Dumbarton Oaks Papers Number 13

Front Matter

Dumbarton Oaks Papers, Vol. 13. (1959) Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C%3AFM%3E2.0.CO%3B2-J Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:35:06 2008

DUMBARTON OAKS PAPERS

THE DUMBARTON OAKS RESEARCH LIBRARY AND COLLECTION

Trustees for Harvard University

Washington, District of Columbia

DUMBARTON

OAKS PAPERS

Dumbarton Oaks Papers were founded in 1941 for the publication of articles concerning late classical, early medieval, and Byzantine civilization in the fields of Art and Architecture, History, Theology, Literature, and Law. Dumbarton Oaks Papers Number 13 contains the following studies and notes: George Ostrogorsky: T h e Byzantine Empire in the World of the Seventh Century Peter Charanis: Ethnic Changes i n the Byzantine Empire in the Seventh Century George Ostrogorsky: Byzantine Cities in the Early Middle Ages Robert S. Lopez: T h e Role of Trade in the Econontic Rcadjustment of Byzantium in the Seventh Century John L. Teall: The Grain S u p p l y of tlze B y zantine Empire, 330-1025 Sirarpie Der Nersessian : T h e Armenian Chronicle of the Constable Smpad or of the "Royal Historian" Andrew Alfoldi : Cornuti : A Teutonic Contingent in the Service of Constantine the Great and its Decisive Role in the Battle at the Milvian Bridge; with a contribution by Marvin C. Ross on Bronze Statuettes of Constantine the Great Paul A. Underwood : Fourth Preliminary Report on the Restoration of the Frescoes in the Kariye Camii at Istanbul by the Byzantine Institute, 1957-1958

T

HE

NOTES Paul A. Underwood: Notes on the W o r k of the Byzantine Institute in Istanbul; 1957 George P. Galavaris: T h e Mother of God, "Stabbed with a Knife" Paul A. Underwood: T h e Evidence of Restorations in the Sanct.uary Mosaics of the Church of the Dormition at Nicaea Cyril Mango: T h e Date of the Narthex Mosaics of the Church of the Dormition at Nicaea Demetrius J . Georgacas: Greek Terms for "Flax," "Linen," and their Derivatives; and the Problem of Native Egyptian Phonological Irtfluence on the Greek of Egypt Byzantium in the Seventh Centuvy. Report on a Dumbarton Oaks Symposium - -

J . J. A U G U S T I N PUBLISHER

LOCUST

VALLEY,

NEW YORK

DUMBARTON OAKS PAPERS NUMBER THIRTEEN

Dumbarton Oaks Papers

NUMBER THIRTEEN

The Dumbarton Oaks Research Library and Collection

Trustees for Harvard University

Washington, District of Columbia

1959

ALL RIGHTS RESERVED B Y THE

TRUSTEES FOR HARVARD UNIVERSITY

THE DUMBARTON OAKS RESEARCH LIBRARY AND COLLECTION

WASHINGTON, D. C.

Distributed by J. J. Augustifi, Publisher Locust Valley, New York

Library of Congress Catalog Card Number 42-6499

Printed in Germany at J. J. Augustin, Gliickstadt

GEORGE OSTROGORSKY

CONTENTS . . . . . . . . . . . . . . .

The Byzantine Empire in the World of the Seventh Century

. . . . . . . . . . . . . . . .

PETER CHARANIS

Ethnic Changes in the Byzantine Empire in the Seventh Century. GEORGE OSTROGORSKY

. . . . . . . . . . . . . . .

Byzantine Cities in the Early Middle Ages

. . . . . . . . . . . . . . . .

ROBERT S. LOPEZ

The Role of Trade in the Economic Readjustment of Byzantium in the Seventh Century JOHN L. TEALL

. . . . . . . . . . . . . . . . .

The Grain Supply of the Byzantine Empire, 330-1025 SIRARPIE DER NERSESSIAN

. . . . . . . . . . . . .

The Armenian Chronicle of the Constable Smpad or of the "Royal Historian" ANDREW ALFOLDI

. . . . . . . . . . . . . . . .

Cornuti: A Teutonic Contingent in the Service of Constantine the Great and its Decisive Role in the Battle at the Milvian Bridge; with a contribution by MARVIN C. ROSS on Bronze Statuettes of Constantine the Great PAUL A. UNDERWOOD

. . . . . . . . . . . . . . .

Fourth Preliminary Report on the Restoration of the Frescoes in the Kariye Camii at Istanbul by the Byzantine Institute, 1957-1958

NOTES PAUL A. UNDERWOOD

. . . . . . . . . . . . . . . 215

Notes on the Work of the Byzantine Institute in Istanbul: 1957 GEORGE P. GALAVARIS

. . . . . . . . . . . . . . .

229

The Mother of God, "Stabbed with a Knife" PAUL A. UNDERWOOD

. . . . . . . . . . . . .

235

The Evidence of Restorations in the Sanctuary Mosaics of the Church of the Domition at Nicaea CYRIL MANGO

. . . . . . . . . . . . . . . . . 245

The Date of the Narthex Mosaics of the Church of the Dormition at Nicaea DEMETRIUS J. GEORGACAS

. . . . . . . . . . . .

Greek Terms for "Flax," "Linen," and their Derivatives; and the Problem of Native Egyptian Phonological Influence on the Greek of Egypt Byzantium in the Seventh Century. Report on a Dumbarton Oaks Symposium

. .

253

271

ILLU STRATIONS (Following Page 183) Andrew Alfoldi: CORNUTI:

A TEUTONIC CONTINGEKT IN THE SERVICE OF CONSTANTINE THE GREAT AND ITS DECISIVE ROLE IN THE BATTLE AT THE MILVIAN BRIDGE

Marvin C. Ross: BRONZE STATUETTES OF CONSTANTINE THE GREAT 1-2. Princeton University, Art Museum. 10. Copenhagen, National Museum. Gundestrup Cauldron, detail. Bronze Statuette of an Emperor, front and side views. 11-12. Stockholm, Statens Historiska Mu3. Florence, Museo Archeologico. seum. Torslunda Plaque. Bronze Medallion of ConstantineI, 13-13 a. Berlin, StaatlicheMuseen. Scale,with obverse and reverse. bronze Statuette of Constantine. 4. Rome, Arch of Constantine. Detail of 14. Moscow, State Historical Museum. Warrior's Head. Bronze Weight. 5-6. Rome, Arch of Constantine. Detail 15. Kherson, Museum. Bronze Weight. of Relief on Pedestal. 16. Washington, D. C., Dumbarton Oaks 7 . London, British Museum. Royal Collection. Bronze Weight. Helmet from Sutton Hoo, detail. Statuette of St. Peter. Location un17. S. Copenhagen, National Museum. known (Photo. Ben Rowland). Scandinavian Helmet. 18-19. Rome, Arch of Contantine, detail 9. Edinburgh, National Museum of (Photos. Deutschen Archaeologischen Instituts, Rome). Antiquities. Torrs Chamfrein.

(Following Page Paul A. Underwood:

210)

FOURTH PRELIMINARY REPORT ON THE RESTORATION OF THE FRESCOES IN THE KARIYE CAM11 AT ISTANBUL BY THE BYZANTINE INSTITUTE,

Key to the Paintings of the Parecclesion: A. Paintings in the Vaults, above Cornice. B. Paintings on the Walls, below Cornice. I . K ariye Camii. The Parecclesion. General view looking East. 2. The Parecclesion. View into the Dome from the Southeast. 3. General View of South Wall, East Bay (250-254). 4. St. George (250). 5. Detail, Head of St. George (250). 6. St. Florus (251). 7 . St. Laurus (252).

1957-1958

8. St. Demetrius (253). 9. Detail, Head of St. Demetrius (253). 10. St. Theodore Tiro (254). 11. Detail, Head of St. Theodore Tire (254). 12. St. Theodore Stratelates (255). 13. St. Mercurius (256). 14. St. Procopius (257) and St. Sabas Stratelates (258).

15. Detail, Head of St. Procopius (257).

16. Detail, Head of St. Sabas Stratelates (258). 17. Unknown Saint (259). 18. Detail, Head of Unknown Saint (259).

ILLUSTRATIONS

19. St. David of Thessalonike (260). 20. Detail, Head of St. David of Thessalonike (260). 21. St. Eustathius (261). 22. St. Samonas (262). 23. St. Gurias (263). 24. Detail, Head of St. Gurias (263). 25. St. Artemius or St. Nicetas (264).

26. Detail, Head of St. Artemius or St. Nicetas (264). 27. St. Bacchus (265). 28. St. Sergius (266). 29. Unknown Military Saint (267) and fragmentary Medallion at right (268). 30. Unknown Stylite Saint (269). 31. Fragmentary Paintings in the Dome of the Prothesis of the Parecclesion.

(Following Page 228) Paul A. Underwood:

NOTES ON T H E WORK O F THE BYZANTINE INSTITUTE I N ISTANBUL:

Kariye Camii, Outer Narthex. Arcosolium of Bay 5 , Tomb of Eirene Raoulaina Palaeologina. 2. Arcosoliun~of Bay 5. Fragment of Fresco Portraits. 3. Arcosolium of Bay 5. Portrait of a Monk. 4. Arcosolium of Bay 5. Portrait of the Nun Athanasia. 5. Arcosolium of Bay 4. I.

1957

6. Arcosolium of Bay 4. Fragment of Fresco Portraits.

7 . Arcosolium of Bay 2.

8. Arcosolium of Bay 2 . Fragment of Fresco. g. Arcosolia of Bays 4 and 5 . Exterior. 10. Window of Bay 6 . Fill partly removed. 11. Window of Bay 6 . Slot in Jamb for Balustrade. 12. Arcosolium of Bay 2. Trace of Window Frame in Arch Soffit.

Illustration in T e x t Page

221.

Inscription accompanying Portrait of the Nun Athanasia.

(Following Page 233) George P. Galavaris: THE MOTHER OF GOD, "STABBED 1-4. Washington, D. C., Dumbarton Oaks Collection. Lead Seals.

WITH A KNIFE"

(Following Page 242) Paul A. Underwood: THE

EVIDENCE O F RESTORATIONS I N THE SANCTUARY MOSAICS O F THE

CHURCH O F T H E DORMITION AT NICAEA

Nicaea, Church of Dormition. Apse Mosaic. I . Reconstruction of Period 11. 2. Areas of Restoration. 3. The Mosaic of the Apse. 4. The Mosaic of the Apse, detail. Head of the Virgin. 5. T he Mosaics of the Bema, North Side, detail. Head of the Angel "Kyriotites."

6. The Mosaics of the Bema, South Side Area of restoration. 7 . The Mosaics of the Bema, South Side. 8. The Mosaics of the Bema, North Side. Area of restoration.

9. The Mosaics of the Bema, North Side.

10. Salonika,Hagia Sophia.Mosaicof theApse. 11. Istanbul, Saint Irene. Mosaic of the Apse.

ILLUSTRATIONS

ix

(Following Page 252) Cyril Mango : THE

DATE OF THE NARTHEX MOSAICS OF THE CHURCH OF THE DORMITION AT NICAEA

I. Nicaea, Church of the Dormition. Narthex, Vault Mosaic. 2. Nicaea,

Church of the Dormition. Narthex, Lunette over Central Door.

3-4. Nicaea, Church of the Dormition. Narthex, looking South. 5 . Nicaea, City Walls. Detail of Tower.

6. Island of Antigoni (Burgaz). Byzantine Ruin, detail.

7. Constantinople, St. Saviour Pantocrator. West Faqade, detail.

8. Constantinople, Odalar Camii. Detail of Brickwork. g. KurSumlija, Church of St. Nicholas. 10. Nicaea, Church of the Dormition. West Faqade. 11. Nicaea, Church of the Dormition. Dome. 12. Nicaea, Church of the Dormition. View from Southeast. 13. Nicaea, Church of the Dormition. West Faqade, Central Door.

14. Nicaea, Church of the Dormition. West Faqade.

PLATES

NOTES

The Byzantine Empire in the World of the Seventh Century George Ostrogorsky Dumbarton Oaks Papers, Vol. 13. (1959), pp. 1-21. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C1%3ATBEITW%3E2.0.CO%3B2-N Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:34:53 2008

THE BYZANTINE EMPIRE

I N T H E WORLD OF

T H E SEVENTH CENTURY

This study is in substance identical with a paper delivered at the Symposium on "Byzantium in the Seventh Century" held at Dumbarton Oaks in May 1957. The author wishes to express his warmest thanks to Mr. Cyril Mango and Mr. John Parker for having translated this study from the Russian.

N discussing some of the characteristic features of Byzantium's position as a world power in the seventh century I do not, of course, intend to make a detailed survey of the history of that period. Its salient features are well known to us. There is no need for me to go through all the stages of the Arab conquests in Western Asia and Africa, the Slavic settlement in the Balkans, or the Lombard invasion of Italy. Here we are interested not so much in the actual progress of these remarkable events as in their consequences. The important question for us is, how were these tremendous upheavals reflected in the Byzantine Empire ? I shall, therefore, attempt to determine the position of Byzantium in the surrounding world after the collapse of the Empire of Justinian, and how it appeared to the Byzantines themselves at that time. After the break-up of Justinian's restored Roman Empire in the West, the mass incursion of the Slavs into the Balkan peninsula, and the very rapid conquests of the Arabs in the East, Byzantine power was confined within territorial limits which were, by comparison, quite modest. What is more, the Empire, thus reduced on all sides, did not present any sort of unity. I t was made up of several clearly differentiated parts, and conditions within these varied so widely that, in order to give any sort of picture of the general position of the Empire, we must first consider separately the state of affairs in each part. We may begin with the most important of all, Asia Minor. Notwithstanding its quite unique importance for the very existence of the Byzantine state, I need say here only a few words about this region. Mesopotamia, Syria, Palestine, and Egypt had all been lost in the first few years of the Arab onslaught. An immediate, urgent threat hung over Asia Minor and over the capital of the Empire itself. But after the historic victory outside Constantinople in the 670's a decisive change came about. The Empire retained its vital center and Asia Minor, too, remained in its possession. By this fact, the further existence of the Empire was assured. I t is necessary to underline this fact as clearly as possible, for Asia Minor was the basis and foundation of medieval Byzantium. Many times subjected to invasion and devastation, though not permanently conquered by the enemy, Asia Minor was the most important and most enduring bulwark of Byzantine strength. Upon it depended the power and the very existence of the medieval Byzantine state. I t was in Asia Minor, too, that the renovation of the state was begun. Here the theme system first came into being - the new military and administrative organization which gave to medieval Byzantium its special strength. Within the frontiers of its much reduced territory the Empire was a more cohesive structure than before, endowed with greater internal solidity. The reconstitution of the state gave it a political strength that the later Roman Empire, gigantic but decaying from within, had never possessed, and also a much greater degree of spiritual unity, for the heterogeneous nature of the earlier Empire had caused continuous internal unrest. I t is enough for us to I*

4

GEORGEOSTROGORSKY

recall the age-old differences that had divided the central regions of the Empire from its Eastern provinces and the never-ending religious controversies in which these differences most tangibly expressed themselves, controversies which it was in fact impossible to settle, as proved by the fruitlessness of all attempts made by the tortuous imperial policy to bring about a pacification. A solution was finally reached only by the loss of the Eastern provinces -a loss which was itself brought about in no small measure by the unresolved conflict between Byzantium and the communities of the East, the Syrians and the Copts. In the East, therefore, Byzantium hat come to rest upon a base which was greatly shrunken, but which was firm and had been strengthened by internal reorganization. Although, here as elsewhere, it was often menaced by hostile invasion, the power of the Empire in Asia Minor remained unshaken. In the Balkans the situation was considerably more complicated and the position of the Empire immeasurably weaker. The stream of Slavic colonization, which had begun in some places at the end of the sixth century, had poured in an irresistible flood over the whole Balkan peninsula early in the seventh, after the failure of the Danubian campaigns of Maurice and the complete collapse of the old system in the time of Phocas. The movement extended from the Danube to Southern Greece, and from the Adriatic to the Black Sea. The former population was either wiped out or fled to the mountains, the coastal regions, and the adjacent islands. The ancient cities in the interior were ravaged, and only a few towns in the coastal areas held out in the midst of the Slavic flood. All the evidence indicates that outside these few strongholds there was no organized Byzantine power in the Balkans during the first half of the seventh century. The Balkan peninsula fell apart into numerous Slavic regions, the "Sclavinias," as the Byzantine sources call them. The full scale of the Byzantine catastrophe in the Balkans has not been generally appreciated. U7hathas led to confusion is that Byzantine authority in this area was not replaced by the organized power of another state, but by a number of separate Slavic tribes or tribal confederations. Insofar as the "Sclavinias" did not constitute a sovereign power, able to substitute itself for that of the Empire, it was possible to maintain the fiction of Byzantine sovereignty in the Balkan peninsula even after the Slavic occupation. But this presumption of authority by no means reflected the actual state of affairs, and with the eruption of the Bulgars into the Balkans there came into being, about 680, a sovereign state which Byzantium was obliged to recognize as such. The Empire was confronted with a strong rival, with which it would have to wage a desperate struggle for Balkan hegemony and for influence over the "Sclavinias." Constantine Porphyrogenitus, describing the migration of the Croats and Serbs from the regions beyond the Carpathians into the Balkans (chapters 29-36 of De administrando imperio), emphasizes with tedious persistence that these two powerful Slavic tribes had always been subject to the Emperor of the Romans, and that "never were they subject to the archon of BulgariaH.l De adm. imp.,ed. Moravcsik- Jenkins, especially c. 31,58and c. 32,146.

BYZANTIUM I N T H E S E V E N T H C E N T U R Y

The migration of the Croats and Serbs represents the second great wave of Slavic colonization in the Balkans, brought about by the decline in the power of the Avars after their disastrous defeat near Constantinople in 626. Whereas the first wave of Slavic colonization had moved forward in alliance with the Avars and under their leadership, the Croats occupied their new lands in the Balkans after a struggle with the Avars, whom they displaced. Seen in this light, the assertion of Constantine Porphyrogenitus that the migration of the Croats and Serbs took place in agreement with the Emperor Heraclius and was accompanied by a recognition of Byzantine sovereign rights does not seem in any way improbable. However, we should not lose sight of the fact that contrasted to the Emperor's persistent repetition of this version of the facts in chapters 29 and 31-36, is the complete absence in chapter 30 of any reference whatever to participation of the Byzantine government in these events. In any case, relations between Byzantium and the "Sclavinias" were shifting and changeable, and varied greatly at different times and in different areas. We are, thanks to the Miracula Sa~zctiDemetrii, rather well informed about the group of tribes in Macedonia which, not later than the beginning of the seventh century, had permanently settled in the neighborhood of Salonica. These Macedonian Slavs often made fierce attacks against Salonica and even raided parts of Greece and the islands, both by land and sea;2 sometimes, however, peace prevailed and some sort of modus vivendi was worked out between the inhabitants of the city and the neighboring tribes3 But even in the few strongholds that remained in their hands the Byzantines had constantly and laboriously to keep up their defenses. Only in 658-more than half a century after the beginning of this catastrophic phase-did the Emperor Constans I1 undertake a counteroffensive against the "Sclavinias," probably in the region of Thrace or Macedonia, where, as the Chronicle of Theophanes briefly notes, he "captured and subjugated many."4 This, it would seem, means that the Emperor was able to compel the Slavic tribes of this area to recognize Byzantine sovereignty, but it also indicates that, up to that time, they had not acknowledged it. What is more, it appears that this recognition was short-lived. In 688/9 Justinian I1 once more campaigned against the "Sclavinias" and Bulgaria. Having warded off the opposition of the Bulgars, he forced his way to Salonica through the lands occupied by the Slavs and made a solemn entry into the city of St. Demetrius. Once again we are told that he, like his grandfather Constans I1 thirty years previously, "subjugated a great multitude of Slavs" and settled them in the theme of Opsikion, in Asia M i n ~ r . ~ But wars with the restless inhabitants of the "Sclavinias" were to go on for a long time after this. Even when Byzantium succeeded in obtaining recognition of its sovereignty, this recognition proved to be transitory and precarious. In the middle of the eighth century the iconoclast Constantine V again fought M i ~ a c u l aS. Demetvii, 11.

I,

11.

2.

Ibid., 11. 4. Theophanes, ed. De Boor, p. 347, 6. Ibid., p. 364, 11-15.

6

GEORGE OSTROGORSKY

the Sclavinias in Macedonia, and again subdued them;6 in the eighties of the same century Stauracius, the logothete of the orthodox Empress Irene, once more moved against the Slavs "with numerous military forces," invaded the area of Salonica and also penetrated into central Greece, where (in the words of the chronicler) "he conquered all the people [i.e., the Slavs] and laid them under tribute," which, evidently, they had not previously been paying. He proceeded further and visited the Peloponnese, whence he carried off many captives.' As the Chronicle of Monemvasia is careful to emphasize, the Slavs in the Peloponnese, an area which, according to this source, they occupied from the end of the sixth to the beginning of the ninth century, were subject "neither to the emperor of the Romans, nor to anyone else."* The question whether Byzantine sovereign rights were recognized by the south Slavic tribes is interesting primarily from the point of view of Byzantine state theory, but to some extent the recognition or non-recognition of this sovereignty did undoubtedly reflect the real balance of forces. As we have seen, Byzantium did not give up its traditional pretensions to supreme authority, but was able to put them into effect only in some areas and for a limited period. However, in order to obtain a clear picture of actual conditions in the Balkans during the seventh century it is important for us to determine how far the governing power of the Empire was effectively exercised, rather than to what extent its nominal sovereign rights were acknowledged. The fundamental point is not whether the more or less theoretical supremacy of the Emperor extended over the Slavs, but whether the real force of Byzantine government did so. That is, were the "Sclavinias" directly subject to the Byzantine administrative apparatus ? To this question a negative answer must be given. We may define the Sclavinias" -that all-important concept in the history of the Balkans from the seventh to the ninth century-by saying that they were the territories occupied by the Slavs, not in themselves constituting any organized state, but separated from the Empire and outside the sphere of its direct administrative authority. We must not forget that in this period Byzantine power in the provinces rested upon the theme organization. Wherever a Byzantine provincial administration existed and functioned, the theme system was also to be found. Where there were no themes, there was likewise no Byzantine administration. This is the one infallible gauge of the actual state of affairs. Wherever Byzantium succeeded in preserving its power through this period of crisis, or was able to overcome this crisis and reconstitute its position, the theme system was set up. Thus it was introduced within the boundaries of Asia Minor as early as the first half of the seventh century. What state of affairs do we find in the Balkans ? First of all it is noteworthy that the theme-organization arose here considerably later than in Asia Minor, and that in the beginning it developed extremely slowly and was "

Ibid., p. 430, 21.

Ibid., p. 456, 25-457, 2.

N. Bbqs, Tb " T l ~ p iK T ~ U E W S-rij$ Mov~@acrias" X ~ O V I K ~ VBvZavris, , 1 ( ~ g o g )pp. , "The Chronicle of Monemvasia," Dumbarton Oaks Papers, 5 (IQ~o), p. 147,

67-68.

P. Charanis,

BYZANTIUM I N T H E S E V E N T H CENTURY

7

confined to a few areas. According to Constantine Porphyrogenitus, the first theme in the Balkan peninsula, the Thracian, adjacent to the capital, was founded at the time the Bulgarians crossed the Danube-that is to say, about 6 8 0 . ~In the famous letter of Justinian I1 to the Pope, dated 17 February 687, among the members of the state council is mentioned the commander of the Thracian army-the one and only strategus of a theme in the Balkan area, together with the strategi of the four themes of Asia Minor then in existence and the Exarchs of Italy and Africa.lo Soon after that, between 687 and 695 the theme of Hellas was founded: Leontius, who dethroned Justinian I1 at the end of 695 was the strategus of this theme.ll So at the end of the century there were two themes in the Balkan peninsula, that of Thrace which probably embraced those regions of the former Thracian diocese which had remained intact after the Slavic colonization and the setting-up of the Bulgarian state, and the theme of Hellas which seems to have comprised Central Greece. For a century this situation remained unaltered; no new themes were set up in the Balkans. Only at the end of the eighth century was the process of organizing themes in this area taken up anew, and then it developed rapidly. This process reflects the gradual reconstitution of Byzantine power after the collapse it underwent in the time of the Slavic migrations. I t gives us a clear picture of the progress of the Byzantine reoccupation-of its successes, but also of its limitations. We cannot here trace all the stages of this process which was at first extremely slow and laborious, but which later became intensive and effective. Its final result was that by the middle of the ninth century the districts of Greece and the coastal regions on both the east and west of the peninsula had been transformed into a series of Byzantine themes under Byzantine jurisdiction. Almost the whole coast-line was girdled with themes, by means of which Byzantium was able to revive its "thalassocracy." The interior of the peninsula, however, remained unaffected by this development, and here Slavic states came into being. With this the period of the "Sclavinias" ends. Where Byzantine administrative authority was restored they dissolved into the theme-organization ; in those areas that remained outside the confines of imperial power they were absorbed by the rising Slavic states. In this way a certain equilibrium was established. Cultural zones were formed which exist even today. The frontiers between the Byzantine and Slavic spheres in the middle of the ninth century correspond fairly exactly to the cultural zones which the eminent Yugoslav geographer Cvijii: has defined in modern times. The region which Cviji6 calls the "Greco-Mediterranean zone" corresponds, in its type of village settlement and habitation, more or less to the region over which the theme-organization extended in the mid-ninth century as a result of the Byzantine reoccupation.12 De thematibus, ed. Pertusi, c. I , 7. I t is possible, however, that the first steps in the organization of this central theme took place a t a considerably earlier date. Already in the 30's of the seventh century there is a mention of Thracian units of the army and of their commander whom, however, the Patriarch Nicephorus, (ed. De Boor, p. 24, 19,)with his customary affectation in the use of titles does not call the strategus but 6 T ~ ~V ~ F ( K IvK~ ~K ~ ~ ~ T E U ~ f&i yT~ pWc bV.V 10 Mansi, XI, 737 B. l1 Theoph., p. 368, 21. l2 J. CvijiC, La peninsule balkanique (Paris, 1918).

GEORGE OSTROGORSKY

However, in the early centuries of the Byzantine middle ages we are still at the beginning of this development. In the seventh and eighth centuries the Balkan peninsula, as we have seen, presented a conglomeration of Slavic settlements which-whether or not they acknowledged Byzantine sovereignty -were a foreign land, outside the Empire's jurisdiction. We have seen that Justinian I1 opened up a road from Constantinople to Salonica by force of arms. To do this he had to transfer cavalry forces from the themes of Asia Minor to Thrace.l3 His entry into Salonica was celebrated as a significant triumph over the Slavs. He made lavish grants to the church of St. Demeti-ius, the protector of the city, in gratitude for his victory "over the common enemy,"14 and it is possible that one of the frescoes in the church depicts his entry into the city.15 In the seventh century Byzantine power in the Balkans had collapsed almost entirely; but in time it was to increase again, though this was a slow process. Moreover, the Balkan regions wrested from the Empire remained open to the influences of Byzantine culture, which, from the strongholds that were retained, was able to penetrate them on an ever-increasing scale. In the West, on the other hand, the Byzantine dominions, which in the seventh century were still extensive, were doomed to slow extinction. The Exarchates of Ravenna and Carthage were indisputably important outposts of Byzantium in the West, but these remnants of Justinian's shattered Roman Empire were moving inexorably towards their liquidation. The African possessions were all finally lost by the turn of the eighth century; the Lombards were to take possession of the Exarchate of Ravenna in the middle of that century and thereafter Rome itself, the spiritual center of the West, was to turn away decisively from Byzantium. Only South Italy, which, together with the Balkan region of Illyricum, was annexed to the Partriarchate of Constantinople by the Iconoclasts in the middle of the eighth century, and which was to preserve its Greek culture, remained for a considerable period thereafter within the orbit of Byzantine influence.16 Here once again what interests us is not the external course of these events, familiar to all, but the question of Byzantium's relations with its Western possessions and with the West in general. We shall attempt to determine l3 Theoph., p. 364, 8. l4

I ff.;

p.

A. A. Vasiliev, "An Edict of the Emperor Justinian 11, September 688," Speculum, 18 (1g43), p. H. GrBgoire, "Un Bdit de l'empereur Justinien I1 date de Septembre 688," Byzantion 17 (1944/j),

11gff.

l5 This was suggested by E. Kantorowitz, "The King's Advent," The A r t Bulletin, 26 (1944), p. 216, note 63, and supported by A. A. Vasiliev, "L'entrBe triomphale de l'empereur Justinien I1 & Thessalonique en 688," Orientalia Christiana Periodica, 13 (1947), p. 3jjff. J. D. Breckenridge, "The Long Siege of Thessalonika, Its Date and Iconography," BZ, 48 (195j),p. 116ff., points out some difficulties in this interpretation. To my mind the only real difficulty is the fact that the city wall is represented as being on fire; however, Breckenridge does not give any more probable explanation of the fresco in question. l6 V. Grumel, "L'annexion de 1'Illyricum oriental, de la Sicile et de la Calabre au patriarcat de Con, 191 ff., has suggested that this annexation stantinople," Recherches de science religieuse, 40 ( ~ g j z ) p. did not take place in the early 730Js, as had been thought, but only some twenty years later, after the fall of Byzantine power in central Italy, during the pontificate of Pope Stephen 11. M. V. Anastos, "The Transfer of Illyricum, Calabria, and Sicily to the Jurisdiction of the Patriarchate of Constantinople in 732-33," Silloge Bizantina i n onore di S. G. Marcati (Rome, 1957)~p. 14ff.,rejects this view and maintains the former date.

BYZANTIUM I N T H E SEVENTH CENTURY

9

what significance the Byzantine government attributed to its Western lands, how far it exerted itself to preserve them, and, on the other hand, how closely connected the Byzantine people felt themselves to be with the Western portion of the Empire, to what extent they interested themselves in the West and what, if anything, they knew about this region. These are two aspects of the problem which, as we shall see, do not coincide. The break-up of Justinian's Empire which was begun by the advance of the Lombards into Italy in 568 compelled his successors to shift the focus of their policy to the East. Above everything else, in order to assure its very existence, it was essential that the Empire's tottering position on that front be strengthened. I t is nevertheless remarkable how active an interest the imperial government showed in its Western possessions for several generations after the reign of Justinian. The loss of most of his conquests in the West and the enforced preoccupation with the East did not mean the renunciation of the idea of universal Empire or the abandonment of the defense of what was left in the West. The foundation of the Exarchates of Ravenna and Carthage in the reign of Maurice, the most notable of Justinian's successors, proves the opposite to have been the case. How hard Maurice tried to maintain the old universal traditions, and how little inclined he was to abandon the West is shown by his famous will. By the terms of this will which, according to Theophylact Simocatta, was drawn up by Maurice during a serious illness in 597 and discovered after the accession of Heraclius, his eldest son Theodosius was to rule over the East from Constantinople and his second son Tiberius, as Emperor of "Old Rome," was to have authority over Italy and the Tyrrhenian islands; to his other sons Maurice bequeathed "the remaining portions of the Roman state."17 Thus Rome, as a second capital, was once more to become an imperial city. The dream of universal hegemony had not been given up, nor had the tradition of dividing the one Roman Empire between members of the ruling dynasty been forgotten. I t is also well known that Heraclius, son of the Exarch of Africa, who was brought to the throne by the power of the African fleet, considered transferring the capital to Carthage.18 Hemmed in by the Persians and Avars, the Eastern part of the Empire at the beginning of his reign seemed too weak to be an effective base for counterattack. This plan was not carried out, but its mere conception is unquestionably a clear indication not only of the extreme difficulty of the position in the East at that time, but also of the attraction which the old Western Roman region had for the imperial government. Even more indicative, in this respect, is the story of Constans 11. Whereas Heraclius had only temporarily considered transferring the center of government to the West, his grandson actually realized this idea. According to Theophanes, his intention was to remove his residence to Rome.19 In fact he l7 Theophylactus Simocatta, ed. De Boor, pp. 305, 25-306, 13. AS Bury, Latev Roman Empive, I1 (1889), p. 94, note 2 , suggests, this probably meant that one of them was to rule over Illyricum and the other over Africa. Is Xiceph., p. 12, 10. l8 Theophanes, p. 348, 5.

10

GEORGE OSTROGORSKY

contented himself with a brief visit to the Eternal City and then established himself in Syracuse, thus taking up a key position between Italy, partly conquered by the Lombards, and North Africa which was menaced by the Arab invaders. In reality, however, historical developments increasingly confined the Byzantine sphere of activity to the East, and the violent end of Constans after five ineffective years in Syracuse merely emphasized this fact. Although Byzantium never gave up its worldwide pretensions and never ceased to insist upon its supreme authority, the dream of a universal empire nevertheless grew faint. I t seems to have appealed less to the successors of Constans 11, and was even more alien to the Emperors of the eighth century. The government's keen interest in Western affairs had not, for a long time, been shared by the average Byzantine citizen. I t is known that Heraclius' African plans aroused much apprehension in Constantinople, and that the departure of Constans I1 for Italy caused open dissatisfaction. Notwithstanding imperial aspirations for a world-Empire, the bonds between East and West in fact grew constantly weaker. The seventh century is an important stage in this process of mutual estrangement, and to some extent marks its turning point. This assertion is, in itself, by no means new. Henri Pirenne pointed out with particular insistence how the ties between West and East were weakened in the seventh and eighth centuries, and, as is well known, he considered the principal cause of this weakening to be the penetration of the Arabs into the central Mediterranean basin.20I t is unnecessary to enter here into a detailed examination of this famous theory. We are familiar with the very telling objections which have been brought against it, objections which demonstrated the insufficiency of its principal argument^.^^ Nobody, of course, would wish to dispute the claim that the Arab invasion furthered a division between East and West, nor would anybody fail to recognize the ingenuity of Pirenne's very provocative ideas. But no-one at the present time could agree that the Arab invasion was the only, or even the principal, cause of the separation between the two halves of the former Roman Empire. I n this connection Professor Dvornik has recently indicated how important was the occupation of the Balkan peninsula by the Slavs.22His observations are most pertinent, although the point here, it seems to me, is not so much the destruction of Christianity in Illyricum, on which he lays chief emphasis, as the actual fact of the Slavic occupation of the Balkans, which erected a new barrier between East and West. This did more to destroy normal relations than did the Arab attacks on the sea. An example taken from a somewhat later source will illustrate this state of affairs clearly enough. I have in mind the life of St. Gregory the Decapolite, composed in the first half of the ninth century, which Professor Dvornik has Cf especially his Mahomet et Charlemagne (Paris-Brussels, 1937). Cf. for instance R. S. Lopez, "Mohammed and Charlemagne: A Revision," Speculum, 18 (1943), pp. 14-38; Anne Riising, "The Fate of Henri Pirenne's Theses on the Consequences of the Islamic Expansion," Classica et Mediaevalia, 13 (1g5z), pp. 87-130 (with complete bibliography). F. Dvornik, The Slavs, their Early History and Civilization (Boston, 1956), p. 44ff. 20

21

BYZANTIUM I N T H E S E V E N T H CENTURY

11

edited. This tells of the Saint's numerous voyages by sea. Many other sources, and in particular hagiographic works, also contain information about sea travel in the Mediterranean in the early middle ages, but we must here confine ourselves to this one very clear example. About the year 820 Gregory the Decapolite, having decided to travel from Ephesus to Constantinople, learned that a large number of merchant vessels in the harbor of Ephesus was ready to put to sea, though their captains were unwilling to weigh anchor because of the Arab pirates who were lying in wait for them.23 Nevertheless, protected by the prayers of the Saint, they reached Proconnesus safe and sound. From Proconnesus Gregory did not continue his journey to Constantinople, but took another ship to Aenus and thence, again by sea, went on to C h r i s t ~ p o l i s ~ ~ where he went ashore and continued his journey by land. However, near a river-probably the Strymon, which was not far distant -he was captured by "Slavic brigands."25 Impressed by the Saint's personality and bearing, the Slavs released him and thus he came to Salonica. There he met a monk with whom he decided to travel to Rome, but, instead of setting out along the Via Egnatia as anyone would have done in Roman times, since it was by far the shorter route, he once more chose the sea route, ignoring sailors' warnings about Arab pirates.26 So, by devious ways, he came, via Corinth, to Reggio, and there took a ship which had just arrived from Naples.27After spending three months in Rome, he once more set sail, this time to Syracuse, thence to O t r a n t ~and, , ~ ~finally-it is not stated by what means-returned to Salonica, where he remained for some time. Later, however, desiring rest and quiet, he decided to journey to "the mountains of the Slavic regions" with one of his disciples. But they had no sooner set out than the Saint, oppressed with terrible premonitions, hastened home, and in fact a few days later a bloody revolt of "the archon of that Sclavinia" broke out; whereupon Gregory told his disciple that he never travelled from place to place without having procured an imperial pass, properly sealed.29 Later on Gregory twice travelled to Constantinople again, undoubtedly by sea. We have dealt at some length with the information provided by this hagiographic work, for it enables us to make a number of significant deductions. First of all, it bears witness to a lively sea traffic over the whole Eastern half of the Mediterranean, although it in no way glosses over the dangers of Arab piracy. Franz Dolger has already pointed out the importance of this evidence and has rightly used it against the arguments of Pirenne.30At the same time the Life of St. Gregory the Decapolite shows -and this has not been sufficiently noted-that travel by land in the Balkan peninsula, even in the first half of F. Dvornik, La vie de saint Grbgoire le DLcapolite (Paris, 1926), chap. g.

Ibid., chap. 10.

25 Ibid., chap. 10, p. 54, 24. Cf. Vizantiski izvori za istoriju naroda Jugoslavije, p. 254. 26 Ibid., chap. 11.

27 Ibid., chap. 12.

Ibid., chap. 13.

20 Ibid., chap 17, p. 61, 20-62, 4.

F. Dolger, Byzanz und die europaische Staatenwelt (1g53), p. 365ff. 25

24

I (Belgrade, 19j5),

GEORGE OSTROGORSKY

the ninth century was almost completely paralyzed. Gregory journeys to and fro across the Mediterranean on his numerous and lengthy voyages; but as soon as he travels the short distance from Christopolis to Salonica by land he falls into the hands of the Slavs. More remarkable, as I have already mentioned, is that he avoids travelling along the celebrated Via Egnatia and makes his way from Salonica to Rome by the lengthy and roundabout sea route. The Slavic penetration thus weakened the link between East and West in a smaller area, but much more seriously than did the maritime incursions of the Arabs. In short, the Life of Gregory the Decapolite shows us that navigation in the Mediterranean continued, whereas communications by land in the Balkan peninsula were still practically paralyzed at the beginning of the ninth century. The causes of the separation of West and East were numerous and varied and cannot by any means be due solely to the difficulties along the lines of communication. The process of mutual estrangement between the two worlds had begun long before these difficulties developed, and was already in evidence in many ways in the early Byzantine period. In the seventh century this process was intensified to a marked degree. I t is not my purpose to go into all of its effects on political, ecclesiastical, and cultural history. But if we wish to determine what conception the inhabitants of the seventh-century Empire had of the world around them, we must note that the average Byzantine showed surprisingly little interest in the West and knew remarkably little about it. For all the attempts of the imperial government to live up to the high traditions of the Roman idea and to cling to the remnants of its former power in the West, this region lay beyond the horizon of most cultured Byzantines at this time. I t is sufficient to refer to our two chief sources for this period to be convinced of this fact. Upon reading the chronicle of either Theophanes or the Patriarch Nicephorus one becomes aware of the pronounced infrequency with which they refer to Western affairs in the seventh century. This indicates that their sources-those seventh-century chronicles which are lost to us but which they used-had little or nothing to say about thewest, for neither Theophanes nor Nicephorus was notable for the independence of his thinking; each merely handed on what he found in his sources. The data which Theophanes supplies in the tables that form a kind of chronological skeleton for his chronicle are very informative in this respect. As is known, the work of Theophanes is distinguished by its detailed and complicated chronological system. The narrative is divided according to years, and at the beginning of each year are indicated, in addition to the date reckoned both by the creation and by the birth of Christ, the years of the reign of the ruling Byzantine emperor and the Persian king-or, later, the Arab caliphand also the years of the episcopates of the pope and the four Eastern patriarchs. These five supreme representatives of the Christian church are, for the early Byzantine period, introduced with absolute regularity; the pope is cited first, after him the patriarch of Constantinople and thereafter the three other Eastern patriarchs. But as early as the seventies of the sixth century the list of Roman bishops is suddenly cut short; from 57415 (A.M.6067) the representa-

BYZANTIUM I N T H E S E V E N T H CENTURY

tives of the Roman Church disappear (in some mss. their years are not indicated after 569170). They are not cited in Theophanes' tables for an entire century and a half -that is to say, all through the seventh and the first quarter of the eighth century, and reappear only when he reaches the period of the Iconoclasts. Similarly, the loss of Syria, Palestine, and Egypt removed the Eastern patriarchs from the Byzantine field of vision. Those of Antioch cease to be commemorated in Theophanes' tables as early as 610111 (A.M. 6102); those of Jerusalem in 63617 (A.M. 6128); and those of Alexandria in 65415 (A.M. 6146). Thereafter the tables contain the names of only the emperors, the caliphs, and the patriarchs of Constantinople. In fact the patriarch of Constantinople had become, in the eyes of the Byzantines, the one supreme authority in the Church, and was the only one of the five patriarchs who was of immediate concern to them.31 The popes cease to figure in Theophanes' tables not in consequence of any papal clash with Constantinople, but purely and simply because they were no longer of interest to the Byzantines, just as the Eastern patriarchs, cut off by the Arab conquests, had ceased to be of interest to them; seemingly Theophanes found no more information about either in his sources. As I have just pointed out, the names of the Roman popes reappear in Theophanes' tables with the beginning of the Iconoclast period. The opposition of the Roman church to the Iconoclasts caused Theophanes, after an interruption of 150 years, to turn his attention once again to its representatives. Interest in Rome was awakened in orthodox iconodule circles, but knowledge of Roman affairs lagged behind. Above his entry for A.M. 6217 (= 725/6), in which he speaks of Leo 111's first measures against the cult of images, Theophanes first notes: The bishop of Rome, Gregory-g years. In fact Gregory I1 had occupied the throne of St. Peter for at least ten years prior to this. What is more, Theophanes was unaware that two popes of this name held the Roman See, one after another. For him Gregory I1 and Gregory I11 were one person, whose pontificate he extends from 726 to 735, whereas in fact Gregory I1 reigned from 715 to 731 and Gregory I11 from 731 to 741. The latter's successor, Pope Zacharias, was in office, according to Theophanes, from 735 to 756 instead of from 741 to 752 that is, for twenty-one, instead of eleven, years. Pope Stephen I1 does not appear at all in his chronological tables, and so on.32None of his data about the 31 AS an exception, Theophanes, from 70718 (A.M. 6199) to 73516 (A.M. 6227)) enters in his tables the name of John, Patriarch of Jerusalem, who, he knows "together with the Eastern bishops" anathematized the Iconoclast Leo I11 (p. 403, 29, A.M. 6221) ; the names of the other "Eastern bishops" seem to have been unknown to him. Similarly, in 74213 there suddenly appear the names of the Antiochene patriarchs Stephen, Theophylact, andTheodore, but they break off again as early as 75516 (A.M. 6247)) in the fifth year of the Patriarchate of Theodore, although, as Theophanes himself notes, he was patriarchforsixyears. I t is interesting to see that the Eastern patriarchs are not listed inTheophanesl tables even in the years of the sixth and seventh Oecumenical Councils, although their representatives took part in these Councils. No mention is made of their participation in the narrative of either Theophanes, or the Patriarch Nicephorus. Nor does either of them refer to the participation of the pope's representatives in these councils. 32 Theophanes' narrative, however, deals in some detail with Pope Stephen's "flight" to Pippin's court, dating it as early as 72415 (A.M. 6216, where the name of the Roman Pope is still not included in the tables)-that is to say, some thirty years too early.

