion,on the basis of which Bakir (1981: 27) restoresHermes. While this is plausible(Hermes is one of Sophilos' favourite characters,especiallyin his early period), we have also seen that for Sophilos, Iris can carrythe kerykeion,as on the Athens Wedding Dinos (B39/15) and the ErskineDinos (B40/16.1). Certaintyis impossible,and there areno furtherclues. The secondfragmenthas at its farright side most of the head of a beardedmale and, partlyobscuredby him, the head of a female. Between these figuresand the single horse behind them is written the name lnooaetov (Poseidon), orthograde, vertical,the letter-basestowardsthe horse. "Poseidon"is an improbablename for a horse, since horse names tend to refer to colours and to equine behaviours.50 The name thus belongs to the divinity. Given the orientationof the letters, it is reasonableto suppose that they name the riderin the chariotbeforewhose horse they arewritten. The occasion and the myth behind the occasion remain obscure for us. If we accept that both fragmentsbelong, the preservedfiguresdo not fit the same patternas the Erskine Dinos and the Athens Wedding Dinos, where all figures move to the right except for the groom, who stands firmly planted, facing his oncoming guests. The presence of Poseidon could make Theseus a candidate for centralfigure in a celebration.The connection of Theseus to (Kekropsand) Pandrososis not so clear. In at least some versionsof the tale Theseus' putative mortal father, Aigeus, was a son of Kekrops'son Pandion, making Theseus Kekrops'(step)-great-grandson.51Chronology in this version does not easily allow Pandrososto be present at her great-nephew'swedding; but we have no reliableactuarialtables for Athens in the Bronze Age, and in myth all things (or nearlyall) are possible. 50For example, Kukapo; (Kyllaros) "Hermit-crab"on Exekias' marvellous Vatican amphora B145/13-the horse may be shy or skittish-or Kakt4opaq (Kalliphoras)"Fair-bearer" or "Beautiful brood-mare"-presumablya name of good omen-on an amphorain Toledo (B146/1.1). The same name is found in the same form on one of Exekias'Berlinplaques(BerlinSMPK, 1820; Beazley 1986: pi. 75.2 vidi and, without final ; (i.e., possibly in the vocative)with Demophon's horse on Exekias' amphorain Berlin (B143/1). Immerwahr(1990: 32, no. 129) suggests that "Kalliphoras" may have been the name of a famous stallion alive at that time. A second horse on the Toledo vase is nluptxoS (Pyrrhikhos),"Fiery,"the name related to that of Akhilles' son Pyrrhos/Neoptolemos:this horse is most probablya roan. 51 See, for example,Graves 1960: 94-95, 320-325.

28

PHOENIX III.

THE ERSKINE DINOS

The Erskine Dinos (B40/16.1) is one of the most complexworks of Sophilos to survive,and, fortunatelyfor us, one of the best preserved. The scene is the wedding of Peleus and Thetis. Starting at the van of the procession,the arrival of the processionat the house of Peleus, we see neXceu (Peleus) standingoutside his house to welcome his guests, his name written in bold letters flanking his lower body (Fig. 1). We can also see the painter-signatureEOt)koS;t'EypcatcJv (with interpunctand Jp) written in much the same position as on the fragmentary dinos with which we began. The first of the guests is Ipt; (Iris), who will stay with Peleus and announce the guests as they arrive(Fig. 1). Just after her come hECTtcl(closed heta, a "reversed":Hestia) written above ActL?ETc[p] (Demeter); then XctptQko(written starting at her face: Khariqlo,the wife of Khiron)52and AcEo (verticalbehind her head, letter bases towards her: Leto) (Fig. 3); then (Fig. 4) Atovuaoo (Dionysos), followed by h3PE(Hebe). Xlpov (Khiron)follows her, carryinggame on a branch and looking back. Next come ?ECLt (Themis) and three Nu)at (Nymphs) (Fig. 5). Next come three ladies whose collective name is lost. These are commonly taken to be the hopat (Hours or Seasons;53 their name is preservedon the FranpoisVase, where they walk behind Dionysos [LIMC 5.2: Horae 45, pl. 348]). The Horai on the FrancoisVase sharea mantle. This does not seem to be the case for the three ladies on the Erskine Dinos. These walk beside the horses which pull hepa (Hera) and Zcu; (Zeus) in their chariot.Next (Fig. 6) come the three XapttcS (Graces)walkingbeside the horses of ri1oactov (Poseidon: his name has to change direction,its last lettersturning the corner and becoming retrograde), in whose chariot is his wife Av4ITpItz (Amphitrite:E placed above T to avoid bumping into the heads of the Kharites). Next come five Moalt (Muses), the central one drawn frontal and playing a syrinx. Williams (1983: 31) suggests that the syrinxaccompaniesApollo, who plays and sings the gamelioshymnos,and that the other Muses "perhapsjoin in the chorus"(though their closed mouths at this moment would allow them only to hum). In the chariotnext in line areAlpo5ttc (Aphrodite)writtenaboveApeq (Ares), generallyconsidered illicit lovers in the extant literarysources,but here treatedapparentlyas a recognized couple. Next come three more Moalt beside the horses that draw hePppE(Hermes) (Fig. 7) and [An]okov (Apollo). Behind 52LIMC s.v. CharikloI, pls. 150-151. She is variouslydaughterof Okeanos (hence an Okeanid), of Apollo, or of Perses (LIMC 3.1: 189-191). 53Sistersof the Moirai (childrenof Zeus andThemis). V. Machaira(LIMC 5.1: 508, Horae 44, pl. 348) acceptsthese as the Horai on the parallelof the Franfois Vase, speculating(reasonablyenough) that their name has been lost to the lacunaabove their heads;so alsoWilliams 1983: 28, pointing out that any remnantsof the letterswere lost when the fragmentwas filed down for an earlierrestoration. The loss of a name is near-certain.What that collective name may havebeen is quite anothermatter. I would arguethat Sophilos'version of this wedding and Kleitias'are not interdependent,and do not derivefrom a single visual source. If this is correct,it is a mistaketo import figuresnamed on one to fill lacunaein the other.