GEORGE OSTROGORSKY

representatives of the Roman church in the eighth century is accurate. His information about the Roman Church in the seventh century is even more meagre.33 This is particularly striking since his contemporary information about the Arab Caliphs is remarkably exact. The years of their reigns are as well known to him as are those of the Byzantine Emperors themselves, and his data here are so faultlessly accurate that they have provided the main basis for elucidating the notorious chronological problem posed by his chronicle -the discrepancy between years and indiction^.^^ An examination of the data provided by the second of our chief sources for the early Byzantine middle ages, the Breviarium of the Patriarch Nicephorus, is no less instructive in this connection. Apart from a brief reference to the murder of Constans I1 in (the history of Constans' reign is not dealt with in his work), Nicephorus mentions the Western possessions of Byzantium only twice in dealing with the seventh century, and then very briefly and in passing, and he has nothing at all to say about Western regions lying outside the imperial boundaries. In the first instance he reports that the sacellarius Philagrius was exiled by Heraclonas "to the fortress called Septas, situated in the West, beyond the pillars of Hercules, in Libya";36 in the second, he reports the capture of the Exarchate of Carthage by the Arabs, doing so, however, only in order to narrate the dethronement of Leontius and the accession of Tiberius Ap~imar.~' And that is all; there is no word about Ravenna and its Exarchate, no mention of Rome and the Roman church, or of Italy in general. I t is with the East, Asiatic and European, that the chronicles of Theophanes and Nicephorus mainly deal, recounting the great events of external policy during that period. Eastern sources, in turn-especially Armenian and Syriac, sometimes Arab and even Ethiopic-also provide much precious information about Byzantium in this period, incomparably more than do Western sources which, nevertheless, devote more attention to Byzantium than do the Byzantine sources to Western affairs. Of first importance in the chronicles of Theophanes and Nicephorus are, naturally, the descriptions of the wars with the Persians and, later, from the thirties of the seventh century onward, with the Arabs. One following the other, these two Eastern powers were the only states that stood opposed to Byzantium on terms of equality-often, in fact, as contenders for a position of superiority-and that the Byzantines themselves tacitly accepted as such. We may recall that in his chronological tables Theophanes regularly comme~

~

33 AS we have seen, the names and years of the bishops of Rome are entirely omitted in his tables during the seventh century, and when he does refer to them in the text, his information is extremely inaccurate. I n contrast to Nicephorus, whose chronicle does not say a single word about the Roman Church in the seventh century, Theophanes mentions, e.g., the Lateran Synod of Pope Martin and dates it correctly to A. M. 6141 (Oct. 649) ; yet he antedates to 6121 (629130) the conflict between Martin and the imperial government and refers to Agathon as Martin's successor (p. 332, 4). Actually Agathon (678-81) is separated from Martin (649-55) by twenty-three years and no less than four popes. 54 Cf. G. Ostrogorsky, "Die Chronologie des Theophanes im 7. und 8. Jahrhundert," Byz.-Neugr. Jahrb., 7 (1930), p. I ff. 35 Niceph., p. 31, 28; on p. 33 he repeats the statement that Constans I1 died in the West. 36 Niceph., p. 29, 15. 37 Ibid., p. 39, 12ff.

BYZANTIUM I N T H E SE V E N T H CENTURY

15

morates, together with the names and years of the Byzantine emperors, those of the Persian king or the Arab caliph, which it does not occur to him to do in the case of any other foreign ruler. The Empire's rivals in the Balkan peninsula constituted the other factor of chief significance in Byzantine external politics during the seventh century. For the Byzantines the Balkans were part of "the West", of "Europe" and indeed, they most frequently applied these terms to the peninsula. This was the European West which was of immediate concern and interest to them. However, the information given by Theophanes and Nicephorus about the Slavs is, once more, poor and scanty. The clashes with the Slavs were mostly of a local character at that time, and escape the notice of the Byzantine chroniclers, whose attention is turned towards the capital and who record only a few of the most important campaigns organized from Constantinople against the "Sclavinias. " The chroniclers have a good deal to say about relations with the Avars at the end of the sixth and the beginning of the seventh century. Here the central place is taken by the imposing figure of the Khagan Bayan whose name, however, is not given by either Theophanes or Nicephorus. After the unsuccessful siege of Constantinople by the Avars and Slavs in 626 and the subsequent decline of Avar power, dealings with them are pushed further into the background. From the year 680 onward relations with the Bulgars come to the forefront. Nicephorus and, more particularly, Theophanes have preserved a fairly detailed account of the appearance of Asparuch's tribesmen on the Danube and of the foundation of the Bulgar state in the northeastern part of the Balkan peninsula -an account which they take from a common source, contemporary with these events.38 Both chroniclers likewise write at some length about relations with the Khan Tervel, who helped Justinian I1 regain the imperial throne, was dignified by him with the title of Caesar, and was enthroned beside him to receive the plaudits of the Byzantine population. Thereafter, relations with the young Bulgarian state occupy an increasingly large part of their narratives. Bulgaria quite clearly becomes the chief rival of Byzantium in "Europe," and has the same central importance in its external policy in the West as does the Caliphate in the East. But, along with those peoples with whom Byzantium had continuous deallings during the seventh century, and with whom it was engaged in an almost constant struggle, the chroniclers also refer to a number of other countries and peoples of Eastern Europe and hither Asia. Some of Nicephorus' information about them is of particular interest. These Eastern peoples were an important factor in the political network surrounding Byzantium, and the story of its dealings with them serves to fill in the picture of its external politics at this time. As early as the second half of the sixth century the Empire had begun relations with the Turkic nomads in Western Asia, and had exchanged diplomatic missions with them in the reigns of Justin 11,Tiberius, and Maurice. Especially notable are the relations with the Bulgaric Onogur principality which 38

Cf. J. Moravcsik, "Zur Geschichte der Onoguren," Ungarische Jahrbiicher,

10 (1930)) p.

70ff.

GEORGE OSTROGORSKY

had arisen in the Kuban region in the first half of the seventh century. I t was without doubt the ruler of this so-called "great Bulgarian Kingdom" who in 619 visited Constantinople and there adopted Christianity. This prince, it would seem, was Orhan (Organas). Heraclius received him with great cordiality and became his baptismal father. Byzantine notables stood godfather at the baptism of his lieutenants, and their wives became godmothers of the converts' wives. All of the visitors were rewarded with gifts and honorary titles, and the Bulgar ruler himself was honored with the dignity of patrician.39Nicephorus calls him 6 TGVO G V V ~TOG V Z ~ V O VK ~ ljplos, but does not mention his name, omitting it apparently through an oversight. There is good reason to believe that the name was given in his source, since further on Kuvrat, the ruler of the Hunnogoundours (6 TGV O ~ O ~ O U V B O K+IOI) G ~ W Vis represented as being the nephew of ~ r h a nthe , latter name being mentioned as though it were already familiar to the reader.40 This nephew is the celebrated Kuvrat, known both to Byzantine and Eastern historians and mentioned in the list of ancient Bulgar rulers, who has in modern times been the subject of a good deal of historical investigation. According to the indications of Nicephorus, Kuvrat revolted against the authority of the Avar Khagan and drove the Avars out of his country ; he apparently succeeded because of the collapse of Avar power after the Khagan's defeat near Constantinople just as, at about the same time, the Croats succeeded in ousting the Avars fr;m the western part of the Balkan peninsula, at the other end of the Empire. Having risen against the Avars, Kuvrat sent a mission to Heraclius and concluded a peace treaty with him "which they observed to the end of their lives," writes Nicephorus; furthermore, the Emperor sent him gifts and conferred on him the title of patrician.41 John of Nikiu reports that Kuvrat while still a child, had been baptised in Constantinople and brought up at the imperial court; apparently he had gone to Constantinople with his uncle Orhan and had been left there as a hostage.42 39 Niceph., p. 12, 20-28. 40 Ibid., P. 24,g. That Kuvrat was the son of Orhan's brother is also stated by John of Nikiu, ed. Zotenberg, p. 460. 41 Niceph., p. 24, 12-15. 4 2 This being so, there is no reason to conflate Nicephorus' references t o two treaties, concluded with two different rulers of the Bulgars-one around 619; the other around 635-by eliminating the later one, as was done by Zlatarski, Istorija nu b&lgarskata dhrz'ava prez srednite vekove, I. I , 33ff. The argument by which he justified this, viz., that after 626 the Emperor would have had no need of a treaty against the Avars, is very weak. On the other hand, there is no reason to suppose that both of the treaties mentioned by Nicephorus were concluded with Kuvrat, as does Grkgoire, "L'origine et le nom des Croates at des Serbes," Byzantion, 17 (194415)~ p. rooff.,for it is obvious that Kuvrat could not have received the title of patrician twice over, and, what is more, it is entirely clear from the narrative of John of Nikiu, upon which GrCgoire himself places such emphasis, that Kuvrat was converted to Christianity before he had become ruler of the Bulgars. J. Marquart, "Die altbulgarischen Ausdriicke in der Inschrift von Catalar und der altbulgarischen Fiirstenliste," Izvestija Russk. Archeol. Inst. v Konst., 15 ( I ~ I I )p., 7, and esp. p. 21, was inclined to identify Orhan with the unnamed ruler of the "Huns" who visited Constantinople in 619, and he was followed in this by J. Kulakovskij, Istorija Vizantii, 3 (1915) p. 91. Marquart himself, however, immediately rejected this very likely identification because it appeared t o him incomprehensible that the name of Orhan, obviously well known t o the Byzantines, should not be mentioned by Nicephorus under 619. But the matter is resolved quite simply if, as we have supposed, the name omitted by Nicephorus appeared in his source under 619, and this was the reason why Nicephorus, under 635, mentions Kuvrat as the nephew of Orhan. How little concerned Nicephorus was t o bring his various references into harmony with each other may be seen, for example, in his later account of the five sons of Kuvrat, when he writes of the latter: K o p p k 6 5 TIS d v o ~ Ka ~ ~ I O

S

BYZANTIUM I N T H E S E V E N T H CENTURY

I t follows from this that Orhan, and not Kuvrat, must be regarded as the founder of Bulgarian power in the Northern Caucasus. I t was also he who initiated the alliance with Byzantium. Kuvrat, when he in turn became ruler, strengthened this alliance by means of a further treaty, directed against the common enemy, the Avars. His personal relations with Heraclius, if we are to believe John of Nikiu, were so close that after the Emperor's death he intervened in the internal struggle that broke out in Byzantium, upholding therights of Heraclius' widow hlartina and her son H e r a ~ l o n a s . ~ ~ The mutual interest of Byzantium and the Bulgar Onogur state in the Northern Caucasus, and the firm and friendly relations between the two powers are very characteristic of this period. Equally characteristic and remarkable are the close personal links which bound the Imperial court to the rulers of this barbarian state, who had associated themselves with the culture and religion of the Empire and had been completely drawn into the orbit of Byzantine politics. Kuvrat, the ruler of the North Caucasian Bulgar State (Koirppa-ro5 or K o p p t h o ~ in Nicephorus, Kpopthog in Theophanes, Kurt in the list of ancient Bulgar rulers, Qetrades in John of Nikiu, Kuvrat in Moses of Chorene) must not, in spite of Gr6goi1-e)~~ appears in the Miracula be confounded with the Kuver ( K O G ~ E ~who ) Sancti Demetrii. The activities of the latter took place in an entirely different area, in the Balkans and P a n n ~ n i aand , ~ ~relate to a later period.46 But Kuver, though he was a vassal of the Avar Khagan, is another example of a barbarian chieftain who had close links with the Empire. Having been appointed by the Khagan to be leader of the Byzantine community which, drawn from all parts of the Balkans, had been settled in the region of Sirmium, in Pannonia, Kuver withdrew his allegiance from the Avar chief and, at the head of the Byzantine population subject to him, moved into Macedonia, and even sought, through a stratagem, to take possession of Salonica. In this he was greatly assisted by his lieutenant Maurus, who, in the words of the Miracula "was skilful1 in all things and knew the Greek, Latin, Slavic, and Bulgar tongues," and whom the EGy1v6p~vo~ -rWv @Awv ~oir-rwv,p. 33. 16.Having rejected the correct identification which he himself had put forward, Marquart preferred the arbitrary supposition t h a t the ruler who visited Constantinople in 619 was the chieftain of the Caucasian Huns or Hepthalites from Varachan in N. Daghestan, into whose country Heraclius was obliged t o retreat in 625; in support of this he refers t o Theophanes, 310, 19, who, V xhpav, without any more precise qualification. however, simply mentions -rjv T ~ Oirvvwv 43 John of Nikiu, ed. Zotenberg, p. 460. 44 GrCgoire, op. cit., p. 104ff. 45 This has been clearly shown b y A. Maricq, "Notes sur les Slaves dans le Pi.loponn&se et en Bithynie," Byzantion, 22 (1g52), p. 345ff., whose conclusions are manifestly correct. 46 Kuvrat died in the reign of Constans I 1 (641-68): cf. Theoph., p. 357, 11; Xiceph., p. 33, 17. Kuver's activity in Macedonia must be, however, dated t o the time of Constantine IV (668-85). Cf. F. BariSi5, Cuda Dimitrija Solunskog kao istoriski izvori (Belgrade, 1g53), p. 126ff. One consideration, in my opinion, is decisive: I n the chapter in which he writes about Kuver (11.5) the author of MiracuZa, I 1 refers to the then emperor as "our Emperor" ( ~ b vpaotABa jpGv, Migne, P G , 116, 1376), "him who is ordained by God t o rule over us" ( ~ b vZrrb 8~oGpaulhaira~vqpiv Aaxbv-ra, ibid., 1365), i.e. t o an Emperor who was still on the throne when Book I1 of the Mivacula was being written. But i t now seems to be generally recognized that this Book dates from the reign of Constantine IV. As for the resemblance between the names of the Bulgar Kuvrat and the Avar Kuver, such a coincidence, taking into account the fact t h a t one common barbarian milieu extended over the whole area of Central and Eastern Europe, is in no way surprising. Without having t o look far for another example, we may cite the eldest son of Kuvrat, who was called Bayan or Batbayan (BaTav6~:Niceph., p. 33, 26; Ba-rpatbv: Theoph., p. 357, 19)-the same name as that of the great Avar Khagan.

18

GEORGE OSTROGORSKY

peror, misled by his apparent submissiveness, honored with the title of consul (hyflatus).47 Similarly the Slavs and their leaders not only made war with Byzantium but, at times, were also in peaceful contact with the Byzantine population and might be found living on very amicable terms with them. The Miracula S . Demetrii tell the story of Pervud, chief of the tribe of the Rinkhinians, who had lived for some time in Sal0nica.~8Naturally, as is explained subsequently, he spoke Greek. On the strength of a slanderous report he was arrested by the Prefect of Salonica and sent to Constantinople; whereupon a deputation made up of both Slavs and inhabitants of Salonica went to the capital and petitioned for the release of the unjustly arrested prince-a remarkable demonstration of this friendly relationship. But -we must reiurn to the barbarian world beyond the Black Sea. The friendly alliance that bound the Empire to the old Bulgar-Onogur Kingdom against the Avars also linked it with the rising power of the Khazars-first against the Persians and later against the Arabs. With the Khazar principality, which was soon to replace the power of "Great Bulgaria" in the North Caucasus and in the lower Don-Volga region, Byzantine relations in the seventh and eighth centuries were e~peciallyfriendl~. As early as the time of his great campaigns against Persia, Heraclius, having s -roirp~ov~irplov,as Nicephorus calls the reached the Caucasus, sent gifts ~ p b -rbv Khazar Khagan, inviting this potentate to join in an alliance against the P e r ~ i a n sThe . ~ ~alliance was confirmed by a meeting between the two rulers. The Khagan came out to meet Heraclius with a numerous suite and, dismounting, both he and his retinue greeted the Emperor with a firoskynesis. Heraclius addressed the Khagan as his son, placed his own crown on the Khazar ruler's head and promised him in marriage his daughter Eudocia, "the Augusta of the Romans." I t is interesting to note that he also showed her portrait to the Khagan, who, on seeing it, was-consumed with love for its " a r ~ h e t ~ ~The e."~~ proposed marriage did not take place because of the Khagan's death.51 But later Justinian 11, who spent some years at the Khazar court during the period of his exile, married the Khagan's sister and after his restoration crowned both her and his son by her, making him his ~ o - r u l e r Constantine .~~ V was also wedded to a Khazar princess and his son, Leo IV, was known as "the Khazar." These princesses were thus the first foreign-born Byzantine Empresses. The fact is noteworthy if we recall the haughty attitude of the ~ ~ z a n t i court n e concerning marriages between the imperial family and foreign dynasties. The alliance with the Khazar Kingdom became a most important factor in Byzantine policy during the early middle ages. The Empire also entertained relations with the Caucasian tribes. At the time

47 Miracula, 11. 5.

48 Ibid., 11. 4.

48 Theophanes, too, considers the I
TOGS ~ o i r p ~ o uEKs ~ i j $ s &asoG5 x a l a p e ~ s6voyCrlouo1 (p. 315, 15; cf. pp. 407, 6, 11, 14; 433, 26).

50 Niceph., pp. 15, 20-16, 20.

51 Ibid., pp. 21, 28-22, 2.

52 Ibid., p. 43, 8; Theoph. p. 375, 28.

BYZANTIUhl I N T H E S E V E N T H CENTURY of his Persian campaigns Heraclius had already attracted a number of Caucasian chieftains into his camp, as both the Byzantine and the Armenian sources make clear.53Like Armenia, the Caucasian regions were a subject of conflict between Byzantium and Persia and, later on, even more vehemently, between Byzantium and the Arabs. In particular, Lazica, the ancient Colchis, was a crucially important area in Byzantine relations with both Persia and the Caliphate as well as with the northern nomadic tribes. The possessions on the south coast of the Crimea, centered around Cherson, were even more consistently important outposts. Although, with the increasing power of the Khazars, the greater part of Taurica came under their control at the end of the seventh century, Cherson remained Byzantine, and gave the Empire direct access to the wandering peoples of the East European plain, over whom a close watch was kept at all times.54 Such was the environment of the Byzantine Empire in the seventh century; such were the partners-rivals and allies-of the Emperors of the Heraclian dynasty. In the East, their most powerful opponents were first the Persian kings and thereafter the Arab caliphs, beginning with Omar the Conqueror, and the great founder of the Umayyad Caliphate, Muawiya; in the Balkans, the Avar Khagans, headed by the terrible Bayan, the princes of the numerous "Sclavinias" who, though frequently at war with the Empire, were also at times in peaceful contact with it, and, from the end of the century onwards, the Bulgar Khans-Asparuch, the founder of the Bulgarian state in the Balkans, and Tervel, the ally and helper of Justinian 11, who was honored by the Emperor with the appellation of Caesar. Beyond the Black Sea, the allies of the emperors against the Avars on one side and the Persians and Arabs on the other were the rulers of the old Bulgar-Onogur principality, Orhan and Kuvrat, baptized in Constantinople and granted the title of Byzantine patrician; and later on the Khagans of the Khazars, linked to the Byzantine court by treaties of alliance and by bonds of marriage as well; and finally, the semi-dependent princes of the many Caucasian tribes and the Armenian Curopalati. Thus Byzantium in the seventh century had many strong ties with the Orient, not only with the world of Islam but also with the barbarian nations of Eastern Europe. This barbarian, semi-nomadic world was subjected to the Empire's cultural influence and brought into the sphere of its political schemes; its chieftains associated themselves with Byzantine civilization, and some of them even adopted its religion and formed ties of kinship with the Imperial dynasty. 53 According to RIoses of Icagankatvaci, Heraclius had dealings with the princes of Albania, Iberia, and Armenia and demanded "that they should go to him of their own accord and serve him with their armies in wintertime" (Russian trans. by K. Patkanian [St. Petersburg, 18611, p. 102; cf. German trans. in A. Manandian, Beitriige zur albanischeva Geschichte, [Diss.Leipzig, 18971, p. 38).Theophanes, p. 3og,14, mentions the Lazi, Abasgians, and Iberians among the allies of Heraclius. 54 A. L. Jakobson, "Vizantija v istorii rannesrednevekovoj Tavriki," Sow. Archeologija, 21 (19j4), p. I jzf., exaggerates a good deal when he speaks of a complete downfall of Byzantine power in Taurica from the seventh to the ninth century (i.e., to the creation of the theme of Cherson). On p. 154 of his interesting and useful paper he himself, however, notes "the balance of power rather favorable to Byzantium, which was established on the north coast of the Black Sea after the arrival of the Ichazars, who afterwards, as is known, became the allies of Byzantium." Cf. also his remark that the activity of the Christian church in Taurica reached "unprecedented proportions" in the eighth century.

GEORGE OSTROGORSKY

Honored with Byzantine titles, they were brought into the hierarchy of the Empire and so were led to acknowledge that ideal sovereignty which resided in the Emperor. Byzantium was, in some sense, indisputably the head of all this diversified congeries of nations, and Constantinople was its center. But in the process of subjecting this barbarian world to its political and cultural influence, Byzantium became more akin to it, and was in turn subjected to its influences, adopted its manners, and took over its styles of dress and decoration. This was pointed out by N. P. Kondakov, who rightly emphasized that the world of the Eastern nomads played a historic role that is deserving of more attention than is usually devoted to it .55 Byzantium, after the break-up of Justinian's Empire, was never again to be a world monarchy. But it continued to exert a very great influence, both political and cultural, on the world around it, an influence which expanded in the East as it diminished in the West. As we observed at the outset, Byzantium in the seventh century was far from being a unified body ; it was made up of a number of clearly distinct parts, with varying outlooks and separate destinies. In Asia Minor the Empire stood firm and unshaken; here was the source of its external and internal strength and the bulwark of its reviving might. In the Balkan peninsula, its power, shattered by the Avar invasions and the Slavic colonization, was confined to a few towns in the coastal area, but from these few remaining centers began the gradual and partial restoration of that power; from them its cultural influence radiated out into the neighboring Slavic lands. In the West both the political power of Byzantium and its cultural influence were in a process of slow but steady liquidation. Finally, on the North coast of the Black Sea the Empire maintained its positions. From here it was able to exert an influence over the barbarian nations of Eastern Europe, with which in this period it had connections incomparably closer and more durable than it had with the alien and far-off West. Byzantium in the seventh century faced eastward. This, however, does not mean that it became an "oriental" state, as is often asserted. We must not forget that catastrophe had fallen on the old Roman world-Empire both in the West and in the East; that it had lost not only its Western possessions but its Eastern provinces as well-these having now entered the orbit of Islam. Byzantium occupied a special place between the Romano-Germanic West and the Islamic East as a Greek state, which it finally became just at this time, in the seventh century. I t did not, of course, become Greek either in the ethnic sense, since it remained multi-national, or in its political ideology, since it jealously preserved its claims to the inheritance of Rome. I t became a state that was Greek in culture and language, and no longer resistant to the natural process of Hellenization, the victory of which had been assured long before by the transfer to the East of the center of the Empire, but which finally triumphed only after the collapse of the restored Roman Empire of Justinian. j5 N . P. Kondakov, "Les costumes orientaux & la cour byzantine,"Byzantion, I ( r g q ) , p. 7ff., and OCerki i zametki Po istorii srednevekouogo iskusstva i kul'tury (Prague, 1929), p. 61 ff. Cf. also J. bloravcsik, "Proischoidenie slova T L I T ~ ~ K I O V , " Sem. Kond., 4 (1931), p. 69ff.

BYZANTIUM I N T H E S E V E N T H CENTURY

21

At once Greek and medieval, seventh century Byzantium was in its general features strikingly different from the half-antique, half-Latin Empire of the preceding epoch, and, at the same time, was separated more and more clearly from the conteinporary Romano-Germanic West. It was not the incursions from without, but the inherent process of cultural and linguistic separation that played the decisive role in the gradual estrangement between the Latin West and the Greek East, an estrangement which, after the unsuccessful endeavor made in the time of Justinian to re-unite the two worlds, found such clear expression in the seventh century.

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: The Byzantine Empire in the World of the Seventh Century George Ostrogorsky Dumbarton Oaks Papers, Vol. 13. (1959), pp. 1-21. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C1%3ATBEITW%3E2.0.CO%3B2-N

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 8

The Chronicle of Monemvasia and the Question of the Slavonic Settlements in Greece Peter Charanis Dumbarton Oaks Papers, Vol. 5. (1950), pp. 139+141-166. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281950%295%3C139%3ATCOMAT%3E2.0.CO%3B2-U

NOTE: The reference numbering from the original has been maintained in this citation list.

Ethnic Changes in the Byzantine Empire in the Seventh Century Peter Charanis Dumbarton Oaks Papers, Vol. 13. (1959), pp. 23-44. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C23%3AECITBE%3E2.0.CO%3B2-V Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:35:31 2008

E T H N I C CHANGES

I N THE BYZANTINE EMPIRE

I N T H E SEVENTH CENTURY

Tlris study is in substance identical with a paper delivered at the Symposium on "Byzantium in the Seventh Century" held at Dumbarton Oaks in May 1957

H E Byzantine Empire was never in its long history a true national state with an ethnically homogeneous population. I t is true that the conquests of the Arabs in the seventh century deprived the empire of great numbers of non-Greek-speaking elements and gave to it an aspect which appeared to be more Greek than had been the case before. Egypt and Syria, where a national consciousness and a literature in the native languages had begun to develop, were lost ; so also was Africa with its Latin and Punic-speaking population. There remained Asia Minor, parts of the Balkan peninsula, the islands of the Aegean, including Crete, certain regions of Italy, and Sicily. Here the Greekspeaking elements were strong, but the ethnic homogeneity which they suggest was more apparent than real. Let us first look at Asia Minor. No doubt, under Hellenistic and Roman domination, the native population of that very important peninsula had been deeply affected by Hellenism, but neither in language nor in culture, particularly in the isolated regions of the back country, was the victory of Hellenism comp1ete.l The evidence for this is scattered and largely circumstantial, but it is unmistakable. The native languages survived long into the Christian era. We know that Phrygian, which in the first three centuries of our era witnessed a true renaissanceI2was still spoken in the sixth ~ e n t u r yThe .~ which, according to Jerome,5 was heard same was true of L y ~ a o n i a nCeltic, .~ in Galatia in the fourth century, survived until the end of the fifth and probably beyond. So we may infer from a hagiographical text concerning a posthumous miracle of St. Euthymius, who died in 487. According to this text a Galatian monk who had lost his speech was cured by the saint, but at first he could 1 For example, in the sixth century the city of Tralles was thoroughly Greek-speaking, but the back country was hardly impregnated by Hellenism, as is shown by the fact that i t still remained predominantly pagan. Agathias, Historiae (Bonn, 1838), 102; E. W. Brooks, Iohannis Ephesini historiae ecclesiasticae pars tertia, CSCO (Louvain, 1936),81,125 (English trans. R. Payne Smith [Oxford, 1860],159,230). 2 IV. M. Calder, "Corpus Inscriptionum Neo-Phrygiarum," Journal of Hellenic Studies, 31 ( I ~ I I ) , 161-215 ; 33 (1913)~ 97-104 ; 46 (1926)~ 22 ff. On page 164 of volume 31 Calder writes: "The existence of over sixty inscriptions of which no two are exactly alike, and all of which exhibit intelligent syntactical variation, is sufficient proof that Phrygian was not a moribund language surviving in a few fixed formulae, but was the everyday language of the uneducated classes at the period to which the texts belong." For a map indicating the Phrygian-speaking zone in Asia Minor about A . D. 250 see W. 31. Calder, ed., Monumenta Asiae Minoris antiqua, VII: Monuments of Eastern Phyrgia (Manchester, 1956),xliv. For a corpus consisting of the Neo-Phrygian inscriptions published up to 1928 see J. Friedrich, Kleinasiatische Sprachdenkmaler (Berlin, 1932)~128-140. Cf. Calder's remarks, Monuments of Eastern Phrygia, p. xxvii. Journal of Hellenic Studies, 31, 165; Karl Holl, "Das Fortleben der Volkssprachen in Kleinasien in nachchristlicher Zeit," Hermes, 43 (1go8),248. We are told concerning an Arian bishop Selinas that his father was a Goth, his mother a Phrygian, and that for this reason he used both languages. He also preached in Greek. Socrates, Historia Ecclesiatica V. 23. Rligne, P G , 67. 648; Sozomen, Historia Ecclesiastica. VII. 16. Migne, P G , 67.1468. Life of St.Martha, Acta Sanctorum, May, V, 413C. Cf. Holl, op. cit., 243-246. For the use of Lycaonian at the time of St. Paul, Acts of the Apostles, 14, 11. Jerome, Commentarium i n Epistolam ad Galatas 11, 3 (Nigne, PL, 26. 357). Cf. F. Stahelin, Geschichte der kleinasiatischen Galater, and ed. (Leipzig, 1go7),104; W. M. Ramsay, A Historical Commenta r y on St. Paul's Epistle to the Galatians (New York, I ~ O O ) , 145-164; J. G. C. Anderson, "Exploration in Galatia cis Halym," Journal of Hellenic Studies, 19 (1899), 316-318.

26

P E T E R CHARANIS

speak only in his native t ~ n g u eThe . ~ continuous use of Mysian is also attested. We are told about St. Auxentius, who lived during the first half of the fifth century, but whose biography was written at the beginning of the sixth century, that, as he had come from Mysia, he was barbarian in language.' In Cappadocia the native language continued to prevail certainly throughout the fourth century, as we learn from Gregory of Nyssas and also from Basil, who says that the Cappadocians were saved from a certain heresy because "the grammatical structure of their native tongue did not permit the distinction between 'with' and rand."'g In Cappadocia, too, there lived a people known as the Magusaeans, who scandalized the Christians by the tenacity with which they adhered to their strange practices, including marriage between brother and sister.10 In Isauria also the native tongue continued to be used. The evidence for this is a hagiographical text written after 596.11 I have found no evidence later than the sixth century attesting the persistence of native languages in Asia Minor. The chances are, however, that these languages, at least some of them, continued to be used long beyond the chronological limits of our evidence, for languages do not die out overnight. The Phrygians, for instance, as we may infer from what we know of the background of Michael 11, seem to have been only semi-Hellenized as late as the beginning of the ninth century. Michael, who is described as coarse, ill-educated, and contemptuous of Hellenic culture, was no doubt typical of the natives of Phrygia, many of whom may not have known any Greek at a11.12 We may suppose, then, with some reason, that there was no complete linguistic homogeneity in Asia Minor in the seventh century. This supposition is strengthened by the persistence of the native heresies, known from both ecclesiastical writers and epigraphy.13 Montanism was widespread in Phrygia, Lycaonia, and perhaps also in Cappadocia, Galatia, and Cilicia.14 Procopius states that A. H. M. Jones, The Cities of the Eastern Roman Provinces (Oxford, 1937), 122.

Vita S. Auxentii, Migne, PG, 114. 1428; Holl, of. cit., 241f.

Contva Eunomium, Migne, PG, 45. 1045.

De SPiritu Sancto, Migne, PG, 32. 208.

lo Saint Basil, The Letters, ed. and tr. Roy J. Deferrari, 4 (Cambridge, Mass., 1939) (Loeb Classical Library), 44-46; Eusebius, P~aeparatioEvangelica, ed. E. H. Gifford (Oxford, 1go3), I :352 (book VI, chap. 10). Holl, 09.cit., 243. I n Cilicia, too, the native language was spoken a t least until the fifth century. We are told by Theodoret, bishop of Cyrrhus, that an ascetic whom he knew personally spoke Greek though he was Cilician by race. We may infer from this that there were natives in Cilicia who did not speak Greek. Theodoret, Religiosa Historia, Migne, PG, 82. 1488. l2 Theophanes Constinuatus, (Bonn, 1838), 49. Cf. J. B. Bury, A Histovy of the Eastern Roman Empire from the Fall of Irene to the Accesion of Basil I (London, 1912), 78. IS W. M. Calder, "The Epigraphy of the Anatolian Heresies," Anatolian Studies Presented to Sir William Mitchell Ramsay (hlanchester, 1gz3), 59-91. For evidence of the persistence of some of the ancient heresies of Asia Minor as late as the seventh decade of the ninth century see C. Mango, The Homilies of Photius, Patriarch of Constantinople. English Translation, and Commentary (Cambridge, Mass., 1958), 279-282; 288-289. Photius speaks further (289) of "the ungodly ideas of those halfbarbarous and bastard clans which had crept into the Roman government." The reference no doubt is to the iconoclasts whom apparently Photius did not consider completely Hellenized. Cf. Mango's note (289, note 16). l4 G. Bardy, "Montanisme," Dictionnuire de thiologie catholique, 10 (1929)~ 2368, for the early period. For a Montanist inscription of the sixth century see H. GrCgoire, "Du nouveau sur la hierarchie de la secte montaniste d'aprhs une inscription grecque trouvBe prhs de Philadelphie en Lydie," Byzantion, 2 (19251, 329-336.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 27 the Montanists in Phrygia destroyed themselves rather than abandon their heresy,l5 but the evidence is that they continued to exist. They are mentioned in the Acts of the Council in Trullo (692), in which they are called Phryges. The same Acts refer to other heresies in Asia Minor, especially in Galatia, as being numerous, and mention some, all of long standing, by narne.le Montanists are known to have existed during the reign of Leo III.17 We also know that early in the ninth century the Paulician Sergius Tychikos corresponded with a certain Leo the Montanist.18 The reference to this correspondence is rather significant, for it indicates that the Montanists, who henceforth cease to appear in history, may have merged with the Paulicians.lg This would explain the apparent increase in the strength of the Paulicians in Phrygia and the consequent apprehensive attitude toward them of the ecclesiastical and imperial authorities of Con~tantinople.~~ Some of the Montanists may have merged with the Athinganoi, another strange sect of considerable importance both I1 is said to have inherited from his in Phrygia and L y ~ a o n i a .Michael ~~ parents the beliefs of the Athinganoi, and Nicephorus I was accused of being friendly to both them and the Paulicians.22 Early in the ninth century the Athinganoi were either exterminated or driven out of their homes, and some of them were settled on the island of Aegina where the natives referred to them as aliens.23This attitude toward them does not prove that their language was not Greek, since the term alien could very well have been applied to newly established settlers from another province. The fact, however, that the Gypsies, descendants of the foreign Zatt who had been settled in the Empire in 855, came to be called Athinganoi may indicate that the latter were distinguished by their strange language.24 There is some basis for believing, therefore, that in the seventh century Procopius, Anecdota, XI. 14; XI. 23. Mansi, X I :984 (Canon 95). 17 Theophanes, Chronographia, ed. C. de Boor, I (Leipzig, 1883), 401. For other texts, J. Starr, T h e Jews in the Byzantine Empire (Athens, 1g3g), 91-92. 18 H. Grkgoire, "PrBcisions gkographiques et chronologiques sur les Pauliciens," Acadimie royale de Belgique : Bulletin de la Classe des Lettres et des Sciences Morales et Politiques, ge SBr., 33 (Brussels, 1g47), 317. 19 Cf. F. C. Conybeare, T h e K e y of Truth. A M a n u a l of the Paulkcian Church in Armenia (Oxford, 1898), LXXIV; CLXXXV. Grkgoire, ibid., 301. There is some evidence to the effect that a community of Cathari continued to exist in Philadelphia, Lydia, as late as the thirteenth century. Grkgoire, "Cathares dlAsieMineure, dlItalie et de France," Mimorial Louis Petit (=Archives de 1'Orient chretien, I ) (Paris, 1948), 144-145. 20 Ignatius, V i t a Nicephori, ed. C. de Boor, Nicephori archiepiscopi ConstantinoPolitani opuscula historica (Leipzig, 1880), 158-159. 21 Theophanes, I:495; Genesius, Historia (Bonn, 1834)~ 32; Theophanes Continuatus, 42. On the Athinganoi one may further consult Joshua Starr, "An Eastern Christian Sect: the Athinganoi, " T h e Harvard Theological Review, XXIX, 2 (1936), 93-106. 22 Theophanes Continuatus, 42; Theophanes, I : 488. Jews also are known to have existed in Asia Minor, as in Constantinople, but they do not appear to have been very numerous. Starr, T h e Jews in the Byzantine Empire, 88ff., 98f.; A. Sharf, "Byzantine Jewry in the Seventh Century," B Z , 48 (195511 111. 23 T h e Life of Saint Athanasia of Aegina, Acta Sanctorz~m,August, 111, 17oE. 24 M. J. De Goeje, Mtmoives d'histoire et de ge'ographie ovientale, no. 3. Mimoire sur les migrations des Tsiganes d travevs 1'Asie (Leiden, 1go3), 75. On the ancestry of the gypsies one may consult, inaddition to the work of De Goeje, A. A. Vasiliev-H. Gregoire, Byzance et les Arabes, I (Brussels, 1g35), 223-224; J. B. Bury, op. cit., 40, note I. 15

16

28

P E T E R CHARANIS

there still remained certain elements of the ancient native population of Asia Minor that had not been completely absorbed by Hellenism, either in language or in culture. But this point should not be too greatly stressed. The native elements were finally absorbed, though perhaps they retained some of their own traits. The administration, the army, the schools, but above all the official Church, with its insistence upon orthodoxy and its use of Greek, were powerful agents of Hellenization. The events of the seventh century, too, may have strengthened Hellenism in Asia Minor. We know that many Christians, the majority of whom were doubtless Greek-speaking, fled from Syria and Egypt when these territories were conquered by the Arabs.25We do not know exactly where they settled, though it is more than likely that many of them settled in Asia Minor. However, the settlement of new peoples, some of whom, notably the Slavs, will be mentioned in the course of this paper, was to complicate the ethnic composition of Asia Minor.26 One of the most important developments in the Byzantine Empire toward the end of the sixth century and the beginning of the seventh was the rise to prominence of the Armenians. They were to maintain this position throughout the seventh and eighth centuries, while in the ninth and tenth centuries they improved it even further. At the end of the sixth century the Byzantine Empire controlled the major part of Armenia,27 but the events of the seventh century, in particular the rise of the Arabs, deprived it of this control,28though it still retained some Armenian-speaking lands. I t was from these lands that the Empire drew its Armenian recruits, but many Armenians who entered its service also came from the Armenian regions under foreign control. Sometimes they came as 25 P. K. Hitti, Ovigilzs of the Islamic State (New York, 1g16), 180: "In the year 49 the Greeks left for the seacoast"; 194: They (the Greeks of Tripoli) "wrote t o the king of the Greeks asking for relief through reinforcement or ships on which they might escape and flee to him. Accordingly, the king sent them many ships which they boarded in the night time and fled away." 195: "He [a certain Greek patrician] made his way together with his followers to the land of the Greeks"; 189: "The fact is that when Damascus was taken possession of, a great number of its inhabitants fled t o Heraclius, who was then a t Antioch, leaving many vacant dwellings behind that were later occupied by the Moslems"; 227 : "At last they[the people and soldiers of Antioch] capitulated, agreeing to pay poll t a x or evacuate the place. Some of them did leave; but others remained, and t o the latter Abu-Ubaidah guaranteed safety, assessing one dipzav and one javib [of wheat] on every adult"; 231 f: "When the Moslem armies reached these towns [the Greek towns of Syria], their inhabitants capitulated, agreeing to pay poll t a x or evacuate the place. Most of them left for the Byzantine Empire"; 348: "Some of its [Alexandria's] Greek inhabitants left to join the Greeks somewhere else." Hitti's book is a translation of the Futuh al-Buldan of al-Buladhuri. 26 We may mention, for instance, the Goths who, in the early centuries of the Empire, were settled in Bithynia, in the territory which later formed the theme of Optimati. They were still there in the eighth century, though they seem t o have been Hellenized. Theophanes, I:385; "Acta Graeca SS. Davidis, Symeonis e t Georgii" Alzalecta Bollalzdiana, 18 (1899), 256. Alans seem to have settled in the Pontic regions of the Empire sometime between 662 and 666. P. Peeters, "A propos de la version armCnienne de l'historien Socrate," Annuaive de l'lnstitut de Philologie et d'Histoire Ovientales, 2 (Brussels, 1934)) 669, note 2. Vandals were settled in Asia Rlinor by Justinian. Procopius, De be110 Vandalico, I1 14, 17. 27 Under the Emperor Maurice the Byzantine frontier in Armenia followed a line extending from Sisibis t o lake Van, Alaku, Dvin, Garni, and Tiflis. Nisibis, Rlaku, Dvin, Garni, and Tiflis did not belong t o the Empire. P. Goubert, Byzance avant Z'Islam, I (Paris, 1g51), 290-295; cf. Ernst Honigmann, Die Ostgvenze des byzantinischen Reiches volz 363 bis 1071 (Brussels, 1g35), 27ff. 28 The Arab domination of Armenia was established in the second half of the seventh century. H. Manandean, "Les invasions arabes en ArmCnie," Byzalztion, 18 (1946-1g48), 190.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 29 adventurers, but more often as refugees. Thus in 571, following an unsuccessful revolt against the Persians, the Armenian Catholicos, a few bishops and numerous noblemen fled to Con~tantinople.~~ The leading men among these refugees, were, besides the Catholicos, Vardan Mamiconian and his retinue. There were also among them some Iberians (Georgians), headed by one Gorgonis, who had joined the Armenians in their unsuccessful revolt.30Vardan joined the Byzantine army; the rest seem to have settled in Pergamum, where an Armenian colony is known to have existed in the seventh century. From this colony sprang Bardanes who, under the name of Philippicus, occupied the imperial throne from 711 to 7 1 3 . ~More ~ Armenians immigrated after Armenia had fallen into the hands of the Arabs. Thus, about 700 a number of Nakharars with their retinues sought refuge in the Byzantine Empire, and were settled by the emperor on the Pontic frontier. Some of these later returned to Armenia but others remained.32 More Nakharars, completely abandoning their possessions in Armenia, fled to the Byzantine Empire during the reign . ~ ~ more came about 790. I t is said they of Constantine V C o p r ~ n y m u s Still numbered 12,000, and they came with their wives, their children, their retinues and their cavalry. They were welcomed by the Emperor and were granted fertile lands on which to settle.%We are not told the location of the lands given to them. This Armenian immigration to the Byzantine Empire was to continue in the centuries to come.35 The Armenians, however, did not always come willingly. They were sometimes forcibly removed from their homes and settled in other regions of the Empire. ~us-tinian had already resorted to this measure, but the numbers involved were small, perhaps a few familie~.~6 Transplantations on a large John of Ephesus, trans. Smith, 125-126, Brooks, 61-62; Theophanes, I:245. Ibid., Smith, 403, Brooks, 231-232; Theophanes of Byzantium, Fragments (Bonn, 1829), 485. 31 H. Gelzer, "Pergamon unter Byzantinern und Osmanen," Abhandlungen de?. Koniglich preussischen Akademie dev TYissenschaften (Berlin, 1903)~42f. Another Armenian colony may have existed a t Pidra in the Anatolikon theme. It is known that Leo V, the Armenian, had immigrated there as a boy. Theophanes Continuatus, 6 ; Genesius, 10, 28. The exact location of Pidra is not known. 32 Ghevond, Histoive des guerres et des conquites des Avabes en Arze'nie, tr. from Armenian by G. V. Chahnazarian (Paris, 1856), 22, 33-34; cf. J. Laurent, L'Arme'nie entre Byzance et Z'Islaz depuis , note 4; J. Muyldermans, L a domination arabe e n la conquite arabe jusqu'en 886 (Paris, I ~ I Q ) 184, 33 Ghevond, 129. Arme'nie. . . (Paris, 1927), 98-99. 34 Ibid., 162. I n reporting this incident Asoghik deplores the fact that, whereas the nobility was able t o flee, the poor had to stay and serve the Arabs. Stephen (Asoghik) of Taron, Histoire z~nivevselle ( ~ partie), e tr. from Armenian by E. Dulaurier (Paris, 1883) (Pzdblications de 1'Ecole des Langues Orientales Vivantes, XVZII), 162. 3= Mention should also be made of the Armenian Paulicians who were driven out of their homes some time before 661 and some of whom settled in the Pontic regions of the Empire, more specifically in the area a t the junction of the Iris and Lycus rivers. Their settlements extended almost as far as Nicopolis (Enderes) and Neocaesarea (Kiksar). These were regions where t h e Armenian element was already considerable. Comana, for instance, is referred to by Strabo (12, 3, 36) as a n emporium of the Armenians. Cf. GrCgoire, "PrCcisions gkographiques et chronologiques sur les Pauliciens," 294f., 298f.; S. Runciman, T h e Medieval Manichee (Cambridge, 1947), 34. Our source for the expulsion of the Paulicians from Armenia is a discourse of the Catholicus John of Odsun (717-728). This event is said t o have taken place during the Catholicate of Nerses who is apparently Nerses I11 (641-661). GrCgoire, op. cit., 300. The discourse of John of Odsun is also cited by Sirarpie der Nersessian, "Une apologie des images du septikme sikcle," B yzantion, I 7 (1944-45), 70-71. 36 Procopius, De bello Golhico, I11 32, 7 ; cf. R. Grousset, Histoire de l ' d r z e ' n i e des origines d 1071 (Paris, 1947))242. Grousset's statement concerning vast transfers of Armenians t o Thrace by Justinian is not borne out by his references. 29

30

30

P E T E R CHARANIS

scale took place during the reigns of Tiberius and Maurice. In 578 ~ o , o o o Armenians were removed from their homes and settled on the island of C y p r ~ s . ~ ' "Thus," says Evagrius, "land, which previously had not been tilled, was everywhere restored to cultivation. Numerous armies also were raised from among them, and they fought resolutely and courageously against the other nations. At the same time every household was completely furnished with domestics, because of the easy rate at which slaves were procured."38 A transplantation on a vaster scale was planned by Maurice, and partially carried out. Maurice, who may have been of Armenian descent, though this is extremely doubtful,39 found the Armenians extremely troublesome in their own homeland. The plan which he conceived called for the cooperation of the Persian king in the removal from their homes of all Armenian chieftains and their followers. According to Sebeos, Maurice addressed the Persian king as follows: The Armenians are "a knavish and indocile nation. They are found between us and they are a source of trouble. I am going to collect mine and send them to Thrace; send yours to the East. If they die there, it will be so many enemies that will die; if, on the contrary, they kill, it will be so many enemies that they will kill. As for us we shall live in peace. But if they remain in their country, there will never be any quiet for us." Sebeos further reports that the two rulers agreed to carry out this plan, but apparently the Persians failed to cooperate. For when the Byzantine Emperor gave the necessary orders and pressed hard for their execution, many Armenians fled to Persia.40 The ~ ~ z a n i i n ehowever, s, did carry out the deportation, though only in part. In ordering this removal, Maurice's real motive was, no doubt, the fact that he needed the Armenians as soldiers in Thrace. Further deportations and settlement of Armenians in the Byzantine Empire, especially in Thrace, are attested for the eighth century. During the reign of Constantine V Copronymus thousands of Armenians and monophysite Syrians were gathered by the Byzantine armies during their raids in the regions of Germanicea (Marash), Melitene, and Erzeroum, and were settled in T h r a ~ e . ~ l Others, also from the environs of Erzeroum, were settled along the eastern frontier. These, however, were subsequently seized by the ~ r a b sand were settled by them in Syria.42During the reign of Leo IV, a Byzantine raiding a7 Theophylactus Simocatta, Histoviae, ed. C. de Boor (Leipzig, 1887)) 143; John of Ephesus,

Smith, 412, 437, Brooks, 236, 252. Cf. Honigmann, op. cit., 23.

Evagrius Scholasticus, Ecclesiastical History, ed. J. Bidez and L. Parmentier (London, 1898))

215. The translation is taken from the English version of Evagrius which appeared in Bohn's Ecclesiastical Library: Theodoret and Evagrius, History of the Church (London, 1854))444. 39 N. Adontz has tried to prove the Armenian origin of Maurice: "Les legendes de Maurice et de Constantin V, empereurs de Byzance," Annuaire de l'lnstitut de Philologie et dlHistoire Orientales, 2 (Brussels, 1g34), 1-12. But see Goubert, op. cit., 36-41. 40 Sebeos, Histoire d'HLraclius, tr. from Armenian by F. Macler (Paris, 1904)) 30-31.Cf. F. Dijlger, Regesten der Kaiserurkunden des ostromischen Reiches, I (Munich, 1g24), p. 13, no. 108. 41 Kicephorus, Opzcscula Historica, ed. C. de Boor (Leipzig, 1880)) 65, 66; Theophanes, I:427, 429; Michael Syrus, Chronique, ed. and trans. J . B. Chabot, 2 (Paris, 1901)) 518, 521, 523; Agapius of Menbidj, Histoire universelle, tr. A. A. Vasiliev, Patrologia Orientalis, 8 (1912)) 544; Ghevond, op. cit., 126-127.

Agapius of Menbidj, 531, 538; Dionysius I of Tell-llahre, Chronique, tr. J. B. Chabot (Paris, 1895)) Cf. A. Lombard, Etudes d'histoire byzantine. Constantin V, empereur des Romains (740-775) (Paris, 1902))35. Among the people involved mere some Alans also. 42

56-57.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 31 expedition into Cilicia and Syria resulted in the seizure of thousands of natives, .~~ 15o,ooo, according to one authority, who were settled in T h r a ~ e These, however, were chiefly Syrian Jacobites, though some Armenians may have also been among them. Nicephorus I used Armenians, along with other nationalities, in his resettlement of Sparta at the beginning of the ninth century.@ Moreover, many individual Armenians are known to have come from Armenia and to have entered the service of the Byzantine state in both the seventh and eighth centuries. Occasionally, however, Armenians were driven out of the Byzantine Empire. We are told that the Emperor Philippicus, himself of Armenian descent, drove the Armenians out of his realm and forced them to settle in the regions of Melitene.45We do not know the reason for this; it may have been religious, but it had no consequence insofar as the role of the Armenians in the Byzantine Empire was concerned. That role was to grow in importance in the centuries to come. Thus, the Armenians were very much in evidence in the Byzantine Empire in the seventh and eighth centuries. They came of their own volition as refugees or were forcibly removed from their homes, and they settled in various parts of theEmpire, particularly inThrace and on the eastern frontier. Also, despite the conquest of Armenia by the Arabs, a conquest which was virtually complete by the end of the seventh century, some Armenian-speaking lands still remained in the possession of the Empire. However, it was as soldiers and officers of the army that the Armenians exerted their greatest influence in Byzantium. I t is well known that the Armenian element occupied a prominent place in the armies of Justinian. Armenian troops fought in Africa, Italy, and along the eastern front. They were also prominent in the palace guard.46Procopius mentions by name no less than seventeen Armenian commanders, including, of course, the great N a r ~ e s . ~But ' the Armenians constituted only one among the different elements that made up the armies of Justinian. These elements included many barbarians: Erulians, Gepids, Goths, Huns, Lombards, Moors, Sabiri, Slavs and Antae, Vandals; a number of Persians, Iberians, and Tzani; and among the provincials, Illyrians, Thracians, Isaurians, and Lycaonians.48 43 Theophanes, I :451-52 ; Ghevond, op. cit., I jo ; Michael Syrus, 3 :2. 44 P. Charanis, "The Chronicle of RIonemvasia and the Question of the Slavonic Settlements in Greece," Dumbarton Oaks Papers, j (1950), 154-155. In 792, following the suppression of a revolt among the Armenians, one thousand of them were removed from the Armeniac theme and were settled in Sicily and other islands: Theophanes, I: 469. 45 Theophanes, I :382 ; Michael Syrus, 2 :482 ; Agapius of Menbidj, 500. 4e Procopius, De bello Persico, I1 21, 2 ; De bello Vandalico, I1 24, 2 ; De bello Gothico, I1 27, 16; I11 6, 10; I11 26, 2 4 ; I11 27, 3, 10; Anecdota 24,16. 47 The names of these commanders are listed in the index of the Haury edition of Procopius under Armenians orpersarmenians. Some of these commanders, as for instance Gilacius, spoke only Armenian: De bello Gothico, I11 27, 24. I n the plot which led t o the assassination of Gontharis, the bodyguard of Solomon, the conspirators communicated between themselves in Armenian: De bello Vandalico, I1 28,16. 48 For Erulians: Procopius, De bello Persico, I 13, 20; I1 24, 13, 41; De bello Vandalico, I1 4, 28; I1 14, 12; De bello Gothico, I1 13, 18; I11 33, 13; I V 26, 13; Agathias, Historia (Bonn, 1828), 57, 79, 148, 184. For Gepids: De be110 Gothico, IV 26, 13. For Goths: De bello Gothico, I 16, 2 ; I11 I, 6 ; De bello Persico, I1 14, 10; I1 18, 24; I1 21, 4. For Huns, De bello Persico, I 13, 20; I 21, 11; I 12, 6 ; De bello Vandalico, I 11, 11-12; I1 I , 5-10; De bello Gothico, I 5, 4 ; I 27, 2 ; IV 26, 13; Agathias, 60-67. For Lombards, De bello Gothico, I11 39, 20; IV 26, 12; IV 33, 2, 3; Agathias, 184. For

P E T E R CHARANIS

Under the immediate successors of Justinian, the composition of the Byzantine army remained very much the same. "It is said," writes Evagrius, "that Tiberius raised an army of 150,ooo among the peoples that dwelt beyond the Alps around the Rhine and among those this side of the Alps, among the Massagetae and other Scythian nations, among those that dwelt in Paeonia and Mysia, and also among the Illyrians and Isaurians, and dispatched them against the Persians."49 The figure given by Evagrius may perhaps be questioned, but the rest of his statement cannot be doubted. It is confirmed by Theophanes, though the figure he gives is much smaller (15,0oo).~~ John of Ephesus reports that, following the breakdown of negotiations with Persia (575-577), a force of 60,000 Lombards was expected in the East.51 The same author states: "Necessity compelled Tiberius to enlist under his banners a barbarian people from the West called Goths. . .who were followers of the doctrine of the wicked Arius. They departed for Persia, leaving their wives and children at Con~tantinople."5~ I n Constantinople the wives of these Goths requested that a church be allocated to them, so that they might worship according to their Asian faith. Thus, it seems quite certain that the ethnic composition of the Byzantine army under Tiberius remained substantially the same as it had been during the reign of Justinian. The situation changed in the course of the reign of Maurice, chiefly as a result of the Avaro-Slavic incursions of the Balkan peninsula. These incursions virtually eliminated Illyricum as a source of recruits and reduced the potential of Thrace. They also cut communication with the west and made recruitment there most difficult.TheEmpire had to turn elsewhere for its troops. I t turned to the regions of the Caucasus and Armenia. In the armies of Maurice we still too,55but the Armenian find some Huns53and also some L ~ m b a r d sBulgars ;~~ element dominates. In this respect Sebeos is once more a precious source. I n connection with the war which Maurice undertook against the Avars after 591 Sebeos writes: Jhlaurice "ordered to gather together all the Armenian cavalry and all the noble Nakharars skilled in war and adroit in wielding the lance in combat. He ordered also a numerous army to be raised in Armenia, an &Ioors, De be110 Gothico, I 5 , 4 ; I11 I , 6 ; De bello Persico, I1 2 1 , 4 ; Agathias, 184. For Sabiri: De bello Gothico, I V 1 1 , 22-26; Agathias, 177. For Slavs and Antae: De be110 Gothico, I 27, 1-2; I1 15, 18, 2 2 ; I11 2 2 , 3 ; Agathias, 186. For V a n d a l s : De bello Vandalico, I1 14, 17; De bello Persico, I1 2 1 , 4. For Persians: De bello Gothico, I11 1 1 , 3 7 ; I V 26, 13. For Iberians: De bello Persico, I 1 2 , 11-13; I 2 2 , 1 6 ; I1 28, I ; De bello Gothico, I 5, 3 ; I11 4 , 10. For T z a n i : De be110 Gothico, I V 13, 1 0 ; Agathias, 109. For Illyrians: De be110 Gothico, I11 1 1 , 1 1 , 1 5 , 16, 111 1 2 , 4 ; 111 39, g ; 1V 26, 1 0 ; De be110 Persico, I1 2 1 , 4. For Thracians: De be110 Persico, I1 2 1 , 4 ; De be110 Gothico, I1 5 , I ; I1 1 1 , 5 ; I11 6 , 1 0 ; I11 1 2 , 4 ; I11 39, 9 ; I V 26, 10. For Isaurians: De be110 Persico, I 18, 5 , 38-40; De be110 Gothico, I 5, 1 2 ; I1 1 1 , 5 ; I11 36, 7 , 1 4 ; Agathias, 184. For Lycaonians, De be110 Persico, I 18, 38-40. Among t h e provincials, especially natives o f Asia Minor, there was a strong dislike for military service; S t . Basil wrote in one o f h i s letters: " . . .a large number o f persons are presenting themselves for t h e ministry through fear o f t h e conscription." Basil, Letters, ed. Deferrari, I : 3 4 4 (Letter 54). a9 Evagrius, 2 o g f . 50 Theophanes, I : 251. 5 1 J o h n o f Ephesus, S m i t h , 407, Brooks, 234. 5 V b i d . , S m i t h , 207, Brooks, 113. 53 Theophylactus Simocatta, 67. 54 Ibid., 104. 5 5 Michael Syrus, 2 : 72.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 33 army composed of soldiers of good will and good stature, organized in regular corps and armed. He ordered that this army should go to Thrace under the command of Musele (Moushegh) Mamiconian and there fight the enemy."56 This army was actually organized and fought in Thrace. Mamiconian was captured and killed,S7 whereupon the raising of an Armenian force of 2,000 armed cavalry was ordered. This force, too, was sent to Thrace.58 Earlier, during the Persian wars, important Armenian contingents under the command of John Mystacon operated on the eastern front.59 In 602 Maurice issued the following edict: "I need 30,000 cavalrymen, as tribute, raised in Armenia. Thirty thousand families must be gathered and settled in Thra~e."~O Priscus was sent to Armenia to carry out this edict, but before he had time to do so the revolution that overthrew Maurice broke out, and the edict apparently was not enforced. I t is interesting- to observe the correlation between the number of cavalry and the number of families that were to be transplanted to Thrace. Each family was obviously intended to furnish one cavalryman, and no doubt each family was going to be given some land. Here we have, perhaps, an indication that Maurice sought to extend the system of military estates in Thrace.60a But, however that may be, it is quite clear that under Maurice Armenia became the principal source of recruits for the Byzantine army.61 The same was true under Heraclius, himself of Armenian descent,62though that Emperor also drew heavily on the people of the Caucasus -Lazi, Abasgians, Iberians Sebeos, 35. Cf. Goubert, 09. cit., I : 197. 58 Sebeos, 36-37. Cf. Goubert, 09. cit., I:200; Dolger, op. cit., 12, no. 94. 59 Theophylactus Simocatta, 205, 216. 60 Sebeos, 54-55. Cf. Goubert, op. cit., I:209; Dolger, 09. cit., 16, no. 137. aoa The widely accepted view which associates the increase of military estates throughout the Empire with the establishment of the theme system and places both of these developments in the seventh century has very recently been questioned: J. Karayannopoulos, "Contribution au probleme des 'thhmes' byzantins," L'hellLnisme contemporain, ser. 2, 10 (1956), 492-501; Die Entstehung der byzantinischen Themenordnung (= Byzantinisches Archiv, Heft 10) (Rlunich, 1959)~71-88. See also Paul Lemerle, "Esquisse pour une histoire agraire de Byzance: Les sources et les problkmes," Revue historique, 220 (1958), 43-70. Karayannopoulos contends that the spread of military estates and the establishment of the theme system were not related, that both developed gradually over a long period of time, and that no one emperor was responsible for either. However this may be, it is very probable, as this passage from Sebeos suggests, that the growth of military estates was connected with the shifting of population from one province to another and the resettlement of immigrant peoples for military purposes. As both of these practices were frequently resorted to in the seventh and eighth centuries, i t is in those two centuries, but most probably in the seventh, that one should put the beginnings of the wide distribution of the military estates. This is not the place to discuss Lemerle's interpretation of the military estates. 6 1 Scholars have long recognized that the ascendency of the Armenian element in the Byzantine Empire dates from the reign of Maurice. M. K. Patkanian wrote in 1866: "A partir de cette 6poque [the reign of Maurice] les chefs des milices arm6niennes, en Thrace, commenckrent B. jouer un r61e important dans I'armCe grecque, parvinrent aux plus hauts grades militaires, et plusieurs d'entre eux monterent m&mesur le tr6ne des empereurs." M. K. Patkanian, "Essai d'une histoire de la dynastie des Sassanides, d'apr8s les renseignements fournis par les historiens armhiens," tr. from Russian by Evariste Prud'homme, Journal asiatique, 7 (1866), 194, note 3. Armenian troops under Armenian officers were also stationed in Byzantine Italy during this period. Cf. H. W. Haussig, "Anfange der Themenordnung," in I?. Altheim-R. Stiehl, Finanzgeschichte der Spatantike (Frankfurt a. M., 1g57), 106, note 76. 6z The father of the Emperor Heraclius, also named Heraclius, who served as general during the reign of Maurice, is said to have been a native of a city located in Armenia. Theophylactus Simocatta, 06. cit., 109-110. John of Nikiu calls the Emperor Heraclius a Cappadocian. Chronique, tr. H. Zotenberg (Paris, 1883), 431. 57

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 35 to 713. Artavasdos, son-in-law of Leo I11 and at one time general of the Armeniacs, also attempted to seize the crown, and for a time was actually master of Con~tantinople.7~ He was ably assisted by other Armenians, his During the cousin Teridates, Vahtan the patrician, and another Artava~dos.'~ brief period when he held Constantinople, he crowned his son Nicephorus, Co-Emperor and made his other son, Nicetas, general of the A r m e n i a ~ sThe .~~ Armeniacs, the vast majority of whom, as has been said, were Armenians, constituted Artavasdos' strongest supporter^.^^ Other eminent Armenians are known to have served the empire under Constantine V Copronymus. Tadjat Andzevatzik, who came to Byzantium about 750, proved to be a successful commander in the course of Constantine's Bulgarian campaigns. Under Leo IV we find him as general of the Bucellarii.78 He subsequently fled to the Arabs. Another Armenian, the prince Artavazd Mamiconian, who joined the Byzantine forces about 771, was general of the Anatolikon under Leo IV.79More Armenians are mentioned during the reign of Constantine VI and Irene. Rardas, onetime general of the Armeniacs, was involved in a conspiracy to have Leo IV succeeded by his brother Nicephorus and not by his son C o n ~ t a n t i n eAnother .~~ Vardas lost his life in the Bulgarian campaign which Constantine VI conducted in 7 9 ~ . Artasaras, ~l or Artashir, was another Armenian general active during the reign of Constantine VI.82 Alexius Musele (Moushegh), drungarius of the watch and later general of the Armeniacs, seems even to have aspired to the throne. At least he was accused of entertaining this ambition and was blinded.83 His family, however, achieved great distinction in the ninth and tenth centuries. Another great Byzantine family of Armenian descent, the Skleroi, made its Theophanes, I :386, 395, 414; Nicephorus, 59. Theophanes, I : 4 1 8 , 419, 420. 76 Ibid., 417. 7? Ibid., 418. 7 8 Ghevond, op. cit., I 50, I 53; Theophanes, I : 4 j 1 . Cf. Laurent, op. cit., 193, note 3. U n d e r Constantine V ,a Constantine, son o f t h e patrician Bardanes was put t o death i n 766 for conspiracy. Theophanes, I : 4 3 8 A Bardanes was general o f t h e Armeniacs i n 772. Theophanes, I : 445. A s t h e name Bardanes i s Armenian, these persons m a y have been Armenians. 79 Ghevond, op. cit., 134, I 50; c f . Laurent, op. cit., 193; Theophanes, I : 4 j r . T w o other Armenians, Varaz-Tirots, general o f t h e Armeniacs, and Gregory, son o f Mousoulak, general o f t h e Opsikion, served t h e Empire under Leo IV. Cf. N . Adontz, "L'2ge e t l'origine de l'empereur Basile I," Byzantion, g (19341, 242. so Theophanes, I :4 54. Bardas' Armenian origin is indicated b y his name. 8 1 Ibid., I : 468. I do n o t k n o w o n w h a t basis Adontz refers t o this Bardas as t h e father o f Leo V t h e Armenian. N . A d o n t z , " R o l e o f t h e Armenians i n Byzantine Science," The Armenian Review, 3 , no. 3 ( ~ g j o )64. , Under Constantine V I , Irene, and h'icephorus I , w e encounter a number o f persons w h o bore t h e Armenian n a m e Bardanes and w h o were probably Armenians: Bardanes, patrician and domesticus scholarum; Bardanes, general o f t h e Thracesians; Bardanes, called t h e T u r k , general o f t h e Anatolicon, w h o made a n a t t e m p t t o seize t h e throne; Bardanes, called Anemas, a spatharius. Theophanes, I : 471,474, 479-80,482. Another Armenian, t h e patrician Arsaber was Quaestor under Nicephorus I . I n t h e unsuccessful plot o f 808 t o overthrow Nicephorus, Arsaber had been chosen as t h e n e w emperor. Theophanes, I : 4 8 3 . C f . J . B . B u r y , A History of the Eastern Roman Empire from the Fall of , T h i s Arsaber was t h e father o f t h e Empress TheoIrene to the Accession of Basil I (London, I ~ I Z ) 14. dosia, w i f eo f Leo V t h e Armenian. Genesius, 2 I . Another Armenian named Bardas a relative (oiryyapppo~) o f Leo V, was general o f t h e Thracesians during t h e reign o f this Emperor. S . Theodori Studitae Vita, PG, 99. 300. C f . B u r y , Eastern Roman Empire, 68, 72. See further N . A d o n t z , " S u r l'origine de L6on V, empereur d e Byzance," Armeniaca, I1 (1927),9-10. 82 Theophanes, I :468, 469. 83 Ibid., I : 4 6 6 , 467,468; c f . Lebeau-St.Martin, Histoire du Bas-Empire, 12 (Paris, 1831))355, n o t e 3 . 74

75

3*

36

P E T E R CHARAN'IS

appearance in Byzantium at this time or soon thereafter. Leo Skleros, governor of the Peloponnesus at the beginning of the ninth century, is the first member of this family known to us.* I t will be noted that most of these Armenians were associated at one time or another with the Armeniac theme. The turbulent, but very energetic, thematic corps of the Armeniacs is very much in evidence throughout the seventh and eighth centuries. I t is the clearest indication of the prominence of the Armenian element in the Byzantine Empire during this period. I t should be pointed out, however, that in general the Armenians who entered the service of the Empire embraced orthodoxy and so identified themselves with the interests of the Empire. Yet there were always Armenians within its borders who sought to maintain their own traditions. The event which, as we have already observed, brought the Armenians into prominence was the collapse of Byzantine power in the Balkan peninsula and the consequent loss of the sources which in the earlier centuries had furnished the Empire with some of its best troops. No doubt the most important ethnic change in the Balkan peninsula since ancient times was brought about by the incursions and the settlement of the Slavs. The circumstances and exact chronology of the Slavic settlements in the Balkan peninsula are still, despite the meticulous work of many scholars, a historical puzzle. The reason for this is, of course, the brevity and chronological vagueness of our sources. This vagueness is best illustrated by the compilation known as the Miracula Sancti Demetri, the most important single text we possess on the settlement of the Slavs in the Balkan No less than three serious studies of this text have been made in the last five but they serve only to emphasize the difficulty of the problem since they offer different solutions to the crucial questions of chronology. The problem has been further confused by the nationalistic bias of certain scholars. The following facts, however, are sufficiently clear. The first appearance of the Slavs in the Byzantine Empire can be dated no earlier than the sixth century.86Throughout this century, beginning with the reign of Justinian, Slavs repeatedly invaded the Balkan possessions of the Byzantine Empire. Not until the reign of Maurice, however, did any Slavs settle in these territories. Between the years 579-587 there took place the irruption of several barbarian waves led by the Avars, but consisting mostly Charanis, "The Chronicle of Monemvasia ...," 145. P. Lemerle, "La composition et la chronologie des deux premiers livres des Miracula S. Demetrii," B Z , 46 (1953)~349-361 ; A. Burmov, "Les sieges de Thessalonique par les Slaves dans Miracula Sancti Demetrii Martyris et leur chronologie," Annuaire de I'Universite' de Sofia. Faculte' de Philosophie et Histoire. Livre I , histoire, 47 (1952) (in Bulgarian), 167-215; F. BariSi6, Miracles de St. Dbme'trius comme source historique (Academic Serbe de Sciences. Monographies CCXIX. Institut d'Etudes Byzantines, 2) (Belgrade, 1953). As I read neither Bulgarian nor Serbian, I have relied principally on the French resume which both Burmov and BariSi6 give of their respective works. I have, however, with the help of my friend George Soulis, consulted certain sections of the Bulgarian and Serbian texts. Cf. also Ep. Chrysanthopoulos, "TaB~pAiaOocvphwv~oB'Ayiou Aqpq-rpiou," O~ohoyia,24 (1g53),597-606; 25 (1954)) 145-152; 26 (1955), 91-106, 293-309, 457-464; 593-619; 27 (1956), 82-94; 260-272, 481-496. This work has now appeared under the same title in book form: Athens, 1958. I t is a serious study, but I find myself unable to agree with its main conclusions. s6 Cf. F. Dvornik, The Slavs. Their Early History and Civilization (Boston, 1956), 34ff. 84

85

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 37 of Slavs. The latter came in great numbers, and, as the troops of the Empire were engaged in the war with Persia, they roamed the country at will. They devastated Illyricum and Thrace, penetrated deep into Greece and the Peloponnesus, helped the Avars to take numerous cities, including Singidunum, Viminacium (Kostolac), Durostorum (Silistria), Marcianopolis, Anchialus, and Corinth, and in 586 laid siege to the city of Thessalonica, the first of a series of great sieges which that city was destined to undergo at their hands.87What is more, they came to stay. "The Slavonians," wrote John of Ephesus in 584, "still encamp and dwell in the Roman territories and live in peace there, free from anxiety and fear, and lead captives and slay and burn."8g The counteroffensive launched by Maurice after 591, following the successful termination of the Persian war, had the effect, on the whole, of checking the repeated incursions of the Avars, who then seem to have transferred their operations farther west beyond the limits of Byzantine territory. The treaty of peace which the Empire concluded with them in 601 (the date is not absolutely certain) fixed the Danube as the boundary line between the two powers, but left the way open for the Byzantines to cross that river and chastise any Slavs that might appear dangerous.89There is no indication, however, that the Slavs who had penetrated into the Empire were forced to retire beyond the Danube, or that they did so of their own accord. The settlement of the Slavs in the Balkan peninsila occurred mainly in the seventh century, more specifically during the disastrous reign of Phocas (602-610) and the early years of Heraclius. For the reign of Phocas there are no specific references in the sources to any Avaro-Slav invasions of Byzantine territory, but a general statement in Theophanes, apparently derived from Theophylact S i m o ~ a t t aleaves , ~ ~ no doubt, despite a recent attempt to minimize its s i g n i f i ~ a n c ethat , ~ ~ the Avars came repeatedly. For the reign of Heraclius our information is more explicit, though it leaves much to be desired, especially with regard to chronology. The Slavs had by now not only reached the Aegean, but also taken to the sea. "It happened. . . ," we read in the Miracztla Sancti Cf. H. GrBgoire, "L'origine et le nom des Croates et des Serbes," B yzantion, 18 (1944-1945), 88-1 18 ; P. Lemerle, "Invasions et migrations dans les Balkans depuis la fin de 1'6poque romaine jusqu'au VIIIe sikcle," Revue historique, 21 I (1954), 281 ff; L. Hauptmann, "Les rapports des Byzantins avec les Slaves e t les Avares pendant la seconde moiti6 du VIe sikcle," Byzantion, 4 (1927-z8), 137-170. The siege of Thessalonica took place on Sunday, 22 September, in the reign of Maurice: Miracula Sancti Demetrii, hligne, P G , 116. 1288. This must have been either in 586 or 597, for these are the only two years during the reign of Maurice when 22 September fell on a Sunday. Considering the position of the Avars in the year 597, it seems unlikely that they could have besieged Thessalonica in that year. The year 586 is, therefore, to be preferred: Charanis, "On the Capture of Corinth by the Onogurs and its Recapture by the Byzantines," Speculum, 27 (1g52), 347; BariSi6, op. cit., 60-64. Some scholars, however, have shown preference for the year 597. For a list of the scholars who have taken a position on this issue one way or another, BariSi6, op. cit., 10. To the list given by BariSi6 we may add Burmov (op. cit., 183-185) and Lemerle ("La composition et la chronologie des deux premiers livres des Miracula S. Demetrii," 354) both of whom adopt the year 597. John of Ephesus, trans. Smith, 432.

Hauptmann, op. cit., 16off.

Theophanes, I : 290; Theophylactus Simocatta, 308.

91 F. BariSi6, "De Avaro-Slavis in Phocae imperatoris aetate," Recueil des travaux de 1'Acad. Serbe des Sciences, X L I X : Institut d'Etudes Byzantines, 4 (Belgrade, 1956) (in Serbian with a Latin summary), 76-86. I have consulted the Serbian text with the help of Dr. hfiloS M. Velimirovi6.

38

P E T E R CHARANIS

Demetrii, that "during the bishopric of John of blessed memory, the nation of the Slavs, a countless multitude, was aroused. This multitude was drawn from the Drogubites, Sagudates, Velegezetes, Vajunetes, Berzetes, and others. Having first invented ships hewn from single pieces of timber, they took to the sea with their arms and pillaged all Thessaly and the islands about it and those about Hellas. They also pillaged the Cyclades, all Achaea, Epirus, and the greater part of Illyricum, and parts of Asia."92 The precise date of this event is not known, although BariSik is probably right in placing it toward the end of 614.93 A year or so later the same Slavs, under the leadership of a certain Hatzon, laid siege to Thessalonica. The city, however, withstood their assault, and they had to turn for help to the Khagan of the Avars. He came two years later, but to no avail. Meanwhile cities of the interior such as Naissus and ~ a r d i c ahad fallen to the barbarians. The narrative of this series of events leaves one with the definite impression that the Slavs who were involved in them had not come from afar, but were already settled in the Balkan peninsula, including the region of Thessalonica. Indeed, if we except the passages that deal with events of the sixth century, we find in the MiracuLa no distinct reference to invasions by Slavs coming from afar. The Slavs involved in the various attacks against Thessalonica were already settled in Macedonia. They had established themselves there in the period between the beginning of the reign of Maurice and the early years of the reign of Heraclius. An e~isodedescribed in the Miracula indicates that other invaders who were not Slavs settled in the region of Thessalonica later in the seventh century. This is the episode involving K0uver,~4a Bulgar whom the Khagan of the Avars had placed at the head of a mixed group under his domination. This group consisted of the descendants of Christian natives whom the Avars had carried away many years previously (about sixty years before, we are told) and the Avars, Bulgars, and other barbarians under the domination of the Khagan with whom these Christians had intermarried. These people dwelt in the region of Sirmium, maintained the traditions of their Christian ancestors, and were anxious to return to their old homes. Kouver, exploiting this desire, induced them to revolt and, after defeating the Avars who tried to check him, directed his followers toward Thessalonica, and then moved them in the direction of Monastir, where we lose sight of them. The date of this event is uncertain, but I am inclined to agree with those who place it toward the end of the reign of H e r a c l i ~ sThis . ~ ~ seems to fit in with what we know of the history of the Avars during this time. Their power in the Balkan peninsula was then in a state of decline, which had begun after their unsuccessful siege of Constantinople in 626.96An attempt has recently been made to identify the followers of Kouver I

92 Miracula S. Demetrii, 1325ff; A. Tougard, D e l'histoire profane duns les actes grecs des Bollandistes (Paris, 1874), I 18-126. O3 BariSib, Miracles de St. Dbme'trius comme source historique, 149. g4 Tougard, op.cit., 187-189. s5 For instance GrCgoire, "L'origine e t le nom des Croates e t des Serbes," I 10f f ; Dvornik, op. cit., 63, note 2. The retirement of the Avars from the Balkan peninsula t o regions farther north is associated by the Miracula S. Demetrii with the successful rebellion of Kouver: Tougard, op. cit., 189. 9a For the latest views on the Avar siege of Constantinople in 626 see BariSi6, "Le sikge de Constantinople par les Avares et les Slaves en 626," Byzantion, 24 (1954, published in 1956))371-395.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 39 with the Croats and Serbs, who also made their appearence at about this time and who contributed decisively to the disintegration of the Avar power in the Balkan peninsula.97This suggestion is tempting, but in view of the obscurity of our sources, which may not have preserved the various names involved in their original form, the identification must be considered doubtful. The Croats and Serbs, representing the last Slavonic wave to reach the Balkans, came with the consent of Heraclius and settled in the upper territory of the peninsula, the Croats in Dalmatia as far as the Sava, the Serbs in the region of the Urbas and the Morava, the ancient M a r g u ~ . ~ ~ The Bulgarg9and Avar invasions of the Balkan peninsula in the sixth and seventh centuries created a demographic crisis. The cities of the interior were plundered and destroyed, while vast stretches of the countryside were left desolate and empty of their inhabitants. Hundreds of thousands of natives, Illyrians, Thracians, and Greeks were deported; thousands of others were killed. Those deported were settled in the regions beyond the Danube, where, as we learn from the text concerning Kouver, they intermarried with the barbarians. Doubtless the vast majority of them were absorbed and lost their identity. Some, however, tried to preserve their traditions and, like the followers of Kouver, made an effort to return to the homes of their fathers. Others no doubt stayed behind. This may provide a clue to the solution of the riddle concerning the origin of the modern Rumanians. South of the Danube the virtual elimination of the native population facilitated the establishment of the Slavs. Their settlements covered the heart of the peninsula and extended to the Adriatic, the Aegean, and the Balkan mountains. They were numerous in the region of Thessalonica, a fact known not only from literary sources, but also from many place-names of Slavic origin.loOThrace, though often devastated by the Slavs, escaped their occupation, but even there they established some settlements, as, for instance, near Vizya.lol The native Illyrians and Thracians of the occupied regions retired into the mountains, where they remained unnoticed till the eleventh century, when they emerged as Albanians and Vlachs. The ethnic composition of the heart of the Balkan peninsula was thus transformed. The coming of the Bulgars into the region between the Danube and the Balkan mountains during the reign of Constantine IV,102 though of great political importance, had hardly any ethnic consequences, GrCgoire, "L'origine et le nom des Croates et des Serbes," I 16ff. But see above, p. I6ff. Constantine Prophyrogenitus, De administrando imfierio, ed. Gy. Moravcsik and R. J. H. Jenkins (Budapest, 1g4g), 122 ff. Cf. GrCgoire, ibid., 88 ff. B9 The depopulation of the Balkan peninsula began with the invasions of the Bulgars (Utigurs and Kotrigurs) during the reign of Justinian. Thousands of inhabitants were deported beyond the Danube. Some of them managed to return. I t is said, for instance, that as a result of the war between the Utigurs and the Kotrigurs, incited by Justinian about 550, "many tens of thousands of Romans," who had been previously captured by the Kotrigurs and transferred to the regions west of the Don, succeeded in escaping and returning to their native land. Procopius, De bello Gothico, IV 19, 1-2. We are also told that Justinian settled two thousand Kotrigurs with their wives and children in Thrace. Procopius, De bello Gothico, IV 19, 7. lo0 Max Vasmer, Die Slawen in Griechenland (Abhandlungen der preussischen Akademie der Wissenschaften. Philosophisch-historische Klasse, nr. 12) (Berlin, 1g41), 202ff. lol Tougard, op. cit., 156. 1°2 Theophanes, I :356-359. 97

98

40

P E T E R CHARANIS

except that perhaps the Bulgars left the imprint of their character upon the Slavs, by whom they were eventually absorbed. Slavs also settled in the Greek peninsula proper. This fact is quite evident, and no serious scholar has ever questioned it. What has been disputed is the precise date and the magnitude of the settlement. The sources, which are lacking in detail, give the impression that the country was flooded by the Slavs and that they overwhelmed every region.1°3 We know that the Velegezetes who took part in the piratical expedition of 614, an expedition to which we have already referred, settled in Thessaly.lo4I t is quite possible that the Vajunet.es, who took part in the same expedition, eventually moved to Epirus, a region which is known from other sources to have been invaded by the Slavs. We also know by name two tribes which eventually settled in southern Peloponnesus. The Slavs likewise penetrated into Attica and into Locris and, we may assume, also into Boeotia, although we are given no specific indication of their settling there in the seventh century. Further, we are told that western Peloponnesus was completely occupied by the Slavs. If we now turn to the place-names of Slavic origin, we find that, according to Vasmer, they are most numerous in Epirus and western Greece (558)) western and central Peloponnesus (387)) and in Thessaly, including Phthiotis (230). They are least numerous in Attica (IS), Argolis (IS), Boeotia (22)) Corinth (24)) and Phokis (45).lo5The Slavic origin of some of these names has been questioned,lo6and some reserve has been shown concerning the historical inferences that may be drawn from them,lo7but even if we makk due allowance for these observations, they remain nevertheless very significant. Indeed, they confirm what we know from the literary sources which, despite their fragmentary nature, clearly indicate that the regions of Greece most affected by the Slavic invasions were Thessaly, western Peloponnesus, and Epirus; those least affected were central Greece, including Attica, and eastern Peloponnesus. Slavs, then, not only settled in Greece, but did so in considerable numbers. Though the date of this settlement has been a subject of dispute, the evidence points to the period which extended from just before the beginning of the reign of Maurice to the early years of the reign of Heraclius. That more Slavs may have come later in no way alters this fundamental conclusion. The settlement 103 For a brief summary of the sources, Vasmer, op. cit., 11-19. See also Charanis, "The Chronicle of Monemvasia and the Question of the Slavonic Settlements in Greece," 141-166. The latest literature is discussed ibid., 164-166. The following works have appeared since: A. Bon, L e Pe'loponBse byzantin jusqu'en 1204 (Paris, 1951)~27-64; P. Charanis, "On the Slavic Settlement in the Peloponnesus," BZ, 46 (1953), 19-103; A. Maricq, "Note sur les Slaves dans le Pkloponbse," Byzantion, 22 (1g52), 337-348; Lemerle, "Invasions et migrations dans les Balkans ...," 305f. I n connection with the Slavonic settlements in Greece there has been considerable discussion concerning the precise geographical meaning of the term "Hellas." On this problem see Charanis, "Hellas in the Greek Sources of the Sixth, Seventh, and Eighth Centuries," Late Classical and Mediaeval Studies in Honor of Albert Mathias Friend, Jr. (Princeton, 1955)) 161-176. Io4 Tougard, op. cit., 166, 176. lo5 Vasmer, op. cit., 20-76; 128-174; 85-110; 120-123; 126-127; 118-120; 123-125; I 13-118. lo6 D. Georgakas, "Beitrage zur Deutung als Slavisch erklarter Ortsnamen," BZ, 41 (1942)~ 351-381 ; ZAaPl~fihri6paun O T ~~ o - r r w v v y ~T~ o~ 'H-rr~ipov, S EIs Mviyqv Xpimov Zo0hq (1892-1956) (Athens, 1956), 149-161. Io7 D. Zakythinos, 01 ZhCrPol Bv 'EAhCrGl (Athens, 1g45), 72-82.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y BYZANTIUM 41 of Slavs in Greece does not, however, mean that the Greek population was completely obliterated. Despite the Slavic flood, the Greeks held their own in eastern Peloponnesus, in central Greece, including Attica (a region which is known to have been a theme as early as 695), and, of course, in the islands. A number of strongholds are known to have remained in the hands of the Byzantines. In the Peloponnesus there was Monemvasia in the south and Corinth in the north.108 In central Greece there was Athens, where, if we may believe a hagiographical text, a Cappadocian conversed with philosophers and rhetoricians in the eighth century;lOg And farther north there was Thessalonica. These strongholds, even Thessalonica, were not great urban establishments in the seventh century, nor for that matter in the eighth, but they were to serve as centers for the pacification, absorption, and eventual Hellenization of the Slavs in Greece. Thessalonica in particular may be called the savior of Greece from the Slavs, for had she succumbed to their repeated attacks in the sixth and seventh centuries, the chances are that Greece would have been completely inundated by them. In the end, the Slavs in Greece proper were absorbed and disappeared from history. Fallmerayer's statement that there is no real Hellenic blood in the veins of the modern Greeks cannot, therefore, be accepted. The Slavic penetration of Greece affected also the ethnography of Sicily and southern Italy. Scholars have noted that whereas about A.D. 600 Sicily "contained a considerable Latin element," by 650 it "had become completely Greek in language, rite, and culture."l1° The explanation for this, it was thought, lay in the influx of a considerable number of Greek-speaking elements from Syria and Egypt as a result first of the Persian and then of the Arab conquests. But for this, with the exception of one or two texts referring to a few individuals, there is no evidence. The evidence that exists is of a different nature.ll1 We 108 Monemvasia was founded by Lacedaemonian refugees at the time of the invasion of the Peloponnesus by the Slavs during the reign of Maurice. Charanis, "The Chronicle of Monemvasia. ..," 148. On Corinth and Athens during the seventh century see Charanis, "On the Capture of Corinth by the Onogurs and its Recapture by the Byzantines," 343-3 50; "The Significance of Coins as Evidence for the History of Athens and Corinth in the Seventh and Eighth Centuries," Historia, 4 (1g55), 163-172. 109 Life of St. Stephen of Surozh, ed. V. Vasil'evskij, Russko-vizantijskija izsledovanija, I1 (St. Petersburg, 1893), 75: BEVITEV~E~S 88 Tq5 rrmpi605 E i s Th5 'ABfiva5 Crri8papEveE ~ X Eyhp irrl&rpiav TOG TpOUT ~ Vvabv T ~ 9~0pfi~opos. S Eirpwv 68 BKE~UE B V ~ Q Y E V TOG E ~TS~ T O W ~ a rrmpiow~ i wvijuat ~ a ~maarrdtuad3ai i

cp~Aoab~ow~ TE ~ aPi-ropa~, l ~ d t v r a s-rrpouopth~uas~ a6t~hExeEi~ i o i r ~bAiya, iv Kwvmavrtvou~r6h~t irriu-rpcya.

L. White, Latin Monasticism in Norman Sicily (Cambridge, Mass., 1938), 17. See my paper "On the Question of the Hellenization of Sicily and Southern Italy during the Middle Ages," T h e American Historical Review, jz (1946), 74-77. But see further 0. Parlangeli, S u i dialetti romanzi e romaici del Salento (Memorie dell'Instituto Lombardo di Scienze e Lettere. Classe d i Lettere, Scienze Morali e Storiche, ser. 111, XXV-XXVI (Milan, 1953). 141f. For a contrary opinion see Stam. C. Caratzas, L'origine des dialectes neb-grecs de l'ltalie m6ridionale (Paris, 1958), 47-61. The arguments of Caratzas against the view expressed here, especially since he accepts the testimony of Arethas of Caesarea, leave me absolutely unconvinced. See also, S. G. Kapsomenos, 'H p a p ~ p i a ~ a?h i a , ll~rrpayyivaTOG 8' AIE~V. TOG kfjthoyiou y ~ h mjv ErriPi~utvTOG EAhf~vtupoGumjv ~ E U E ~ P ~'II V Bulavr~voh..Iwv~6piow, 3 (Athens, 1958)~299-324. Besides the lexical material, which constitutes the

basis of his work, Kapsomenos examines also the historical evidence, but his examination is very superficial. The question of the survival of Greek in southern Italy is briefly touched upon by E. Pulgram, T h e Tongues of Italy. Prehistory and History (Cambridge, Mass., 1g58), 50. He is inclined to agree with those who claim the continuity of ancient Greek, but refers also (50, note 7) to B. Migliorini who, in his as yet unpublished work, T h e Italian Language, suggests a compromise: "the two factions are not really so far apart since even Rohlfs admits that the hold of Greek had become very tenuous by the time i t was in fact invigorated through fresh Byzantine immigration." Cf. my paper (75, 84) where I make this observation.

42

P E T E R CHARANIS

know that at the time of the great Avaro-Slav invasion of the Peloponnesus during the reign of Maurice many Peloponnesians fled and sought refuge elsewhere. We are specifically told that among these Peloponnesians, many Lacedaemonians settled in Sicily, while the people of Patras found new homes in the territory of Rhegium in Calabria. And although documentation is lacking, it is not improbable that other Greeks, too, from Epirus, central Greece, and western Peloponnesus went to Sicily or Italy at that time. As the Slavs occupied virtually all the western part of the Peloponnesus, the Peloponnesians who succeeded in fleeing could find no nearer haven than Sicily or Italy. That Greekspeaking elements from Syria and Egypt may also have settled in Sicily and southern Italy cannot, of course, be ruled out, but such evidence as there is clearly indicates that the bulk of the settlers came from Greece, particularly from the Peloponnesus, during, and as a result of, the great Avar and Slav invasions of the late sixth century and perhaps later. I t is interesting, too, to observe that as a result of the changes which took place in the Balkan peninsula and in Italy during this time, the effective jurisdiction of the papacy was reduced to lands where the Greek-speaking element was very considerable. This fact explains the predominance of Greek-speaking orientals among the popes of the seventh and eighth centuries. I t is well known that of the thirteen popes who occupied the pontifical throne from 678 to 752 eleven were Greek-speaking. The ethnography of Asia Minor also was to some degree affected by the coming of the Slavs. In their various raids the Slavs touched upon Asia Minor,ll2 but there is no evidence that they settled there of their own volition. They were brought to Asia Minor by the Byzantine emperors for political and military reasons; political, because the emperors wanted to reduce the pressure that the Slavs were exerting in the Balkan peninsula, especially in the region around Thessalonica; military, because they wanted to enroll these Slavs in their armies. There are for the seventh century two references in our literary sources to the establishment of Slavic colonies in Asia Minor. The first tell us that in the course of an expedition which the Saracens made into "Romania" in 665, five thousand Slavs went over to them and were settled by them in Syria.113 "Romania" means Asia Minor in this context, and although we are not explicitly told that the Slavs in question were settled there, the chances are that they formed a military colony which had been established in those parts. The second reference is more explicit. We are told that in 688 Justinian I1 "made an expedition against Sclavinia and Bulgaria . . . and sallying forth as far as Thessalonica, seized many multitudes of Slavs, some by war, others with their consent.. . and settled them in the region of the Opsikion theme," i.e., in Bithynia.l14From among these Slavs Justinian raised an army of 30,000, which he led against the Arabs (A.D. 692). Twenty thousand of these Slavs-probably an Tougard, op. cit., I 18. Theophanes, I :348. 114 Ibid., 364; Nicephorus, op. cit., 36. Justinian's expedition to Thessalonica is also attested by an inscription that has been edited by A. A. Vasiliev, "An Edict of the Emperor Justinian 11, September 688," Speculum, 18 (1g43), 1-13. But cf. GrCgoire, "Un Cdit de l'empereur Justinien," Byzantion, 17 'I3

(1944-1945),

119-124a.