SOPHILOS' VASE INSCRIPTIONS

3 Leto, Khariklo, Hestia, and Demeter (Williams 1983: fig. 26).

29

Hebe, Dionysos,andLeto (Williams1983: fig. 27).

I 5 Themis and three Nymphs (Williams 1983: fig. 28).

6 Amphitrite, Poseidon, and three Graces (Williams1983: fig. 30).

7 Hermes (Williams 1983: fig. 32).

8 Hephaistos(Williams1983: fig. 34).

Figs. 3-8: Scenes from the wedding processionon the Erskine Dinos (B40/16.1). All photos arecourtesyof the BritishMuseum, London (afterWilliams 1983).

PHOENIX

30

them come three M[o]tpat (Fates) beside the horses of AO0vata (Athena) and ApreiLt (Artemis). Athena drives the chariot, and Artemis wears the pretty clothes. OKeavo; (Okeanos) comes next, and, just a little behind his human foreparts,two female figures,OEc0u54(Tethys) and htXc0ua55(Eileithyia)with h[Ec]ataTro; (Hephaistos)on his donkeybringingup the rear(Fig. 8). Noticeablefor her absenceis the bride,Thetis. There aretwo plausiblereasons for this. The bride, before the processionbegins, is in the house of her father, Nereus. At the end of the procession, she is in the house of her new husband, Peleus. It follows that she does not appearin the procession-and the one house stands in for both her formerhouse and her new maritalresidence. Kleitias,on the FrancoisVase, placesThetis (named)inside the building,with just her seated legs showing (Wachter1991: 90, no. 89). There also the single house likely serves as both beginning and end of the procession,though Thetis' seated position fits ratherbetter in the maritalresidence,where she is hostess, than it does in her father'shouse, where we might reasonablyexpect her to be preparingfor her move.56 IV.

ORTHOGRAPHY,PHONOLOGY,AND CULTURALLITERACY

Sophilos'own literacyin the basic sense of that word has been demonstrated in the earliersections of this paper.It shows in his abilityto write his own name and the requisiteform of the verbypape*tv.He is also able to write the names of a good assortmentof divinities,demigods,and heroeswith consistency.We have seen a numberof names in which Sophilos'spellingis significantlydifferentfrom the "standard" spellingof survivingliterarytexts, and some examplesof othervase painters'spellingsof the same names. For the most partI have reportedspellings of near contemporaries,but in some instanceshave includedexamplesas late as the earlyfifth century. It is eminently clear that Sophilos'spelling is consistent within its own apparentrules and that he uses spellingsconsistentwith those we find in most Attic and relatedvase-paintersof his time. More interestingthan basicliteracyis what we might call culturalliteracy:that is, familiaritywith the myths which characterizethe society in which the person lives. Sophilos clearlydemonstratesthat he has this. He seems to assume that 54This is surelyTethys.I haveno questionthat Sophilos'upsilonis intentional.In black-and-white photographs,the letteringis difficultto distinguishfrom the serpent'sforkedtongue; in colour, or in person, there is no difficulty.A closely similarspelling appearsabout seventyyearslaterfor a kindred name in the tondo of Peithinos'Berlin name vase R115/2 (vidi), where the nymphThetis (wrestling with nIleEu)) is called cEtS;,the adjacentaspiratedsyllablesan apparentviolation of "Grassman's law." 55We find the same divinity as hAkuOua on Euthymides'fragmentedand scatteredcup R29/19. Clearlythe initial aspiratehad not been droppedat that date, late in the sixth century,at least to the earsof Euthymides. 56A graduatestudent at Exeterpointed out (une 1996) that in Catullus'epyllion, the gods come not for the wedding itself, but on the following day. Could this also be what Sophilos had in mind?

SOPHILOS'VASEINSCRIPTIONS

31

his clients also have it. Thus, the mythologicalnames placedwith the figuresare expected(once the mythologicalsetting has been established)to be not so much new informationas confirmation.In this context, it is a help to realizethat the attributesof mythologicalcharacterswere still in the processof establishmentin the visualarts.In fact, Sophilos appearsto have made importantcontributionsto the developmentof mythologicaliconography(as also did Kleitias). This makes his inclusionof names all the more understandable. The issue of literacy can also be approachedfrom a different angle. In the recent past, under the influenceof McLuhan and others,we have come to think of writing primarilyin terms of its usefulnessin communicatingcomplex matter (facts, ideas, attitudes,etc.) ratherthan at a more basic level. Alphabeticwriting, when firstintroduced,presentedto potentialreadersa groupof signswhich, when soundedout accordingto convention,allowedthem to speakthe words intended. Each letterhad a limited groupof soundsfor which it might standon its own and a still limited but much largergroupof soundswhich it could representwhen used in combinationwith other letters.5 When we look at this from the standpointof people earlyin a society'sliteracy,we must recognizethat they can have accessto only a limited tradition.Writing must often presentto them challengesof types which become rarerin a culturewith a long traditionof literacy.When there is no establishedtraditionof orthographyfor a particularword, the writermust call on the rulesof sound ("phonology")for generaldirection,but also must often try something less easily characterized.The writermust attemptto make the letters tell any readeras nearlyas possiblehow to say the word. This is difficultenough in a languagesuch as Latin in which a chartof soundvaluesversusletter forms is relativelysimple. English is a much more difficultchallenge(one need only cite the conglomerate"ough"to demonstratethis: see Powell 1991: 75, 115). Greek stands somewherebetween Latin and English in the complexityof sound versus symbol.The challengewas greaterfor the Attic writerin the Archaicperiodthan for the writer in the later fifth centuryor after, who had a largeralphabetwith greaterflexibilityand a longer tradition to follow. We should perhapsbear in mind that Sophilos may be among the first Athenians (i.e., residentsof Attika: we do not know either his birthplaceor his citizenship)who tried to put some of these names into written form and others he may never have seen written. He should not be condemnedfor illiteracybecausepeople at a substantiallylaterdate used a differentand more flexiblesystemfor writingthe samewords. Stewart (1983) and, more recently, Wachter (1991) have shown that on the FranpoisVase Kleitias' so-called errors in spelling are not best explained as "copyingerrors." In fact, for most "errors"poor copying provides a weak explanationfor the variantsin spelling. Using their work as a starting point, 57I make no attempt here to trace the steps by which Greek people first reached this set of conventions: the attached bibliographyincludes severalworks which discuss the matter. Svenbro (1993 [1988]: 2) saysthat "Greekwritingwas firstand foremosta machinefor makingsounds,"which is not a bad characterization.