E T H N I C C H A N G E S I N S E V E N T H - C E N T U R Y B Y Z A N T I U M 43 exaggerated figure-deserted to the enemy, an act of betrayal which so angered Justinian that he killed the remaining ~ o , o o otogether with their wives and children. Formerly I expressed the view that Justinian had destroyed the entire Slavic colony in Bithynia,l15but a more attentive reading of the text, as A. Maricq has pointed out,ll6 does not bear out this conclusion. The Slavic colony in Bithynia not only survived,ll7 but was, in the following century, augmented by another great settlementlls and perhaps by others besides.llg In the beginning of the ninth century a Slav of Asia Minor very nearly ascended the throne; the view, however, that his uprising was an expression of Slav nationalism is a figment of the imagination.120The Slavs of Bithynia still existed in the tenth century,121though they were eventually absorbed and lost their identity. But let us return to the Balkan peninsula. The settlement of the Slavs in that area virtually eliminated the Latin-speaking element from the Byzantine Empire. The Latinized Illyrians and Thracians were killed or deported, or else retired into the mountains, where they lived unnoticed for centuries. I t is true that the Empire still clung to Ravenna, Rome, and Naples, had a foothold in southern Italy, controlled all of Sicily, and did not lose Carthage until the very end of the seventh century. Here the Latin-speaking element was dominant, although in Sicily and southern Italy Greek had begun to gain the upper hand. But these were peripheral regions which did not play a significant role, in spite of the importance that the Byzantine emperors attached to retaining them. I t had been otherwise with Illyricum and Thrace. Illyricum had been for a long time the best recruiting ground for the Byzantine army. Some of its ablest officers had come from there as well as from Thrace. The loss of Illyricum meant the elimination of the most important Latin-speaking element of the Empire. In the central regions of the Empire there was, thenceforth, no significant segment of the population that spoke Latin, and Latin had to surrender its position as the language of the administration and of the army. Under Heraclius Greek became the official language of the state. Latin ceased to be studied and was eventually f0rg0tten.l~~ An emperor of the ninth century refered to it as a "barbarous Scythian language."l* In the meantime developments in the west gave to the papacy a western orientation, and so there evolved the conditions which were to bring about the separation of the Latin and the Greek worlds. 115 Charanis, "The Slavic Element in Byzantine Asia Minor in the Thirteenth Century," Byzantion, 18 (1946-1g48), 74. According to Michael Syrus (2: 470) the number of Slavs who deserted to the Arabs numbered only about seven thousand. Cf. Maricq, ofi. cit., 349. Maricq, loc. cit. "7 I am now inclined to agree with Ostrogorsky that the seal which refers to the Slavs in Bithynia dates from 694195 and not from 650 as I had formerly thought. Cf. G. Ostrogorsky, Geschichte des byzantinischen Staates (Munich, 1952)~107, note I. Nicephorus, 68f. ; Theophanes, I: 432. llD Theophanes Continuatus, 50. But cf. Charanis, op. cit., 73. 120 Charanis, 09. cit., 79-80. lZ1 Charanis, op. cit., 80-81. 122 On the status of Latin in the Eastern Roman Empire one may consult the important work of H. Zilliacus, Z u m Kampf dev Weltsprachen im ostromischen Reich (Helsinki, 1935). I t was only gradually, however, that Latin was eliminated as the language of the army. I t was still in use at the end of the seventh century. Cf. A. Pertusi, "La formation des thkmes byzantins," Berichte z u m X I . Internationalen Byzantinisten-Kongress (Munich, 1958), 2 5-26 and note 129. Michael I11 in a letter to Pope Nicholas I : Nicolai Pupae Epistolae et Decreta, Migne, P L , I 19.932.

44

P E T E R CHARANIS

Among the many scholars who have attempted to determine the causes of this estrangement only a few have given due weight to the occupation of the Balkan peninsula by the Slavs. In reality this was one of the most important causes.12* In his account of the revolt of Thomas the Slavonian, the historian Genesius, himself of Armenian descent,lZ5lists a variety of peoples from whom the army of the rebel had been drawn; Saracens, Indians, Egyptians, Assyrians, Medes, Abasgians, Zichs, Vandals, Getae, Alans, Chaldoi, and Armenians, as well as adherents of the heretical sects of the Paulicians and A t h i n g a n ~ iEven . ~ ~ ~if the identity of all these nations is not entirely clear,127 the mere enumeration of them illustrates vividly the multi-racial character of the Byzantine Empire. I speak here of the ninth century, but the same could be said of both the preceding and the following periods. Greeks, including the Hellenized natives of Asia Minor, Armenians, Slavs, peoples from the Caucasus, obscure tribes such as the Mardaites whom Justinian I1 removed from Lebanon and settled in the Empire (probably in the region of Attalia),l28 remnants of the Huns, Bulgars, and Turks-all these nationalities were represented in the population of the Empire. The Greeks no doubt predominated, but some of the others, as, for instance, the Armenians and the Slavs, were both important and numerous. But, despite the multi-racial nature of the Empire, two forces tended to give it unity. The first was orthodoxy; the other was a common language. Both were Greek, and to the extent that they were Greek theEmpire also was Greek. But in another sense the Empire was neither Greek nor Roman. I t was above all Christian, and in it, if we may use the words of St. Paul, there was "neither Jew nor Greek," but "all one in Christ Jes~s."l2~

Prof. F.Dvornik has repeatedly emphasized the importance of this factor. See his recent work, Slavs. Their Early History and Civilization, 44-45. Cf. also above, p. off. lZ5 C. de Boor, "Zu Genesios," BZ, 10 ( I ~ o I )62-65. , lZ8 Genesius, 33. 127 Vasiliev-GrBgoire, Byzance et les Arabes, 1:31, note 2. Cf. F. Hirsch, Byzantinische Studien (Leipzig, 1876), 131. lZ8 Theophanes, 1:363,364; Agapius of Menbidj, 497; Michael Syrus, 2 : 469. Cf. Honigmann, op. cit., 41; J. Morelli, Bibliotheca manuscripts graeca et latina, I (Bassano, I ~ O Z ) ,217. I n the ninth and tenth 124

The

centuries we find Mardaites also in the Peloponnesus, at Nicopolis in Epirus, and Cephalonia: Theophanes Continuatus, 304, 311; Constantine ~ o r ~ h y r o g e n i t u~e s , ceremoniis, I ( ~ o n n , - 1 8 2 ~665. ), Gal. I11 27, 28.

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: Ethnic Changes in the Byzantine Empire in the Seventh Century Peter Charanis Dumbarton Oaks Papers, Vol. 13. (1959), pp. 23-44. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C23%3AECITBE%3E2.0.CO%3B2-V

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 2

Corpus Inscriptionum Neo-Phrygiarum W. M. Calder The Journal of Hellenic Studies, Vol. 31. (1911), pp. 161-215. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281911%2931%3C161%3ACIN%3E2.0.CO%3B2-K 5

Exploration in Galatia cis Halym. II J. G. C. Anderson The Journal of Hellenic Studies, Vol. 19. (1899), pp. 280-318. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281899%2919%3C280%3AEIGCHI%3E2.0.CO%3B2-6 21

An Eastern Christian Sect: The Athinganoi: To the Memory of Prof. Andréas Michael Andréadès (1877-1935) Joshua Starr The Harvard Theological Review, Vol. 29, No. 2. (Apr., 1936), pp. 93-106. Stable URL: http://links.jstor.org/sici?sici=0017-8160%28193604%2929%3A2%3C93%3AAECSTA%3E2.0.CO%3B2-U 44

The Chronicle of Monemvasia and the Question of the Slavonic Settlements in Greece Peter Charanis Dumbarton Oaks Papers, Vol. 5. (1950), pp. 139+141-166. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281950%295%3C139%3ATCOMAT%3E2.0.CO%3B2-U

NOTE: The reference numbering from the original has been maintained in this citation list.

Byzantine Cities in the Early Middle Ages George Ostrogorsky Dumbarton Oaks Papers, Vol. 13. (1959), pp. 45-66. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C45%3ABCITEM%3E2.0.CO%3B2-%23 Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:35:45 2008

BYZANTINE C I T I E S

I N T H E EARLY MIDDLE AGES

This study is based on a paper delivered at the Symposium on "Byzantium in the Seventh Century" held at Dumbarton Oaks in May 1957. The author is most grateful to Mr. John Parker for his translation of this paper.

MONG the fundamental problems of Byzantine history it would be hard to name one that has been studied less than has that of the cities. This is not surprising in view of the extremely limited information about them to be found in the sources. Byzantine authors, with their attention steadily directed towards the capital and the imperial court, make comparatively few references to other towns, and when they do so it is, in the majority of cases, to mention them only in passing. The evidence provided by surviving official documents is likewise very sparse, and exists only for the later periods. Yet an understanding of the development and vicissitudes of urban life in Byzantium does, in many respects, provide a key to a better appreciation of the fortunes of the Empire itself; and this is especially true of the stormy period of the early Middle Ages. For Byzantium the transition from the ancient to the medieval epoch was marked by a series of tremendous shocks, both internal and external. Its agonies at the turn of the seventh century are to some extent the pendant of those that preceded the fall of the Western Empire in the second half of the fifth. I n the case of Byzantium, the established forms of imperial government were maintained, and the old political and cultural traditions lived on; none the less, the state went through the same profound crisis, the crisis of the entire Roman Empire, and overcame it only at the cost of abandoning a large part of its ancient heritage. So it was that during the seventh century the Byzantine Empire underwent far-reaching changes and emerged from these bearing substantially different traits. In the provinces new agrarian conditions came into being, and with them new social relationships. The old aristocracy of great landowners was, to a considerable extent, replaced by a growing class of small proprietors. I t was a time which saw the formation of a new economic and social order, a new administrative system and a new military organization; in a word, a time in which, after the collapse of the old system, the renovation of the Byzantine state was accomplished. I t is natural to ask why Byzantium, despite all the upheavals through which it passed, and all the changes it underwent, was able to preserve its political structure and to survive the Roman Empire in the West by a thousand years. The answer usually given is that its economic wealth was greater, its rule stronger and more durable. But this, far from explaining anything, only raises further questions. Why was the Byzantine state stronger ? What were the elements of its greater durability ? What were the sources of its more ample wealth ? Any examination of the elements of continuity in Byzantine history must soon lead to a consideration of the development of Byzantine towns and of their position in the transitional period of the early Middle Ages. I t will not, of course, be possible to present here any final solution of this complex and rarely studied problem, nor is it my intention to undertake a detailed and comprehensive investigation of Byzantine urban conditions during this epoch. My aim is a

A

GEORGE OSTROGORSKY

more limited one; that is, merely to indicate in what way and by what methods such an investigation might be carried out. I n calling attention to the insufficiency of the literature about the history of Byzantine cities, I do not have in mind the fairly numerous and often very valuable studies dealing with the so-called Book of the Prefect, since such works are devoted only to Constantinople, and, furthermore to Constantinople in the tenth century. A work of greater scope is a most instructive little book by Bratianu, the first serious attempt to trace the development of Byzantine cities through the centuries.1 In recent years a number of valuable investigations into the history of Byzantine towns have appeared in Soviet Russia, and fortunately some of them are devoted to the early Middle Agesla. Yet, in spite of the real merits of these works, the most important questions about the development of Byzantine towns still remain open. Such is the problem which here particularly concerns us: to what extent did the old cities survive the turbulent times of the early Middle Ages ? Learned opinion is sharply divided over this question. Some scholars believe that the late antique cities continued an uninterrupted existence into medieval , ~ was earlier times. This view is held by E.E. LipSic and M. J. S j u z j ~ m o vand advanced most emphatically by A. P. Rudakov, according to whom the Byzantine Empire, in the seventh century as in older times, was "an aggregate of Others, on the contrary, have concluded that there was a comcities (p~leis)."~ plete decline in the city life of Byzantium in the early Middle Ages. A. P. Kaidan,4 who takes this view, points out that the assumption of an uninterrupted development of ~ ~ z a n t i cities n e was based on evidence of intense urban life taken from sources either of the sixth century on the one hand, or of the tenth and following centuries on the other, whence it was deduced that a similar situation existed also in the early Middle Ages, about which our information is so scanty. In the absence of direct evidence about the conditions of Byzantine cities in the early Middle Ages, scholars have turned to numismatic material. I t is chiefly on this basis that Kaidan supports his conclusions about the decline of Byzantine cities from the seventh to the ninth century. With the help of the catalogues of the great coin collections, especially the one by Wroth of the British Museum collection, and the still more comprehensive catalogue of I. I. T ~ l s t o y , ~ Kaidan attempts first of all to establish certain data regarding the relative size of monetary issues in different periods of Byzantine history. He fully realizes G. I. Bratianu, PriuilBges et franchises municipales duns 1'Empire byzantin (Paris, 1936). la We may note two important papers on the Byzantine city which appeared after the present study had gone to press: E. Kirsten, "Die byzantinische Stadt," Berichte zum XI. Internationalen Byzantinisten-Kongress (Munich, 1958); F. Dolger, "Die friihbyzantinische und byzantinisch beeinflusste Stadt," Atti del 3O Congresso internazionale di studi sull' alto medioezfo (Spoleto, 1958)~p. I ff. E. E. Lip&;, "K voprosu o gorode v Vizantii VIII-IX vv," Viz. Vrem., 6 (1953), pp. 113-31; M. Y. Sjuzjumov, "Rol' gorodov-emporiev v istorii Vizantii," ibid., 8 (1956), pp. 26-41; "Problemy ikonoborteskogodviienija v Vizantii," UCenye zapiski Sverdlovskogo Gosudavstvennogo PedagogiCeskogo Instituta, 4 (1948), p. 58ff. A. P. Rudakov, OCerki vizantijskoj kul'tury Po dannym greteskoj agiografii (Moscow, 1g17), p. 71 ff. * A. P. Kaidan, "Vizantijskie goroda v VII-XI vekach," Sozfetskaja Archeologija, 21 (1g54), pp.

164-83.

W. Wroth, Catalogue of the Imperial Byzantine Coins in the British Museum, I and I1 (London,1908); I. I. Tolstoy, Vizantijskie monety, I-IX (St.Petersburg, 1912-14).

BYZANTINE CITIES

49

the fortuitous element in the composition of every coin collection, and he duly emphasizes this. Yet, with all necessary reservations in mind, it must be admitted that the evidence of the collections, which number several thousand coins, is nevertheless significant. On the basis of Wroth's catalogue, Kaidan gives the following statistic^:^ from the time of Anastasius I to that of Maurice (491-602) the catalogue has 1,349 coins, i.e. an average of 12.3 coins per year; from Phocas to Constantine IV (602-685) 1,134, or 13.7 per year; from Justinian I1 to Michael I1 (685-829) however, there are only 423 coins, i.e. 2.9 coins per year ; from Theophilus to Nicephorus I1 (829-969) there are 226, or 1.6 per year; from John Tzimiskes to Nicephorus I11 (969-1081) 283 coins or 2.5 per year, and from Alexius I Comnenus to Alexius I11 Angelus (1081-1195) 349 coins, that is, an annual average of 3.0. Thus, from the end of the seventh century (and even, in fact, from the middle of that century) there appears to have been a drastic reduction in coin issues. Fundamentally the same picture is given by Tolstoy's catalogue, which is fuller but covers a shorter period. Kaidan then goes on to discuss coin finds made in the course of excavations. As regards the cities of Greece, especially Athens and Corinth, such numismatic material has been adequately published and investigated.' I t shows in general a drastic diminution in the number of coins found dating from the middle of the seventh century onwards (this is especially clear at at hen^).^ The subsequent increase in their number begins in the middle of the ninth century at Corinth, and not earlier than the second half of the tenth century at Athens. These facts have already been the subject of a number of investigations, and have provided material for some c o n t r o ~ e r s y . ~ On the other hand, there have been very few publications or studies of coin finds made in Asia Minor. This is unfortunate, since such material, assuming the validity of numismatic evidence for our purpose, would have much greater importance for the history of Byzantine cities. The decline of the cities of Greece, even if we were to regard it as established by the above statistics, would not K a i d a n , op. cit., p. 166. Margaret T h o m p s o n , The Athenian Agora. Results of Excavations Conducted by the America?z School of Classical Studies at Athens, 11: Coins from the Roman through the Venetian Peviod (Princeton, 1 9 5 4 ) ; A. R. Bellinger, Catalogue of the Coins Found at Corinth, 1925 ( N e w Haven, 1930); J . I<. Finlay, "Corinth i n t h e Middle Ages," Speculum, 7 (1932),p. 494; Katharine M . Edwards, " R e p o r t o n t h e Coins in t h e Excavations a t Corinth During t h e Y e a r s 1930-35,'' Hesperia, 6 ( 1 9 3 7 )p.~ 241 f f . ;J . M . Harris, "Coins Found a t Corinth. Report o n t h e Coins Found i n t h e Excavations a t Corinth during t h e Y e a r s 1936-39," Hesperia, 10 ( 1 9 4 1 )p.~ 1 4 3 f f . ;c f . also A. Bon, Le Pe'ZoponnBse byzantin jusqu'en I204 (Paris, 1951), p. 5 3 ; P. Charanis, " T h e significance o f Coins as Evidence for t h e History o f Athens and Corinth i n t h e Seventh and E i g h t h Centuries," Historia, 4 ( I Q S ~ )pp. , 165-6. T h i s diminution is not evident f r o m t h e table given b y M . T h o m p s o n , " S o m e Unpublished Bronze Money o f t h e Early E i g h t h Century," Hesperia, 9 ( 1 9 4 0 ) p. ~ 7 3 , n o t e 33, which w a s used b y Icaidan, op. cit., p. 170, as t h e coins are there grouped b y century, and o n l y t h e s u m total o f all seventh-century coins is given. I t can b e clearly seen, however, i n t h e excellent later publication o f t h e same scholar (see previous n o t e ) w h i c h shows t h a t a large number (232) o f t h e coins o f Heraclius w a s found in t h e Athenian Agora, and a still larger number (817)o f Constans 11's issues, b u t only v e r y f e w o f those o f Constantine IV (30) and a quite negligible number o f t h e coins o f his successors (op. cit., pp. 70-1). C f . B o n , 09. cit., and " L e problkme slave dans le PBloponnhse i la lumihre de l'archCologie," Byzanlion, 20 ( I Q ~ o ) p, . 1 3 f f . ;Charanis, op. cit., p. 1 6 3 f f . 4

GEORGE OSTROGORSKY

necessarily imply that all cities of the Empire suffered a similar fate. However, in the few cases from Asia Minor where we do have the necessary data at our disposal-as, for example, a t Pergamum, Sardis, and Priene-we obtain, in general, the same picture: a fairly large number of coins of the sixth and first half of the seventh century (to Constans I1 inclusive), and a very small number, or even a complete lack of coins in the second half of the seventh century and later. Only at Pergamum is there further evidence of a considerable increase at the time of the Comneni.lo On the basis of the coin finds at Pergamum, their investigator K. Regling lost no time in sketching the historical development of that city and asserted that the material published by him exactly reflected its changing fortunes.ll Essentially the same is done by Kaidan, admittedly on the basis of much more extensive material, but also drawing much wider inferences. I t would seem, however, that the conclusions of his interesting article, which at first sight appear so convincing, are based upon a serious misunderstanding. The crucial point is this: all the finds referred to above, both in Greece and Asia Minor, consist very largely, if not almost exclusively, of bronze coins. The marked decrease of their number, therefore, general as it was after Constans 11, means only that, beginning with the middle of the seventh century, there was a cut back in bronze coinage, and not necessarily a diminution of all coin issues.12 This becomes clear if we examine the collections of Byzantine coins with this reservation in mind. In compiling his statistics from the catalogues of Wroth and Tolstoy, Kaidan did not differentiate between the various kinds of coins, and therefore, did not notice that the decrease in the seventh and eighth centuries was limited to bronze, and that it did not affect the basic coinage, which was, of course, gold. If we make a count of the gold coins published in t h e same 10 Cf. K. Regling, "llliinzfunde aus Pergamon," reprinted from Blatter fur Munzfreunde, 1914 = Perpp. 355-63, 329-36; H. 'IV. Bell, Sardis. Pzcblications of the American Society for the Exgamon, I (1913)~ cavation o f Savdis, X I , Coins, I , 1910-14, (Leiden, 1916), p. 76ff. ; K. Regling, Die M u n z e n von Pviene (Berlin, 1927). See also S. RlcA. Jlosser, A Bibliography of Byzantine Coin Hoards (New York, 1935), where various chance finds in a number of other towns in Asia Minor are briefly recorded. The finds of Byzantine coins a t Antioch are also relevant to this subject. I t would be natural to suppose that the Arab conquest had a great influence in determining their chronological distribution, yet we are confronted with the striking fact that here, too, coins of Constans I 1 were found in fairly large numbers, while those of subsequent emperors down t o the tenth century were completely lacking. See Dorothy RI. Waage, Antioch on the Ovo+ztes,IV, pt. 2 : Greek, Roman, Byzantine and Cvusaders' Coins (Princeton, 1952). A. P. Icaidan, op. cit., p. 167f., included in his investigation coin finds made beyond the boundaries of the Empire, mainly relying for this purpose on the data given in Mosser's catalogue. On the basis of these data it is apparent that, both in the West and in the Eastern European countries, and in Scandinavia no less than in the Danubian regions, the number of Byzantine coin finds of the seventh century suddenly declines sharply everywhere, becomes still smaller in the eighth and ninth centuries, and increases once more in the tenth. But if we are to see in this an indication of the decline of Byzantine trade with other countries, or the adoption of a system of barter, we should not seek for all the causes of such a change only in Byzantium itself; even more compelling causes are certainly to be found in the conditions of those other countries. l1 Regling, "RrLiinzfunde aus Pergamon," p. 329. l2 Kaidan, op. cit., p. 172, points out that the only exception to the presumed general scarcity of coins from the eighth and ninth centuries is the find made a t Lagbe in Pamphylia, where coins of precisely this period were discovered. But this seeming anomaly is easily explained: the hoard found here consisted of gold coins, which is indeed an unusual phenomenon. I t contained 102 coins of all theEmperors from Leo I11 t o Theophilus, those Emperors, that is, whose coins are so scarce in the ordinary bronze hoards. Cf. E. T. Newell, T h e Byzantine Hoard of Lagbe (New York, 1945).

51

BYZANTINE CITIES

catalogues, leaving aside both bronze and silver, we shall obtain very different results. Wroth's catalogue will then give the following figures for gold coins: From Anastasius I to Phocas (491-610) : 197, an average of 1.66 per year From Heraclius to Justinian I1 (610-711) : 380, an average of 3.76 per year From Philippicus to Irene (711-802) : 149, an average of 1.64 per year From Nicephorus I to Leo VI (802-912) : 95, an average of 0.86 per year From Alexander to Basil I1 (912-1025) : 45, an average of 0.39 per year From Constantine VIII to Nicephorus I11 (1025-1081) : 85, an average of 1.52 per year. Let us now compare these figures with Tolstoy's catalogue which includes the data of Sabatier and Wroth and also provides material from his own collection and from the Hermitage. Here we have: From From From From From

395 to 491 to 610 to 711 to 802 to

491 : 250 gold coins, an 610 : 241 gold coins, an 71 I : 595 gold coins, an 802: 213 gold coins, an 867: 91 gold coins, an

average of average of average of average of average of

2.60 per 1.87 per 5.89 per 2.34 per 1.40 per

year year year year year.

Thus, we observe from the above statistics that gold issues not only did not diminish in the seventh century, but on the contrary, increased significantly.13 I might not have insisted particularly on this fact and might have ascribed it to chance, which, of course, greatly influences the composition of every coin collection,l* had I not had the opportunity of becoming acquainted with the rich coin collection at Dumbarton Oaks, the principal part of which (the Hayford Pierce Collection) contains about half as many Byzantine coins as are to be found in the British Museum.15 I t is most instructive to find that this Collection, when compared with the data in Wroth's and Tolstoy's catalogues, gives on the whole a similar picture of the coin issues in different epochs. Once more, it shows a strong falling-off in bronze coins after the reign of Constans 11, and at the same time not only no reduction, but even an increase of gold coins in the seventh century, when compared to the previous epoch. The figures are : l3 With regard to the number of gold coins, there is no difference to be observed between the first and second halves of the seventh century that would even remotely suggest a parallel t o the very marked diminution, noted above, in the number of copper coins after ConstansII. Wroth's catalogue lists 136 gold coins of Heraclius (a yearly average of 4.4), 103 of Constans I1 (= 3.81 per year), 47 of ConstantineIV (= 2.76 per year), 71 of the two reigns of Justinian I1 (= 4.44 per year), and 33 of Tiberius Apsimar (= 4.7 per year). Tolstoy's figures are: Heraclius, 180 (= 5.8 per year), Constans 11, 199 (= 7.37 per year), Constantine IV, 84 (= 4.94 per year), Justinian II,82 (= 5.125 per year), and Tiberius, 50 (= 7.14 per year). l4 I t is undoubtedly t o chance that we should ascribe the very small number of gold coins of the ninth, and particularly of the tenth, century in the British Museum Collection. 16 In the table given below I have analyzed only this basic part of the Dumbarton Oaks Collection. Later acquisitions and smaller individual collections a t Dumbarton Oaks are not included in the table. These would not, in fact, materially alter the general picture, since the composition of small collections is usually fortuitous and reflects the personal tastes and interests of the collectors.

4*

52

GEORGE OSTROGORSKY

From 491 to 610: 109 gold coins, an average of 0.99 per year From 610 to 711 : 212 gold coins, an average of 2.1 per year From 711 to 802: 65 gold coins, an average of 0.71 per year From 802 to 912: 96 gold coins, an average of 0.87 per year From 912 to I025 : 70 gold coins, an average of 0.62 per year From 1025 to 1081 : 152 gold coins, an average of 2.71. per year. The Dumbarton Oaks Collection does not show the striking reduction of gold coins in the ninth and especially in the tenth century, that is found in the catalogue of Wroth. This confirms the view, expressed above, that the apparent reduction suggested by the British Museum figures is a matter of chance. ]lie also find here, for the eleventh century, a greater increase of gold coins than that indicated by Wroth's material. But -and this is what most concerns us at present-the Dumbarton Oaks figures once more show a very marked increase of gold coins in the seventh century. This is a phenomenon that deserves the attention of numismatic scholars. We do not know why it was that from the middle of the sixth to the middle of the seventh century so much bronze was issued, whereas from the second half of the seventh and the following centuries so little has been preserved, or why, for that matter, the minting of silver coins was so restricted in Byzantium. Further numismatic investigations may some day provide the answers to these questions, which still remain open.16 Perhaps it will be shown, among other things, what influence reminting had in preserving the currency of given periods. In any case, numismatic material can have only a secondary value in dealing with the problem that confronts us now, namely, the development of urban life and urban economy. Such evidence, however, far from supporting the theory of the decline of the Byzantine cities and urban economy in the seventh century, only confirms the continued existence of a monetary economy in Byzantium, a fact which is sufficiently documented by other sources. We have, however, further evidence, also indirect, it is true, but much more reliable, which may shed some light on the existence of cities in early medieval Byzantium. First of all, there are lists of bishoprics existing at that time. I t is well known that ecclesiastical organization was founded on the administrative organization of the Empire, which in turn was based on the city, the 9oZis. Sees were naturally established in cities, and, as a rule, in every city of some importance. In Justinian's Codex, a law dating back to the time of Zeno i . . C-rrio~o-rrov proclaims, 176oav -rrbh~v.. . EXEIV i~ -rrav-rbs -rpo-rrou Crx6p1mov ~ a 'i61ov. 0 ~ o - r r i l o ~ ~ vThe . l ' episcopal list ascribed to St. Epiphanius always begins the enumeration of bishoprics in individual regions by indicating that such and such a metropolitan see had under it so many cities or episcopal sees la As one of the most distinguished of contemporary numismatists, A. R. Bellinger, has emphasized, the problem of the abundance or scarcity of the issues, "the basic question about imperial policy and the coinage" has so far received no answer. A. R. Bellinger, "The Coins and Byzantine Imperial Policy," Speculum, 31 (1956)) p. 81. l7 Cod. Just., I. 3.35.

BYZANTINE CITIES

fin'a6-rj v ~ 6 1 ~~T1O ~~ IT ~ B K O T T.I8 ~ SThe ) concept of a see thus coincided with that

of a city. This so called Ecthesis of St. Epiphanius, the most ancient of surviving Notitiae episcopatuuwz, belongs, according to Gelzer, to the time of Heraclius.19 Whether it was in fact drawn up under that Emperor or somewhat laterJ20 it does not, unfortunately, provide any data which might help us to determine what sees did in fact exist within the confines of the Empire in the seventh century. As Gelzer himself has pointed out, this Ecthesis does not introduce any appreciable changes into the order that had been established under Justinian.21 An even less reliable source is the Notitia episcopatuum that is next in seniority, usually called the Notitia of Leo 111.22 This is a compilation made from older lists, including, as if nothing had happened, provinces that had long been lost, and presenting, furthermore, instances of obvious negligence.23 In a word, this Notitia, like many other documents of the same kind, does not reflect the real facts, and it would be a mistake to draw from it any conclusions about the number of cities actually existing on imperial territory at that time.24 Quite different in character are the lists of bishops found in conciliar acts. Is H. Gelzer, "Ungedruckte und ungeniigend veroffentlichte Texte der Notitiae episcopatuum," Abh. der Buyer. Akad. der Wiss., 21 ( I ~ o I )p., 534f. Is Gelzer, op. cit., p. 545. 20 If it is true that the town of r o p 8 0 ~ 8 p p ain Bithynia, mentioned in this document (Gelzer, p. 538) owes its name t o the settlement of Serbs in that province, i t follows that this work can hardly have been composed as early as the time of Heraclius and must be referred t o the reign of one of his successors. The first reference of certain date t o this town is to be found in the acts of the Trullan Council of 692, one of the signatories of which was 'loi6wpos drva(~os i - r r i m o ~ o srop8ooippwv ~ i j sBievvijv i-rrapxias (Mansi, X I , 996B). 21 Gelzer, op. cit., p. 545. 2%d. C. de Boor, "Kachtrage zu den Notitiae Episcopatuzwn," Zeitschr. furKirchengeschichte, 12(1891), pp. 519-34. Cf. E. Gerland, Covpus Notitiavum Episcopatuum Ecclesiae Orientalis Graecae, I , I (Kadikoy, - r r aOIKOV~EVIKOO i ~ ~ m p i a p ~ ~ iaoi iv " T ~ ~ I S " 1931) ; G. Iconidares, Ai ~ ~ T ~ O T T ~~ ~a dEir Ip xS ~ ~ ~ ~ c s ~ oTOO alj-rijv, I. I , Texte u. Forschungen zur byz.-neugr. Philol., 13 (Athens, 1934). Unlike the Notitia of PseudoEpiphanius, this document includes the sees of Illyricum and must, consequently, have been composed some time after these had been placed under the Patriarchate of Constantinople. But this does not in any way prove that it dates from the reign of Leo 111: it merely shows that it cannot be earlier than his reign if, as is generally thought, the separation of Illyricum from Rome took place in 733. However, according t o V. Grumel, "L'annexion de 1' Illyricum oriental, de la Sicile et de la Calabre au patriarcat de Constantinople," Recherches de science veligieuse, 40 (1952), p. 191 ff., this separation took place under Constantine V. All of this, of course, does not exclude the possibility that this notitia originated a t a still later date. The view of N. Bees, "Beitrage zur kirchlichen Geographie Griechenlands," Oviens Chvist., X. S. 4 (1915), p. 238, that this notitia was composed "sicher vor dem Slaveneinbruche in Griechenland und zwar vor dem J. 723," rests on an obvious misunderstanding of the facts. I t is curious that 1LIgr. Sophronius (Eustratiades) who, apparently unaware of De Boor's Edition 1891, published the text of , 556f., 577f.), considered i t to be earlier than the this notitia a second time (NBa Ztbv, 26 [ I ~ z I ] pp. notitia of Pseudo-Epiphanius, and referred i t t o the period before the Arab conquest. 23 &lore than sixty years ago L. Duchesne ("Les anciens Cvkchiis de la Grkce," Me'langes d'avchkol. et d'hist., 15 [1895], p. 379 ff.) made some very penetrating observations about this notitia. Cf. also V . Laurent, Byzantion, 7 (1932)~p. 521. 24 AS, for instance, did A. P. Rudakov in his most useful book (op. cit., p. 73), where he estimated the number of towns in the Empire during the seventh century on the basis of the notitia of Pseudo-Epiphanius. See the justified criticism of A. P. Kaidan, op. cit., p. 165. I t should be added, however, that the significance of the Notitiae Episcopatuum as historial sources, and in particular that of the so called iconoclast notitia, was also greatly over-estimated by those scholars who contributed so much to the understanding of documents of this kind: i.e., H. Gelzer, "Die kirchliche Geographie Griechenlands vor dem Slaveneinbruche," Zeitschr. f u r wiss. Theol., 35 (1892), p. 419ff.; E. Gerland, op. cit.; G. Iconidares, op. cit., esp. p. 3 , not to mention Sophronius Eustratiades, op. cit.

54

GEORGE OSTROGORSKY

Here we have, not compilations from older sources, but official lists of persons who represented their city-bishoprics at councils held at precisely known dates. Often we have both the official list of participating bishops and a list of their signatures confirming the conciliar decisions. At no time, of course, did all the bishops living at a given time take part in a council, so that these lists cannot give us an altogether complete picture. Besides, such lists as have come down to us have not always been preserved in a satisfactory condition, and a great number of synodal acts, including those of the seventh century, have not yet been critically edited. In spite of all these shortcomings, the lists of bishops to be found in conciliar acts constitute a source of first-rate importance and, generally speaking, of great reliability. If they do not provide a complete inventory of all the sees, they do at least enumerate a large number of them, often the great majority of the dioceses and towns existing within the imperial borders at a given time. A comparison of the lists of bishops present at councils, with their actual signatures -in which the possibility of inadvertent omission is very small and which are, for this reason, especially important - makes a mutual check possible and is of great help in correcting chance errors. These conciliar lists have often been used by historians in other connections,25but have not hitherto been exploited as evidence for the history of Byzantine towns. For the study of Byzantine cities in the seventh century, the episcopal lists of the Sixth Oecumenical Council of 680 and those of the Quinisext or Trullan Council of 692 are of particular interest. To judge from our somewhat unsatisfactory editions, the acts of the former were signed by 174 bishops and those of the latter by 211, each of whom represented his own episcopal city.26 These are very sizable figures, which cannot in any way be made to square with the supposed disappearance of Byzantine cities in the seventh century. Furthermore, they do not represent anything like the total number of all the episcopal cities existing at that time. This is especially evident from the different number of signatories who were present at two councils chronologically so close to each other. In addition, not only are many signatures found in the acts of 692 that are absent from those of 680, but also many sees were represented at the earlier council and not at the later one. Theophanes, although his figure may not be quite correct, asserts that 289 bishops were present at the Sixth Oecumenical Council,27i.e., over IOO more than in the preserved lists. At any rate, the episcopal lists of 680 and 692, despite their incompleteness, do give us a valuable and wholly factual enumeration of episcopal cities existing on Byzantine territory at the end of the seventh century. The trustworthiness of these lists will become quite clear when we examine more closely the names of the cities mentioned therein and see how they differ from those mentioned in earlier documents of this kind. 25 Cf. for instance F. Dvornik, Les Slaves, Byzance et Rome a u I X e siicle (Paris, 1926)) where the author makes extensive use of conciliar episcopal lists and also of the various Notitiae Episcopatuurn as sources for the history of ecclesiastical organization in the Slavic regions of the Balkans. 26 Mansi, XI, 640-53 and 988-1005. Cf. Hefele-Leclercq, Histoire des conciles, 111, I ( I ~ o I )p. , 508, 575. 27 Theophanes, p. 360, 2.

BYZANTINE CITIES Unfortunately, the acts of the previous council, the Fifth Oecumenical Council of ~onstantinoplein 553, have come down to us in such a mutilated condition that they can be used for our purpose only to a limited extent.2s A more dependable criterion for purposes of comparison is provided by the acts of the Fourth Oecumenical Council at Chalcedon in 451, which enumerate 340 bish0ps.2~These acts may be supplemented in some cases by the episcopal lists of the Council of Ephesus in 431 which, though it was attended by a much smaller number of bishops, nevertheless affords evidence of a number of sees not found in the acts of C h a ~ e d o nand , ~ ~ also by the signatures on letters of individual bishops in answer to the encyclical of Emperor Leo I in 4 ~ 8 . ~ ~ Furthermore, we may use by way of comparison the enumeration of cities within the Empire in the well-known Synecdemus of Hierocles, probably composed in the early years of Justinian I,32though it must be borne in mind that not all the towns cited by Hierocles were episcopal sees, so that the differences between his list and the episcopal lists of theconciliar acts are not always significant .33 A comparison of the episcopal lists of 680 and 692 with earlier lists clearly shows the changes that had occurred in the intervening period. These changes cannot be attributed only to fortuitous circumstances, i.e. either to the absence of certain bishops from a given council or to omissions from the acts. The fundamental and really significant difference consists in the fact that the bishops of those regions which had been lost by the Empire in the first half of the seventh century, or in which Byzantine authority had, in fact, been suspended, either do not appear at all at the councils of 680 and 692, or appear in quite insignificant numbers. From the territories that had fallen under Arab domination only the three Oriental patriarchs themselves were represented, since, according to tradition, an oecumenical council was hardly possible without their participation. From the Balkan lands there was only a handful of metropolitans and bishops, largely from cities which, like Thessalonica and a few coastal towns of Thrace and Greece, had been able to survive the Slav invasion. At both councils the only city represented from the interior of the Balkans was Stobi3*I t is curious that the representatives of the most important Balkan sees 28 Mansi IX, 173f. Signatures of the bishops, pp. 389-96. 28 E. Schwartz, Acta conciliorum oecumenicorum, 11,vol. I-6.All the episcopal lists found in the acts of this council arecollected byischwartz in hiswork "Uber dieBischofslisten der Synoden vonchalkedon, Nicaea und Konstantinopel," Abh. der Buyer. A k a d . der Wiss., Phi1os.-hist. Abt., N . F. Heft 13 (1g37), p. 15ff. 30 E. Schwartz, op. cit., I, I , 2 , pp. 55-64 and I, I , 7, pp. 111-17 (lists of signatures). Cf. also I, I , 2. pp. 3-7 and I, I , 7. pp. 84-8 (lists of those present). 31 Mansi, VII, 523-622. 32 E.Honigmann, Le SynekdBmos dlHieroklBs et l'opuscule giographique de Georges de Chypre (Brussels, 1939).F or the date of composition see Honigmann's introduction, p. I ff. The list of George of Cyprus, composed about 600, is unfortunately of no assistance as it has come down to us in a very incomplete form and the sections most relevant to this enquiry have been lost. 33 The much-disputed question whether the Synecdelnzts of Hierocles is based upon ecclesiastical documents or upon secular and administrative sources is decisively answered by Honigmann, op. cit., p. 2 f . in favor of the second alternative. 34 Mansi, XI, 673C and 993B. St. P. Kyriakides, T h e Northern Ethnological Boundaries of Hellenism (Salonica, 1955), pp. 22, 26, maintains that, in addition to Stobi, Serdica and Scupi remained uninterruptedly in Greek hands through the seventh and eighth centuries. I n the case of Serdica he does not

GEORGE OSTROGORSKY

mentioned in the acts of 692, namely, the metropolitans of Thessalonica, Heracleia, and Corinth, were not actually present at the Council, and that a blank space was left for their signatures, as was done also for the signatures of representatives from Rome, Sardinia, and Ravenna.35 This might perhaps be explained as an intentional absence on the part of the metropolitans of Thessalonica and Corinth, who were at that time subordinate to the Roman church,36 but this explanation cannot apply to the case of the metropolitan of Heracleia. In addition, the fact of being subordinate to Rome did not hinder the metropolitan of Gortyna in Crete from attending the council of 692 and signing its decisions as the representative of "all the synod of the Holy Roman Chur~h."~' As far as can be ascertained, only twelve bishoprics from the whole Balkan peninsula were represented at the Council of 680: Thessalonica, Heracleia, Corinth, Selymbria, Mesembria, Stobi, Byzae, Sozopolis, Panion, Argos, Athens, Lacedaemon. In 692 there are again twelve signatures, including the three metropolitan sees, the representatives of which, as I have said above, were not actually present, but the list is not identical with that of 680: Thessalonica, Heracleia, Corinth, Selymbria, Mesembria, Stobi, Uzusa, Ainos, Philippi, Amphipolis, Edessa, D y r r h a c h i ~ mOf . ~ ~the eighteen Balkan provinces listed by Hierocles, only six (Europa, Rhodopa, Macedonia I , Macedonia 11, Hellas, Epirus nova) sent a few individual representatives, while the cities of the remaining twelve provinces (Thracia, Haemimontus, Moesia 11, Scythia, Thessalia, Epirus vetus, Dacia mediterranea, Dacia ripensis, Dardania, Praevalis, Moesia I , Pannonia) were not represented at all at either council. All those cities, offer any evidence in support of this statement. For Scupi he refers to the signature of John, Bishop of Nova Justinianopolis in the acts of 692, but does not indicate the grounds for identifying Scupi, the modern Skoplje, with this town. To these two places he also adds Castoria, "which we know was Greek and used during Irene's reign as a place of exile for conspirators." (op. cit., p. 26). This rests on the statement of Cedrenus, 11, 24, 10, that Irene, prompted by the counsels of the logothete Stauracius, exiled the Patrician Theodore Camulianus, together with a number of other dignitaries, Bv Ka'io~opia (i.e. i.12 quaestovio,) which Kyriakides simply altered to Iv Kamopiq. Theophanes refers to the same Camulianus more clearly and in greater detail (whereas Cedrenus rather unsuccessfully abbreviates their common source)-and from his narrative i t is clear that Camulianus was not sent to Castoria, but held under arrest in his own house (Theoph. 465,6; cf. also 464,23), and soon after that we find him occupying the position of sfrafegus of the Armeniac theme (ibid., 468, 24). The problem of Nova Justinianopolis, referred to above, deserves more serious study.It is remarkable that the bishop of this town should have signed the acts of 692 immediately after the patriarchs and before the representatives of all the most ancient and powerful metropolitan sees (see Mansi, XI, 989A). This unusual distinction, which was not to be repeated at any other council, suggests that the town in question enjoyed the special favor of JustinianII. I n any event, it was not the Bithynian Nova Justinianopolis or Nova Justiniana, founded by Justinian I (see below, p. 57, n. 43). The bishop of the latter town also signed the acts of 692, together with the other bishops from Bithynia, and is very characteristically styled i-rrimo-rro~T ~ A E W S via5 ' l o u m ~ v ~ a vGi j~~u ~ i p aTGV s B10uvGv i ~ a p x i a s(Mansi XI, p. gg6D). 35 Mansi, XI, 928BC and g8gAB. 36 At the council of 680 (Mansi, XI, 640B and 641A) they both figure as vicars of the Roman see. 3i Alansi, X I , 928C and 989B. 3s The bishops from the islands are not counted here. But i t is curious to see that a t the council of 680 three representatives from Crete were present, and a t that of 692, four. The church of Crete belonged to the diocese of Illyricum; its head, the Metropolitan of Gortyna, was present a t both councils. Inview of the extremely small number of representatives from Illyricum itself, this is striking and does not suggest that communication by sea was particularly difficult a t that time. I t was also probably by sea and not along the Via Egnatia, that the bishop of Dyrrhachium came to Constantinople in 692. He had not been present a t the council of 680.

BYZANTINE CITIES

57

as also the majority of cities of the six poorly represented provinces, must either have disappeared or have been cut off from Constantinople. When we compare next the towns of Asia Minor mentioned in the acts of 680 and 692, with those mentioned in the acts of preceding centuries or even in the very detailed list of Hierocles, we obtain a completely different picture. Here the majority of episcopal towns continued to exist and to send their representatives to Church synods. I t is for this very reason that the councils of 680 and 692, in spite of the almost total absence of bishops from the Balkans, were nevertheless fairly well attended. The overwhelming majority of those present came from Asia Minor. Even in the case of Asia Minor, of course, when the lists of the seventh century are compared with more ancient ones, they are not found to coincide completely, if for no other reason than that the representatives of all the bishoprics existing at a given time could not all have been present at any one council. However, the differences which we notice here are quite insignificant. In fact, when we compare the conciliar lists of the seventh century or even of later councils with the lists of the preceding centuries or with Hierocles', we are struck not by their differences, but by their similarity. The degree of this similarity becomes very apparent in the tables given by Ramsay in his fundamental work on the historical geography of Asia I shall quote only a few telling examples. In the province of Galatia I1 (Salutaris) out of the seven bishoprics represented at Chalcedon in 451,six appear at the councils of 680 and 692, i.e. all except for the insignificant town of Petenis~us;~O on the other hand, at both councils of the seventh century the bishop of Celanion was present, whereas this town was not represented at Chalcedon, but is mentioned by Hierocles.41 In Bithynia Hierocles lists sixteen towns (the list in the acts of Chalcedon is here quite incomplete) : yet fourteen of them appear at the councils of 680 and 692, so that only two are missing, whereas there appear four new bishoprics,42 which were presumably-some of them certainly -founded in the intervening period.43Finally, the most amazing instance of continuity is the following: in the province of Paphlagonia, both Hierocles and the Councils of the fourth, fifth, and sixth centuries mention fourteen towns, and we find the same fourteen towns in the episcopal lists of 680,692, and 787, and even in a number of Notitiae e ~ i s c o ~ a t u ~ r n . ~ ~ Thus, while the great majority of Balkan bishoprics known to us from early Byzantine sources are absent from the episcopal lists of 680 and 692, the bishop39 W. &I. Ramsay, T h e Historical Geography of A s i a M i n o r (London, 1890). Cf. the less exhaustive tables in A. H. M. Jones, Cities of the Eastern R o m a n Provinces (Oxford, 1937), pp. 51off. 40 On this town see Ruge in Pauly-\lTissow-a,Realencyclopadie, XIX (1938), p. 11z7f. 41 Ramsay, op. cit., p. 223. 42 Ramsay, op. cit., table facing p. 197. 43 Of these only Nova Justiniana-Xova Justinianopolis, Justinianopolis Mela, S o v a Justinianopolis Gordi-appears in 553. Three others-Cadosia, Linoe, Gordoserbon-are unheard of before 680. 44 Ramsay, op. cif., ad p. 197. I n some of his tables Ramsay does not provide data from the lists of 680 and 692, but conflates the lists of earlier councils with the lists of the Council of Nicaea in 787. This in no way diminishes the evidential value of his data for our purposes here. The strong resemblances between the lists of the fifth and eighth centuries for the provinces of Asia and Hellespontus are particularly notable: Cf. Ramsay, o f . cit., ad p. 104, and p. 152.