32

PHOENIX

and factoringin the researchof Threatte on phonologyand Immerwahron letter forms, I would say that most of what has been characterisedas error in the inscriptions of both Kleitias and Sophilos stems from the fact that the sixthcenturypainterswere workingwith an alphabetdifferentfrom that used in Attika in the laterfifth centuryand onward,andwith a quite differentset of conventions of spelling. Archaic Attic inscriptionsuse, for the most part, an alphabetof only twenty-one letters. In their Ionic forms, these would be writtenA, B, F, A, E, Z, H (heta), E, I, K, A, M, N, O, nI, P, X, T, Y, D (phi), X (khi). 9 (qoppa/koppa) they use as an occasional adjunct (Threatte 1980: 19). The advantageoffered by the small number of letters was apparentlya close link between each symbol and the sound(s) it denoted. The Ionian alphabet(which enters the repertory primarilyafter the Persian wars, as far as the evidence of writing on pottery indicates, and was not adopted for formal public inscriptionsuntil the end of the fifth century) gave two new vowel symbols, Q for long O (omega), H for long E (eta). The letter-formH had for over a centuryrepresentedthe aspirate, mostly at the beginning of words.58 Use of the sign H as the vowel eta meant that the form which had been expressinginitial aspiratescould no longer be used for that purpose. At this time, the expedient adopted was simply to rely on the readerto provide initial aspirateswherevernecessary:the "roughbreathing" mark was a much later (Hellenistic) invention. Two furtherconsonants were added: y ("ps")replaced4a and 4 replacedXa. The archaicAttic forms which the "doubleconsonants"y and 4 replacedseem to have representeda somewhat differentpronunciation:the Ionian letters do not, in their native region, include the aspiratewhich both 4o ("phs")and Xa ("khs")include. The consonants4, 0, X were aspirates,not fricatives.As Allen (1987: 19-20) argues, "Otherclear evidence comes from the languageitself. When a voiceless unaspirated plosive (i7, t, K), as in, e.g., OUKor elided an', KaT', stands before an

aspiratedvowel (i.e., initial [h]), it is changedto ),0, X;which can only mean that ), 0, X here standfor aspirated[ph], [th], [kh], and not for fricatives."We should now examine some diagnostic cases from the corpus of Sophilos' inscriptions, and from those of his (later)near-contemporaryKleitias,in the hope of eliciting phonologicalimplications.Of principalinterestarethese aspiratedconsonants. In the earliersixth century,the time of Sophilos, the word eypacxoevwas apwith some degreeof separation parentlypronouncedsomethinglike "egraphsen," and distinctnessfor the consonantsoundsrepresentedby the aspiratedconsonant 58Threatte(1980: 25-26) lists exceptions-rh, mh, lh in mid-word. To Threatte'sexamplesfor gh as the initial consonantgroup we should add the word MhovoSfrom the Amasis Painter'sAigina tripodpyxis (B157). On ph, see also Allen 1987: 40-45. The name Aheov(A&ov,Leon = Lion) occurs on an unattributedBF cup in Athens, NM Akr 1632 (Carpenter1986: pl. 15A), where it is probably a dog name, in a battle of gods and giants. I owe this referenceto Heather Loube, a doctoralstudent at the Universityof Ottawa. The same spelling is found on a band cup with the poietes signatureof Neandros (and by the Neandros Painter), this time the label for a lion attackinga boar: Ahaov Tov Kcaupov (verb"killing"velsim. to be understood).

SOPHILOS' VASE INSCRIPTIONS

33

D combined with X. The later spelling `ypacev need not in its first adoption representa change in pronunciation.It simply uses a letter-formborrowedfrom Ionia, which is conventionallyacceptedto representthe same grouped-consonant sounds "phs."It is more difficult to pronounce the separatedconsonant sounds: naturalhuman laziness could be expected to lead eventually to some form of simplificationof the consonantalsounds. We can see a similar process in the shift to 4 from xS, though this pairing is not representedin extant words from Sophilos' hand. Kleitias, a younger contemporaryof Sophilos, displaysXc for this sound-sign cluster, for example, in the name of a hunter in the Kalydonian Boar hunt, ?opaXg; (Thorakhs: B76/1; Wachter 1991: 87, no. 4) and in the name of one of the youths in the scene of Theseus and the young Athenians, {3}EuxcnapcxoS;,Eukhsis(t)ratos, (B76/1; Wachter 1991: 88, no. 35).59 pronouncedapparentlyas "Eukhsis(t)ratos" The dog qopacx (Qorakhs) in the Calydonian Boar scene displays the same spellingfor this sound cluster(Wachter 1991: 87, no. 6; for the qoppa,see above, writes his name 13). The potter/painterExekias,in all of his extant "signatures," EXoCTeKI;.60 This spelling probably represents a pronunciation something like