58

GEORGE OSTROGORSKY

rics of Asia Minor continue for the most part to be represented. This fact is particularly interesting, since even in early Byzantine times the towns of Asia Minor were much more numerous than those of the Balkans and were always represented in much greater numbers than the latter at the early Councils. In the seventh century, after the collapse of Byzantine power in the Balkans, this long-established distinction became particularly marked. Our investigation into the fate of Byzantine cities in the early Middle Ages has thus completely confirmed our general view of the Byzantine Empire in the seventh century, as regards both Asia Minor and the Balkan peninsula.45Asia Minor continued to be covered with a network of cities, as in earlier times. I t might, of course, be asked whether some of the bishops who attended Councils were not titular dignitaries, whose sees had in fact ceased to exist, and who lived on a pension either in the capital or in other dioceses. This question has already been raised -very justifiably -about the bishoprics of the Balkan peni n ~ u l a . 4We ~ may suspect, for example, that Bishop Andrew of Amphipolis, who was present at the Council of 69zj4' and perhaps the Metropolitans John and Margarites of Stobi, present at the Councils of 680 and 6 9 ~were ) ~such ~ titular bishops, since neither for that time nor for later periods do we have any positive evidence about the existence of their sees.49There can be no possible doubt, however, in connection with the cities of Asia Minor which, for the most part, are otherwise attested, and the representatives of which reappear at the Councils of the eighth and ninth century. If we go further, and compare the episcopal lists of the Councils of 680 and 692 with those of later councils, we shall see that by the time of the Seventh Oecumenical Council held in Nicaea in 787, the number of episcopal cities in the Byzantine Empire had increased considerably. 319 bishops are recorded as having been present at this assembly.50This imposing number not only emphasizes the fact that the attendance of bishops at the Councils of 680 and 692 was by no means complete, but also bears witness to the foundation of new sees. According to Theophanes 338 bishops attended the iconolast council of 7 ~ 4?Ve . ~do~not, of course, possess any lists of those who participated in this council, but taking into account the similar number of bishops at Nicaea in 787, the figure given by Theophanes does not seem unrealistic and he certainly had no reason to exaggerCf. G. Ostrogorsky, "The Byzantine Empire in the World of the Seventh Century," supra, pp. 3 ff. Cf. Fanula Papazoglu, "Eion-Amfipolj-Hrisopolj," Zbornik radova Viz. inst. Srpske akademije nauka, 2 (19531, P. 14. 47 Mansi, XI, gg3B. 48 Ibid., PP. 645A and gg3B. 48 Such suspicions are particularly well grounded with regard to Amphipolis, in place of which the town of Chrysopolis is later found. Cf. F. Papazoglu, loc. cit. In the case of Stobi, which was represented a t the councils of 680 and 692 by two different persons (from which i t follows that the second of them, Margarites, was appointed after the earlier council),it is not possible to be so categorical. Archaeological investigation has shown that the town's most flourishing period was from the fourth to the sixth century; there are no signs of building activity after this, but certain monuments of painting and sculpture are referred to a later date, though only tentatively, by the investigators of the site. Cf. E . Kitzinger, "A Survey of the Early Christian Town of Stobi," Dumbarton Oaks Papers, 3 (1946), pp. 81-161. 50 See their signatures in Mansi X I I I , pp. 380-97. 51 Theoph. 427, 30. 45

46

BYZANTINE CITIES

59

ate the numerical strength of this iconoclast gathering. In iconodule circles it was maintained that the Emperor Constantine V had founded new bishoprics so that, by appointing his own partisans to them, he would be able to increase the strength of the iconoclasts at the council. There is probably some truth in this accusation, but this was certainly not the only motive for founding new sees. In any case, they were not all established in mere villages, and their foundation presupposes the existence of a corresponding number of towns, small though these may have been. A study both of the acts of the Nicene Council of 787, and of those of the councils held in Constantinople during the ninth century-the anti-Photian assembly of 869 and more particularly the Photian Synod of 87g52-shows that new sees had also been founded in considerable numbers in the Balkan peninsula. Gradually consolidating its very shaky position in this area, the Byzantine government built new towns or revived old ones, wherever it succeeded in strengthening its position and, in consequence, it founded new sees. As early as the Council of 680 the bishops of the towns of P a n i ~ nMesembria54 ,~~ and Sozop 0 1 i s ~had ~ appeared, and present at the Council of 692 were those of Edessa,56 U ~ u s and a ~A ~ i n ~ s ~ ~ - p l a c ewhere, as far as can be seen, no bishoprics had pres viously existed. A whole series of new Balkan sees appears in the acts of 787, mainly situated in Thrace and, to a considerably lesser extent, in Greece.59 SqMansi, XVI, 189-95, and XVII, 373-7. Mansi, XI, 643B Hierocles (ed. Honigmann 632, 5) mentions ndrvlov among the towns of the province of Europe, but no bishops of such a see are t o be found a t the councils of the fifth or sixth centuries. Considering the proximity of this town t o the capital, it would seem that its omission cannot have been fortuitous. The town, indeed, was not represented in 692; but a t the council of 787 its bishop was once more in attendance. (Mansi, X I I I , 388C). 54 Mansi, X I , 653C This town is not mentioned by either Hierocles or in the acts of the earlier councils. I t figures, however, in the episcopal lists of not only 680, but 692, 787, and 879. 55 hlansi, X I , 645C and 653C. Peter, Bishop of Sozopolis, signed the decisions of this council, and, for some reason, did so twice. Subsequently, the bishops of Sozopolis took part in the councils of 787 and 879. Sozopolis does not figure in Hierocles or in the episcopal lists of the Council of Chalcedon. ~ ~ U K O TTO ~~KUTU S Aouih~ov But the acts of the Council of Ephesus in 431 were signed by 'ABavao~o~ ~ a Zw16-rroh1v i &yia$ . . E K K ~ T ) Q ~ U $ (E. Schwartz, Acta conc. oecunz., I , I, 2, p. 62, n. 160) who was probably bishop of both Develtus and Sozopolis; i t seems that a t this time the rights of the bishop of Develtus, who is also found in the acts of 451 (and later in those 787 and 879) also extended over Sozopolis, which was not yet an independent see. 56 Mansi, X I , 993B: 'Iui6wpo~ C ~ & X I U T O $ I-rrimo-rro$ 'EG~aoqvCiv -rr6h~w$.I t is obvious t h a t the Macedonian Edessa (Voden) is referred t o here, since the signature of its bishop is immediately preceded by those of the Bishops of Philippi, the Island of Lemnos and of Amphipolis, and is followed by t h a t of the Bishop of Stobi. Edessa is listed by Hierocles (Honigmann, 638, 8), but does not appear in the episcopal lists of the councils of the early Byzantine period. 57 hlansi, XI, g p B : r ~ h p y ~ oEA&x~mos s ET~IJKOITOS 001oiruq~~ i Ojp q~ Q vxhpas. The name of this town may not be given here correctly. I t s identification presents a difficult problem, unless we assume t h a t it was Ostudizus, which was the same as the Thracian Xicaea or Kice. (Cf. C. JireEelr, D i e Heerstrasse von Belgrad nach Constantinopel u n d die Balkanplisse [Prague, 18771, pp. 49 and 100). Xicaea-Yice is found in Hierocles and in the conciliar lists of the eighth and ninth centuries: see note 59 below. 58 George, Bishop ~ i Aivl~Qv j ~ 1~6hsw5Opq~QvxGpa5 (Mansi, X I , ggzE) is probably t o be identified as a bishop of Ainos. The town was known to Hierocles (Honigmann, p. 634, 5), but does not appear in the ancient lists of sees. I t s bishop took part in the council of 879. 59 A list of these towns follows here, but we cannot vouch for its completeness, since any indication that might facilitate the work of identification is almost never found in the episcopal lists of this council. For convenience the towns are grouped under the metropolitan sees t o which they are assigned in the Notitia episcopatuunz of the time of Leo VI. (H. Gelzer, Ungedruckte u n d ungeniigend veriigentlichte T e x t e der Notitiae episcopatuum, p. 549ff). Directly subject to the Patriarchate of Constantinople: 53

..

.

60

GEORGE OSTROGORSKY

Among them are towns which, apparently, are mentioned only in conciliar lists and notitiae efiiscopatuzwz,but there are also others whose names are well known in Byzantine history, such as Rhaedestus, Tzurullon, Apros, Bulgarophygon, Kallipolis, Monemvasia. Besides these towns, some of which continue to feature in the acts of the councils of the ninth century while others do not, a very considerable number of new Balkan dioceses was represented at the Photian Synod of 879, and a number of these had earlier appeared in the acts of the anti-Photian Synod of 869. Once again these towns were situated mainly in the East of the peninsula. Some of them are scarcely known from other sources ;other bore historic namesa61 Neapolis (Christopolis, ICavalla), Derkos (Delkos, cf. Oberhummer, Pauly-lVissowa, IV, zqq.;), Sicaea; under the Metropolis of Heraclea in the province of Europe: Theodoropolis, Rhaedestus, Chersonesus, Chariopolis, Chalcis, Pamphylos, Lizica, Metrae, Tzurullon, Kallipolis, Apros, Koila (these last two are not in Leo's Notitia); under the 3Ietropolis of Corinth: 3lonemvasia, Troezen, the Island of Aegina; under the hletropolis of Athens: Oreos (Histiaia, in Euboea), Porthmos (also in Euboea); under the Metropolis of Trajanopolis in the province of Rhodope: Anastasiopolis; under the 3Ietropolis of Adrianople in the province of I
BYZANTINE CITIES

61

A number of other important towns which, having featured in the conciliar acts of the fifth and sixth centuries, do not appear in those of the seventh and eighth are once more in evidence, such as Larissa (represented in 869)) Demetrias, Naupactus, Trajanopolis and Anchialus.62 I t is impossible not to observe how close is the agreement between the lists of the Council of 879 -and, to a lesser extent, those of 787 -and the list of metropolitan sees, archbishoprics and bishoprics in the Notitia of Leo VI's reign. Naturally, only a part of the sees enumerated in the notitia was represented at the council. But the coincidences are none the less remarkable. The Balkan provinces most remote from Constantinople are sparsely represented in the acts as is to be expected. The provinces near the capital, such as Europe and Haemimontus were much more in evidence at the Councils of 787 and 879, and that of Rhodope was fully represented at the latter council, all eight bishops mentioned in the ~zotitia,headed by their metropolitan, being present. This creates an impression that the lists in this notitia correspond fairly closely to the actual situation. This impression would be further strengthened if a similar comparison were to be made in the case of the provinces of Asia Minor. Thus-exercising all the caution required in dealing with this kind of document -it may be assumed that the lists of the Notitia of Leo VI reflect to a considerable degree the actual state of affairs. Therefore, it may be interesting to note that they enumerate no less than fifty-one metropolitan sees, forty-nine archbishoprics subject to Constantinople, 388 episcopal sees in Asia illinor and eighty-three in the Balkan peninsula, as well as twenty-nine in Calabria and Sicily, ten in Rhodes and five in Lesbos.63 The conclusions we have reached on the basis of conciliar acts are fully corroborated by our narrative sources, meagre though they are. Let us consider the principal one, which is the Chronicle of Theophanes. In describing the events of the seventh and eighth centuries, Theophanes mentions a considerable number of towns in Asia Minor, albeit they are only those that happen to concern his narrative.64On the other hand, the number of Balkan towns mentioned by the Metropolis of Nea Patras and, finally, Ochrida (which lay outside the imperial boundaries and is not mentioned in the hTotitia episcopatuum of Leo VI). Not one of these sees is mentioned in the conciliar episcopal lists of the preceding centuries, including those of Xicaea, 787. I t should be noted, nevertheless, that the bishops of Patras, Messene, and Chalcis (Euripos) had signed letters in reply to the encyclical of the Emperor Leo I in 485 (cf. Mansi, VII, 612BC). Of these towns only Pharsalos, Rlothone (= Methone) and Patras were known t o Hierocles (Honigmann, 642, 13; 647, 17; 648, 3 ) . Of course, such absences could be entirely fortuitous. Thus, for instance, there is the striking fact that, after the Council of Chalcedon in 451, the representatives of Philippolis were not present a t any council, including that of 879, when, as has been seen, three bishops subject to this metropolis attended. Gelzer, Ungedruckte und ungeniigend vev6fientlichte Texte dev Notitiae episcopatuum, pp. 550-559. 64 We subjoin a list of the more important towns of Asia Minor mentioned by Theophanes in his Chronicle after the beginning of the seventh century. Cf. De Boor's excellent index and also the very useful one to the German translation of a part of the Chronicle by L. Breyer, Bilderstreit u n d Avaberstuvrrz in B y z a n z . D a s 8. Jahvhundert (717-813) a u s der Weltchronik des Theophanes, B y z . Geschichtsschveibev, hrsg. E von Ivhnka, Bd. VI s.a. (1957). Abydos, Adramyttion, Akroinon, Amasia, Amastris, Amicla, Amorion, Ankyra, Antiocheia (Pisidiae), Ataleia, Caesarea, Chalcedon, Charsianon, Chrysopolis, Cyzicus, Dorylaion, Edessa, Ephesus, Germaniceia, Iconium, Martyropolis, Melitene, Mopsuestia, Myra, Xacoleia, Nicaea, Ficomedia, Pergamon, Perge, Prusa, Pylae, Samosata, Sardis,

GEORGE OSTROGORSKY

him for this period is rather limited.65 Many famous cities of the Balkan peninsula that figure in his narrative of the preceding period, such as Salona, Sirmium, Singidunum (Belgrade),Naissus (NiS), do not appear after the end of the sixth century. The towns of Asia Minor, on the other hand, cited by him for the early period continue, with a few fortuitous exceptions, to be mentioned later. Kaidan tries to support his view of the complete decline of Byzantine cities in the early Middle Ages by a reference to Ibn-Khordgdhbeh who, he claims, knows of only five cities in Asia Minor: Ephesus, Nicaea, Amorium, Ancyra and Samala (in addition to Nicomedia which lay in ruins).66 In fact, Ibn Khordgdhbeh refers to these cities by a word which his learned editor de Goeje translated as "ville," but he also gives for each theme he mentions the number of fortified places ("places fortifikes," "forteresses" according to de Goeje's translation), which in only a few rare cases he calls by name. According to his data, at the time of his visit to the Empire, i.e. in the forties of the ninth century, there were ten fortified towns in the theme of Thrace, as against three in the theme of Macedonia (the other themes of the Balkan peninsula are not described by Ibn-Khordgdhbeh), which he does not name, while in the themes of Asia Minor there were all together 128 cities and fortified places.67Here again we see the same contrast between the number of Balkan and Anatolian towns. In the light of what has been said above, we must not only reject the notion that there were only five cities in Asia Minor at the time of Ibn-Khordgdhbeh, but we must also suppose that their number was in fact greater than the sum total of fortified places mentioned by the Arab geographer. I t should be noted, too, that the old towns not only continued to exist, but were subsequently fortified and rebuilt, and, what is more important, that new towns were being founded. In the Chronicle of the Patriarch Nicephorus there is a statement that the Emperor Constantine V, around the year 755, '(began to build towns (-rrohiopa~a)in Thrace" for the Syrians and Armenians whom he had brought over from Melitene andTheodosiopolis, and whom he had "provided with every necessity and generously endowed."68 We may assume that such migrations, frequent in Byzantium between the seventh and ninth centuries, were in other cases, too, accompanied by the building of cities and forts. Such was the migration of Slav population to Asia Minor as well as that of people from Asia Minor to the deserted and ravaged regions of the Balkans. It is temptSebasteia, Sebastopolis, Sinope, Smyrna, Syllaion, Synada, Tarsus, Theodosiana, Theodosiopolis, Trapezus, Tyana. To these might be added a list of smaller localities and fortresses : Amnia, Andrasos, Apollonias, Artake, Atroa, Bryas, Darenon, Herakleokastron, Kamachon, Kelbianon, Kopidnadon, Krasos, Libos, Malagina, Malakopea, Masalaion, Melon, Rlodrina, Phoenix, Siderokastron, Sision, Sophon, Sykekastron, Taranton. 65 Apart from Constantinople itself, the following are the more important Balkan towns referred to by Theophanes after the beginning of the seventh century: Adrianople, Anchialus, Arcadiopolis, Athens, Barna, Beroea (Thrace), Develtus, Heraclea (Thrace), Marcianopolis, Mesembria, Pllonobasia (= Monemvasia), Odyssus, Philippi, Philippopolis, Selymbria, Serdica, Thessalonica, Tomis, Tzurullon, and the following lesser settlements and fortresses: Abrolebas, Lithosoria, Marcellae, Medeia, Meleona, Probaton, Versinicia. Kaidan, op. cit., p. 165.

67 Ibn-Khordtidbeh, Bibl. geogra+horztm arabicorum, ed. M . J . de Goeje, VI, pp. 77-80.

aa Niceph. 66, 11. Cf. Theoph. 429, 26.

BYZANTINE CITIES

63

ing to suppose that it was emigrants from the Anatolian Chrysopolis who built the town of Chrysopolis at the mouth of the Strymon that replaced destroyed Amphipolis.69 On the other hand, the migration of a Serbian group to AsiaMinor probably caused the foundation of the town Gordoserbon in Bithynia, which, as we have already noticed, is first mentioned in synodal acts of the end of the seventh ~ e n t u r y . 7The ~ renovation of old Thracian towns by the Empress Irene is directly mentioned in the Chronicle of Theophanes, which says that Irene, after visiting Thrace in 784, ordered the rebuilding ( o i ~ o t i o ~ ~ 8KE~E~uCXUCX) ~va1 of Beroea and later that of A n ~ h i a l u s The . ~ ~ same chronicler reports that the WellEmperor Nicephorus I rebuilt ( ~ K T I U E )Ancyra, Thebasa, and Andra~os.'~ known inscriptions, re-edited and studied by Grkgoire, speak of the fortification of Ancyra and Nicaea by Michael III.73Even if the building activity knownfrom such occasional and incomplete data was primarily determined by military considerations, it attests not to the decline but to the growth of towns. In particular, as far as the Balkan peninsula is concerned, it reflects the process of the gradual re-establishment of Byzantine authority in certain regions, and this is confirmed, as we have seen, by the appearance of new bishoprics in conciliar acts. We have, unfortunately, almost no information about the character and life of Byzantine cities in ourVperiod.The little that we do have concerns only cities like Constantinople, Thessalonica, and the most prominent centres of Asia Minor, about whose importance there is absolutely no question. Thus, for example, a stray remark in the Chronicle of Theophanes tells us that at the end of the eighth century the fair of Ephesus yielded IOO lbs. of gold in commercial taxes.74The sum is enormous, and possibly exaggerated. Or again, in the beginning of the ninth century Gregory the Decapolite saw in the harbor of Ephesus a multitude of ships (-rhqOoS -rrhoIyov) engaged in commerce.75Nicaea, according to the Continuator of Theophanes, was a city of ancient wealth and large population (-r6hlS &pxalo-rrhov-ros~ a -rr0hirav6~o~).~~ i There is no reason to doubt that many other cities, both the coastal ones and the more important centers of the hinterland, retained their importance, not only as military and administrative posts, but also as centers of trade and the crafts. Otherwise the continuance of a developed monetary economy in Byzantium would have been inconceivable. That the economy was a monetary one is confirmed, as we have seen, both by Cf. F. Papazoglu, op. cit., p. 22. Cf. St. StanojeviC, Vizantija i Srbi (1go6), p , 41; Istorija naroda Jugoslavije, I (1g53), p. 229. 71 Theoph. 417, 6-11. The Thracian Beroea (Stara Zagora) which is mentioned by Hierocles and the bishop of which was present a t Chalcedon in 451, does not feature in later conciliar acts and reappears only a t the anti-Photian Council of 869 (Mansi, XVII, 1948). I t is certainly the Thracian Town which is here referred to, for the signature of its Bishop is followed by those of the Bishops of Tzurullon and Heraclea. Anchialus, known to Hierocles and appearing in the Acts of the Fifth Oecumenical Council of 553, is found again only in the Acts of the Photian Council of 879 (Mansi, XVI, 376-77). SVe do not find the bishops of either of these towns a t the Council of 787. 7 2 Theoph. 481, 9. 73 H. GrCgoire, "Inscriptions historiques byzantines," Byzantion, 4 (1927/8), p. 437f. 74 Theoph. 469, 31: ~b ~ w p Q p ~ TOG ~ o v~ a v q y u p l o vB hispwv xpvoiov dv. Vie de Saint Grbgoire le Dbcapolite, ed. Dvornik, p. 53, 19. 7 6 Theoph. Cont. 464, 8. Cf. Kaidan, op. cit., p. 184ff., where he has assembled other evidence from narrative sources. 89

O'

64

GEORGE OSTROGORSKY

numismatic evidence, and by a multitude of data in Byzantine and non-Byzantine sources, of narrative as well as of juridical character. We learn from them that taxes, the salaries of officials, and even the wages of both city and village labourerswere paid in money. The Farmer's Law often speaks of the daily wageof village laborers, communal herdsmen, and guards,77and in one passage explicitly states that this daily wage consisted of 12 f01les.~~ The same wage is mentioned in papyri and in hagiographic documents.79Monetary fines appear both in the Farmer's Law and in the Eclog~.~O The ninth-century Arab geographer IbnKhordadhbeh indicates the cash pay of Byzantine soldiers and officers of different ranks.81 The highest received the enormous sum of 40 lbs. a year, a figure that appears likewise in a somewhat later official pay list of the strategi of the themes dating from the reign of Leo VI.82 The same Arab writer reports that a hearth-tax of 6 dirchem was levied in Byzantium on each household.83 This again is confirmed by the statement of a Byzantine chronicler that in the twenties of the ninth century each provincial taxpayer contributed z miliaresia by way of k~pnikon.~~ The developed state of the Byzantine monetary economy in the early Middle Ages appears most clearly in the well-known account by Theophanes of the Here we are told of various taxes financial measures of Nicephorus I (8oz-811).~~ and duties, among them a fee of z keratia that was charged for the issue of official acquittances; of a duty of z !nomismata on each imported slave; of the obligation of village communities to contribute a given sum (18 ~ / nomismata) z for the upkeep and equipment of indigent soldiers; of the considerable loans (12 lbs. of gold) which the Treasury granted at heavy interest to merchants and shipowners. Although these data refer to the early years of the ninth century, we may apply them without reservation to the preceding period as well, since they express nothing new, but merely mark the introduction of a stricter financial control and a more stringent exaction of government revenues within an existing framework. Theophanes' account of the loans that the Emperor Nicephorus gave to traders beyond the sea should be compared with the so-called Rhodian Sea Law (vbvos V ~ V T ~ K presumably ~ ~ ) , composed in the seventh or eighth century.@This juridical compilation of the early Middle Ages reflects a fairly intensive maritime trade. I t enumerates different types of merchandise, among them gold and silver. The value of merchandise as also the income of the captain shipowner (varj~hqpos),his assistants, and sailors are all expressed in monetary units. Ed. SV. Ashburner, "The Farmer's Law," Journal of Hellenic Studies, 30 ( I ~ I O ) , para. 22, 25, 34. 78 Ibid., para. 22. 79 Cf. G. Ostrogorsky, "Lohne und Preise in Byzanz," BZ, 32 (1932)~ p. 295ff. so Farmer's Law, para. 22, 62. Ecloga, XVII 29. Cf. Ostrogorsky, op. cit., 305 for monetary fines in the legal documents of the preceding and following periods. s1 Ibn KhordAdbeh, p. 84. s 2 De Caerim., 696, 13-14. s3 Ibn KhordAdbeh, lac. cit. Theoph. Cont. 54, 4-7. 85 Theoph. 486-7. W. Ashburner, The Rhodian Sea Law (Oxford, 1909). 77

33,

84

BYZANTINE CITIES

65

Trading operations are often said to be conducted on credit, and we are told about loans and interest. Here, too, cash fines are appointed for certain offen~es.~' All these data, which it would be easy to multiply, attest to the preponderant role of a monetary economy in early medieval Byzantium. A developed monetary economy implies in turn a developed urban life and a considerable number of towns. Granted, however, that the continuous existence of a great many cities is proven, it does not follow that the Byzantine town did not differ in many ways from the ancient polis. The ruralization of towns was a common phenomenon in the Middle Ages, and it also affected Byzantium. Even in the largest cities a part of the population was engaged in cultivation and viniculture. The Miracula S. Demetrii tell us that during one of the Avaro-Slavic attacks on Thessalonica many of the inhabitants were outside the city walls, having gone out to work in their fieldss8 I t is not true that in the early Middle Ages Byzantine towns were in complete decay and had nearly ceased to exist. No more is it true that they preserved intact the organization of the ancient polis, since a certain decline of city life had already started in late Roman times. The greatest authorities on ancient history see in this decline one of the most important aspects of the general decay of the Empire.89Perhaps in this, too, there is a certain degree of exaggeration, since the cities as such did not cease to exist; yet there is no doubt that even in the late-Roman period the old municipal organization began to decay. This process was not only continued, but intensified in early Byzantine times, until, with the introduction of the theme system, municipal government was altogether destroyed.g0The most important towns of the Byzantine provinces became the seats of the strategi, and centers of their military and civil organization. In becoming the nerve center of the Empire, the new military and bureaucratic machinery stifled the last remains of city autonomy. The results of this process may be observed in the peripheral regions of the Empire; at Cherson on the one side, in Dalmatia on the other, where we see how, with the introduction of the theme system, imperial power replaced local city rule.g0aI n the central regions of the Empire this change must have occurred even more rapidly and radically. Municipal organization had long been dead when its final abolition was formally decreed by Leo VI, whose well-known novella proFor a fuller treatment of this point see E. E. LipSic, "K voprosu o gorode v VizantiiVIII-IX vv.," the Sea Law in connection with this problem is rightly emphasized. 88 Miracula Sancti Demetrii, 11, 2. no. 170. Cf. M. Rostovtseff, The Social and Economic History of the Roman Empire (Oxford, 1926),p. 478ff. W. M. Ramsay (The Cities and Bishoprics of Phrygia, 2 vols. [Oxford, 1895-18971) has very clearly shown how certain ancient cities declined as the old Roman routes lost their predominant position: the case of Apamea is especially instructive in this connection. But even this pehenomenon, brought about by the transfer of the center of communications to Constantinople, goes back to the early Byzantine period. Furthermore, just as the older focal points declined in importance at this time, so, on the other hand, did the towns situated on the routes connected with the new capital become more prominent. Cf. Bratianu, op. cit., p. 83ff. 90a Cf. J. Ferluga, Vizantiska uprava u Dalmaciii (Belgrade, 1957) (with a French summary: L'administration byzantine en Dalmatie, p. 154-62). 87

Viz. Vrem., 6 (1g53),p. 117ff., where the importance of

GEORGE OSTROGORSKY

claims that henceforward all authority is vested in the emperor and his admin i s t r a t i ~ nThe . ~ ~ triumph of all-embracing centralism more than anything else explains why the subsequent development of Byzantine urban life was marked by a certain impotence, why it did not produce that class of traders and craftsmen who in the West were the forerunners of new times, and why, finally, Byzantine towns were left so far behind the rising power of the Italian cities. But that is another subject, and a very large one too. To come back to the early Middle Ages: by surviving the crisis that assailed the imperial domains as the ancient world gave place to the medieval, the Byzantine cities, and especially those of Asia Minor, assured the continued existence of a monetary economy and, in this way, safeguarded the proverbial wealth of the Empire. This, it would seem, was the chief element of stability that preserved the traditional framework of the state. At the same time, however, the provinces underwent a violent upheaval and agrarian relations were completely changed, thereby altering and renovating the whole social structure of Byzantium; and in this renovation was the source of new strength which determined the subsequent evolution of the Empire. 91

NOV.Leonis XLVI, ed. Noailles-Dain (Paris, 1944))pp. 183-5.

http://www.jstor.org

LINKED CITATIONS - Page 1 of 2 -

You have printed the following article: Byzantine Cities in the Early Middle Ages George Ostrogorsky Dumbarton Oaks Papers, Vol. 13. (1959), pp. 45-66. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C45%3ABCITEM%3E2.0.CO%3B2-%23

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 7

Report on the Coins Found in the Excavations at Corinth during the Years 1930-1935 Katharine M. Edwards Hesperia, Vol. 6, No. 2. (1937), pp. 241-256. Stable URL: http://links.jstor.org/sici?sici=0018-098X%281937%296%3A2%3C241%3AROTCFI%3E2.0.CO%3B2-9 7

Coins Found at Corinth Josephine M. Harris Hesperia, Vol. 10, No. 2. (Apr. - Jun., 1941), pp. 143-162. Stable URL: http://links.jstor.org/sici?sici=0018-098X%28194104%2F06%2910%3A2%3C143%3ACFAC%3E2.0.CO%3B2-C 8

Some Unpublished Bronze Money of the Early Eighth Century Margaret Thompson Hesperia, Vol. 9, No. 3, The American Excavations in the Athenian Agora: Eighteenth Report. (Jul. - Sep., 1940), pp. 358-380. Stable URL: http://links.jstor.org/sici?sici=0018-098X%28194007%2F09%299%3A3%3C358%3ASUBMOT%3E2.0.CO%3B2-G 49

A Survey of the Early Christian Town of Stobi Ernst Kitzinger Dumbarton Oaks Papers, Vol. 3. (1946), pp. 81-162. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281946%293%3C81%3AASOTEC%3E2.0.CO%3B2-4

NOTE: The reference numbering from the original has been maintained in this citation list.

http://www.jstor.org

LINKED CITATIONS - Page 2 of 2 -

77

The Farmer's Law Walter Ashburner The Journal of Hellenic Studies, Vol. 30. (1910), pp. 85-108. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281910%2930%3C85%3ATFL%3E2.0.CO%3B2-V 78

The Farmer's Law Walter Ashburner The Journal of Hellenic Studies, Vol. 30. (1910), pp. 85-108. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281910%2930%3C85%3ATFL%3E2.0.CO%3B2-V

NOTE: The reference numbering from the original has been maintained in this citation list.

The Role of Trade in the Economic Readjustment of Byzantium in the Seventh Century Robert S. Lopez Dumbarton Oaks Papers, Vol. 13. (1959), pp. 67-85. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C67%3ATROTIT%3E2.0.CO%3B2-8 Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:36:07 2008

T H E ROLE O F TRADE I N T H E

ECONOMIC READJUSTMENT O F

BYZANTIUM I N T H E SEVENTH

CENTURY

This study is in substance identical with a paper delivered at the Symposium on "Byzantium in the Seventh Century" held at Dumbarton Oaks in May 1957

'1IP

he peculiarity that first strikes an economic historian as he looks at early mediaeval Byzantium is the persistence of the features it inherited from the Greco-Roman world. Again and again, he comes across institutions, methods, and even physical objects that bear the mark of classical antiquity deeply etched below the incrustations of later times. As late as the tenth century, the coinage was virtually the same as in the age of Constantine, the legal rates of interest were hardly changed, and at great banquets the fish sauce was prepared with the very recipe ordered by the Digest to be concealed from barbarian scullions. What is more, the basic patterns that had been woven into the texture of the ancient Mediterranean society, while somewhat faded, remained clearly recognizable down to the tenth century and later. The state maintained its grip on such economic activities as seemed relevant to its own security and glory. The towns preserved both their leading functions as residences of aristocrats, bureaucrats, and prelates, and their social stratification under which the merchant was subordinate to the landowner. And, whereas in Western Europe the predominence of serfdom almost eliminated the gulf between free men and slaves, in the Byzantine Empire slavery continued to play a significant role, and freedom at the lowest level survived.l Yet a closer scrutiny reveals a great deal of flexibility and progress which had to be blended with conservatism, so that the Empire could survive. The fundamental, multisecular trends at work in other parts of the hemisphere from England to China affected the Byzantine economy as well. A headlong depression and deflation which lasted to the mid-eighth century was followed first by a period of uneasy stabilization until the mid-tenth century, and finally by a phase of rapid economic growth which carried over into the later middle ages. In addition there were innumerable storms and squalls of more restricted scope, but often of greater violence, such as wars, revolutions, and famines. The relations between central government and local units, agriculture and trade, masters and dependants underwent substantial modifications. New contracts, new ships, new wares, new routes came into being, and even what was preserved owed its survival to skillful adaptation. To be sure, the Byzantine Empire never became "mediaeval" to the same extent as did Western Europe-and this was a mixed blessing, for the advantage of a less profound decadence was eventually offset by the disadvantage of a less abundant and extensive recovery. Still, readjustment there was, often of such novelty and imagination that Byzantium was an inspiration to both Catholic Europe and Islam, and well deserved the praise of a chronicler who said: "The Empire, this old lady, looks like a young girl attired in gold and precious stones." -

1 A good basic bibliography on Byzantine finances and economic life, by P. Charanis, is appended t o the American edition of Ch. Diehl, Byzantium: Greatness and Decline (New Brunswick, N. J., 1g57), 341 ff. The outline of Byzantine economic development which opens the present paper has been drawn more fully in my lecture "Un borgne au royaume des aveugles: la position de Byzance dans I'Cconomie europkenne du haut moyen %ge,"Bulletin de 1'Association Marc Bloch de Toulouse, V-VI (1953-55), 25-31.

70

ROBERT S. L O P E Z

The seventh century, with which this paper is particularly concerned, was the crucial period, the time of decisions which set the Byzantine economy on the half-conservative, half-progressive course that was to be its distinctive characteristic. Most of these decisions are obscure to us, not only because at this critical juncture the chronic poverty of economic evidence increased to the point of destitution, but also because no broad planning was feasible while the continuous state of emergency required everyone to take whatever immediate action seemed at a given moment to be expedient and possible. And perhaps the most important changes were not the results of conscious decisions, but merely the outcome of the graiual adjustment by individuals and the masses to changing conditions. Whereas the reconstruction of the imperial guilds was very probably the act of authority of an individual prince, and whereas the reorganization of controlled foreign trade was the patchwork of many administrators, the new contracts and pacts of sea trade and communal agriculture were the ultimate product of everyday practice, the invention of hundreds of unknown seamen and peasants. it-is my purpose here to describe the fate of sailors, merchants, and artisans, as I see it in the dim light of seventh-century sources or through the distant glow of later sources reaching back into that century. I shall extract what information I can out of political chronicles, lives of saints, legal texts, seals, and other archaeological evidence, for at that time nobody wroteexpressly of sordid economic matters. I cannot begin, however, without mentioning briefly three general problems that conditioned all efforts for economic readjustment. First, there was the demographic crisis. This was not a new affliction, but it wasmade worse by the plague of 541,with its many recurrences down to 618, and was greatly aggravated by the Avar, Slav, Bulgarian, Germanic, and Persian invasions. Minor raids, famines, and monastic celibacy made the affliction chronic for the remainder of the seventh century and the first half of the eighth.2 Some of the provinces, especially in Asia Minor, suffered less than others, but on the whole the molecular structure of the Byzantine Empire became almost as loose as that of the barbarian West. I n the West extensive uninhabited territory separated the surviving clusters of agricultural villages, and trade gradually deserted the deteriorated roads. In the Byzantine Empire the effect was somewhat different, because nearly every of agricultural units included a city where the inhabitants continued their trade while growing most of their own food, and because most cities were on the sea coast. Hence, the disruption of land routes, documented by the virtual disappearence of the postal service and of road liturgies except on the main Anatolian highway, left greater opportunity and necessity for intercourse by sea and often by river.3 To be sure, the letters of 2 One does not detract from the gratitude owed to A. Andrkadks' pioneer works on Byzantine demography by noting that his estimates were far too high. A good deal of useful material is collected in John Teall's dissertation (see below, pp. 92f., ~ o o f . )on the grain trade in the Byzantine Empire. For Constantinople and the Balkans see also my report on "East and West in the Early Middle Ages: Economic Relations," Relazioni del X Congresso Internazionale d i Scienze Storiche, I11 (Florence, 1 9 j j ) , 119ff., and P. Lemerle, "Invasions et migrations dans les Balkans depuis la fin de 1'8poque romaine jusqu'au V I I I e si&cle," Revue historique, CCXI (19j4), 281ff.; additional data and references will be found in the papers of Ostrogorsky and Charanis in the present volume. Cf. L. BrBhier, Les institutions de l'empire byzantin (Paris, 194g), 324ff. ; R. S. Lopez, "The Evolution of Land Transport in the Middle Ages," Past and Present, I X (1956), 17ff.; Lemerle, 09.cit.., 273ff.

T R A D E IN SEVENTH-CENTURY BYZANTIUM

71

Gregory I already indicate adiminution in the frequency and size of long-distance navigation, but the reduced population did not call for more. The Miracles of Artemius -to quote only one of several seventh-century hagiographic sources speak of regular navigation between Constantinople and Rhodes, and of ships calling at other ports of the Aegean and Black Seas. Connections with Sicily and Sardinia were less satisfactory, but were never seriously threatened.4 The coming of the Arabs, of course, did not make things easier; but no-one any longer believes that it had catastrophic consequences for trade.5 To fulfill their task of ensuring communications by every available anchorage and channel, and to dodge or outrun the hostile ships that lurked along every route, Byzantine sailors had to improve their techniques. I t was probably at this period that a revolution in the steering of ships was completed; the lateen or triangular sail, which had been known, though infrequently used, for several centuries finally p r e d ~ m i n a t e dSwifter .~ and slimmer ships were also experimented with. The Life of St. John the Almsgiver mentions a new name, the 86p~c3vor gazelle. Was she the forerunner of the yahaia or "swordfish," first mentioned by Leo VI, and destined to become queen of the sea, suitable for either trade or war? We do not know, but all this nautical zoology suggests vessels emphasizing agility and speed at the expense of size, and this can only mean a departure from the traditional three-deckers and other behemoths of ancient fleets.' To what extent this was a direct answer to the small and fast Slavonic monoxyle and Arab dhow, and to what extent it depended on the fact that commercial cargoes were no longer very large, we cannot At any Cf. G. Mickwitz, "Der Verkehr auf dem westlichen Mittelmeer um 600," Festschrift A . Dopsch (Leipzig, 1938)~74ff.; Miracula Artemii, ed. Papadopoulos-Kerameus in Zapiski Istor.-Filol. Fak. Imper. S. Petersburgskago Universiteta, XCV (1909)~33, 39, 55-57; G. Zanetti, "I Longobardi e la Sardegna," Atti del I Congresso Internazionale di Studi Longobardi (Spoleto, 1951)~yjff. The above statement sums up the views that I expressed in my report at the Tenth International Congress of Historical Sciences, where they were not challenged. Apparently the controversy oni'Mohammed and Charlemagne," after providing excellent ground for discussion during the last twenty-five years-more than a hundred works are quoted in the rich but incomplete list of A. Riising, "The Fate of Henri Pirenne's Theses," Classica et Medievalia, XI11 (1952)~ 87ff.-has reached the point of diminishing returns. I t does not seem useful to dwell on this problem here. To the bibliography in L. White, "Technology and Invention in the Middle Ages," Speculum, XV (1940)~145 and notes 5-6, we now must add L. Casson, "The Sails of the Ancient Mariner," Archaeology, VII (1954)) z ~ q f f . notable , for its reproductions of Hellenistic bas-reliefs of the second or third century representing lateens or spritsails. These antedate by more than five hundred years the Byzantine miniatures of 886 which previously had been regarded as the earliest clearly dated repre. sentations; cf. H. H. Brindley, "Early Pictures of Lateen Sails," Mariner's Mirror, XI1 (1926)~12-13. Between these extremes there are a few literary references and other data that lead one to think of a gradual diffusion of the lateen in the period under consideration in this paper. Leontius, Vita S. Johannis Eleemosynarii, ed. Gelzer (Freiburg i. Br. and Leipzig. 1893), and cf. the translation of E. Dawes and N. H. Baynes, Three Byzantine Saints (Oxford, 1948)~chaps. 10, 13; Leo VI, Tactica, chap. 19; further bibliography concerning the galley in R. S. Lopez, "Les Influences orientales et 1'Cveil Bconomique de I'Occident," Journal of World History/Cahiers d'histoire mondiale, I (1954)~600 and note 14. One might object that the Life of St. John mentions a "gazelle" carrying no less than 20,000 bushels of grain, but even if the figure is accurate, it would not impair the suggestion of a change in proportions, i.e. a more elongated type. Galleys, too, carried heavy loads if they were long enough. Although there is no direct reference to an influence of foreign ships in the seventh century, the following quotation from Leo VI's Tactica deserves attention: "Thou shalt equip small and large vessels according to the character of the hostile peoples. For the barbarian Saracens and the so-called northern Scythians do not use the same sort of ships. The barbarians use larger and slower vessels and the Scythians smaller, lighter, and faster boats .. . ." A similar statement is found in Nicephorus Uranus, cf. A. A. Vasiliev, "The SecondRussian Attack onconstantinople," DumbartonOaks Papers, 6 (1g51), 183.

R O B E R T S. L O P E Z

rate, we can detect an indication of improved seamanship in the fact that the problems of shipwreck and jettison, though still prominent in hagiography, lost much of their prominence in legal sources. Jettison was the only aspect of maritime law discussed in the Digest, but in the Nomos Nautikos, compiled between the seventh and the ninth century, it yielded first place to ordinary business matter^.^ The second general problem, interwoven with the first, was the military crisis. The Empire lost Syria, Egypt, and large parts of Africa, Italy, and the Balkans. I t is true that shedding an entire province, liabilities and all, was not altogether damaging in an age of poor communications and shrunken states. Moreover, the loss of the Egyptian granary may have stimulated Thracian and Anatolian agriculture,just as the lossof Syrian textile workshopsmust have fostered thegrowth of the silk industry in Constantinople and Greece.lo Further, the loss of African olive oil accelerated the conversion to better lighting by wax candles and better cleansing by fat soap. Seventh-century sources have little to say on this subject, except for a mention of the "sacred" soap that Constans I1 was using when he was murdered in his "sacred" bath, but the contrast between the harassed guild of Neapolitan soapmakers in 599 and the exalted guild of soapmakers in tenthcentury Constantinople is striking.ll Though the use of wax for illumination is as old as that of oil, the profession of ceriolarius is mentioned only once in the Corpus Inscriptionurn; but by the time of Nicephorus I, a keroularios was the choicest victim of imperial extortions. A typical nouveau riche, he paid a thousand pounds for the unsought honor of sitting at the imperial table. By 930 we hear of an entire street of wax sellers burned down in a fire at Constantinople.12 If the core of the Empire did not suffer too much from amputations at the periphery, readjustment was hard for those maritime towns whose hinterland fell under barbarian rule. Thessalonica, Cherson, Carthage, and Naples were the most notable among such towns in the seventh century ;later Venice eclipsed all of them. If a city that had been thus hemmed in accepted its fate passively, it had no alternative than to depend on doles from other parts of the Empire, or to shrink to a size that could be fed by the harvest of its own citizens. In other words, it had to become what was hardly more than an agricultural village. For some cities, however, the handicap became a stimulus for befriending their foreign neighbors. They traded the salt produced in their own lands and the refined products of Byzantine workmanship for foodstuffs and raw materials of the barbarian hinterland. The Miracles of St. Demetrius and an inscription of Justinian I1 show the beginning of this process in seventh-century Thessalonica. Gradually the Cf. W. Ashburner, T h e Rhodian Sea-Law (Oxford, ~ g o g )p. , ccliiff. That this fifty year old work is still of fundamental importance indicates the admirable scholarship of its author, but it also shows that the Byzantine sea law has not been adequately reconsidered in recent times. lo On grain the most recent work is the thesis of John Teal1 published herewith; on silk see R. S. Lopez, "Silk Industry in the Byzantine Empire," Speculum, XX (1g45), I ff. l1 Gregorius I, Epist., IX, 5 ; Leo VI, Book o f the Prefect, X I I ; further references in A. Stockle, Spatronzische u n d byzantinische Ziinfte (Leipzig, I ~ I I )39ff. , l2 Corpus Inscviptionum Latinarum, 111, 2112 (and see also cerarius and ~ q p o ~ r h h inq Plautus, ~ Miles glor. 111, I , 102; Gloss. Gv. L a t . ) ; Theophanes, ed. De Boor, p. 487-488; Theoph. Cont., VI, 30; Leo VI, Book of the Prefect, XI. I am preparing a paper on the conversion from oil to butter, fat, and wax in JVestern Europe and the Byzantine Empire, to appear in Byzantion. @

T R A D E I N SEVENTH-CENTURY BYZANTIUM

entire Italian, Dalmatian, Crimean, and Anatolian fringe of the Empire became studded with thriving cities, which acted as commercial intermediaries between Byzantium and its adversaries not only in time of truce, but also during wars.13 This leads us to the third general problem: political control. Not unlike the barbarian monarchs, the Byzantine emperor found it increasingly hard to exercise his authority far from his proper residence. Heraclius, who disliked leaving the capital, twice lost most of the provinces; Constans 11,who wandered through Italy, lost even his life. But a political solution was brought about in two opposite ways: by consolidating the provinces into themes, and by releasing the forces of self-government and self-defense in the outer areas, as in the exarchates. A similar transformation occurred in the control of foreign trade. As long as the Empire was a large, compact, and almost self-sufficient state, its principal preoccupation was not to promote international exchanges, but to prevent foreigners from spying into state secrets, exporting war materials and essential commodities, or evading the payment of customs duties. A central official, the Count of the Commerces, channeled and supervised all external trade through special international fairs held periodically at a small number of checkpoints just inside the border. No alien merchant was allowed to trespass beyond the fairs, or to trade except in the presence of a representative of the central government .I4 This tightly knit system, however, proved untenable when the solid frontier was broken and when foreign trade became a vital part of the economy. By the time of Heraclius, the single Count of the Commerces had been replaced by a host of K O ~ ~ E P K ~ & P I and the intermittent fairs of the border areas had OI, been supplemented by a multitude of permanent bt-rro3fj~al TGV Pa5lhlKGv K O ~ ~ E P K ~ in O V every important harbor and road terminus of the Empire. All this we know only through seals of officials and offices, but it seems safe to assume that the change was a response to the breaching of the frontier. Like the old fairs, the new local bureaus were not intended to promote international exchanges but to control them; they were safeguards of commercial defense, even as themes were safeguards of military defense. Yet international exchanges in a freer atmosphere were indispensable to the survival of those outlying possessions that were encircled by alien land. Here the Empire lifted its control, allowing local authorities to make what arrangements they could to maintain '3 T h e economy o f t h e Byzantine border towns could provide t h e subject for a highly interesting essay, b u t so far there are only local studies, m o s t o f t h e m old and primarily concerned w i t h political history, such as M . Schipa, I1 vnezzogiorno dlItalia anteriormente alla monarchia (Bari, 1923); R. Cessi, Venezia ducale, 2 vols. (Padua, 1928-29) ; L. M . Hartmann, Analekten zur Wirtschaftsgeschichte Italiens (Gotha, 1904); Ch. Diehl, L'Afrique byzantine (Paris, 1896); 0. Tafrali, Thessalonique des origi~zesa u X I V e sidcle (Paris, 1919); A. A. Vasiliev, The Goths i n the Crimea (Cambridge, Mass., 1936), etc. O n salt, i n particular, see also Vasiliev's " A n Edict o f t h e Emperor Justinian 11," Speculum, XVIII (1943). T h e largest selections f r o m t h e scattered, yet fairly abundant material o n t h e subject are found i n t w o works devoted t o other topics, t h e dissertation b y J o h n Teal1 herewith, and G. P. Bognetti, Note per la storia del passaporto e del salvacondotto (Pavia, 1933). l4 O n t h e problems discussed i n this paragraph and t h e following one, t h e most detailed work is m y o w n " D u march6 temporaire & la colonie permanente," Annales (e'conomies, socie'te's, civilisations), IV ( 1 9 4 9 ) 3~8 g f f . T o i t s bibliographic references one m a y add P. Duparc, "Les cluses e t la frontikre des Alpes," Bibliothhque de 1'Ecole des Chartes, CIX ( 1 9 5 1 )and ~ F. L . Ganshof, Le Moyen Age (Histoire des relations internationales), I (Paris, 1953), chaps. 2 and 3.

74

R O B E R T S. L O P E Z

and increase their trade across the border, even as they had to rely upon their own resources to forestall enemy occupation. In the sixth century Rome made informal agreements with the Lombards. In 715 the local Byzantine authorities of Comacchio concluded a formal commercial pact with the Lombard authorities of the Po valley. The Arab-Byzantine treaty of 688 included a peculiar modus vivendi for the island of Cyprus : the inhabitants paid tribute to both states, promised to disclose to each party any hostile preparations of the other party, and traded freely on both sides of the fence. I t is outside the limits of this paper to follow up the further drifting apart of the outlying provinces from the rest of the Empire. Eventually Venice and other Italo-Byzantine cities emerged as independent states, where international trade was supreme if not unrestricted, and where the entire social and economic structure was utterly unlike that of the Byzantine hard core. In the Byzantine Empire proper, on the other hand, the recovery of compact stretches of agricultural land restored the traditional predominance of agrarian interests. Diffidence towards foreign trade found new expression by confining alien merchants not at the points where they were furthest from vital centers, but where control of them was easiest, that is, in the compulsory lodging houses (pl-ra-ra) of the capital. The enterprising spirit of Greek merchants was eventually stifled by the convenience of trading with foreigners without leaving the protected areas of the fatherland. Nor was this the only field where the resurgent conservatism of the ancien re'gime killed the seeds of evolution and progress that had been planted in the soil of the seventh century. Conservatism, however, showed its better side in backing up the basic lubricant of trade, currency. This was partly due to the fact that the unalloyed gold besant of full weight was singled out as a fitting symbol of the unalterable integrity of the imperial power. The seventh century began with what I believe was a law of Heraclius to enforce a new discipline on mint workers and a new penalty for counterfeiting; it ended with a war waged by Justinian II to punish the Arabs for issuing the first competitor of the besant, the dinar. To be sure, the aims of this policy were not primarily economic, nor were the methods entirely sound; but the result was a coin that propped up the purchasing power of Byzantine merchants at home and abroad,15 and though not all of the besants which have been found in hoards all the way from England to Siberia are indisputable tokens of direct commercial relations, they seem to indicate that what commerce lost in volume was partly compensated for by gains in geographic expansion. To confirm their message we find occasional references to Greek merchants in Western chronicles and documents, the reception of Byzantine monetary legislation in Germanic law books and Islamic custom, and archaeological evidence of variable reliability and precision, ranging from the buckles, whose Byzantine origin has recently been reaffirmed after they had been taken as proofs of Bulgarian penetration, to the impressive Cf. R. S. Lopez, "The Dollar of the Middle Ages," Journal of Economic History, XI (1g51), zogff., and its bibliographic references. More recently, C. M. Cipolla, Money, Prices, and Civilization in the Mediterranean World (Princeton, N. J., 1956), chap. 2, has dealt with the same subject.

T R A D E I N SEVENTH-CENTURY BYZANTIUM

silver plates bearing Byzantine hallmarks.16 The praise bestowed by Bede upon a British princess, whom he found as pure as the untarnished besant, parallels the celebration of British tin in the Life of St. John the Almsgiver, wherein the tin is miraculously transformed into marked silver plate. This miracle was lately "proved" to the St. Thomases of our age, when they were able to see and touch a Byzantine silver plate in the seventh-century shipburial of Sutton Hoo. The least one can say is that there were Britishers who by trading with the Greeks turned tin into silver. On the other side of the hemisphere, Chinese reports of four embassies from the Byzantine Empire that reached China between 643 and 719 may lend support to the often doubted story of Zopatros, the Byzantine merchant who, according to Cosmas Indicopleustes, demonstrated the superiority of his emperor over the Persian autocrat by showing to the King of Ceylon the unmatched splendor of the besant.17 The reasons that lead one to call the besant the "dollar of the middle ages" have been explained on another occasion, but here it is proper to point out that even the prestige of the dollar in international trade rests upon the foundation of a prosperous internal economy based on quarters and dimesnot to mention nickels and pennies, which have recently fallen victims to inflation. Similarly, in the Byzantine Empire internal trade continued to play a large role in total amounts if not in concentration of capital. Less glamorous than the unique stability of the besant, yet probably no less significant, was the survival of silver coinage and the uninterrupted availability of copper coins, at the very period when barbarian Europe ceased to strike the small change which alone can serve the daily transactions of the common man. It would be too much to expect that the collapse of local monetary exchanges and the total disappearence of divisional coinage in Western Europe had no parallel in the East; there are indications that the Byzantine mints, too, diminished their output of silver and bronze in the course of the seventh century,ls Is The economic historian can make little headway in this direction without the collaboration of the archaeologist, the numismatist, and the historian of law. Written and archaeological evidence relevant to the economic history of the seventh century is effectively combined in a number of works: J. Ebersolt, Les arts somptuaires de Byzance (Paris, 1923); V. Laurent, "Bulletin de numismatique byzantine 1940-49,'' Revue des e'tudes byzantines, I X (1951); L. Matzulewitsch, Byzantinische Antike (Berlin and Leipzig, 1929) ; N. Pigulevskaia, Vizantiia nu putiakh v Indiiu (Moscow and Leningrad, 1951); J. Werner, "Byzantinische Giirtelschnallen des 6. und 7. Jahrhunderts aus der Sammlung Diergardt," in Kolner Jahrbuch fur Vor- und Friihgeschichte, I (1955)~p. 36ff. l7 On the missions to China cf. C. P. Fitzgerald, China, a Short Cultural History, and ed. (London, 1948), chap. 15; on commercial relations with England, see the bibliographic references in R. S. Lopez, "Le problkme des relations anglo-byzantines du septikme au dixikme sikcle," Byzantion, XVIII (1946-48), 139ff. ls Professor Ostrogorsky's interesting remarks on the comparative rarity of copper coins of the seventh century in numismatic collections-whereas gold coins are fairly well represented, see above, pp. 48 ff.,-suggest the following comments: A . Obviously local and petty trade suffered more than long distance commerce, and the scarcity of copper in the Byzantine coinage reflects this trend; yet some copper was struck. The crisis was less serious than in the West, where gold alone was struck for a long period, and in some instances the function of divisional coinage was taken over by such substitutes as bread loaves of a standard size; cf. G. P. Bognetti, "I1 problema monetario dell'economia longobarda," Archivio Storico Lombardo, LXIX (1944)~112 ff. B. Granted that copper coins seem to have been struck in smaller quantities than ever before, their rarity in numismatic colletions partly depends on the fact that they did not attract modern collectors as much as gold coins, if only because of their shabby

76

R O B E R T S. L O P E Z

but did not discontinue it. Indeed, the Heraclian dynasty is to be credited with the issue of an unusually heavy silver coin, the hexagram. In contrast to this, its copper follis became gradually smaller and was badly struck. Yet the ugly Heraclian folles must have changed hands rapidly in the shops of the city butchers, bakers, and candlestick makers-all of them mentioned in the sources of that period-and in such local fairs as that of Trimithus in Cyprus, where Theodore of Paphos bought coarse clothing and blankets while collecting information for his Life of St. Spyridon. No doubt the fair of Trimithus was a modest commercial gathering when compared to the great -rravqyirp~ovof Ephesus, which by 787 was assessed a lump tax of a hundred gold pounds, that is, 7200 gold besants. But today besants circulate only on the numismatic market, and ruined Ephesus has lost its wealth and its pride, whereas the fair of Trimithus is still held every year on the feast of St. Spyridon, as it was in 625 -a reminder that nothing can be as enduring as local tradition and humble trade.lg One Byzantine industry and trade, that of silk, did better than to endure the troubled seventh century: it entered a period of unprecedented growth. Not unlike currency, it provided glittering symbols of imperial power, which claimed the very best products for its exclusive use. Still other silk products formed the most valuable part of Byzantine exports, and there was a good deal left for internal consumption. The silk industry's most serious crisis had occurred in the age of Justinian I when the imperial monopoly had been extended to cover every stage in its production and trade, and when raw materials had become scarce, prices had skyrocketed, and many craftsmen had run away to Persia. However, that age had also been blessed by the first, unwilling contribution by China to the industrial potential of Europe: the silkworm was introduced to the Byzantine Empire. There followed an expansion which cannot be reconstructed in detail, for documents are too few and often impossible to date, but which may perhaps be imagined by the report that in

appearence and careless striking. Moreover, we cannot expect that copper was hoarded as much as was gold or silver. C. Above all, the decline of copper in a period of economic contraction or stagnation is quickened by the heavier incidence of minting costs and the slower obsolescence of old coins in comparison to the costs and obsolescence of gold and silver coinage. Because of the high value of the metal, the latter can usually bear the minting cost and leave a profit to the mint even if the difference between face value and metallic content is relatively small; on the other hand, an old and worn coin will not easily be accepted at par with a coin of full weight. Therefore it is indispensable and profitable to issue new gold or silver coins a t frequent intervals, even if the economy is not expanding. But the cost of minting a copper coin is too high in proportion to the value of the metal to leave a profit to the mint unless the difference between face value and metallic content is large; on the other hand, an old and worn coin is often accepted at par with a coin of full weight. Therefore it is neither indispensable nor profitable to issue new copper coins unless an expanding economy calls for an increased supply; yet if the demand for copper coins is sufficiently sustained, the government, even in a stagnant or contracted economy may try to freshen up old coins by the cheaper method of re-striking, that is, superimposing a new mark on an old piece. This method was often used in the early seventh century; old copper coins also were often used without re-striking. Cf. W. Wroth, Catalogue of Imperial Byzantine Coins in the British M u s e u m , I (London, 1908)) 237ff., 370ff.; E. Leuthold, "Monete bizantine rinvenute in Siria," Rivista Italiana d i Numismatics, ser. 5, I (1952--53),31ff. l9 Theophanes, p. 469; Theodore of Paphos, ed. P. van den Ven, L a lkgende de S . Spyridon Lvtque de Trimithonte (Louvain, 1953)~chap. 23, and cf. van den Ven's introduction, 114and note 2.

T R A D E I N S E V E N T H - C E N T U R Y BYZANTIUM

768 Constantine V was able to ransom all prisoners taken by Slavic pirates on three Aegean islands by the single gift to the captors of 2500 silk robes. An article in the Bulgarian-Byzantine commercial treaty of 716, providing specifically for the export of textiles and leather, points out the importance of silk trade with the closest neighbors of the Empire. A textile from Liege, with the monogram of Heraclius, is dated evidence of the diffusion of Byzantine fabrics to the more distant countries of Western Europe. I t matters little that the textile may have been a present of the Emperor to the Belgian church, for, if the gifts of precious silks from the imperial factories to foreign rulers and ecclesiastic institutions were trump cards in the diplomatic game, they also served to advertise the slightly less precious textiles that Byzantine private industry was in a position to sell to foreign merchants.20 The progress of the silk industry and trade was certainly connected with the diffusion of sericulture in the seventh century and with the elaboration of imaginative patterns by Byzantine designers. Yet probably the most significant change-in terms of human dignity, at least-was the collapse of the late Roman regimentation of virtually all craftsmen and traders in hereditary guilds, collectively responsible for a production quota or for the payment of a lump sum into the treasury. Historians have often pointed out that this bondage was already slackening under Justinian, and that by 599 a letter from the soapmakers of Naples to Pope Gregory I casts some light on the imminent breakdown of one compulsory guild. But it is only for the silk industry that we can observe the transition in some detail and place the decisive turn in the late sixth and early seventh centuries-if my attribution of the undated law Tl~pi ~ E T ~ Cto~ Sthe years immediately following Justinian, and of certain articles in the Basilics to Heraclius is correct. These texts show that both the imperial monopoly of silk manufacturing and trade and the restricting of silk workers and their descendants to their guilds lost their rigidity as the supply of raw material and the availability of manpower increased. The principle of heredity was preserved only for the 6qp6o1a ochpa-ra, the reorganized imperial guilds which manufactured in state-owned buildings the precious silks reserved for the emperor and the court. Privilege replaced coercion since all members were allowed to quit a guild but no man who was not of a member's family could join an imperial guild. Apart from textiles reserved for the emperor and his craftsmen, the silk industry and trade were thrown open to the free competition of private guilds and of individual entrepreneurs assisted by their own dependents. The government continued to see that legal restrictions of foreign trade were obeyed, but otherwise limited its intervention to the maintenance of fair trade practices. I t helped the guilds enforce their own by-laws and apportioned opportunities between the different groups of producers, "so that 20 Nicephorus, ed. De Boor, Breviarium, 76; Theophanes, 497; Lopez, "Silk Industry," plate IVa (where the textile is reproduced upside down). The latter paper contains a fuller discussion of the entire problem and bibliographic references, to which we may add F. E. Day, "The Tiraz Silk of Marwan," Archaeologica Orientalia in memoriam Ernst Hersfeld (Locust Valley, N. Y., 1g5z), 39 ff., an important study of the relations with the young Islamic silk industry of the seventh century.

R O B E R T S. LOPEZ

men. . . should not shamelessly trample upon one another," to quote a later statement of Leo VI in the Book of the P r e f e ~ t . ~ ~ I do not propose here to discuss in detail such scattered information on other craftsmen and merchants as can be obtained from the historical and hagiographic sources of the seventh century. The little we know seems to indicate a similar transition from qualified bondage to qualified freedom of employment. Eventually the improved status of craftsmen and shopkeepers 21 Here I have summed up an argument which was developed a t much greater length in three earlier papers, "Byzantine Law in the Seventh Century and its Reception by the Germans and the Arabs," Byzantion, XVI (1942-43), 445 ff. ; "Silk Industry" (above, note 10); "Harmenopoulos and the Downfall of the Bezant," Tomos Konstantinou Armenopoulou (Salonika, 1g51), 111ff.; cf. also "An Aristocracy of Money in the Early Middle Ages," Speculum, XXVIII (1953), 5-9. My tentative reconstruction was well received by some scholars, but repeatedly challenged by Franz Dolger, Byzantinische Zeitschrift, XLII (1g50), 244, and XLVI (1953)~235, 472, and by his disciple B. Sinogowitz, Saeculum, IV (1953)~321. Their case seemed to me less impressive than their erudition would have led me to expect: they merely insisted that the Basilics are nothing but a compilation of the Corpus Juris and cannot possibly contain any material of the seventh century. This, however, is a gratuitous assumption. Although by far the largest :part of the Basilics consists of reshuffled and often obsolete Roman law, we know for certain that this monumental work also contains many passages borrowed from the Procheiros Nomos and the full text of the Nomos Nautikos, the latter being a compilation datable somewhere between the seventh and the ninth century. I t is true that in the Basilics there is no trace of the handful of laws concerning ecclesiastical problems which is all that survives of original seventh-century legislation; but there is no a priori reason to exclude the possibility that other laws, of which the originals are no longer extant, may have left a trace. The circumstantial evidence that led me to postulate certain laws of Heraclius has already been listed in the three papers mentioned above. But it may be useful to recall very briefly the basic documents which were used for my reconstruction: A . I n the fourth century a series of laws, collected in Cod. Theod. X, 20 and Cod. Just. XI, 7, "froze" in their employments all manufacturers and dyers of precious textiles, mint workers, and other craftsmen attached to imperial factories, forbidding even their descendants to change profession. B. A law of 426 (Cod. Theod. X, 20, 16 = Cod. Just. XI, 7, 13) repeated the same provisions for all corpora. . . ad Divinas Largitiones nexu sanguinis pertinentia, but opened a small loophole by allowing individual workers to withdraw from their guilds if they found a substitute acceptable to the Count of Sacred Largesses. C. Justinian brought the entire silk industry under state monopoly (Procopius, Secret History, chap. 25; schol. 6 ad Basil. XXIII, 3, 74), but opened another loophole by discharging some workers of the guild of purple-dyers, which had become overcrowded (Nov. Just. XXXVIII, 6). D. An undated ordinance, found in a collect~onof legal sources later than the Corpus Juris (Peri metads), relaxed certain obligations of the guild of importers of raw silk; the mention of the kommerkiarioi and other features of the ordinance suggest a date between the death of Justinian and the accession of Heraclius. E. A letter of the soapmakers of Naples (Greg. I, Epist., IX, 5) shows that by 599 the regulations "freezing" the workers of that guild, whose status was similar to that of the imperial guilds, were no longer strictly enforced. F. An undated law, discovered by Cujas in a manuscript of the Basilics which is no longer available (Basil. LIV, 16, 16), indicates that the conditions of employment in the "public" or "imperial guilds" had changed to the exact opposite of what they had been in the fourth century. Now only the descendants or relatives of existing guild members were allowed to seek admission to the guild, and admission was granted only if there were vacancies and after an examination supervised by the provincial governors; in other words, the government strove not to keep unwilling workers in understaffed guilds, but to keep unnecessary workers out of overstaffed guilds. The fact that the Count of Sacred Largesses is no longer mentioned while the provincial governors still are, as well as other features of the law, suggests a date within the reign of Heraclius. Perhaps it was issued after the emergency minting of precious metals donated by the Church to fight the Persians (631), an act which must have caused a temporary overcrowding of the mints. G.The penalties of Roman law for dishonest moneyers and forgers were abandoned in favor of the new penalty of cutting off a hand; in Byzantine law this penalty is first mentioned in the Isaurian Ecloga XVII, 18, but the simultaneous introduction of the same penalty in Visigothic, Lombard, and Arab law a t about the time of Heraclius suggests the probability that there was a law of Heraclius and that the neighbors of the Byzantine Empire adopted it. H. Later Byzantine law draws a clear distinction between "private" guilds, whose regulations are codified in the Book of the Prefect, and "public" guilds, whose regulations, stemming from Roman law, are preserved in the Basilics with such modifications as were needed to bring them in harmony with the spirit of the laws mentioned under F. and G .

TRADE I N S E V E N T H - C E N T U R Y B Y Z A N T I U M

manifested itself in the constitutional history of the empire as the t p y a o ~ q p i m o i of the capital, together with the senate, the army, and the higher ranking citizens, were invited to take an oath of loyalty to Constantine V. That a sovereign should lower himself to demand a declaration of allegiance from ordinary laborers would have been unthinkable in Western Europe, where most craftsmen at that time were serfs and had no part whatsoever in the governing of the body politic.22 Lest we excessively idealize the Byzantine democracy, however, we must remember that there slave workers were more numerous than in Western Europe. As late as the tenth century they were employed in many professions, and did not attain the status of masters except in the lesser If we recall that maritime trade often provided the only link between isolated limbs of the Byzantine territory, we shall not be surprised that the class which made the greatest gains during the seventh century was that of the v d ~ A ~ l p oor i sea merchants and captains. The burdensome obligations which the Late Roman Empire had imposed upon its navicularii disappeared with state transportation itself, although a case for the partial survival of grain liturgies may perhaps be made out of a passage of the Miracles of St. Demetrius; but on this subject my former student and present colleague, John Teall, has much more to say than I have.24At any rate the gradual adoption of smaller, one-decked vessels in the place of the old three-deckers and four-deckers was a powerful factor of democracy, if I may be permitted to use this term, which may seem anachronistic here though it was first used by the Greeks. The trireme called for slave oarsmen rowing in a floating prison, but on the single deck of the galley (and, perhaps, of its forerunner, the dorkon) all sailors shared the dangers and the opportunities of a common venture in the invigorating air of the open sea. I t was also easier for a seaman to own a smaller ship, although any ship could be divided into shares small enough for the income bracket of a sailor. Thus it was that the sources of the seventh century very frequently mention a naukleros as an independent trader, if not necessarily as a well-to-do man. Indeed, it would have been hardly possible for the naukleroi to support maritime themes or to codify their self-imposed by-laws in the Nomos Nautikos except in an atmosphere of freedom. Both the Nomos Nautikos and the earliest maritime theme most probably go back to the seventh century. At the beginning of the ninth century Nicephorus I, while paying the naukleroi the undesirable honor of singling them out for two out 2 2 Theophanes, 449; the same chronicler already mentions the ergasteriakoi in 703 as one of the classes from which Justinian I1 exacted payment for his vengeful expedition against Cherson (377). Prevailing conditions of mobility and free employment are also indicated by the fact that Constantine V hired skilled building workers (&v6pasT E X V ~ T1€ ~5 S 01~080pfi~ Ep~r~Ipou~) from all parts of the Empire t o restore Valens' aqueduct in Constantinople, cf. Kicephorus, 76. 23 Cf. A. Hadjinicolaou-Marava, Recherches sur la vie des esclaves dans le monde byzantin (Athens, 1950), 4zff.; M. Siuziumov, "Remeslo i torgovlia v Konstantinopole v nafale X veka," Vizantiiskii IV (1951)~r ~ f f . t; he problem, however, has not yet been adequately studied. Vremennik, N.S., 24 Miracula Sancti Demetrii, AA.SS. October IV, chap. 4, 175; but on this see Teall below, p. 122. This problem may receive further clarification in Professor Lemerle's forthcoming study of the Miracula.

80

ROBERT S. L O P E Z

of ten of his so-called fiscal "iniquities," made no attempt to curtail their welldeserved liberty.25 There is definitely a parallelism between the seamen of the Nomos Nautikos and the peasants of the Nomos Georgikos. Both groups had a hard life in the troubled seventh century, but both gained in status because they were the backbone of the armed forces and because they played an indispensable role in the economy of an underpopulated empire. The role of the seamen, however, was more exalted and offered far greater opportunities to profit from larger investments. I t is no wonder that some of the most impressive miracles in Byzantine hagiography were performed on their behalf. Although the ship of 30,000 bushels that 300 men could not ease into the sea, mentioned in the Sfiiritual Meadow, may be a mere figment of the fertile imagination of John Moschus, the matter-of-fact Nomos Nautikos speaks of ships of many thousand bushels. A bushel was forty pounds, and a ship's worth was reckoned at fifty pounds of gold every thousand bushel. Then there was the cargo, which could include gold, silver, whole silk, pearls, wine, oil, or grain, each specifically mentioned in the Nomos Nautikos and in other sources. Slaves, too, were a rewarding, if often illegal and morally objectionable merchandise. Alone or associated with others, with capital partly owned and partly borrowed, the sea captains conducted innumerable ventures; and every sailor was entitled to a share of the profits according to the rules of the Nomos N a ~ t i k o s . ~ ~ None of the new ideas of the seventh century is more important than the first references in the Nomos Nautikos and in the saurian Ecloga to what proved to be the most popular mediaeval contract to pool capitals in sea ventures. These sources call it xp~o~o~vwvia, more preK E P ~ O K O I V W V ~ Uor, , cisely, GQaov [Cxi] ~ o t v a v i a(loan in partnership); the late mediaeval West, which developed and exploited the contract more fully, called it collegantia or ~ o m r n e n d aAs . ~ ~the Byzantine name indicates, it combined the advantages 25 On the "iniquities" cf. G . I. Bratianu, Etudes byzantines d'histoire e'conomique et sociale (Paris, 1938), ~gjff., with further references; in general on ships and ship captains see our references in footnotes 7 and 9 above. 26 The scholar who undertakes a fresh study of the Nomos Nautikos-may this happen soon!-will find a great wealth of collateral evidence not only in John Moschus' Spiritual Meadow (the ship mentioned in our paper appears in chap. 83; cf. Nomos Nautikos, ed. Ashburner, 11, 16), but also in the Miracles of St. Demetrius and those of Artemius, the Lives of St. John the Almsgiver, St. Theodore Sikeotes, St. Spyridon, St. Nilus, St. Luke the Younger, St. Peter of Argos, St. Gregory Decapolites, and other religious literature. Indeed, for two or three centuries the ship captain and the sailor occupy in Byzantine hagiography virtually the same central position as the landowner and the peasant in Western hagiography; the contrast reflects all the difference between a maritime economy with a fairly lively trade and a land-bound economy with narrow horizons. Two pioneer papers are based upon sources of this kind, B. S. Nelson and J. Starr, "The Legend of the Divine Surety and the Jewish Moneylender," Annuaire de Z'lnstitut de Philologie et d'Histoire Orientales et Slaves, VII (1g39-44), 289ff. and G. R. Monks, "The Church of Alexandria and the City's Economic Life in the Sixth Century," Speculum, XXVIII (19j3), 349ff. ; they are samples of what a methodical exploration with an eye to economic information might yield. 27 N O ~ ONautikos S 111,g and 17; Ecloga X, 5 ; for the cornmenda see the selected contracts translated in R. S. Lopez and I. W. Raymond, Medieval Trade in the Mediterranean World (New York, 1gj5), chap. 9, and the introduction to that chapter. The origin of the cornmenda is one of the most disputed points in economic and legal history; i t would not be proper here to go over the various theories that have been suggested. My opinion is close to that of G. Astuti, Origini e svolgirnento storico della commenda fino a1 secolo X I I I (Turin, 1933)-this does not seem to have been seriously damaged by the

T R A D E I N SEVENTH-CENTURY BYZANTIUM

81

of a loan with those of a partnership. Like a loan, it entailed no liability for the investor beyond the sum of money or the quantity of merchandise handed over to the manager, and like a partnership, it divided the risks and profits of an investment between investor and manager. A longer description of this flexible, equable contract, which is justly regarded as having been a sinew of mediaeval sea trade, would be tedious for all but specialists in the history of mediaeval business. Still, it was gratifying to me, as I was preparing this paper, to encounter a passage in the Life of St. Spyridon containing most of the formulary of such contracts although it does not yet use the technical term, the nomen juris. Perhaps I may be forgiven if I quote here a literal translation of this early seventh-century formulary and a translation of the formulary of the earliest extant Venetian cornmenda contract, of 1073. Certain awkward expressions which had always surprised me in the Latin of Venice become obvious as a translation of the Greek of Trimithus. The Life of St. Spyridon: A sea captain of Trimithus . . . asked that he be given as a G a v ~ ~ oav small sum of money in order to throw an investment in his ship, so that by putting it to work he might support his household . . . .28 The Venetian contract: I , Giovanni Lissado . . . received in collegantia from you . . . zoo pounds Venetian, and I threw in it IOO pounds, and-out of this investment we have two shares in the ship of which Gosmiro da Molino is captain . . . . I promise to put to work all of this and to strive the best way I can....29 The cornmenda left the managing party, that is, the borrower, in virtual control of the investment ( ~ v s ~inK Byzantine ~ sources of the seventh century, objections o f Silberschmidt and Condanari-Michler-although I would attach greater importance t h a n h e does t o t h e influence o f Muslim contracts (themselves influenced b y Byzantine contracts) a n d , above all, t o spontaneous evolution i n t h e W e s t . I believe, however, t h a t Professor A s t u t i is right i n regarding t h e chreokoinonia o f t h e Nomos Nautikos and t h e Ecloga as t h e m o s t important and direct foreign contribution t o t h e origin o f t h e W e s t e r n cornmenda. T h e seventh-century source I quote here seems t o add some weight t o t h i s view. 28 Theodore o f Paphos, chap. 2 1 , p. 92 v a n d e n V e n : "-rrpovcnjKAqp6~-r1~ Tp1p10oOvrwv ... -rrapmdLjv , P&hq PAqoi61ov (variant: EvSfi~qv)Pv -r@ 609ijva1 a h @ Pv 6 a v ~ 1bhiyqv T I V ~xpuoiov ~ o o 6 - r q ~ aIva -rrhoiv &oG, ~ T W S 61h ~ i ~apt-rroufioew~ j ~ &oG 61a9p6yg ~ b BmoG v OIKOV .. ." (See o n his passage t h e philological comments o f H . GrCgoire, " L e s m o t s i3hqoi61ov e t o~op-ri61ovdans la vie de S. Spyridon," Nouvelle Clio, IV [1g52],324-25). There follows a brief description o f t h e activity o f t h e sea captain w i t h t h e investment which could fit a cornmenda merchant o f t h e later mediaeval W e s t : t o trade h e goes f r o m port t o port (txyopaoas, which corresponds t o t h e familiar "causa mercandi' o f t h e comrnenda contracts), sells t h e investment, and carries back t h e proceeds which h e delivers t o t h e investor. There is n o sharing o f profits, however, because t h e hagiographer wishes t o point o u t t h a t S t . Spyridon is merely interested i n extending charitable loans; indeed, t h e Saint does n o t even protest w h e n t h e captain fails t o return t h e capital! 28 R. Morozzo della Rocca and A. Lombardo, Documenti del commercio veneziano nei secoli X I - X I I I , I (Turin, 1940), 12 : Accepi ego quidem Iohannes Lissado . . . i n collegancia de te . . . libras denariorum ducentas, et ego ibidem iactavi libras denariorum centum, et de isto habere habemus sortes duas in nave ztbi nauclerus est Gosmirus da Molino . . hoc totum laborare et procertare promitto in quo melius potuero . . . . (There is a full translation o f t h i s document i n Lopez and R a y m o n d , op. cit., 176-77, b u t i t is n o t as literal as t h e translation o f t h e excerpt i n t h e present paper). I t m u s t b e noted t h a t t h e Venetian contract o f 1073 is a bilateral cornmenda, t h a t is, t h e person traveling also contributes some o f t h e inv e s t m e n t ; whereas t h e Byzantine example comes closer t o t h e unilateral comrnenda, as i n Morozzo della Rocca and Lombardo, ibid., 59.

.

82

R O B E R T S. L O P E Z

entica again in many Italian contracts of the later Middle Ages).30 If the investor wanted a tighter hold on the enterprise, there was the commission contract. Thus, in the Doctrina Jacobi nuper baptizati, we read of an Anatolian Jew who took for sale in Africa a few pieces of clothing and other merchandise belonging to a rich man. He was to receive not a share of the profits, but a fixed commission (M~a965) of 15 besants. Even so, the investor could hardly keep an eye on the manager once the latter had sailed off to put the capital "to work." As a matter of fact, the story tells us that the Jew stealthily sold the merchandise to his coreligionists in Carthage for a trifling sum.31 The rich man himself might have done better if he had taken a fixed interest while letting the agent trade at his own risk and profit from the investment. This could have been arranged through the "sea loan," known from classical antiquity and fairly popular down to the later Middle Ages, owing to its special clause that relieved the borrower from the obligation of returning capital and interest if the ship or the cargo were lost at sea. Roman law had established for such loans a legal interest rate which changed but little throughout Byzantine history. Unlike the Western Church, the Eastern Church made no sustained effort to outlaw as sinful all interest-bearing loans, and this was undoubtedly a boon to Byzantine trade.32 Yet there are indications that such contracts did not meet with the same approval as the commenda and the commission. Had not the Gospel preached the lending of money out of love, without hope of any reward? The relevant verse of Matthew significantly recurs in the Life of St. Spyridon and in that of St. John the Almsgiver, both times to point out the inspiring example of a saint who had extended free loans to needy naukleroi, and had not even complained when the borrower failed to return the capital 30 Isolated terms, of course, do not by themselves prove any similarity of legal forms. ' E v 9 ~ j ~isq often found in Byzantine sources not only for the chreokoinonia and other contracts which seem to bear a resemblance to it, but also in commission agreements and loans: Nomos Nautikos 111, titles 8, 39, 40; chaps. I I, 21, 32; Ecloga X, 5 ; Theodore of Paphos, 21, 92 var. ; Leontius, Vita S. Johannis, 18, ed. Gelzer; Doctrina Jacobi nuper baptizati, ed. Bonwetsch, p. go; additional references in Ducange (the word is post-classical). I n turn, hentica is the technical term for the investment in Pisan commenda contracts (see, for instance, Lopez and Raymond, op. cit., 180)~but entigum is the nomen juris and entega the investment in another type of contract in Ragusa (Dubrovnik; cf. Ashburner, ccxlviii ff.); in Venice, the investment is called habere or capetanea, etc. The same can be said for such terms as taxedium, nauclerzts, catabolus, all of them of Greek origin but widely used in the Latin West as well. Indeed the term commenda was used indiscriminately in the Middle Ages for contracts which modern jurists classify under other headings; on the other hand, what modern jurists call commenda was designated by other names in Venice and in some other towns. I t is obvious that we cannot expect uniformity of official or learned jurisprudence in legal customs which grew almost spontaneously as the collective invention of thousands of obscure men. The development of mediaeval maritime law in the Mediterranean is to some extent comparable to that of vernacular languages; an ancient form was modified along the same lines in various centers, but i t was some time before these changes were codified and given uniform rules valid for more than one locality. 31 Doctrina Jacobi nuper baptizati, ed. N . Bonwetsch, Abhandlungen der Koniglichen Gesellschaft , On the economic der Wissenschaften z u Gottingen, Philo1.-Histor. Klasse, N.S., XI1 ( I ~ I O )89-90. activities of the Jews see J. Starr, T h e Jews in the Byzantine Empire, 641-1204 (Athens, 1939) and its abundant bibliographic references, which do not, however, include the Doctrina. For this reference I am indebted to Professor Dolger. 32 Cf. G. Cassimatis, Les intdrkts duns la ZLgislation de Justinien et duns le droit byzantin (Paris, 1931); Nelson and Starr, zggff.; S. Runciman in Cambridge Economic History, 11, 108-09; Lopez, "East and West," p. 156-7.

TRADE I N SEVENTH-CENTURY BYZANTIUM

83

after his safe arrival home. Similarly, the Spiritual Meadow stresses the moral advantage of lending without asking either for a reward or for colI s it necessary to point out that even in the seventh century neither the average business man, nor the ordinary church administrator, was a saint ? A papyrus of 541, concerning a banker in Constantinople and his correspondent in Alexandria, indicates that a priest had to pay the legally allowed 8 per cent interest, plus 4 per cent "to refund expenses," on a short-term loan of twenty nomismata. A passage in the Life of St. John the Almsgiver informs us that the church of Alexandria kept in its archives all sorts of memoranda (.rru~~a~ua) of business transactions in which it was involved either as a borrower or as a lender of money, so that it is safe to assume that informal loans extended by word of mouth, such as are mentioned in other passages of the same book, were quite exceptional.34 Perhaps the most significant text of the period is the legend of the kindly Jewish money lender and the image of the Antiflhonetes Savior, which has come down to us in a later version, but refers specifically to the time of Heraclius. I t has many elements in common with an episode in the Life of St. John the Almsgiver and was retold through the years in many countries undergoing changes with each retelling until finally the generous Abraham became Shakespeare's Shylock. To recount the tale only briefly, since it is a legend familiar to Byzantinists, -Abraham was originally said to have lent a substantial sum to a Christian sea merchant, who had no other surety to offer than the icon of Christ set in the Chalkoprateia, and had, therefore, been able to find no-one else willing to risk capital with him. When the merchant's ship was wrecked and the entire investment lost, Abraham did not lose faith, but granted him a second loan. The miraculous intervention of the icon enabled the merchant to return this loan and would have yielded great financial profit to the lender had he not decided to become a Christian, to devote his riches to the building of a chapel enshrining the icon, and to serve as a priest in that Clearly, the legend has a double edge. By presenting a Jewish money lender under a most favorable light, it indicates that neither the sect nor the profession was the object of indiscriminate hatred. But, by having the Jew give away his 33 Theod. of Paphos, chap. 21; Leontius, Life of St. John, chap. 35; cf. Ioh. Moschus, chap. 185; these examples could be multiplied. I n the opinion of K. E. Zachariae von Lingenthal, Geschichte des griechisch-romischen Rechts, 3rd ed. (Berlin, 1893)-accepted by E. Freshfield, A Manual of Roman Law, the Ecloga (Cambridge, 1926), g6ff.--religious considerations were paramount in the peculiar i av~iov).T he chapter makes no mention of interest on loans, but arrangement of Ecloga X ( r ~ p G a more precisely, of the KIPGOUSK O I V W V ~ ~ , describes the profit-sharing arrangement of the ~ o ~ v w v i(or, specifically mentioned in the title of Ecloga X by one of the manuscripts). This suggestion is plausible, but we must not forget that actually the chreokoinonia is a modified loan or "loan in partnership." Its inclusion in Ecloga X may have helped the lawgiver to de-emphasize interest bearing loans, but did no violence to legal classifications. 34 Papyrus Maspero, 67126; Leontius, Life of St. John, chap. 46; further comment in G. Mickwitz, "Die Organisationsformen zweier byzantinischer Gewerbe im X. Jahrhundert," Byzantinische Zeitschrift, XXXVI (1936)~63ff.; Monks, op. cit., 359. 35 Nelson and Starr give full details on this legend and its various ramifications, and provide abundant bibliographic references. Add now Lopez, "Relations anglo-byzantines," 145ff. ; E. Kitzinger, "The Cult of Images in the Age before Iconoclasm," Dumbarton Oaks Papers, 8 (1g54), 102 and note 63, especially important for the religious background of the legend.

84

R O B E R T S. LOPEZ

honestly gathered riches and enter ecclesiastic life, it exalts spiritual treasures as contrasted to hazardous commercial gains. John Moschus, too, issues a similar warning: a merchant learns that the sailors of the ship on which he is embarked are plotting to murder him in order to steal his pearls, and decides to throw the pearls into the sea, thus acquiring thegreater wealth -peace of mind.36 To be sure, religious writers in the Byzantine Empire were no more hostile in their attitude toward Mammon than were the writers of Western Europe. There, too, the world of business had to wait for Aquinas to give it a qualified absolution, and for Calvin to make it a blessed vocation; and this change came only after the rise of towns had created a new nobility based on trade and new states dedicated to economic success. No such change occurred in the Byzantine Empire. The bias of the Church coincided with the views of the state and the lay aristocracy, which ever since classical antiquity had looked down upon the merchants and had placed agriculture well above trade. Why, then, did the star of the naukleroi rise so rapidly during the seventh century ? It might not be too bold to suggest that there was a certain influence of the Muslim world, where trade and traders had always been held in highest esteem. Muhammad himself had been a merchant, and not only the Koran, but also the early manuals of Muslim ethics pay sounding praise to the quest for wealth and the profession of ~ommerce.~' More direct reasons, however, can easily be found in the Byzantine empire itself. The physical survival of entire provinces hinged on the success of the seamen in bringing food, weapons, and other supplies. Moreover, the seventh century saw the liquidation of the old aristocracy and the breakdown of large estates in most of the provinces. There was a vacuum of power, which the naukleroi temporarily filled. For a moment it looked as though the land of Minos and Ulysses was to become once more a thalassocracy: the long series of soldiers and bureaucrats who had been exalted to the imperial throne made room for an admiral, Tiberius 111. But soon thereafter, the restoration of Justinian I1 -the probable author of the Nomos Georgikos -marked the end of the maritime interlude. A new landed aristocracy arose in the place of the old and drove the adventurous people of the sea back into comparative obscurity. In the centuries that followed Constantinople and her distant daughter Venice went their separate ways. Venice embraced the destiny which her lagoons had prepared for her, and made the commenda a driving force for a great future. Constantinople did not shun the sea, but showed a growing preference for the land. Her merchants gradually shortened their journeys and tarried in the mitata of the capital. Her seamen could win no social status unless they took the old advice of Cicero-sell your ship as soon as you have earned enough, and live like gentlemen on a country estate. Thus it was that in the ninth century, while the Venetian Doge Giustiniano Partecipazio invested a large proportion Ioh. Moschus, chap. 203; many other examples could be given. Cf. G. E. von Grunebaum, Medieval Islam, 2nd ed. (Chicago, 1953)~215ff.; S. D. Goitein, "The Rise of the Near-Eastern Bourgeoisie in Early Islamic Times," Journal of World HistorylCahiers d'histoire mondiale, I11 (1957)~583 ff ., with bibliographic references. 36

37

T R A D E I N SEVENTH-CENTURY BYZANTIUM

85

of his capital in sea ventures, the Emperor Theophilus ordered the destruction of a ship chartered by his wife, for, he said, "God has made me Emperor, and you should not transform me into a naukleros !"38

38 The well-known story of the "justice of Emperor Theophilus" is related by several Byzantine chroniclers: Theoph. Cont., Theoph., 4; Genesius, ed. Bonn, 75-76; Zonaras, XV, 25 (357-358 Bonn). At an informal meeting in the Byzantine Seminar of the University of Brussels, Professor GrBgoire agreed with me that Zonaras' is the best of the parallel accounts; as often happens, the chronicler is later, but has used better sources. The will of Giustiniano Partecipazio is translated in Lopez and Raymond, op. cit., 3gff. and commented upon in my paper on "Venezia e le grandi linee dell'espansione commerciale nel secolo XIII," L a civiltd veneziana del secolo d i Marco Polo (Florence, 1955)~4zff.; for its historical background see G. Luzzatto, "Les activitks Bconomiques du patriciat vknitien," Annales d'histoire e'conofizique et sociale, I X (1937)~25ff., now reprinted in Luzzatto's Studi d i storia econornica veneziana (Padua, 1954).

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: The Role of Trade in the Economic Readjustment of Byzantium in the Seventh Century Robert S. Lopez Dumbarton Oaks Papers, Vol. 13. (1959), pp. 67-85. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C67%3ATROTIT%3E2.0.CO%3B2-8

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 3

The Evolution of Land Transport in the Middle Ages R. S. Lopez Past and Present, No. 9. (Apr., 1956), pp. 17-29. Stable URL: http://links.jstor.org/sici?sici=0031-2746%28195604%290%3A9%3C17%3ATEOLTI%3E2.0.CO%3B2-H 15

The Dollar of the Middle Ages Robert Sabatino Lopez The Journal of Economic History, Vol. 11, No. 3, Part 1. (Summer, 1951), pp. 209-234. Stable URL: http://links.jstor.org/sici?sici=0022-0507%28195122%2911%3A3%3C209%3ATDOTMA%3E2.0.CO%3B2-C 35

The Cult of Images in the Age before Iconoclasm Ernst Kitzinger Dumbarton Oaks Papers, Vol. 8. (1954), pp. 83-150. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281954%298%3C83%3ATCOIIT%3E2.0.CO%3B2-%23

NOTE: The reference numbering from the original has been maintained in this citation list.