"ekh-sekias."Had the aspiratenot been part of the name as it was spoken, the paintersurelywould have written it with the appropriateletter: the simple velar kappa, or the guttural qoppa.61 I refer the reader to Svenbro'sdictum (1993 wouldusecrp is commonin the sixthcentury:seeThreatte 59Useof p wherelaterorthography 1980:571-572. 60Asmallcupin Munich,whichI haveexaminedandphotographed, mayappearto contradict this:B147/3(top).Betweenpalmettes,on eachside,"signatures" (Beazley1971correctsthe reading given in Beazley 1956). (A) EXaeKtXaqS:E7otv. The omikron is small and solidly filled, iota of cIotEV blobby. The middle horizontal line of e is sometimes the longest of the cross-strokes. (B) Xce&KiKa;:?e7CO completefore and aft: the name neverhad its initial epsilon, and the second kappais unmistakable.The verb is left incomplete. The qualityof the potter-work,and the black slip used to decoratethe stem and interiorand to applythe broadstripe on the exteriorof the bowl, areconsistent with Exekias;the palmettesareof moderatequality.The inscriptionsarewrittenwith thin, ratherpale slip, ratherthan the rich blackwith significantreliefwhich is Exekias'norm. The letteringis careless, orthographyand grammarshaky. Exekiasdoes not misspell his own name elsewhere. Although the letter-formshave, for the most part, reasonableresemblanceto those Exekiasuses, I do not believe that these were writtenby Exekias.Nor does Immerwahr(1990: 36, no. 148). Immerwahr(1990: 32-36), following Beazley (1956: 143), consistently transcribesthis potterpainteras though he wrote his name 'E1Kica;. As with the name of Sophilos, it seems to me that morecautionwould be appropriate;thus I have retainedthe archaiclettersfor him, without presuming to add accentsor changevowel values. See also Kirchner1966: 314, nos. 4729-32, all fourth-century, all the same family. Pape and Benseler (1959: 361-362) cite only the vase-painter,and appearto connect the name to the verbE,tKo.

61A readerforPhoenixremarks thatExekiashadno choice,as thatwasthe convention.My first

responseto this is to point out that archaicAttic has a very high correlationof speech to the written word. With veryfew exceptions,what you see written gives a very preciseidea of how you areto say it aloud. I would also point out that Exekiascould, if he had chosen, have used Ionic scriptratherthan Attic. Should someone then ask what evidence there is that Exekiasknew Ionic letterforms, I would point out that he certainlyknew the Chalcidianscript (which has many more differencesfrom Attic

34

PHOENIX

[1988]: 2) that "Greekwriting was a machine for making sound." The Amasis painter,working at about the same time as Exekias,spells the name of Akhilles' older companion, Phoenix, Ootvtx; (B152/27), which suggests a pronunciation "Phoinikhs"ratherthan the "Foinix"suggestedby the laterspelling4otvti. In all of these, later Attic orthographywould use the double consonantE (ksi) rather than the aspiratedconsonantX (khi) plus sigma. The letter0 (theta) bearssome relation.If we look, for example,at the spelling Sophiloschose for the marinegoddessTethys, we findher on the ErskineDinos as GeOu;.As Sophilos wrote it, it should be pronounced"The-thys,"the t emphatic in both positions. Kleitias'spelling of the word for nurse,later spelledTp64oS,as Opo4o;62-which might be taken simplyas parallelto Sophilos'Oe0u;, if the rho did not present complications-may mean any one of three things: there really was an aspiratein the initial consonant cluster, making it "throphos";or there could be difficulty,at least for some individuals,in distinguishingT-aspiratefrom unaspiratedT in the spoken language;63or the aspiraterequiredfor pronouncing the letter rho64was expressedin the aspiratedconsonant0 ratherthan by writing the consonant sounds with three separateletters. It might be argued that Grassman'slaw militates againstthese spellings (the objection must stand against both Sophilos'O?0u; and Kleitias'Opo4oq),which thus must represent errorsby the vase-painters. This descriptivelaw says that when two successive syllables of a Greek word begin with aspirates,the first syllableloses its aspiration.The inscriptionscited here suggestthat, for Athenians of the early sixth century, Grassman'slaw was not yet universallyapplied. That this state of affairscontinued, at least for some writers(and, I shouldthink, some speakers),into the early fifth century, the Douris cup (R444/421) suggests and the ostrakawith the name of Themistokles seem to me to confirm (see above, note 30). On this questions Rudoph Wachter writes (pers. comm. to Eleanor Dickey): 'The cases of throphos, hekho etc. are a trickymatter. (1) They could be cases which passed by Grassman, a law which is never quite complete in its than Ionic does) at least well enough to copy a beautifulinscription-one of his most complex-in it. At least one of his pots, in the British Museum, appearsto incude H as eta (London B 209, Immerwahr1990: 33, no. 134; the inscription is otherwise also problematic). On the Chalcidian inscription,see Immerwahr1990: 34, no. 142: "I do not know whether the inscriptionswere written before or afterfiring, but they appearto me to be by Exekias."M. Robertson(1991: 6 and fig. 2a-c) suggeststhat it is a post-firing incision (as far as I canjudge from the photograph,that is correct).He on the same firmlystates (1991: 6-8) that the same hand wrote this as wrote Exekias'poietes-signature vase; and is equally certain that Exekias wrote both. The vase in question is the dinos in the Villa Giulia (50599), B146.20. 620n the Francois Vase, see Wachter 1991: 88, no. 44. The same word in this same spelling is also found on the cup by Arkhikles and Glaukytes,Munich 2443, B163/2. The context is closely related-Ariadne and her nurselook on at the combatof Theseus and the minotaur. 63On this, see 20, n. 30. 64For discussionof this, see also Threatte 1980: 25-26.