The Grain Supply of the Byzantine Empire, 330-1025 John L. Teall Dumbarton Oaks Papers, Vol. 13. (1959), pp. 87-139. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C87%3ATGSOTB%3E2.0.CO%3B2-F Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:36:23 2008

THE GRAIN SUPPLY OF

THE BYZANTINE EMPIRE, 330-1025



This study is drawn from a dissertation submitted in 1955 to the Graduate School of Yale University in partial fulfillment of the requirements for the degree of Doctor of Philosophy. I should like to thank Professor Robert S. Lopez for his advice and encouragement during its preparation. Much of the research for the original work was completed with the assistance of a Junior Fellowship at the Dumbarton Oaks Research Library and Collection of Harvard University during the academic year 1954-1955 To the faculty of that institution I should like to express my deep gratitude for their help so willingly given. In particular, Dr. Cyril Mango made available to me whatever materials in the Russian language I was able to use. For any errors of fact or interpretation I am, of course, responsible. No attempt has been made, particularly in the second section of this study, to compile a complete bibliography of secondary literature. The reader is referred to that prepared by Peter Charanis for Charles Diehl, Byzantium: Greatness and Decline, trans. N. Walford (New Brunswick, N. J., 1g57),esp. pp. 341-349,For analysis and discussion of recent publications on rural life, see P. Lemerle, "Esquisse pour une histoire agraire de Byzance," RLvue historique, CCXIX (1958),pp. 32-74> 254-284;ibid., CCXX (1993,pp. 43-94.The most recent study of the Byzantine city, that of E. Kirsten, "Die byzantinische Stadt," Berichte zum XI. Internationalen Byzantinisten-Kongress,Munich, 1958, V, 3,was published without footnotes which are, however, shortly t o appear. References of interest for the present study will be found also in A. Pertusi, "La formation des th&mes byzantins," ibid., I, esp. pp. 14-16.

I. INTRODUCTION : T H E PROBLEM

FAT

hile man may not live by bread alone, few this side of Asia have found satisfactory ways to replace it in their daily diet. And how to replace bread must rarely have seemed more crucial a problem than it did in the seventh century when first the Persians, then the Arabs, conquered Egypt, the major source of grain for Constantinople and for the expeditionary armies of the later Roman Empire. Loss followed upon loss as Islam conquered North Africa in the eighth century, and Sicily and its granary in the ninth. I t seems almost incredible that an imperial grain economy geared as closely as it was to predictable deliveries from such "food belts" could have survived their loss. But survive it did. Even when, in 618 and 626, the emperors terminated the greater part of their grain distributions, the "many" stayed on within the walls of Constantinople. They complained, they grumbled, they protested in the Godguarded church, but somehow they fought off the invaders, somehow they battled with a new spirit, and somehow they found the means to replace first the temporary, then the final, loss of the Mediterranean grain supply. When John Seismos, the prefect of the city, threatened at the height of the crisis of 626 to raise the price of bread from three to eight folleis, God did not allow him to do s0.l The sources have little more to say directly on the matter, rendering it all the more difficult to answer the three questions of fact posed by the loss-above all-of the Egyptian granary. Of what significance was Egyptian supply to the grain economy of the later Roman Empire ? How absolute was the disruption of this and other Mediterranean shipments during the seventh and eighth centuries ? If the loss was really grave, and if little could be expected from provinces within the world of Islam, what adjustments were forced upon a reduced empire, now none too rich in grain-producing regions ? These questions the historian can answer only by analyzing development over a wider span of time than he ordinarily cares to consider. He must glance backward in time to the fourth century, and forward to a date not far short of the Crusades. To settle upon the year 330 as a point of departure is not at all difficult. The foundation of Constantinople, important as it was in a military sense and in a political as well, had an impact no less fundamental on what men chose to produce and trade in the Mediterranean. By creating, in effect, a new deficit area, the establishment of the capital gradually forced upon Egyptian exports a new direction and a new organization. The importance of Egypt's role begins with the first Constantine. Long before 1453, when the last Constantine fell in defense of his city, the problems of the seventh century had been solved and replaced by others of a different nature. Later centuries, then, offer nothing so dramatic as a city's birth to serve as a terminal date, a point when old difficulties are surmounted and new ones only begin to appear. While a specific terminal date cannot be found, there is much to be said for considering-as a whole-the years 820Chronicon Paschale, ed. Bonn, I, p. 716.

J O H N L. T E A L L

867, the generation of the Amorian dynasty, the immediate predecessors of the Macedonians (867- 1056). In fact, it has lately become apparent that many characteristics of the flourishing society over which the latter house ruled can be traced into the earlier years of the ninth century and even further back. The commercial vitality and urban growth implied in the sources for the reigns of Nicephorus I (802-811) and Theophilus (829-842) presuppose an agricultural surplus stored up or in the hands of men willing and able to export their grain. Even the obscure reign of Constantine V (741-775) may not have lacked achievement. One year at least witnessed remarkably low grain prices, attributed by some to high production. I n the second half of the eighth century ships were at hand to mount expeditions in the Black Sea, while reserves of grain cushioned intermittent crises in Sicilian relations. Thus it is by no means incautious to assert that the Macedonians could build upon firm foundations in agriculture and commerce laid by their predecessors. The crises of 619 and 641 seem to have been met, and met successfully, generations before Basil the Macedonian set out on the adventures destined to win him a t h r ~ n e . ~ To meet these crises, and others no less challenging, demanded of the men of the seventh and eighth centuries the greater part of their energies, and, with little left to expend on discussing what they had actually accomplished, the records they kept were scant and unsatisfactory. Two biased and laconic chronicles, a textbook of military strategy, a few saints' lives : this brief enumeration all but exhausts the list of literary materials wherein the answers to our questions must largely be found. Thus the nature of the adjustment must be traced with the aid of materials from the ninth and tenth centuries, but our examination of them will be based upon understanding that the major shifts in demand and supply were long past, that a pattern already shaped now became fixed and ~isible.~ To examine systematically the period after the death of the last effective representative of the Macedonian dynasty in 1025 would raise new problems beyond the scope of the present essay. The Empire survived the loss of Egypt by developing, in part fortuitously, in part deliberately, a system of supply consonant with new needs and demands. This system rested upon a certain social pattern in the village that, with Basil 11's death, lost its last great defender. As control over sources of supply slipped from the emperor's hands, the means of distribution, too, began to change. In the eleventh and twelfth centuries, the markets reflect the interests not only of a potent land-owning class, but of foreign merchants, engaged in what was developing into a truly Mediterranean grain trade.4 See below, pp. 105, 131, 137f. From the fourth to the sixth centuries the grain economy was largely predictable, in a state of equilibrium. After the death of Justinian and until the latter half of the ninth century, the Eastern Empire fought off invaders from literally all points of the compass. The grain economy reflected the continuous emergencies of chronic political upheavals. With the gradual cessation of the worst of the invasions, demand and supply reached a new equilibrium comparable to the old, but bearing the marks of centuries of crisis. 4 A. R. Lewis, Naval Power and Trade in the Mediterranean (Princeton, 1951)~ p. 245, and references. 2

3

T H E GRAIN SUPPLY OF T H E BYZANTINE E M P I R E

91

Within our chronological framework, the economic categories of demand, supply, and distribution require a systematic exposition, although the condition of the sources obviously will determine (and in this instance will hamper) the completeness of any such treatment. In general, it is far easier to analyze qualitative aspects of change than it is to assess the quantities involved in the adjustment of the Byzantine grain economy. I t is possible, for example, to discover the kinds of grain consumed by the urban proletariat and the civil and military servants of the emperor, but fluctuations in the size of these groups can be determined only relatively. Since it requires knowledge of the activities of a merchant class largely ignored by the sources, it is far more difficult to explain export variations than to analyze demand. Fortunately, the political threat inherent in famine gave the emperor sufficient reason for interfering, throughout the period under review, in the equilibrium of supply and demand. Such interference frequently manifested itself in laws dealing with market organization and with waste land, and these are valuable indices of both the impact of the loss and the subsequent success of the agricultural adjustment.

A careful reexamination of the sources on the grain economy of the fourth to the sixth centuries confirms a generalization familiar from previous studies of the subject. Until the very end of the period, a steady, predictable supply of grain from Egypt was absolutely essential to satisfy the needs of Constantinople and the expeditionary army in the eastern province^.^ This striking dependence of the great city on its Egyptian granary was caused, in the first place, by what an economist would term an almost complete inelasticity of demand for bread. As Procopius remarked, the low income of the urban masses permitted them little or no choice among the possible foods they might c ~ n s u m eBread . ~ remained an essential item in their diet, and their choice among the grains to be used for baking it was similarly limited, by pressures of cost and income, to barley or to the cheaper, inferior extractions of wheat.' From these 5 See the comments of G. Bratianu, "Etude sur l'approvisionnement de Constantinople," Etudes byzantines d'histoire Pconomique et sociale (Paris, 1938), pp. 129-181, and for analysis of sources, G. Rouillard, L'administration civile duns I'Egypte byzantine, 2nd ed. (Paris, 1928), pp. 121-124, and A. C. Johnson and L. C. West, Byzantine Egypt: Economic Studies (Princeton, 1949).To the sources discussed there may be added a paragraph on Egypt, relevant to the fourth century, from A. A. Vasiliev, ed. and trans., "The Expositio Totius Mundi," Seminarium Kondakovianurn, VIII (1936), pp. 12 f., 23: "In such a way the soil [of Egypt] yields every year and is profitable t o all the provinces. For Constantinople of Thrace is for the most part fed by Alexandria; likewise, the eastern regions, especially on account of the army of the emperor and the war with the Persians, because no other province can suffice but the divine Egypt." There were, of course, other resources available t o Constantinople and the army, upon which both drew. See below, Appendix B. Procopius, Anecdota, xxvi, 19, ed. and trans. H. R. Dewing, VI (London, New York, 1914-1940), pp. 308f. Hereafter, Procopius (ed. Loeb). See also Theophanes, Chronographia, ed. de Boor, I, 230. See, in general, the study of Ph. Koukoules, 'Ov6~crra ~ a €760 i & p ~ w v ,' E-rra-rqpi~ 'E-ralp. BwZ. t ~ o w 6 G v ,V (1gz8), pp. 3off.,esp. 44f. (Hereafter, Koukoules, "Kinds of Bread"), and his later republication of the same material in B w l a v r l v i j v P i o ~~ a -rroh1~1op65, i V (Athens, 1948-52), pp. 47-62. (Hereafter, Koukoules, Byzantine Lije), and for the economics of the "competition among grains" the excellent study of N. Jasny, The Wheats of Classical Antiquity (Baltimore, 1944), as well as

92

J O H N L. T E A L L

the bakers produced popular or "dirty" bread, carefully distinguished in the sources from the finer loaves consumed by the wealthier ~ l a s s e s . ~ With increased demand, the result of a phenomenal population growth, supply from Egypt became ever more essential in fourth-century Constantinople. Thanks to imperial pressure, the natural attraction exerted by a wealthy senatorial class, and the advantages of metropolitan living in the later Roman Empire, immigration had enlarged Constantinople by about the year 400 to a size of probably 500,ooo souls within the walls.9 Shortly thereafter, however, the trend was reversed. During the fifth and early sixth centuries the virtually insoluble problems of administering and controlling the metropolis thus created forced the emperors to abandon their policy of fostering urbanization and to try instead to check the growth of their great city. Inasmuch as the population of a pre-industrial city increased significantly only when immigration from without compensated for a lower birthrate among the urban classes, it is probable that the late fifth and early sixth centuries were years of stagnation, or, at most, of much slower growth.1° A third period of absolute decline may be dated from 542, when the great bubonic plague or "pandemic" that struck Constantinople inaugurated what comparative materials and direct evidence alike lead us to suspect were generations, even centuries, of a surplus of deaths over births.ll In one important respect, then, namely the diminution of demand, the later sixth century witnessed a significant alteration in the structure of the Mediterranean grain economy, a change that would render it easier for Constantinople to subsist from less abundant granaries once the loss of Egypt forced it so to do. How the Empire managed to survive becomes even more comprehensible if we examine the changing character of military demand at the end of the sixth century, in particular, certain crucial developments concerning the size and location of the forces as well as the administration of the services of supply upon his more popular "Competition among Grains in Classical Antiquity," American Historical Review, XLVII (1941-1942), pp. 747f. HOWimportant grain-either wheat or barley-was in the popular diet may be deduced from the account of the riots of 555 in Theophanes, ed. de Boor, I, 230. Millet was particularly favored by the Slavs, but held in contempt among the Byzantines : Maurice, Strategikon, XI, 5 ; IX, I , 3, ed. J. Scheffer (Uppsala, 1664), pp. 273, 204f., z14f.The low estate of barley, in contrast to wheat, may be seen in Palladius, Dialogus de vita S. Ioannis Chrysostomi, trans. H. Moore (London, 1921), p. 142. Koukoules, "Kinds of Bread," pp. 38-41. See, for comparison, Vita Symeonis Stulti, chaps. 49, 55, edited in Ioannes Bollandus, et al., Acta Sanctorum..., 3rd ed., January I (Brussels and Paris, 1863-); July I, 144, 147. (Hereafter, Acta SS.);also Vita Porphyrii ep.Gazelzsis, c. x, ed. and trans. H. Grkgoire and M.-A.Kugener, Marc le diacre: Vie de Porphyre, e'vdque de Gaza (Paris, 1930))p. 10. See Appendix A. lo On immigration as essential to urban growth, see W. S. Thompson, Population Problems (New York, 1953))p. 75. On overpopulation, see L. BrChier, La civilisation byzantine (Paris, 1950))pp. 81 f.; Les institutions de 1'Empire byzantin (Paris, 1949)) p. 1g4f. l1 Sources cited and discussed by E. Stein, Histoire du Bas Empire, I1 (Paris and Brussels, 1g4g), pp. 756-761.Particularly revealing are the statements of John of Ephesus, in W. J. Van Douwen and J.P. N. Land, Commentarii de beatis orientnlibus et historiae ecclesiasticae fragmenta (Amsterdam, 1889)~ pp. 227-240, and of Evagrius, Historia ecclesiastics, IV, 29, ed. Migne, PG., 86, cols. 2752f. The latter stresses the repetitive and continuing nature of the plague. For the significance of this aspect, see Y. Renouard, "ConsCquences e t intCr6t dkmographique de la Peste Noire de 1348," Population, I11 (1948), pp. 459-466; J. Saltmarsh, "Plague and Economic Decline in England in the Later Middle Ages," Cambridge Historical Journal, V I I (1941)~pp. 23-41; C. F. Mullett, The Bubo& Plague and England (Lexington, Ky., 1956)~pp. 12-41.

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E EMPIRE

93

which they depended.12 It was with difficulty that Tiberius or Maurice found men for their armies.13 Even the scattered figures given in our sources suggest that the manpower shortage, so threatening t o the success of imperial policies, stemmed not only from a more pressing need for soldiers, but from a quantitative decline in the expeditionary forces as well. In determining the impact of military demand for grain, the location of the troops would be of equal importance to their number. Thus it was fortunate for Byzantium, when the crisis came, that military demand was decentralized. While it yet contained some federates and their dependents, Constantinople had certainly ceased by the late sixth century t o resemble, as it once had, a great armed camp grouped about a warrior emperor.14 Finally, from the very political weakness of a Tiberius or a Maurice the people of Byzantium were ultimately to derive a certain economic a d v a n t a g e . Unlike predecessors such as Julian, Anastasius, and Justinian, neither Emperor seems consistently to have exerted the authority necessary to ensure inter-provincial transport of grain, and to support frontier armies from distant centers of supply in the Mediterranean provinces. In East and West the field armies of the early seventh century drew upon the localities in which they found themselves and were not, therefore, seriously or directly threatened by the loss of Egyptian supply.15 l q h e age of Justinian may be taken as a starting point although Procopius' presentation of statistics should encourage skepticism: G. Downey, "The Persian Campaign in Syria in A.D. 540," Speculztm, XXVIII (1953)) pp. 340-348, esp. 343ff. A number of converging indices suggest that the size of Justinian's armies generally ranged from I 5,000 to 2 5,000 men, although smaller forces were frequent and larger hosts seem to have been collected only with effort. See, e.g., Procopius, Bellum Persicum, I, xiii, 23, xviii, 5 (ed. Loeb, I , 108, 160); Idem, Bellum Vandalicum, I, xi, 2 (ed. Loeb, 11, 102); Idem, Bellum Gothicum, I, v, 2 (ed. Loeb, 111, 42); Agathias, 11, 4 ; 111, 24 (ed. Dindorf, 11, 185, 281). Cf. J. R.Bury, History of the Later Roman Empire from the Death of Theodosius I to the Death of Justinian (A.D. 395-565)) I1 (London, 1g23), p. 78 (hereafter, Bury, Roman Empire 395-565). 13 S o t e the implications of Menander, frg. 15, 19; Excerpta historica iussu Constantini Porphyrogeniti confecta I : Excerpta de legationibus, ed. de Boor, pt. I (Berlin, 1903))pp. 208, 217. I n 583-584, Maurice's troops were estimated a t 5,000 by John of Ephesus, Historia ecclesiastica: pars tertia, 111, 13, ed. and trans. E. W. Brooks, Corpus scriptorum orientalium, ser. 3, t. 111, versio (Louvain, 1936)) p. 102; but contrast the calculations of P. Goubert, Byzance avant l'lslam, I (Paris, 1951)~p. 45, who does not seem to give any valid reason for preferring the larger figures of Theophanes and Joannes Riclarensis to those of John. I n 587, in Thrace, Comentiolus commanded a garrison of IO,OOO, of whom 6,000 were fit for active fighting: Theophylactus Simocatta, Historiae, 11, 10, 9, ed. de Boor (Leipzig, 1887), p. go. For additional sources, see below, note 24, and for general discussion, E. Stein, Studien zur Geschichte des byzantinischen Reiches (Stuttgart, 1919), pp. 118-121. l 4 I n general, R. Grosse, Romische Militargeschichte von Gallienzts bis zum Beginn der . . . Themenverfassung (Berlin, 1920), pp. 6off. References to the quartering of troops in the city during the fourth and fifth centuries: Codex Theodosianus, ed. Th. Mommsen and P. Meyer (Berlin, ~ g o j )VII, , 8, 13; I ; 3; 5. (hereafter, C. Th.). Under Justinian, scholariz and domestici are found in various provinces of Asia Minor: Edictum Iustiniani (ed. Th. Mommsen, et al., in Corpus Iuris Civilis: Institutiones, Digesta, Codex, Novellae [Berlin, 1889-19281, VIII, 3, 3 (hereafter, C. J., Ed. J., Nov. J.). Also, Procopius, Anecdota, xxiv, 24f. (ed. Loeb, VI, 289); Theophanes, ed. de Boor, I, 236. There were still barbarian federates or allies within the wall: Procopius, Anecdota, xxviii, 24 (ed. Loeb, VI, 277). I n 578, the wives and children of Gothic federates remained in Constantinople while the fathers of the families went off to frontiers in Persarmenia and Africa: John of Ephesus, 111, 13 (ed. Brooks, 102). Maurice, significantly, relieved the citizens from their burden of quartering troops: Novellae Constitutiones Imperatorum post Iustinianum, Coll. I, nov. 19, ed. K. E. Zacharia v. Lingenthal, Jus Graeco-Romanum I11 (Leipzig, 1856-1887), p. 32 (hereafter, Zacharia, Jus Gr.-R.). Possibly the increasing military significance of the popular parties is to be explained by a numerical decline of the professional military garrison: G. Manojlovi6, "Le peuple de Constantinople," Byzantion, XI (1936))pp. 631-634. l5 Support assured by ecclesiastical authorities: Evagrius, Historia ecclesiastica, VI, 11 (Migne, PG, 86 bis, col. 2860); Gregorius I, Epist. IX, 27; MGH, Epist. 11, 60. I n 602, Maurice ordered Peter to lead his forces into winter quarters in enemy lands across the Danube so that the granaries within

J O H N L. T E A L L

Indirectly, however, the loss of Egypt constituted a true crisis from the point of view of military demand. No longer able, after the seventh century, to draw upon the resources of that province, the emperors must have been seriously limited in their choice of an alternative source for the support and reward of their armies. With the province gone, it became all the more difficult to supply the militia praesentalis with the annona to which law and custom entitled it.16 If Constantinople were ever again to shelter a large armed force, its support would obviously have become a problem of the first order for the emperor. The maintenance at Constantinople of foederati and their families, such as those Tiberius had imported within the walls,17would then have been difficult if not impossible. Finally, the emperor could not have promised, as easily as had Maurice, to nourish the veteran, once his term of service was past, in the city's homes for old men.18 We shall see in what ways Heraclius and his successors avoided these problems. Patterns long-established are not easily shattered, and the habit of dependence on the subject province of Egypt was no exception to this rule. The decline and readjustment of civilian and military demand remained unmatched by any concomitant reorganization of the supply system, by any attempt to exploit to the full other producing areas and thus to end an often dangerous dependence upon one source of supply.19While costs, political difficulties, and administrative weakness may have made Egyptian grain deliveries even more unpredictable under Maurice and Heraclius, the events of the seventh century had nonetheless Inasmuch as these crises a profound effecton the provisioning of Con~tantinople.~~ forced the Emperors to exploit more consistently and thoroughly resources to which, in the past, they had turned only occasionally and under duress, they brought into focus problems that more favorable circumstances had enabled the predecessors of Heraclius to avoid. From another point of view, these same problems may be considered deficiencies in the supply system of the later Roman Empire. They had widened the gap between the quantity of demand and the amount produced and distributed to the Roman province might be relieved of public distributions: Theophylactus Simocatta, V I I I 6 (ed.de Boor, p. 295).This action may be compared with a passage from the Strategikon attributed to Maurice, wherein the author advises that the troops take the enemy harvest and emphasizes the abundance to be found in the lands of the Slavs: Strategikon, IX, I (ed. Scheffer, pp. 204f.);cf. ibid., IX, 3, p. 214f. According to the same source, troops in winter quarters were ordinarily to be given leave to search for their food within the provinces: I, 7, p. 37. These practices contrast with arrangements during the fourth, fifth, and sixth centuries: see, e.g., Ammianus hiarcellinus, Res Gestae, XVIII, 2, 3 ; C. Th. VII, 4, 13; 17; VIII, I , 10; and Malalas, Chronographia, ed. Bonn, p. 467. For an interesting change a t Rome during the Gothic Wars from water-borne supply to local subsistence, cf. Procopius, Bellum Gothicum, I, xiv, 17; 11, xxiv, 14 (ed.Loeb, 111, 148, IT', go), with ibid., 111, xv, 9 ; 111, xxxvi, 2 (ed. Loeb, IV, 278, V, 2). ls If u ~ ~ q p i u ~may o v be taken in the narrower sense of grain distributions, we may have in 612 one of the last instances of a new grant of military bread. I n that year, Heraclius enrolled soldiers drawn from the bucellarii of Priscus into the domestics and gave them TO 2c E0ovs u~-rqpiu~ov (Nicephorus, ed. de Boor, p. 6).Distributions of this type of bread ceased in 626: Chron. Pasch., ed. Bonn, I, P. 715. l7 Above, note 14. 18 Theophanes, ed. de Boor, I, 274. l9 See below, Appendix B. 20 Nicephorus, ed. de Boor, p. 12.

T H E G R A I N SUPPLY O F T H E B Y Z A N T I N E E M P I R E

95

satisfy the demand. Among these problems, sheer physical risk assumed increased importance toward the end of the period under review. The attacksactual and threatened-of Avars, Slavs, and other peoples upon the Mediterranean ports led frightened merchants to avoid, and even to abandon centers such as Rodosto or Thessalonica.21To physical risk of this nature may be added the more abiding consideration of high transportation costs. Significant longdistance export of grain seems to have existed only insofar as public authority would direct it, forcing the taxpayer to assume its unwelcome burden. Thus at Edessa in 503, the soldiers whom Anastasius had collected for an offensive against Persia consumed bread furnished from Egypt. The civilian population, in striking contrast, was threatened with tar vat ion.^^ Transportation costs were among the factors preventing smaller cities situated at some distance from the sea from drawing extensively upon Egyptian grain supplies. Only where size and the degree of specialization necessitated it, only where transportation facilities permitted it, did the local demand represented by the smaller cities seriously diminish the surplus available for Constantinople and the army. A great city such as Thessalonica, fulfilling because of its geographical location and its internal economy the conditions just mentioned, upon at least one occasion did compete with Constantinople for Mediterranean supply.23In contrast, the smaller centers probably affected Constantinople only insofar as their cumulative demand diminished the surplus which the surrounding province could export. The evidence is too fragmentary to permit us to trace a curve and assess fluctuations in local demand for local resources from the fourth to the sixth centuries. Certainly it did not diminish. I t may even have increased as new cities were founded and the old were stimulated to new growth after the establishment of Constantinople. Whatever the direction of the curve, the lack of data necessary to delineate it may not be so serious as it seems. As long as Egyptian grain remained available, transportation costs and physical risk were equally crucial administrative problems.24 The great plague of 542 introduced another problem to harrass the officials administering Constantinople's grain supply. A serious decline in manpower left its mark alike in the chronicles, on imperial legislation, and on the military

"

As concerns Rodosto, see Procopius, De Aedifciis, IV, ix, 17-21 (ed. Loeb, VII, 296). For Thessalonica: Miracula S. Demetrii, sec. 65 ( A c t a SS., Oct. IV, 127f.). Chronicle of Joshua the Stylite, cc. lxx, lxxvii, ed. and trans. W. C. Wright (Cambridge, 1882)~ PP. 58, 62. 23 Miracula S . Demetrii, secs. 69f. (Acta SS., Oct. IV, 129-131). On the high cost of transportation, see C. Th. XI, I , 22 ; Gregorius Nazianzenus, Oratio XLIII, cc. xxxv, xxxvi (Migne, P G , 36, cols. 544f.), and-above all-Ioannes Lydus, De magistratibus populi romani, 111, 16, ed. Bonn (1837), pp. 255f. 2"conomic and social aspects of the cities from the fourth to the seventh centuries are discussed by Kirsten, "Die byzantinische Stadt," pp. 10-17, and documentation for the growth of cities along the routes to Constantinople will be found in E. Gren, Kleinasien und der Ostbalkan in der wirtschaftlichen Entwicklung der romischen Kaiserzeit (Uppsala, 1941). The provisioning of the cities has been ) , 260f. discussed by A. H. hI. Jones, T h e Greek City from Alexander to Justinian, (Oxford, I Q ~ O pp. His views on the relation of the city and the village should be compared, however, with those of P. Petit, Libanius et la vie municipale ri Antioche (Inst. Franq. d'Arch. de Beyrouth: Bibliothhque Arch. et. Hist., t. Ixii), (Paris, 1955)~p. 307f., and should also perhaps be revised in the light of such sources as the capital description of the market a t Amida in the sixth century, to be found in T h e Syriac Chronicle of Zachariah of Mitylene, VII, 5, trans. F. J. Hamilton and E. W. Brooks (London, 1899), p. 162.

J O H N L. T E A L L

policy of Tiberius and Maurice, forcing these two Emperors to value the winning of men more highly than the possession of land. Their policy was essential for both agricultural and military considerations. Given the conditions of soil and climate in Mediterranean lands, and the consequent need for the close and continuing union of man and land, any temporary or permanent shortage among the peasant population seriously threatened the successful completion of the "Farmer's Year" and the production of abundant harvests. Thus manpower shortage may be considered the final factor augmenting a discrepancy between demand and supply already created, on the one hand, by the needs of the consuming groups, on the other, by such determinants of production and distribution as high transportation costs, competition for grain among the larger cities, and physical risk.25 To eliminate this discrepancy, to close the gap between what was needed and what the economy could produce, the emperors of the fourth, fifth, and sixth centuries applied to the system what remedies they could. The peasant was bound to the land, the shipmaster to his transport With an effort limited alike in its scope and in its success, the emperors in the east attempted to settle barbarian tribes upon waste or unproductive lands.2' Masterless or abandoned land was assigned by them to those who would work it and pay the taxes due on it. Particularly in the third and last of these measures it has been possible to find an indication that the ratio of man to land within the productive economy remained at a perilously low level in view of the demands made upon it for taxes and fo0dstuffs.2~ 25 Evidence of a critical manpower sho tage stemming from the plague of 541-544 may found in Nov. Just., cxxii, pr.; Procopius, Anecdota, xxiii, 19-21 (ed. Loeb, VI, 274-276); for subsequent effects, see Evagrius, V, 19 (Migne, P G , 86, col. 2833) ; John of Ephesus, 111, vi, 27 (ed. Brooks, 252). During the negotiations with the Persians in 580, the envoy of Tiberius I1 was quite willing t o surrender Persarmenia to the enemy, while insisting that the inhabitants who had fled to Roman authority be kept within the boundaries of the Roman state: Menander, frg. 19 (ed. de Boor, p. 217). For other, albeit less telling references, idem, frg. 20, 31 (ed. de Boor, pp. 220, 475). 26 For an unusually clear survey of the problem of the colonate in its juridical aspects, see Lemerle, "Histoire agraire," Revue historique, CCXIX, pp. 37, 45-48, who points out, "le souci d'approvisionnement ou celui du recrutement de 1'armCe. . .s'effa~entdevant le souci du rendement de l'imp8t." Small peasant proprietors who were free survived even into the seventh century; see the data from hagiography gathered by A. P. Rudakov, Ocherki Vyzantiiskoi kul'tury Po d a n n y m grecheskoi agiogvafci (Moscow, 1g17), pp. 175f. The navicularii deserve a detailed study which would consider the issues raised by G. Mickwitz, Die Kartellfunktionen der Ziinfte und ihre Bedeutung bei der Entstehung des Z u n f t z ~ e s e ~ z s (Finska vetenskaps-societeten: Commentationes humanarum litterarum, VIII, 3), pp. 200ff. HOWthe guild worked in the fourth century has been described by A. Piganiol, L ' E m p i r e chre'tien : 32.5-395 (Paris, 1947), p. 289f. For evidence of activity late in the sixth century, see John of Ephesus, 111, i, 33 (ed. Brooks, pp. 30f.) possibly Miracu2a S. Demetrii, loc. cit., note 23, above. 27 Constantine provided for agricultural settlement of veterans in C. Th. VII, 20,3 (320),but the law is not preserved in Justinian's code and Maurice had to promise support for veterans in the gerocomia: above, note 18. On the limitanei, see S. Mazzarino, Aspetti sociali del quarto secolo, (Rome, 1951), pp. 330-344. I n believing that barbarian settlements were, on the whole, few, restrained, and unsuccessful, I have adopted the point of view of P. Lemerle, "Les invasions et migrations dans les Balkans," Revue historique, CCXI (1954), pp. 265-309, esp. 281. The problem deserves a fuller investigation. 28 For a summary of recent work on the epibole or forced attribution of waste lands in the "proto-Byzantine" period, see Lemerle, "Histoire agraire," RHist., CCXIX, pp. 37f. To the literature there noted, add the helpful and suggestive study of J. Danstrup, "The State and Landed Property in Byzantium," Classica et Medievalia, VIII (1947), pp. 222-262. See also below, pp. 138-139.

T H E GRAIN SUPPLY O F T H E BYZANTINE E M P I R E

97

The political events of the seventh, eighth, and ninth centuries were destined to make these problems even more critical. Well before the end of that period new centers of consumption within the world of Islam laid claim to the grain produced in Egypt, North Africa, and Sicily. The produce of Egypt in particular seems to have soon been lost to Byzantium for several reasons, of which political hostility is only one and perhaps not the most important. Production in the first centuries after 641 declined along the Nile.29 The grain the land did produce had first to satisfy the needs of a warrior aristocracy situated in the garrison cities. New markets for the exportable surplus were found in the holy cities of the Hedjaz and in Nubia in exchange for slaves.30 The destination of North African and Sicilian grain is a problem rather more difficult to solve. In neither area was warfare during the early stages of the conquest so destructive as to ruin agricultural productivity or to terminate shipping. North African and Sicilian ports remained in contact with Constantinople, while sporadically the emperors manifested in their western provinces an interest that led to more effective defense and tighter administrative control. Thus grain may well have been exported at different periods from Carthage, Catania, or Syracuse to C ~ n s t a n t i n o p l eAt . ~ ~the same time, such shipmentsif indeed they existed --were in all likelihood occasional, small in quantity, and organized under private a~spices,~2 for during the sieges of 674-678 and 717-718 Constantinople had to exist without dependence upon western supplies; indeed, throughout the earlier half of the eighth century shipments from Sicily 29 See C. H. Becker, "Grundlinien der wirtschaftlichen Entwicklung Aegyptens im Mittelalter," Islamstudien, I (Leipzig, 1924-1932), pp. 201-217; M. de Bouard, "Sur 1'8volution monktaire de 1'Egypte mCdikvale," L'Egypte contemporaine, XXX (1939)~pp. 427-459, esp. 436f.; D. C. Dennett, Conversion and the Poll T a x in Early Islam (Cambridge, Mass.), esp. pp. 110-115, who offers an explanation differing from Becker's for the agri deserti and the flight of the peasant; Lewis, Naval Power and Trade, pp. 83f., 87-97, who suggests that a true decline in prosperity was not visible until the eighth century; corrections to Lewis' use of the evidence on seventh-century Alexandria will be found in P. Kahle, "Zur Geschichte des mittelalterlichen Alexandria," Der Islam, XI1 (1922), pp. 29ff. 30 Exports to the Hedjaz and Nubia: de Bouard, "Evolution monktaire," pp. 436ff.; G. F. Hourani, Arab Seafaring (Princeton, 1951), p. 60; G. Wiet, L'Egypte musulmane de la conqu&te arabe jusqu'd la p. 113; Kahle, "Alexandria," p. 31. conqzdte ottomane in Pre'cis de l'histoire de I'Egypte, I11 (Cairo, 1932)~ 31 See below, Appendix C. 32 In the absence of indisputable evidence to the contrary, it seems a reasonable assumption that with the loss of Egypt there vanished the interlocking systems of forced transports through shipmasters' guilds and bureaucratic exploitation of a subject province for its grain. So P. Charanis, "On the Social Structure of the Later Roman Empire," Byzantion, XVII (1944-1945)) p. 49. The outlay in money and men necessitated by these systems may have encouraged the impoverished Byzantine state to abandon such policies in the seventh century. For an estimate of costs in the administration of Egypt, see A. C. Johnson, Egypt and the R o m a n Empire (Ann Arbor, 1951)) p. 153. But the obscure tax nauticatio, levied by Constans I 1 in the West, needs more elucidation than it has received from, e.g., J. K. Danstrup, "Indirect Taxes a t Byzantium," Classica et Medievalia, VIII (1946)) p. 167. See Paulus Diaconus, Historia Langobardorum, V, 11, ed. G. Waitz, M G H , Script. Rer. german. (Hanover, 1878), p. 192, and the Liber Pontificalis, ed. L. Duchesne, I , p. 346. Throughout the Byzantine possessions in southern Italy-where often the government was in no position to organize regular deliveries of the annona-the grain seems to have been turned over to local needs; so J. Gay, L'Italie mkridionale et l'empire byzantin (Paris, rgoq), pp. 178-18 I , 209,232 f. For a t least one case of coemptio used, oddly enough, for purposes of resale to the Arabs in Sicily: Cedrenus, Compendium historiae, ed. Bonn, 11, 357.

7

98

J O H N L. T E A L L

must have been precarious at best.33 Thus it is far from surprising that the loss neither of North Africa nor of Sicily had any discernible impact upon the grain economy of the Empire in the east. In the latter region the significant adjustments, whether of demand or of supply, had already been undertaken, and it is to them that we now turn. I.

The Demand for Grain

Bread, young Epiphanius is made to remark in the Life of St. Andreze! the Fool, is something we eat each day.34 So it was during the tenth century in a family of considerable means; and so it would have been in popular circles about the year 600. For upon the quality of demand, or what men chose to eat, the loss of the Mediterranean granaries had no appreciable effect. Bread, supplemented by fish and vegetables, remained the dietary staple, with flour from the wheat of the naked emmer family still preferred as the ingredient for the best of the several varieties. Such are the conclusions reached after reading the lives of the eighth- and ninth-century saints, but not until the twelfth century does Byzantine literature in the satirical poetry of Theodore Prodromus provide a tolerably complete account of popular diet.35 Yet, since the information to be found therein coincides with both the materials from the sixth century and the fragmentary offerings of hagiography in the ninth, the poetry at least reinforces the conclusion that the Byzantine period saw no essential changes in what men chose to eat and drink at Constantinople. Prodromus spoke mainly about food in his iong poem addressed to the Emperor Alexius protesting the evil state of the writer and man of letters in Comnenian society: had he chosen a trade even so generally scorned as that of embroiderer, his cupboard might be full of salt fish, bread, wine, tunny, and other delicacies ;36 or, consider the fine estate of the shoemaker. A neighbor of his, Prodromus continues, who was a cobbler or pseudo-shoemaker lived each day on tripe, Vlach cheese, and delicious stews;37in fact, at one time he himself had tried to become a shoemaker, not in order to win the first quality of bread, but only "that of the middle" (-rb ~~~oo~CtBapov), but had wounded himself on the job and had had to spend time in the hospital.38The baker's man, of course, lives very well indeed, as do the vendors of sour milk and pepper mills together with the dyers of silk.39As for another neighbor, a sieve-maker, 33 For the two sieges: Theophanes, ed. de Boor, I, 353ff., 384ff.; Nicephorus, ed. de Boor, pp. 32f., 52 f. For analysis of these and the eastern sources as well, see M. Canard, "Les expeditions des arabes contre Constantinople," Journal asiatique, N.S., CCVII (1925-1926), pp. 40-102. Dennett, "Pirenne and Muhammad," p. 171, does not believe that all communication between Constantinople and the West was cut off from 674 t o 678. 34 Vita Andreae Sali, sec. 47 (Acta SS., May VI, 27f.). See, e.g., ibid., sec. 23, pp. 14f.; Vita Stephani iunioris in Migne, P G , 100, col. I 124 ; Vita Basilii iunioris, sec. 33 (Acta SS.,Mar. 111,App. p. 27) ; Theodore Prodromus, "To the Emperor," ed. D. C. Hesseling and H. Pernot, Podrnes prodromiques en grec vulgaire (Amsterdam, I ~ I O ) , pp. 72-83. 36 Ibid., lines 25-28, p. 74. 37 Ibid., lines 45-63, pp. 75f. Ibid., lines 79-89, p. 76. 88 Ibid., lines 97f., p. 77, lines 109-113, pp. 77f.

THE G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E

99

he eats unbelievably well, for through the thin partitions of his dwelling Prodromus can smell his fish frying and his meat cooking.40 But Prodromus himself ventures to ask only for bread. Theodore found his protector, but the "Byzantine Villon" remained unsatisfied. In another poem, addressed to Andronicus Comnenus, he grudgingly thanked his benefactor for his gift of twelve medimnoi of wheat four times a year,41 complaining that it barely satisfied his household of thirteen for one month, and where, indeed, were the other things the well-appointed household needed: the clothing, the household goods, the many spices, the vegetables ?42 For the purchase of these his own income was quite inadequate, and he entreated his patron for more assistance.43 Finally, he recounts how his own wife had once all but reduced him to starvation, when, having been locked out by the shrew, he had gained admission to his house and board, which was burdened by a great steaming stew, only by disguising himself as a beggar and speaking in the tongue of the B ~ l g a r s Thus, . ~ ~ except for instances rare enough to become material for the satirist's pen, the lower ranks of Byzantine society seem to have subsisted almost entirely on bread, fish, and vegetables, and, of these, bread seems to have been the fundamental element in the diet. There is no reason to doubt either, that wheat maintained its superior position among the cereals. Millet, according to the eleventh-century medical encyclopedist Symeon Seth, should be eaten only when more satisfactory grains are not at hand.45Since medical writers all too often drew their observations from their more esteemed predecessors, Symeon's statement should not be accepted without the confirmation of actual popular attitudes, which, fortunately, can be found in the Alexiad of Anna Comnena. The armies of Bohemund, the princess reports, fell sick at Dyrrachium when they ate millet, oats remained the food for a cereal she considered ~ n s u i t a b l e Apparently .~~ animals, while barley was still considered to be qualitatively inferior to wheat.47 Thus the rebel Bardanes Turcus gave up wheat and ate barley when the collapse of his political ambitions forced him to adopt the habit of a monk.48Nicon the ~ e t a n i i t e ,while still in his monastery near the boundary of Pontus and Paphlagonia, At hanasius the Athonite, and Luke of Hellas, all alike rejected iip-ro~~dteapo~ and consumed barley bread to manifest their humility.49 Barley remained the ingredient often found in the poorest type of "dirty bread" or f i u ~ r a p 6 and, ~ , ~ ~ inasmuch as barley prices during the Byzantine 40 Ibid., lines 130-144, pp. 78f. Theodore Prodromus, "To the Sebastocrator," ed. Hesseling-Pernot, Pobmes prodromiques, 11, lines 26-28, p. 41. 42 Ibid., lines 29-61, pp. 41-45. 43 Ibid., lines 63-72, p. 45. Prodromus, "To the Emperor," op. cit., I, lines 178-181, p. 34, lines 240--2jg, p. 36. 45 Symeon Seth, Syntagma, ed. B. Langkavel (Leipzig, 1868), p. 137. 48 Anna Comnena, Alexiad, XIII, 2, ed. and trans. B. Leib, 111, pp. 93-95. 47 Symeon Seth, op. cit., p. 137. Theophanes Continuatus, De Leone Armenio, c. iii, ed. Bonn (1838), p. 10. Vita Athanasii dthonitae, ed. in Analecta Bollandiana, XXV (1906), p. 16; Vita Lucae Helladis, c. vi, ed. in ibid., XI11 (1894), p. 84. Cf. Vita Niconis, ed. Sp.Lambros in NIos 'Ehhqvopmjpwv, I11 (1906), p. 136. 60 Koukoules, "Kinds of Bread," p. 43f. 7*

100

J O H N L. T E A L L

period maintained the 2 : 3 ratio to wheat that is attested by late Roman evidence, probably neither the demand for, nor the supply of, this inferior grain changed in relation to wheat on the market at C ~ n s t a n t i n o p l eWithin .~~ the provinces there may quite possibly have been important regional differences. For Greece, the classic land of barley production, the author of the life of Nicon the Metanoite reports an instance of barley bread consumption, in no way remarking upon it as anything out of the 0rdinary.~2On the other hand, even when the grip of poverty lay most heavily upon them, the family of Philaretos the Merciful continued to eat the wheat so abundantly produced in Paphlagonia.53 Among the consumers, hulled spelt continued to find less favor than did the naked subspecies of wheat. Spelt, Symeon Seth had to remind his readers, is quite fit for bread, and, in fact, even more nourishing than barley,54 while Nicetas Choniates seems to reflect a popular aversion to the subspecies of wheat .55In the writing of Symeon and in the twelfth-century satires of Theodore Prodromus, semidakites remained a better type of bread, still to be distinguished from the "dirty" variety.56 Thus the diet of the inhabitants of Constantinople never ceased to be that of a Mediterranean people. For Comnenian society, this assertion can be made with considerable certainty, and there are no references to foods available in the Macedonian period that would contradict it. Rye and millet, so important in the western and northern centers of post-classical agricultural development, made no inroads upon what remained a Mediterranean cereal economy based upon wheat and barley. I t was not in the quality of demand, but in its quantity and in the exploitation of new resources that the Byzantine Empire made its significant adjustment to the loss of its provincial granaries. During the age of iron, ushered in by the reign of Heraclius, famine and plague, together with the curtailment of civic benefits, probably all but halved the population of Constantinople and its suburbs. Only if one turns for comparison to contemporary Rome or to descriptions of the eastern city itself on the eve of its final defeat in the fifteenth century, do the results of famine, plague, and other disasters seem in the seventh century less than absolute, and the history of the early medieval city not one of complete catastrophe. Three outbreaks of plague have been recorded in the sources of the seventh and eighth centuries, but only one seems to have equalled in scope and intensity the initial disaster of 541-544. About the plague accompanying the loss of Egypt in 619, we have only the notice that it occurred.57Of the visitation of G. Ostrogorsky, " L o h n e und Preise i n Byzanz," Byzalztinische Zeitschrift, XXXII ( I Q ~ ) p., 320. Vita Niconis, ed. Lambros, p. 154. 63 Vita Philareti eleemosynarii, ed. and trans. M . H . F o u r m y and H . Leroy, " L a vie de Philarkte le misCricordieux," B yzantzon, I X (1g34),pp. 13 I , 133. 54 S y m e o n Seth, op. cit., p. 136. 65 Nicetas Choniates, Historia: De Andronico Comneno, I, 9 , ed. B o n n (1835), p. 396. Cf. Koukoules " K i n d s o f Bread," p. 44, note 7. 5 6 Prodromus, " T o t h e Emperor," op. cit., IV, 96-108, pp. 77f., and "Stichoi," ed. Hesseling-Pernot, 111, 316, p. 62. 57 Nicephorus, ed. de Boor, p. 12. 51

62

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 101 698, Nicephorus or his source chose to say simply that it lasted four months and was occasioned by the cleansing of the Neorion harbor.58 Thanks to the literary activities of their children and grandchildren, much more is known about how the generation of the mid-eighth century suffered from the great plague of 746. Egypt and Syria, as in 541, were the first to suffer. The cities were harder hit than the country, and the peasant grew grain for his own mouth alone since there were none to buy it from him.59In one respect the disaster of 746 took on a new and significant form: its course ran not directly through Syria to Anatolia and thence towards Constantinople, but indirectly to Africa, Sicily, and the Peloponnesus. In 747 it reached the imperial city, giving Theophanes a chance to try his hand in imitation of Thucydides and Procopius. Apparitions shook men's minds as the houses of rich and poor alike were stilled. Revelry ceased, and the hands of the living proved inadequate to the task of burying the dead.G0 After making due allowance for literary embellishment to his lurid account, enough evidence remains to suggest that this was a pandemic of extraordinary virulence. Both maternal grandparents of Theodore the Studite fell victim to the buboes, leaving the education of their little daughter, Theoctiste, to the care of a charitable house.G1The actions of Constantine V seem to prove that such a tragedy was not unusual in the annals of other families, for the Emperor brought Greeks from the mainland and the islands into the unoccupied sections of the city.62 While disease did its work within the walls, the enemy without fell upon the suburbs. In the early years of the seventh century the European side of the Bosphorus suffered from Avar raids, culminating in the great offensive by land and sea during the campaign season of 626. The Persians and Avars were succeeded by the Bulgars and the Arabs, and according to Nicephorus the latter harassed the Hebdomon daily from their fleet in the sea of Marmora during the great siege of 674-678. At the end of the century civil war combined with enemy action to bring about one of the darkest hours in the city's life. In view of the history of the preceding twenty years, marked as well by the depredations of Thracian brigands that remained unchecked until the reign of Constantine V, there is little reason to doubt the report of an Arab eyewitness on the conditions of the suburbs in 717:63 Ibid., p. 40; cf. Theophanes, ed. de Boor, I, 370. Michael the Syrian, edited and translated by J. B. Chabot, Le chronique de Michel le syrien, I1 (Paris, 1899-I~IO), pp. 506f. His comment may be compared with the observations of TVilliam Hamilton on the effects of the plague of 1837 in Asia Minor. See L. S. Stavrianos, The Balkans since 1453 (New York, 1958)~pp. 134f. 60 Theophanes, ed. de Boor, I, 423. 61 Theodorus Studites, Oratio XI, sec. 3, 4 (Migne, P G , 99, cols. 805-808). 62 Theophanes, ed. de Boor, I, 429; Nicephorus, ed. de Boor, p. 66. A clearer idea of the economic impact, and probably of the objectives as well, of this measure and the peasant settlements also undertaken by Constantine V may be gained by a comparison with the strikingly similar policies adopted by Mehmed I1 after 1453. See Critobulus, De rebus gestis Mechemetis, 11,i, I, xxiii, 1-3 ;III, xi, 1-7; ed. K. F. Miiller Fragmenta historicorurn graecorum, V (Paris, 1883)~ pp: 105, 118, 126f. 6a From the Khitab a1 Uyun, ed. M. de Goeje, Fragmenta hzstoricorum arabicorum, I (Leipzig, 1869)~ and trans. E. TV. Brooks, "The Campaign of 716-718 from the Arabic Sources," Journal of Hellenic Studies, XIX (1899))p. 23. See below, note 182.The period as a whole is covered in G. Ostrogorsky, History 58

59

J O H N L. T E A L L

"And when Maslama had encamped at Kustantiniyya, he blockaded the inhabitants and attacked them with siege engines; and he collected the provender and corn, and they were conveyed to him from the outlying and exposed lands of the Romans; and they came to him in wagons, until that which was brought him became like mountains and those stores abounded in his camp ; and he excluded the inhabitants of Kustantiniyya from all gainful occupation by land and sea. And the district of Marakiya [Thrace] was at that time laid waste, having been laid waste in that civil war; but at the present time it is well-peopled. And this was in their time one of , the greatest weaknesses of al-Kustantiniyya. . . . 9

As Alexandria already had done early in the seventh century, and as Carthage was to do a few decades later, Constantinople might have made up its population deficit in city and suburbs from the refugees circulating in the Mediterranean during the seventh century. There were some in fact, particularly among the clergy, who sought refuge behind the Theodosian walls. In 681 the Council in Trullo expressed the hope that the wandering ecclesiastics lodged in the city might return to their proper stations once tranquility had been restored.64 Possibly such refugees found conditions none too enticing in a city bereft of the important urban privileges it had once enjoyed, among which was the right to "political bread."65 In any case, their numbers failed to compensate for those whom plague and warfare had struck down. During the seventh and eighth centuries that part of Constantinople which was within the walls gave every evidence of a crucial shortage in manpower and of an increase of unoccupied land in parts of the urban precinct that had once been heavily populated. When a series of earthquakes levelled parts of the city walls in 7 3 2 , Leo I11 issued an edict to the "people," informing them that they were not competent to undertake the repairs. Instead, he ordered his tax collectors to add a surcharge of two keratia to the canon. This tax was paid by the Empire (fi puaih~ia),a nd it was the Empire that completed the reconstruction of the walls.66Unable to muster the necessary manpower from the ranks of the popular parties as had Theodosius I1 in the fifth century, Leo increased the taxes in order to hire workmen from outside the walls. Even more indicative of a decline in the city's population are the conditions under which Constantine V reconstructed the aqueduct of Valens in 766. Built late in the fourth century, the aqueduct was essential to the supply of baths and cisterns within the city. After it was destroyed during the Avar attack of 6 2 6 , of the Byzantine State, trans. J . Hussey (New Brunswick, N. J., 1958), pp. 79-162, and to the literature there cited may now be added F. BariSi6, "Le siege de Constantinople par les Avares," Byzantion, XXIV (1g54),pp. 371-395. The Scamars or Thracian brigands are mentioned by Theophanes, ed. de Boor, I, 430. 84 Sacrorum conciliorum nova et amplissima collectio, ed. J . D. Mansi, XI, col. 951. 135 The termination of popular and military bread in Chron. Pasch., ed. Bonn, I, pp. 711, 715. For the hypothesis of a general diminution of urban privileges, see G. I. Bratianu, PrivilBges et franchises mzcnicipales (Paris, 1g36), pp. 95 ff. Theophanes, ed, de Boor, I, 412.