SOPHILOS'VASEINSCRIPTIONS

35

actions and moreoverworked at different periods [i.e., episodically]. (2) They could be restoredforms, by analogy(hekso, threpso)or by chance (assimilation). I don't see how we could tell with certainty."I would add to this that in archaic Attika the futuresof the verbs exItvand Tp?e8tv would have been written hscoo (hekhso)and Opecso (threphso)with the double aspirationclearlydenoted. In particular,it looks as though some individualsfound the letters T and 0 difficult to differentiate. This may suggest that the aspirationwas not always stronglymarkedin the spokenword. One could offer the modernpartialparallel (more markedin the U.S.A. and Canadathan in Britain)of the loss of the initial aspiratein words such as "which"and "white,"which manypronounceas though they were "wich"and "wite"(or "witch"and "wight"!).London'sCockney dialect offersa morelocalisedparallel,with both abandonmentof opening aspirate(when comparedto "standardEnglish")and addition of aspirateto opening vowels not aspiratedelsewhere. LaterGreek orthographyplacesthe "roughbreathing"mark,a sign indicating aspiration,before rho whenever rho is the first expressedconsonant of a word. This system presumablyreflectedactualpronunciationof the sound represented by the letter rho. In our period, lambda and mu as initial consonants are commonly pairedwith heta (see above, note 58). Rho within words (especially when geminate)often seems to have carriedwith it some degreeof aspiration.65 Kleitias' spelling Opo4o; could, perhaps, simplify an earlier spelling which included heta as part of the opening consonant cluster. We might also look for parallelsto later spellingsof complexwords. For example,the word for "hair"in the nominativeis Opit (thrix);the genitive of this, once Grassman'slaw works on it, is TptX6o (trikhos). The aspirateis importantenough that it changes its associationin the word between the z and the K. That the rules (or customs) for visual expressionof aspiratesremained complex may be gleaned from later examples, such as Douris' cup (R444/421) in which the second person singular imperativeof Xctivis written as h??c (hekhe) not EXe(ekhe). Its date is after 500.66 CONCLUSIONS

The conclusionsto be drawn from this study, however modest they may be, are not to be taken as coveringall archaicAttic vase painters:they apply directly only to Sophilos. His contemporariesand near-contemporarieswill have been in 65Allen 1987: 41-45. Fossil examples of aspiratedliquids might include words such as XpCopc (not Kcpoua),Xpop6S(not KXOp6O;),the name XXIo (not KX6ro).Unfortunatelyfor the etymologist, LinearB did not expressaspirates,which makes it particularlydifficult to trace the earlierhistoryof these words. My suggestionhere is admittedlysheer speculation. 66At Kilmer1993: 83 (R577*)I quoted the inscriptionfrom Dover 1978, misguidedly"correcting" both this form and Douris' "misspelling"of fiouxoS as hcatxoq. I would now take heatcxo as an acceptablevariantspelling.

36

PHOENIX

a similarposition, but we shallbe able to make securejudgmentsaboutthe ranges of similarityand of variationonly when an adequatecorpusof vase inscriptions has been assembled and studied. Since this articlewas first submitted, one of us (Kilmer)has collected a large numberof inscriptions(photographs,drawings, and transcriptions)from archaic Attic pottery. These are being preparedfor worldwideaccessin the reasonablynearfuturethroughthe Internet. On the matterof internalconsistency,the few nameswritten more than once in the extant oeuvre of Sophilos (for example, Demeter, Khariqlo) are spelt consistentlywhere complete; and the preservedletters in partiallypreservedbut certainnames match the spelling of their better-preservedparallels(for example, compare Poseidon on the Athens Wedding Dinos [B39/15] with that on the Erskine Dinos [B40/16.1; Fig. 6]). Sophilos is also consistent in his spelling of the one verb he repeats:eypcracav. The complex sound clustersrepresentedin classicalAttic Greekby 4 and V, Sophiloswritesconsistentlyas Xa and (a. When other archaicAttic painterswrite a word Sophilos has written, the spellings are, as a rule, either the same or so closely similarthat we ought to have no difficulty in understandingthem as minor variants-rarely as far apartas standardBritish English and standardU.S. spellingsoften are. Furtherdemonstrationof this must awaitfuturepublications. In the matterof culturalliteracy,we suggest that Sophilos was fully at home with the mythology of his time. We would suggest that he knew more about mythologythan many of his contemporariesknew. He illustratescharactersand episodes from myth on many of his survivingworks. Those myths which we also know from literarysourcesrevealmany variantsin their written forms even within the Homeric, Archaic, and Classical periods. Where Sophilos' version may differ from what is often taken to be canonicalform, we think it ill-advised to criticize him for "not knowing"the version usually consideredauthoritative. Our guess is that he knew the Homeric stories very well (whether in the Iliad and Odysseyor not) and that he knew other literaryversionseven of well-known stories, or versions preservedas folk-tale ratherthan as more formal literature. Scholarstake it for granted that writershad "permission"to make up their own variationson familiarstories. Euripidestold of individualepisodes, in different plays,in wayswhich are impossibleto reconcilewith one another;and his version of a tale is often substantiallydifferentfrom the "same"storyas told by Aiskhylos or by Sophokles. Why should not paintersalso have had the freedomto change elements of a story?That this is inconvenientfor the modern scholaris true, and in some ways unfortunate. It is by no means a reason to devalue the work of Sophilos. DEPARTMENT OF CLASSICS AND RELIGIOUS STUDIES UNIVERSITY

OF OTTAWA

OTTAWA, ONTARIO

KIN 6N5

[email protected]

SOPHILOS' VASE INSCRIPTIONS

37

APPENDIX: INSCRIPTIONSON ARCHAICVASES

In this Appendix, each inscriptionis given in upper-casetranscription(left column), its equivalentin classicalGreek (centralcolumn), and in English translation(right column). ABBREVIATIONS

ABV ARVAdd ARV2 ARVPara

Beazley 1956 Carpenter1989 Beazley 1963 Beazley 1971 PROTO-ATTIC

B-W1 BerlinA21: standedovoid kraterfrom Aigina Morris 1984: 74 and pl. 19; CVABerlin 1, pls. 10-11. Vidi [No inscription]Heraklesand Nessos? BLACK-FIGURE

NettosPainter B4/1 Athens NM 1002 ABV 4.1, 679; ARVPara 2.6; ARVAdd2 1; Immerwahr1990: 20.55 NETOS1 Nessos N?ooos/N6?Tto HEPAKAEE2

'HpaKfij;

Herakles

KX Painter B27/36 Athens Agora P10507 ABV 27.36; ARVAdd28; Moore and Pease Philippides1986: pl. 15.126; Immerwahr 1990: 21.60. ?? HEKA[.....] 'EK[a.....] Sophilos B39/15 Athens NM Akr 15165 (587): dinos (frr):Athens Wedding Dinos ABV 39.15, 681; ARVAdd2 10; Immerwahr1990: 21.61; Bakir1981: A2 pls. 3.5-5.9, 89.187-190.94; Beazley 1986: pi. 15.2; LIMC s.v. Amphitrite 53 pl. 580 and s.v. CharikloI pl. 150. Vidi ESOIAOSErPAOSEN

2o?txoS; 6ypapev

AETO +API9AO AE[M]ETEP HEETIA

"Sophilos drew"

ArijTc XapiKro Ar?lfrlTp 'Eoarci

Leto Khariqlo/Khariklo Demeter Hestia

1Three-bar sigma. 2 four-barsigma;stemmedrho. Retrograde;

PHOENIX

38 MoT.E[IAON]

no=stKov NAApvrpt"rlj

AN[~dITPITE] NU^Cra NYEAI HEP[.... 'Hpax,`HijLvyz;, 'EpCtti; Many possible supplements; best contenders 'Hp(XKXfi; HEPA,HEDAI1TOZ, HEPMEE,HEPAKAEX

Poseidon Amphitrite Nysai/Nyssai ?Hera, Hephaistos, Hermes, or Herakles

B39/16 Athens NM 15499: dinos (fr.) from Pharsalos ABV 39.16, 681; ARVPara 18;ARVAdd2 10; Immerwahr1990: 21.62, fig. 13; Bakir 1981: pl. 6 A3. Vidi 1O0VIAO0XMErPAD- XootXo;pi'Eypctacv "Sophilosdrew me" ZEN the games of Patroklos nATPO9AYXi ATAA naTp6cXuK &OXc A+IAEX Akhilles, Akhilleus 'AXtkkc'uq 1o44txo;q PCSi'Rofjaev] 1Ob(I0RA MEnOI"Sophilosmade me" ECEN]3 or 1O44iAOE MEFPAD- Zo44tXoo q'p"ypayev] "Sophilosdrew me" SEN] or Name ending in "ios."4 if orthograde....]IOX B40/16.1 London 1971.11-1.1 (once London, Erskine):dinos with stand;ErskineDinos ABV 40.16bis;ARVPara 19, 523; ARVAdd2 10, 1974: 24; Immerwahr1990: 21.65; Bakir 1981: Al pls. 1-2, figs. 1-4; Williams 1983: 13-34, with additionalfragments. Vidi EOCNAOE1MErPAD- Eo4tko; ji' ?cypwjicv "Sophilosdrew me" EEN Peleus rlEAEYX fihxc6; Iris 'Ipt; IPI1 Hestia HEITIA 'Ecatkt Demeter AEMETE[P] A?~ilp Khariklo Xap{Kxo +API9AO Leto AETO Ailz6 AIONYXO, At6vuao; Dionysos Hebe HEBE "H3rj Khiron +IPON Xipwv0/XCfpov Themis EEMIZ Og4u; NYDAI N'cpac Nymphs Hera HEPA "Hpa IEYX Zeus ZAU; +APITEE Kharites Xa6pvIs; Poseidon IrIOEEIAON Floa&i6&w 3Thereadingverydubious. suchasphi,maybe substituted fortheI. 4Anotherletterwithhastaextendedupward,

SOPHILOS' VASE INSCRIPTIONS ANMDTPITE MOXAI A4DPOAITE APEX MOXAI HEPMEX [ArI]OAON M[O]IPAI AGENAIA APTEMIX OKEANOX OEOYX HIAEOYA HEDAIYXTOX

'Agt4wrpiti Mo6oyat 'A0po6i?rii "Aprl; Moi3act 'Epgij; 'AiTco'kow Moipat 'A0ivWt/'A0i/vM "AptreCt; 'DYeav6;

39

Amphitrite Muses Aphrodite Ares Muses Hermes Apollo Moirai/Fates Athenaia/Athene Artemis Okeanos, Ocean

u; &tiO6k/7Th0

Thethys/Tethys

Elket{0uta "H`atatroq

Eileithuia5

Hephaistos

B40/17-18 Athens NM Akr 585a-b: lipless dinoid vessels (Beazley);kotyle krater(Bakir) (frr.)

ABV 40.17-18; ARVPara 18;ARVAdd2 11; Immerwahr1990: 21-22.66; Bakir1981: A17 pl. 35.64-65, 36.67-68; text 26, fig. 17 (drawing). Vidi Kastor KAXTOP Ka'r(op Pandrosos IIANAPOXOX fl&v6poao; Poseidon FIOXEIAON loaCet6&6v B40/20 Izmir 3332: "NuptialLebes,"standed("lebesgamikos"),from Old Smyrna ABV 40.20, 714; ARVPara 18; ARVAdd2 11; Immerwahr:22.67; Bakir 1981: A21 pls. 39-45, figs. 71-82, text 34-37, figs. 32-38 (drawingsof A, B) K]AXTOP rIOAYA[E]YK[E1X]

K6aoaop

Kastor

HoDU6E&KT1;

Polydeukes

HEAEN[E]

'EE6vr

Helen6

B40/21 Athens NM 15942, 15918 (2035.1-2): Standed krater,Ausgusskessel, louterion, from Menidi ABV 40.21, 42.36; ARVPara 18; ARVAdd2 11; Immerwahr1990: 21.63-64; Bakir 1981: A20 p. 11.7 (drawing),8.15-10.20. Vidione of the two frags. OEE [XoCbtXo; [X0D]IAOX MErI "Sophilosmade me" ~ t'F'.ino{ii or [Yo40D]jA0o [Xo4]tko;P'ciypctie "Sophilosdrew me" MEFPAVXE ... ]EPO)[..

... ]CPO[...

... ero...

KETA[.. or

KVT4[..?

?? or

KETA[YPOI]

Ksvcraupot

centaurs[pluralnot certain]

into the earlyfifthcentury,thoughthereis muchvariancein 5The aspirateformis standard vowels. Compare HEAEIeYA:TyrrhenianAmphora, Berlin (F1704) B96/14 (Kyllenios Painter); HIAEITHYA:Tyrrhenian amphora B96/13 (Prometheus Painter); HIAYEYA: Euthymides cup R29/19.

6Initial heta rounded,perhapsmiswrittenas theta?A variantletter form seems to me the better explanation-Sophilos' hetas and thetas are more than usuallyvariable.

PHOENIX

40

B40/24 Athens, NM 12578 (N911): column-krater ABV 40.24; ARVPara 18; ARVAdd2 11; Bakir1981: A15 pls. 18-23, figs. 33-44, text 21-22, figs. 12-16. [No inscription] Circleof Sophilos B43/4 Athens Agora P18567: Athens, Agora fragment ABV 43.4 centre;ARVAdd2 12; BakirB19 83.171; Moore and PeasePhilippides1986: p1.121.1921. Vidi ...]+0Z ...khos ...]XoS Kleitias B76/1 Florence4209: FrancoisVase ABV 76.1, 682; ARVPara 29; ARVAdd2 21; Immerwahr1990: 24-25.83, figs. 18-19; Stewart1983: pls. 4.1-4.3. For the inscriptionstranscribed,see Wachter 1991 [withoutillustration]. B77/2 Basle, Cahn: fr. of volute krater Johnston 1991: pl. 88a: from Egypt; Shapiro1994: 55, fig. 34; Bothmer attributesthe Moscow fragmentwith Perseus (Pushkin Museum 2986; ABV 77.2; ARVAdd2 21; Beazley 1986: 2, pl. 30.1) to the same vase:see Beazley1986: 33; Shapiroloc.cit. OAYTEYX 'OXUTc6/'O6uociSU Olyteus/Odysseus IIEPEYS Perseus Fnepace6 Nearkhos B82/1 Athens NM Akr. 15155 (611): Largekantharosor kantharoidkrater? ABV 82.1; ARVPara 30, Add2 23; Immerwahr1990: 26.94. Vidi NEAP+OZMEFPA()DEN N&apXo;pt'Eypayev Kac[7niolr?v "Nearkhosdrew me [and made me]."7 KA.[IOIEYEN TyrrhenianAmphora B96/13 Paris E852 ABV 96.13; ARVAdd2 25; Immerwahr1990: 40.172. NEEOS No6ao;/N6TTo;

Nessos

Exekias B143/1 Berlin 1720 (PergamonMuseum): Neck amphorawith spirals ABV 143.1, 686; ARVPara 59; ARVAdd239; Beazley1986: pl. 62.1; 63.1-3, Bothmer 1985: 29, fig. 16 (lip); Immerwahr1990: 32.132 B145/13 Vatican344 ABV 145.13, 686; ARVPara 60; ARVAdd2 40; Immerwahr1990: 33.136; Beazley 1986: pls. 64-66.4 7Beazley'ssupplement on the analogyof Exekias' double"signatures."

SOPHILOS' VASE INSCRIPTIONS

41

Exekian B146/1.1, Group E?, Toledo, Ohio, 80.1022A ABV 146.1bis,ARVAdd2 391, 400; True 1987: 115.69 (lid); Immerwahr1990: 32.129. B147/3 (top) Munich 2125 (.25) ABV 147.3 (top);ARVPara 61. Vidi A) E+EEKIAE:EnOIEN 'EcK{itC; ToOiEV

"Exekiasmade"

B) +EEKIKAZ:EnOEs8

"Xekikas ????"

ECKIKclS?i6Eq

AmasisPainter B152/27 Boston 01.8027 from Orvieto, neck amphora(shouldered,specialmodel) ABV 152.27; ARVPara 63; ARVAdd2 44; Beazley 1986: pl. 49; Immerwahr1990: 37.160, figs. 37-39 Phoinix 4)OINI+ (D)ovtM B155/s.n. New York 56.11.1 ARVPara 66; ARVAdd2 45 (155); Bothmer 1985: 64, 182-183, figs. and colour pl. 4 (no. 47). B157/A Aigina: Tripod pyxisfrom Aigina M. Ohly-Dumm in Bothmer 1985: 236-238; Immerwahr1990: 36-37.152 Arkhiklesand Glaukytes B163/2 Munich 2443 ABV 163.2; Shapiro1989: pl. 66b EPO()O1 Txp60o;

nurse

SoklesPainter B173/1 Berlin SMPK F1737 ARVPara 72 (ABV 173.1);ARVAdd2 49; Boardman1974: 122. Vidi NeandrosPainter B176/3 Boston 61.1073 ARVPara 69-70.1 (to ABV 167.3, 168.1);ARVAdd221; Immerwahr1990: 49.229 Swing Painter?PrincetonPainter? B307/59 London 1849.11-22.1 [B144] ABV 307.59, attributedto the Swing Painter;Bohr 1982: 110, attributedto the circle of the PrincetonPainter;Immerwahr1990: 57.298. Vidi

8 Completeforeandaft.

42

PHOENIX BIBLIOGRAPHY

Allen, W.S. 1987. VoxGraeca. Cambridge. Amyx, D. 1989. CorinthianVasePainting of theArchaicPeriod.CaliforniaStudies in the History of Art 25. Berkeley. Bakir,G. 1981. Sophilos:ein Beitragzu seinemStil. Mainz. Barber,E.J. 1992. "The peplos of Athena,"in Neils 1992: 103-117. Beazley,J. D. 1956. AtticBlack-FigureVasePainters.Oxford [= ABV]. - 1963. Attic Oxford [=ARV2]. Red-FigureVase-Painters. 1971. Paralipomena: Additionsto Attic Black-FigureVase-Painters and to Attic RedOxford [-ARV Para]. FigureVase-Painters. - 1986. The Ed. D. von BothmerandM. B. Moore. DevelopmentofAtticBlack-Figure2. Berkeley. A Handbook.London. Boardman,J. 1974. AthenianBlackFigureVases: -1978. GreekSculpture: ArchaicPeriod.London. Mainz. Bohr, E. 1982. Der Schaukelmaler. Bothmer, D. von. ed. 1985. TheAmasisPainterandHis World.Malibu. Buck, C. D. 1955 [1933]. ComparativeGrammarof GreekandLatin. Chicago. Carpenter,T. H. 1986. DionysianImageryin ArchaicGreekArt. Oxford. - 1989. BeazleyAddenda2.Oxford [=ARVAdd2]. - 1991. Art and Myth in AncientGreece.London. London. Dover, K. J. 1978. GreekHomosexuality. Frel, J. and S. Morgan (eds.). 1983. GreekVasesin the GettyMuseum.OccasionalPapers on Antiquity 1. Malibu. Graves,R. 1960. GreekMyths.Harmondsworth(orig. publ., 1955). Harris,W. V. 1989. AncientLiteracy.Cambridge,Mass. Henrichs, A. 1987. "Myth Visualized:Dionysos and His Circle in Sixth-CenturyAttic Vase-Painting,"in True 1987: 92-124. Immerwahr,H. R. 1990. Attic Script:A Survey.Oxford. Revisedwith supplementby A. W. Jeffery,L. H. 1990. TheLocalScriptsofArchaicGreece2. Johnston. Oxford. Johnston,A. W. 1991. "GreekVases in the Marketplace,"in Rasmussenand Spivey1991: 203-231. Kilmer,M. F. 1993. GreekEroticaonAtticRed-FigureVases.London. Berlin (orig. publ. Berlin 1901). Kirchner,J. 1966. ProsopographiaAttica. Revisedwith additionsby R. A. Tomlinson. Lawrence,A. W. 1983. GreekArchitecture4. Pelican History of Art. New York. Moon, W. ed. 1983. AncientGreekArtandIconography. Madison, Wisc. Moore, M. B. and M. Z. Pease Philippides. 1986. Attic Black-FiguredPottery. The Athenian Agora 23. Princeton. Morris, S. P. 1984. TheBlackand WhiteStyle:AthensandAiginain the OrientalizingPeriod. New Haven and London. Neils, J. ed. 1992. Goddessand Polis:ThePanathenaicFestivalin AncientAthens.Hanover, New Hampshireand Princeton. - ed. 1996. Panathenaiaand Parthenon.Madison, Wisc. WorshippingAthena: Ortiz, G. 1994. In Pursuitof theAbsolute:Art of theAncientWorldfromthe GeorgeOrtiz Collection.ExhibitionCatalogue:RoyalAcademyof Arts, London. Berne.

ERRATUM

Figure 2 in Martin F. Kilmerand RobertDevelin, "Sophilos'Vase Inscriptions and CulturalLiteracyin ArchaicAthens,"Phoenix55 (2001) 9-43, at 16, should have appearedas follows:

Fig. 2: Fragmenti of AthensWeddingDinos (B39/15)with inscription namingtheNyssai.Photo:M. Kilmer.

SOPHILOS' VASE INSCRIPTIONS

43

Pape, W. and G. Benseler. 1959. Worterbuch dergriechischen Eigennamen.Graz (orig. publ. Braunschweig1884). Powell, B. 1991. Homerand the Originsofthe GreekAlphabet.Cambridge. Rasmussen,T. and N. Spivey(eds.). 1991. Lookingat GreekVases.Cambridge. Robb, K. 1994. Literacyand Paideiain AncientGreece.Oxford. Robertson,M. 1991. "Adoptingan Approach,"in Rasmussenand Spivey 1991: 1-12. 1992a. TheArtof Vase-Paintingin ClassicalAthens. Cambridge. 1992b. "The Pioneersin Context,"in Wehgartner1992: 132-139. Shapiro,H. A. 1989. Art and Cultunderthe Tyrantsin Athens.Mainz. 1992. "The Marriageof Theseus and Helen," in H. Froning,T. Holscher, and H. Mielsch (eds.), Kotinos:FestschriftfirE. Simon.Mainz. 232-236. 1994. Myth intoArt: Poetand Painterin ClassicalAthens. London and New York. Simon, E. 1992. "Euphroniosund die Etrusker,"in Wehgartner1992: 90-103. 1996. "Theseusand Athenian Festivals,"in Neils 1996: 9-26. Snodgrass,A. 1998. Homerand theArtists.Cambridge. Stewart,A. 1983. "Stesichorosand the FrancoisVase,"in Moon 1983: 53-74. An Anthropology Svenbro,J. 1993 [1988]. Phrasikleia: of Readingin AncientGreece.Ithaca and London. Thomas, R. 1992. Literacyand Oralityin AncientGreece.Cambridge. Threatte, L. 1980. TheGrammarofAtticInscriptions1: Phonology.Berlinand New York. True, M. ed. 1987. Paperson theAmasisPainterandHis World.Malibu. Vanderpool,E. 1953. ReviewofJ. D. Beazley, TheDevelopmentofAtticBlack-Figure.AJP 74: 321-333. Various. 1993. Die Etruskerund Europa. Exhibition Catalogue: Staatliche Museen PreussischerKulturbesitz.Berlin. Wachter, R. 1991. "TheInscriptionson the Fran:oisVase,"MusHelv48: 86-113. Wehgartner,I. ed. 1992. EuphroniosundseineZeit. Berlin. West, M. 1978. Hesiod,Works and Days. Oxford. Williams, D. 1983. "Sophilosin the BritishMuseum,"in Frel and Morgan 1983: 9-34.

A~~~~~~~~~~

O~~

Plate 1. Athens Wedding Dinos; fragmentwith retrogradeverticalinscription Photo: M. Kilmer,with permissionof the National ArchaeologicalMuseu

i,

~~~~~~-~i; r:~ iiIli~

,f5~~~~~~~~~~~~

-: L;

I

:l

}J kmt? Ill' i ~~~~~~~~~~I i

IL

LI

I

[

pi --

with chariot chariotrace raceand and retrograde from Pharsalos; 2. Dinos Dinos t}omn Plate 2. fa fragmentwvith Pharsalos;fragment retrogradeinscriptions inscriptionsX-o4nXoc Plate Zo4>.thoc ' !.t with of the National Photo: M. Kilmer, Museum permission Archaeological Photo:M. Kilmer,withpermission of the NationalArchaeological Museu