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 103 the ruins outside the walls lay unused until a particularly severe summer's drought forced Constantine to undertake repairs. Gathering artisans from Asia, Pontus, Greece, the islands, and Thrace, he completed the work and The episode is remarkably revealing. The brought water into Con~tantinople.6~ city, often critically short of water in normal times, could, with a decreased population, survive the loss of an aqueduct for more than a century. Repairs, it should be noted, became necessary in something less than a generation after Constantine V had repopulated the urban precincts. No less essential to the city's life than its aqueducts were its cisterns. Built, according to Procopius, to contain the winter's excess of water against the summer's shortages, and designed also to aerate the supply, "not a reign" went by in the earlier history of the city when cisterns did not have to be constructed to answer vital and continuing needs.6$ The seventh and the eighth, together with the thirteenth and the fourteenth, centuries seem to have been the exception to this rule.69Dramatically suggestive of conditions on the eve of theEmpire's collapse, when the city had in effect dissolved into a series of separate villages, are the descriptions Nicephorus and Theophanes provide of cisterns dried up and unused at the time of the great plague. As the corpses piled up, exceeding the number of carts and animals available to carry them off, the living dumped the bodies of the dead into fields, vineyards, gardens, and unused cisterns.70 More specific information is available concerning the cisterns of the imperial proasteion at Hieria and those of the Great Palace. According to an etiological tale, Heraclius' inordinate fear of death by drowning induced him to fill up the receptacles, converting them into vegetable gardens and parks, and the structure at Hieria remained in such condition until the ninth century. ". . .the famed Emperor Basil, seeing the region was endowed with sufficient places suitable for the construction of parks, but that it was scant in water that was potable and clean, straightway set many hands and much zeal to carrying out the earth, and changed the meadow, shortly before conspicuous in its verdure of growing things, back into its former state and made a receptacle of bounteous, flowing water in place of the park." As in the case of the aqueduct of Valens, the absence of the cisterns was apparently none too deeply felt until an expanding population made it imperative that they be recon~tructed.~l Serious as the population decline became, it was never absolute. Even in the darkest of days, none of the city's harbors seems to have been decommissioned Ibid., 440; Xicephorus, ed. de Boor, p. 75. Meyer, Byzantion, p. 5. K. C. Dalman, Der Valensaquadukt (Berlin, 1933); P. Forchheimer and J. Strzygowski, Die byzantinischen Wasserbehalter Konstantinopels (Vienna, 1893), p. 2. The importance of the cisterns is explained by Procopius, De Aedificiis, I, xi, 14f. (ed. Loeb, VII, 90-93). For the difficulties that might result when water supplies were low, see Procopius, Anecdota, xxvi, 23 (ed.Loeb, VI, 308), and Malalas, Chronographia, ed. Bonn, p. 492. I n the tenth century water shortages were again a problem; see Liudprandus, Relatio de legatione constantinopolitana, c. i; ed. J. Becker, Liudprandi opera, (MGH: in usum scholarum), 3rd ed. (Hanover, 191j), p. 175. e3 For comments on the city just prior to its collapse, see materials cited in note 79 below, and the references in BrChier, Civilisation b yzantine, pp. 79 f. 70 Theophanes, ed. de Boor, I, 423, 425; Nicephorus, ed. de Boor, p. 63. Theophanes Cont., Vita Basilii, c. 92 (ed. Bonn, pp. 337f.). 87

J O H N L. T E A L L

permanently. Among the ports of the Golden Horn, the Neorion was dredged by Le0ntius.~2The sources occasionally mention even the installations on the southern shore, less favored by wind and tide; so the adherents of Phocas found it necessary to defend the harbors in both the third and twelfth regions later, the harbor of the twelfth against the attack of H e r a ~ l i u sA . ~century ~ region was reconditioned by Agallianus, turmarch of the Helladic theme under Leo III.74 The chroniclers and hagiographers make it clear, too, that the ports were used for commercial, as well as military, purposes. A trade treaty with the Bulgars distinguished the otherwise obscure reign of Theodosius III.75Composed around the middle of the seventh century, the Miracles of Artemius tell of merchants from Chios, Amastris, and ~onstantino~le-together with an African from Carthage and numerous unidentified nautae-who sought relief from disease and other afflictions at the shrine of the healer saint. Most significant of all, a merchant from Rhodes asked his family to permit him to remain in the city while he prayed for a miracle. On what seem to have been normal, periodic voyages, his sons twice returned to inquire whether he had yet been cured of his malady. The merchants mentioned in the Miracles usually owned their own ships, a detail suggestive of how slight was the need to specialize, how relatively anemic the volume of trade.76 To judge from the actions of the government in time of siege, grain may well have bulked large among the objects of trade such merchants brought to Constantinople, for a state of siege usually implied danger from starvation in the seventh and eighth centuries. The Emperor Artemius, following the precepts of the Strategicon of Maurice and the Tactica of Leo, ordered all those who could not lay in a three years' supply of grain to leave the city before the Arabs attacked.77 So, too, in 742 the siege undertaken by Constantine V forced Artavasdus to despatch^ boats beyond Abydus in frantic search for sup~1y.7~ In this very need to import grain lies an important clue to the numerical strength of Constantinople's inhabitants during the period in question. In the seventh and eighth centuries the population never sank to a level permitting it-as it did in the fourteenth and fifteenth centuries-to exist on its own resources. A description of the city in the year 1437 states that " . . .there can be living therein but barely, as I believe, forty thousand men. These can live from their own vines, fields, and other necessary things within, as is often proved." Since the population for the fifteenth century has been estimated to 72

73

Nicephorus, ed. de Boor, p. 40; Theophanes, ed. de Boor, I, 370.

See, in general, R. Guilland, "Les ports de Byzance sur la Propontide," Byzantion, XXIII (1g54),

PP. 225f.

T. Preger, Scriptores originum constantinopolitanarum, I1 (Leipzig, 1go7), 257. Theophanes, ed. de Boor, I, 497. 76 Miracula Artemii, ed. A. Papadopoulos-Kerameus, Z a p i s k i Istoriko-Filolog. Fakulteta I m p . St. , 3, 5, 9, 33, 39, 55-57. Petersburgskago Un.iversiteta, fasc. 95 (St. Petersburg, ~ g o g )pp. 77 Theophanes, ed. de Boor, I, 384. 78 Ibid., I, 419. 74

75

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 105 have been, at most, 180,000 persons, it is probably safe to assume for the period under discussion a somewhat larger maximum figure of z jo,ooo.79 With the waning of the eighth century came a gradual demographic recovery at Constantinople. After 760, the city had many reasons to boast of an excess of births over deaths and of immigration over emigration. Most important among them, the plague of 746-747 seems to have foreshadowed the pest of seventeenth-century London in quickly passing and in leaving behind it for the next three centuries -so far as the sources indicate -no endemic infection as a heritage for succeeding generations. Stimulating demographic growth was an agricultural surplus, ensured by resettlement and colonization during the eighth century, and fostered by almost two generations of relative order under the Isaurians. To the west of the city, within the Long Walls, grazed the cattle of emperor and patricians, whence they were brought to the market established by Constantine V on the Forum ~ a i r i . 8 Arab 0 visitors at the beginning of the ninth century testify to a flourishing agricultural life in the European suburbs and provinces nearest the imperial city.sl So, too, do the administrative measures of the Amorian emperors. Before 850, they had organized new themes about Dyrrachium and Thessalonica whence grain ships sailed, unless major civil disorder prevented it.82 During one such period of civil war, the revolt of Thomas the Slav, the surplus in the city's granaries was of sufficient quantity not only to assure its survival, but even to support refugees and seditious monks, brought within the walls that their hostility to the ruling dynasty might not make them ready adherents of the rebel. Quite in constrast to its conduct early in the eighth century, during a major siege Constantinople accepted, rather than expelled, men who were worthless from the military point of view.s3 If successful wars and the restoration of order provided the peasant a breathing space in which to cultivate and lay up his surplus, for the large proprietor and urban folk, too, they meant greater quantities of cheaper slaves on the market. Since it was the good fortune of the Empire to carry on many of its wars with peoples outside the Christian world, there was little moral 79 Quotation from "Memorial on the Greek Lands and Churches," ed. S. Lambros, Nlos 'Ehhqvopvipwv, VII ( I ~ I O )p., 361. Compare Ruy Gonzales de Clavijo, Embassy to Tamerlane, trans. G. Le Strange (London, 1928), pp. 87f., and Pero Tafur, Travels and Adventures, 143j-1439, trans. Malcolm Letts (London, 1926), pp. 146f. For a general discussion of the population of Constantinople in the last centuries of the Empire, see P. Charanis, "A Note on the Population and Cities of the Byzantine Empire in the Thirteenth Century," T h e Joshua Starr Memorial Volume (New York, 1953)~pp. 135-149. so Ibn a1 Fakih in E. \V. Brooks, "Arabic Lists of the Byzantine Themes," Journal of Hellenic Stztdies, X X I ( I ~ o I )pp. , 72f. According t o the patriographs, Constantine V moved the cattle market from the Golden Horn t o the Forum Tauri. A. Van Millingen, Byzantine Constantinople : T h e W a l l s of the C i t y and Adjoining Historical Sites (London, 18gg), p. 226. 81 The Khitab a1 Uyun, trans. Brooks, "Campaign of 716-718," op.cit. p. 23. See also Harun ibn Yahya, trans. A. A. Vasiliev, "Harun ibn Yahya and his Description of Constantinople," Annales de l'lnstitut Kondakov, V (1g32), p. 162. Cf. note 182 below. 82 On the foundation of these themes, Ostrogorsky, Byzantine State, pp. 17zff., which is based upon pp. 8f. For commuF. Dvornik, Les lhgendes de Constantin et de Me'thode vues de Byzance (Prague, 1933)~ nications and travels within the region, see the latter author's L a vie de Saint Gre'goire le De'capolite (Paris, 1926), pp. 30, 54. 83 For the flight of refugees, including monks, t o Constantinople during the revolt: V i t a Theodori Studitae, A, c. xxviii (Migne, P G , 99, col. 148); V i t a Theodori Studitae, B, c. lxi (ibid., cols. 317, 320).

106

J O H N L. T E A L L

objection to the enslavement of men whom generals captured in battle in the ninth and tenth centuries.84 To swell their numbers, merchants brought rural slaves from the lands bordering on the Black Sea and others from Mediterranean regions to the south. While the greater number were doubtless destined to work estates possessed by such as the widow Danielis and the prototype of the master of Andrew the Fool, others served as house slaves in, among others, the bourgeois establishment of Theoctiste, mother of Theodore the Studite, and in the aristocratic houses that sheltered Andrew and . ~ ~patriographs attributed the foundation of the Saint Basil the Y o ~ n g e rThe slave market at the Staurion to the Emperor T h e o p h i l u ~and , ~ ~in his continual complaints against members of the monastic order who purchased slaves, Theodore the Studite suggests that the rule against ownership was more honored in the breach than in the observance.87 The slaves to be found at Constantinople constituted a special class of immigrants ; and it is far from surprising that a wealthy class, able to purchase them in such great numbers, attracted free men as well. Such visitors might find temporary food or lodging in the xenodochia or hospitia constructed by or, should these be fully occupied, saintly men such Irene and her successors,~~ as Basil the Younger might offer the wanderer a place at the table of a great house.g9 Shelter could always be found at least in the churches, along the arcades, or in the theater.90 Two of those who migrated to the city left their mark on subsequent events: Thomas the Slav first appears in the records of the chroniclers when a search for the essentials of life brought him into the service of one of the leading citizens.91 An alliance with a member of the Martinakes family made it possible for Basil the Macedonian to migrate.92 By the end of the ninth century, in consequence, the city administration found itself faced with many of the same problems of overpopulation that had confronted the eparchs of Justinian's age. The streets of Constantinople as described in the Life of Saint Andrew the Fool, were narrow, crowded, and filled with dungg3In one of his novels Leo the Wise revived the old provisions on the distance to be maintained between houses and even between house balconies. In another, a most significant law, he regulated the building plans of those who intended to build their houses on land hitherto assigned to the 84 See A. Marava-Hadjinicolaou, La vie des esclaves dans I'Empire byzantin, Collection de 1'Institut Franqais dlAthbnes,XLV, (Athens, ~ g j o )pp. , 40f. 85 Danielis and her hosts of slaves are mentioned in Theophanes Cont. (ed. Bonn., p. 228). See also Vita Andreae Sali, c. xi (Acta SS.,May VI, App., p. 5) ; Vita S. Basilii Iunioris, Acta SS., Mar. 111, 670; Vita et officiumTheophanis, Migne, PG, 108, col. 49; Theodorus Studites, Oratio XIII, 5 (Migne, PG, 99,

col. 888).

Patria Constantinopolis, 11, 64 (ed.Preger, 11, 185).

The rule of Theodore the Studite in his Testamentum, Migne, P G , 99, col. 1817.

88 A full catalogue of the major foundations in R. Janin, La gkographie ecclksiastique de lJEmpire by-

zantin, pt. I, vol. 3: Les kglises et les monastEres (Paris, 1953)~pp. 564-582.

as Vita S. Basilii Iunioris, Acta SS., Mar. 111, 670.

Vita Andreae Sali, c. xvii (Acta SS., May VI, App., p. 12).

91 Josephus Genesius, Regna, ed. Bonn (1834), p. 35.

86

s7

sa s3

Ibid., p. 110.

Vita Andreae Sali, c. lxix (Acta SS., May VI, App., p.

35).

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 107 plow, the olive tree, or the vine.94Rents were high and Romanus Lecapenus became popular by cancelling debts owed the landlords.95During a particularly severe winter a paternalistic government had to cover the apertures in the arcades so that the poor who lived there would not die from the cold.96The compilers of the Epanagoge included among its provisions Justinian's novel ordering the quaestor to regulate visitors and put beggars to work.97 With the greatest of energy the emperors strove to control and restrict both short-term visitors and hopeful permanent migrants. How effectively the fifth chapter of the Epanagoge could be enforced Liudprand of Cremona discovered as he waited, a supplicant for admission, outside the gate.98One of the objectives of the proposals for the commercial treaty of go7 with Russian merchants was the limitation and control of those who came to trade in the vicinity of the city.99 The legitimate merchant received his ration but if "Russes come here without merchandise, they shall receive no provisions. . . . Such Russes as arrive here shall dwell in the Saint Mamas quarter. Our government will send officers to record their names, and they shall then receive their monthly allowance.. . ." Palestinian monks were directly told and brutally shown that the Emperor Theophilus did not welcome their presence.loO The need to maintain peace and good order, together with a certain ingrained suspicion of the "stranger at the gate," usually accounted for the vigor with which emperors such as Theophilus enforced their laws. To watch and control the movements of peoples, Cecaumenus advised his son, was one of the basic principles of administration. Look out for the stranger to the province and scrutinize carefully the affairs of the city's corporations so that "whenever anything is afoot, you may know it."lol Politically motivated though they may have been, the measures the emperors took could hardly have failed to have a demographic effect, and in some instances the state seems to have acted frankly from the desire to prevent the creation of an impoverished urban proletariat, without bread, without homes, without even the hope for either. If Irene brought men to Constantinople, they took the place of those she had 94 Novellae Leonis, lxxxi, cxiii, ed. P. Noailles and A. Dain, Les novelles de LLon le sage (Paris, 19-14), PP. 256f.P 372-374. g5 Cedrenus, ed. Bonn, 11, 318. 96 Theophanes Cont., De Romano Lecapeno, c. xxvii (ed. Bonn, pp. 417f.). 97 Epanagoge V, ed. J. and P. Zepos, Jus Graeco-Romanum, I1 (Athens, 1931), pp. 244-246. g8 Liudprandus, Relatio de legatione, cc. i, ii (ed. Becker, MGH, pp. 17jf.). ee The Russian Primary Chronicle, ed. S. H. Cross, Harvard Studies in Classical Philology, XI1 (Cambridge, Mass., 1930), pp. 150, 160--163. lW Leo Grammaticus, Chronogra$hia, ed. Bonn (1842), p. 2 2 6 ; Vita S. Theodori Grapti, Migne, PG, 116, cols. 672-680. Both the first and second phases of Iconoclasm depopulated the monastic communities and the laity as well. When Theodore the Studite's uncle Plato turned to Constantinople after the

persecutions of Constantine V, he is supposed to have found only twelve monks still at Studion although their numbers are alleged t o have increased to seven hundred under Nicephorus. Vita Theodori Studitae, c. xxix (Migne, PG, 99, col. 145).The monastery of the Chora suffered a similar fate: Vita Michaelis Syncelli, Izvestia Russkago Arkheologicheskago Instituta v Konstantinopole, VII (1go6), p. 251. When Irene came to the city after the death of Theophilus she is supposed t o have found the monastic community in ruins: Vita Sanctae Irenes, cc. viii, x (Acta SS., July VI, 604f.).For the laity: Vita Stephani Iunioris, Migne, P G , 100, col. 1088.Theophanes Cont., De Theophilo, c. x (ed.Bonn, p. 10). lol Cecaumenus, Strategicon, c. x, ed. V. G. Vasilievsky and V. Jernstedt, Zapiski Istoriko-Filolog. Fakulteta Imp. St. Petersburgskago Universiteta, fasc. 38 (St. Petersburg, 1896), p. 5.

J O H N L. T E A L L

forced to leave.lo2Refugees from the Bulgars in 826 were dispatched by Theophilus to their proper homes in the provinces once they had reached the haven of Constantinople.lo3 When, as did the Phrygians of the fourth century, starving peasants from Paphlagonia left their villages during a famine, they were -unlike their predecessors-sent back home where the Bishop of Ancyra ministered to their needs.lo4 Whether the emperors were able to maintain the population at a level sensibly below that of the early sixth century, we cannot tell. They were successful at least to the degree that large tracts of land, suitable for agricultural purposes, remained uninhabited within the walls.lO5 A consideration of military need will demonstrate why their policies became ever more vital to the grain economy in the course of the tenth century. Until then mediation between urban and military demand, while sometimes necessary, did not become a problem of crucial and continuing importance. In the first place, the small size of the Byzantine armies precluded much competition between the two consuming groups. To begin with the period immediately subsequent to 619, there is no reason to believe that the armies Heraclius and his immediate successors commanded were any larger than the contingents Maurice had led. Rejecting fantastically large estimates of ~oo,ooo and 120,000 to be found in Sebeos, Nicephorus, and Theophanes, it is possible to accept for the expeditionary forces such figures as the 8,000 fighting under Heraclius at Antioch in 614 or the 3,000 men with whom Valentinus besieged Constantinople in 642.106 In 715 or 716, when both urban and rural population were probably at their lowest ebb, the future Leo I11 at the head of 300 men conferred with the Arabs at Amorium; later he retired to Pisidia while entrusting the defense of the beleaguered city to a complement of 800 men.107 The armies Constantine V collected for his Bulgarian wars bespeak a healthy increase among the ranks of the themes, but it is impossible to do more than merely estimate a maximum figure, and that for the cavalry alone. Towards the north in 763, Constantine led an army on foot through Thrace, while 800 ships carried twelve horses each to the mouth of the Danube.losAssuming a man to each horse, the sum in question would indicate a maximum cavalry force of 9,600. The number was almost certainly smaller, inasmuch as ~ ~ i a n t i n e armies maintained extra horses for remount.i0g The reign of ~ i c e ~ h o r uushers s"~ Theophanes, ed. de Boor, I, 462. Leo Grammaticus, ed. B o n n , pp. 231-233, c f . p. 215. For discussion, J . B. B u r y , A History of the Eastern Roman Empire from the Fall of Irene to the Accession of Basil I (London, I ~ I Z ) pp. , 370f. Cedrenus, ed. B o n n , 11, 499. T h e Phrygians: Socrates, Historia ecclesiastica, IV, 16 (Migne, PG, 67, C O ~ .501). lo5 Theophanes Cont., Dc! Romano Lecapeno, c. s l v (ed. Bonn, p. 431). 106 T h e command o f large armies is ascribed t o Heraclius and Justinian I1 b y e.g., Sebeos, Histoire d'Heraclius, trans. Rfacler, p. 8 1 ; Theophanes, ed. de Boor, I , 377; Nicephorus, ed. de Boor, p. 43. T h e figure given i n t h e last t w o sources seems particularly questionable since i n t h e same year t h e Emperor had t o increase his regulars w i t h a n ad hoc peasant militia. Theophanes, loc. cit., Nicephorus, p. 43. B u t for t h e suggestion o f a differentinterpretation o f t h i s passage, see Lemerle, o f . cit., CCXIX, p. 72, whose skepticism I share. Contrast t h e figure given b y Sebeos, p. 67, for Heraclius' a r m y a t Antioch and for Valentinus, hTicephorus, p. 30. lo7 Theophanes, ed. de Boor, I , 387f. lo* Ibid., I, 432. 109 W h e n winter set i n at Kanzaka, 628, Heraclius ordered his cavalrymen t o quarter t h e greater part o f their complement o f horses i n t h e houses o f t h e t o w n , each m a n keeping w i t h h i m b u t one m o u n t i n t h e camp. Chron. Pasch., ed. Bonn, pp. 727-734. 1°2

lo3

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 109 in a period for which figures are at once more reliable and more frequently found. Kodama thought that the Byzantine army of the ninth century included 120,ooo men; if infantry, thematic, and tagmatic troops are included, the estimate should not be summarily rejected.l1° Kodama's figures indicate that the total armed force fell short of Justinian's military reserves by 30,000 men at the time Agathias wrote, while the ideal tactical force seems to have been smaller than those Belisarius commanded.ll1 From the Strategicon of Maurice, Leo VI adopted into his Tactica the basic principle that wars were best carried on by small, well-disciplined units. He contemplated a themal army of 4,000 to 5,000 men, a figure that is repeated in the works on military practice.112 The anonymous author of the De velitatione bellica, describing the warfare on the Taurus frontier conducted by successive members of the Phocas family, felt that a great deal had been accomplished there by only a few troops.l13Even to repel a massed hostile force, a good general needed only 5,000 horse in addition to the help of God.l14 If a general collected all the themes of the East to combat the enemy in full force, he might, Leo believed, muster an army of 30,000 men.l15 Even in the ninth and early tenth centuries, then, normal warfare seems to have been conducted on a rather small scale. I t seems unlikely that a strategos ever found himself, save in the most extreme of emergencies, at the head of a force greater then 5,000 horse associated with an indeterminate infantry complement. Leo VI believed that the armies of the Byzantine state were smaller in size than armies once had been, and other writers seem to prove his contention. The ideal total force of 16,000 envisaged for an expedition in the tenth-century treatise De castrometatione is smaller than the 25,000 Procopius states Belisarius commanded in 532.116 The active fighting force on the expedition to Crete in 911 -8,000 -is smaller than the active fighting force of 15,000 Procopius believed to have participated in the Vandal expedition.l17 The 16,527 men enumerated for the second Cretan expedition of 949 includes both oarsmen and the disembarkation force, and it inaugurates an age of large-scale warfare when an eye-witness such as Leo the Deacon did not hesitate to speak of armies numbering 40,000 men.118 In the tenth century, when the civilian population once more abounded, the emperors maintained, even furthered, that decentralization of forces already a 11° So Bury, Eastern Roman Empire, p. 226, note I , on the basis of the figures t o be found in Kodama, Khitab a1 Kharadj, ed. and trans. 14. de Goeje, Bibliotheca geographorum arabicorum, VI (Leiden, 1889), pp. 197-199. See above, note 12. Leo, Tactica, constitutio iv, sec. 62 ; xiii, sec. 32 (hligne, P G , 107, cols. 713, 816). 113 See the summary and discussion of this type of warfare by G. Schlumberger, Un empereur byzantin au dixiBme sihcle: Niciphore Phocas, 2nd ed. (Paris, 1g23), pp. 142f., 286-291. 114 Anonymous, De velitatione bellica, c. xvii, ed. Bonn (1828), p. 17. 115 Leo, Tactica, const. xviii, secs. 142-153 (Migne, P G , 107, cols. 981-989); cited and discussed by F. Lot, L'art militaire et les armies au moyen dge, I (Paris, 1946), pp. 27-74? esp. 67f. Leo, Tactica, const. xx, sec. 72 (Migne, PG, 107, col. 1013) ; Scriptor Incertus, Liber de re militari, cc. i, viii, ed. R. VAri (Leipzig, I ~ O I ) , pp. I, 17.For the date and background of this pamphlet, see the review of Vgri's edition by J. Kulakovskii, BZ, XI (1902),pp. 547-558, itself an important contribution. Constantinus Porphyrogenitus, De cerimoniis, 11, 44, ed. Bonn, I, (1829)~654. 11* Constantinus Porphyrogenitus, De cerimoniis, 11, 45 (ed. Bonn, I, 664-667). Leo Diaconus, Historiae, IV, 2 ; V I I I , 4 ; VI, 12 (ed. Bonn, pp. 56, 132, 109).On the changing character of warfare and the larger scale of the expeditions, see note 142, below.

J O H N L. T E A L L

feature of the military establishment late in the sixth century.llg Of the four tagmatic divisions attached to the emperor's person, the most important had diminished in numbers while, on the contrary, its military effectiveness increased. I n contrast to Justinian's eleven scholae of 500 mediocre warriors each, tenth-century sources mention thirty elite banda of fifty each, a reduction at least in theory from 5,000 to 1,500 men. The excubitores, it is true, had increased to 700 men, but the augmentation of 200 failed to offset the decrease in the more important scholarian division. The Ikanatoi numbered 476 in 949, but for the fourth division, the arithmos, no numbers are to be found. The function of this tagma, that of guard for the imperial palace and tent, suggests that its ranks were none too large.120 As did their predecessors in the later Roman Empire, the palatine forces of the Byzantine period were often garrisoned not in the city itself, but in dispersion throughout the provinces. Thus, detachments of the scholarians, according to the De cerimoniis, were to be found in Thrace and Macedonia, while the domestics of the scholae joined the Emperor on expedition at the great camp near Malagina on the Sangarius river.121 In one respect decentralization had been carried even further some time after the accession of Heraclius. No longer were large numbers of foederati accomodated, as had been the Goths of Tiberius and Maurice, with their families in Constantinople itself. During the long, slow process of the elaboration of the themes and the military settlements, they were given homes in Lycaonia and Pisidia whence some of them were drawn backwhether occasionally or regularly we do not know-to perform garrison duty in Constantinople. In similar fashion, although not in settlements on the land, the Hetaireia or guard of foreign mercenaries so important in the sources of the tenth and eleventh centuries, were sometimes to be found scattered through Bithynia, sometimes in winter quarters in the Thracesian therne.lz2 Thus it is not surprising that, at the time of the Tornicius revolt, Constantine I X (1042-1055) could defend himself, against troops from the western themes, with little more than the city walls.lZ3When Isaac Comnenus (1057-1059) acceded "with all due legal sanction" to the throne of Michael VI, one of his first acts was to dispatch his-army from the city to their homes in the ~ 0 u n t r y . l ~ ~ As he watched the men depart, Psellus breathed a sigh of relief, and with reason, for, although the higher ranks sometimes forgot it, often to their own regret, the proper abode in time of peace for the Byzantine themal soldier and his officer was at home in the country. One cannot be, observed Cecaumenus, See above, p. 93, note 14. For the Ikanatoi, De cerimoniis, I I , 4 5 (ed.Bonn, I, 666); on the other units, BrBhier, Institzttions de I'Empire byzantin, pp. 353-355; J. B. Bury, The Imperial Administrative System in the Ninth Century, British Academy, Supplemental Papers, I (London, I ~ I I ) pp. , 53f. But Bury wondered if he had not overestimated the reduction: Eastern Roman Empire, p. 228. lal De cerimoniis, I, App. (ed.Bonn, I, 445). I follow here the emendations suggested by H. Gelzer, Die Genesis der byzantinischen Themenverfassung, Abhandlungen d. konigl. sachs. Gesellschaft d. Wissenschaften XVIII, 5, (Leipzig, I Sgg), pp. 108-1 14. 122 On the foederati, see the arguments of Stein, Studien, p. 140.The Hetaireia: Cedrenus, ed. Bonn, 11, 508f., 606, 617. Michael Psellus, Chronographia, VI, 105, ed. and trans. E. Renauld, 11, (Paris, 1928), pp. 18f. Ibid., VIII, 45 (ed.Renauld 11, p. 111). '19

lZ0

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 111 both a man of the city (TTOAITIK~S) and a general.125If the Byzantine army was small, if it was decentralized, it was also predominantly a rural force; though it may not have been, it is true, a peasant militia composed of those who fought part-time and farmed part-time. I n clarifying the distinction between the rnpa-r~cj-rq~ (he who holds land obligated to support military service) and the o - r p a ~ ~ u 6 p ~(he v o ~who physically performs the service), a recent study has ~ ~ l may ~ h not necessarily have been lands suggested that the ~ p a ~ l u-rfi~a-ra held and worked by the very men who fought to defend them. At the same time "the condition of the soldier and that of the peasant are very close"; the soldier was a familiar figure on the rural scene.126The chronicles and tactical treatises of the tenth century suggest that the soldiers' quarters, either during the intervals between campaigns or upon retirement after the term of service, were usually in the villages. What tenth-century materials clearly reveal is more dimly reflected, as we shall see, in the sources of an earlier period. The implications for the supply system are obvious. In time of peace, military demand was dispersed, thereby lessening the burden on any one locality. An emperor no longer had to promise accomodations for veterans in urban homes for old men or to require a provincial garrison to take up winter quarters in the lands of the enemy.127 IVhen rumors reached him of the death of Romanus 11, Nicephorus Phocas found his plans checked and his position weakened by the absence of his troops, who were at home whither he had sent them.128When the same Emperor returned from his expedition of 964, he despatched his Cappadocians to their villages, warning them to maintain their horses and equipment against the call to arms in the spring, and keeping by him only a small 6lite group to take part in military exer~ises.12~ If the emperor wished to undertake an expedition or defend a province against Arab or Bulgar, probably this same 6lite group (or "cadre") was responsible for establishing camps and searching out those whose names were listed in the military catalogue. I t was then, reported the anonymous author of the treatise De castrometatione, that a check was made on the condition of horses, equipment, and estates.130 "Before invading enemy lands, trustworthy men of the emperor's own should muster up the entire army, that they may know how many are present to bear arms for his sacred empire; how many establishments, in contrast, have fallen to ruin and how many men have fled; in addition, how Cecaumenus, Strategicon, c. xxxviii, ed. Vasilievsky and Jernstedt, p. 20. Lemerle, "Histoire agraire," RHist., CCXX, p. 59, and, in general, 43-70; cf. ibid., CCXIX, pp. 70-73, where Lemerle, commenting on a passage from the Vita Philareti (see below, note I ~ I ) , believes that it shows a garrison "installCe dans un district rural," where "les soldats sont peut-&re etroitement melCs & la vie des paysans." 12' Theophanes, ed. de Boor, I, 415. u8Leo Diaconus, Historiae, 11, 1 1 (ed. Bonn, p. 31). Is8 Ibid., 111, I I (ed. Bonn, p. 53). 13O Scriptor Incertus, Liber de re militari, c. xxix (ed. VBri, pp. qgf.). Is5 Is6

J O H N L. T E A L L

many have let themselves slip through inertia and how many men have died; further, which ones have kept their equipment and horses in good order. Nothing ought to be overlooked; it should be discovered which are fulfilling their obligations and which through sluggishness, have preferred their own affairs before matters that touch the general interest." The adnumion, or muster system, was enforced with all the rigidity the author advised, and it was older by far than his treatise of the tenth century. In the ninth-century Lives of Saint Philaretus the Merciful and Saint Eustratius Augarus, frantic soldiers, fearful of the punishment in store for them, received from the saints horses to replace their own mounts.131 Armed, then, by his own efforts or with the assistance of a kindly saint, the soldier left his home to join the other members of the theme in preparation for the campaign. I t was obviously at this point, with men heavily concentrated in certain localities, that military demand might threaten most seriously the normal pattern of the wheat economy. Thus Leo VI urged that once the muster had taken place, once the camp had been formed and the condition of men and horses investigated, then this great body should split up and carry on its exercises in more manageable g r o ~ p s . 1Such ~ ~ a concentration of men, Leo stated, could be politically dangerous, and there is reason to believe that he was well aware of the chaos it could bring to the equilibrium of supply and demand. Repeating the provisions of the Strategicon of Maurice, Leo advised his general to avoid urban agglomerations and to treat the peasant's fields gently. He even added a suggestion of his own, urging that the expeditionary army be led as quickly as possible beyond the frontier that it might live on foreign rather than on domestic a g r i ~ u 1 t u r e . l ~ ~ Now that they could no longer draw upon Egypt in emergencies, the emperors strove -with a decentralized, dispersed, and rurally based army -to avoid incidents similar to the famines suffered at Antioch under Julian or at Edessa under Anastasius. Yet it was impossible completely to avoid such emergencies. While the predominant pattern of warfare in the ninth and early tenth centuries (when population, and hence urban demand, was in the ascendancy) may have been one of limited action to which relatively few troops were committed, these years were by no means unmarked by larger enterprises both offensive and defensive in character. Even if unsuccessful, the expeditions mounted under Leo VI represented a large expenditure of military effort. Frequently it was necessary to displace and concentrate the themal armies on only one of the many frontiers. Armies from Asia Minor accompanied Constantine V and Nicephorus I against the Bulgars to the north.134A passage from the life of Michael I11 in Theofihanes continuatus suggests that the opposite was often true on those occa131 Vita S. Philareti, ed. Fourmy and Leroy, p. 127; Vita Eustratii, hegumeni Augavi, c. xiii, ed. A. Papadopoulos-Kerameus, ' Avahema I s p o u o h u y ~ ~m~a~~i uj ~o h o y i a IV, ~ , (St.Petersburg, 1891-18g8), P. 377. Leo, Tactica, const. ix, sec. 4 (Migne, PG, 107, col. 768).

133 Ibid., const. ix, secs. 15, 23 (Migne, loc, cit., cols. 772f.).

134 See above, note 126, and for Nicephorus: Theophanes, ed. de Boor, I, 48gf.

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 113 sions during the ninth century when the caliphate organized an offensive to penetrate deeply into the Empire's heartland.135 Upon such occasions, when the themal armies could be as much armies of movement as the exercit~scornitatensis of the later Roman Empire, the emperor provided annona for his troops. I t was only because of special arrangements, clearly distinguishing him from the majority of his comrades, that Luke the by Ibn KhorStylite supplied himself from his "paternal h ~ u s e . " A l ~statement ~ dadbeh, which gives a contrasting picture, probably refers to the akritae, or frontier warriors, alone : "There is," he remarks, "no market in the Roman camp. Each soldier is obliged to bring from his home (manzil) the bread, oil, wine, and cheese that he needs.'' Fighting, in a more literal sense than did the themal soldiers, to defend their own land, carrying provisions for fifteen days, and rotated from their posts at the end of that time, the akritae may well have been the contingents Ibn Khordadbeh had in mind.13' For the bulk of the army the logistic problem was quite different, and responsibility for assuring supply had, obviously, to lie with the emperor's agents. How they carried out their duties during the wars on the northern frontiers it is impossible to tell, nor do we know in general how the armies of the seventh and early eighth centuries supported themselves when on campaign within the boundaries of the Empire. In alllikelihood they followed a system of ad hoc requisitions much like that upon which the armies of Maurice had depended. In 813 the troops of hlichael Rangabe, for example, are supposed to have caused more damage in Thrace than had the Bulgars, and it may have been only the reduced number of the contingents that prevented the scope and burden of their demands from becoming even greater.13* Generalization is possible for the wars in Asia Minor alone, and then only when the frontier had stabilized itself in the latter half of the eighth century. I t is, in fact, probable that stability was necessary before new principles could be elaborated and each functionary in the chain of command assigned a definite role in the support of the mobile field units. The document from which we must draw the greater part of our information on the services of supply for the expedition -the ceremonial observed when the emperor left on campaign -attributes the first elaboration of organization to the Isaurians.l39 To judge from Constantine VII's description, the imperial bureaucracy attempted to distribute the burden of demand as widely as possible. As the emperor passed through the great camps or assembly points on the way to the 135 .for when the Bulgars are at peace, it was the rule that they [soldiers from Thrace and hfacedonia] go on campaign and share dangers with the Anatolics." Vita Michaelis, c. xxv (ed. Bonn, ' I . .

p.

181).

136 Vita Lucae Stylitae, c. vi, ed. H. Delehaye, Les saints stylites (Brussels, 1923)~ p. 201. 1S71bn Khordadbeh, Khitab a1 masalik, ed. M. de Goeje, BGA, VI, 85; English translation by Dr. Irfan Kawar. For the conditions of service, see sources and discussion, note 113, above. On this, and the preceding passage from the Vita Lucae Stylitae, cf. Lemerle, op. cit., CCXX, pp. 64f. lS8Theophanes, ed. de Boor, I, 500. Under Heraclius, the troops seem t o have depended largely upon the supply they could find along the route of march. This factor was an important consideration in the choice of strategic alternatives: Theophanes, ed. de Boor, I, 312. 139

8

De cerimoniis, I,

App. (ed. Bonn, I, 457f.).

J O H N L. T E A L L

Syrian or eastern frontiers, he was met by successive themal armies, and as the imperial host approached a particular camp, it was the duty of the protonotary of the respective theme to furnish the emperor with whatever the latter might need. Thus if the system worked well, the burden would fall upon each of the successive themes as a whole, avoiding excessive demands on the resources of any one region through which the troops passed.140 For the sea-borne expeditions to Crete, the Thracesian and Anatolic themes provided bread as well as men when the ships set sail from Attalia.141 Despite the best of planning and organization, the great expeditions undertaken by land and sea late in the tenth century were a grievous burden on the supply system. I t may be noted that chronic short supply and high prices at Constantinople are first noticed in the chronicles at the point when large expeditions mounted to capture and hold enemy cities and lands became the rule rather than the exception.l42 In July 960, when he launched the third and most successful attack on Crete, Nicephorus Phocas brought his troops into a land already suffering from a bad harvest. For many months the siege was slow and indecisive; on the urgent appeals of his general, the Emperor Romanus I1 sent further supply to his invasion army in the winter of 960-961. In October, 960, Constantinople began to suffer from a famine so severe that Joseph Bringas had to bring speculators to heel and search for supplies in "East and \Vest."143 This famine of 960 -the first the sources record since the bad winter of 927-928, and the second or possibly the third since the siege of 742-may of course have had causes other than the initial shipment to Crete in July. I t was apparently a bad Ibid., I, App. (ed. Bonn, I , 451 f., 477, 489). Ibid., 11, 44 (ed. Bonn, I, 658f.). 142 On the character of warfare in the tenth century: Bury, Eastern R o m a n Empire, p. 249; S . Runciman, T h e Emperor Romanus Lecapenus and H i s Reign (Cambridge, ~ g z g )pp. , 120-150, esp. 146f.; Ostrogorsky, Byzantine State pp. 256-258. A good idea of the conflict, from the Arab side, along the frontiers in Asia Minor may be gained from the sources translated by E. W. Brooks, "Byzan, 728-747. tines and Arabs in the Time of the Early Abbasids," English Historical Review, XV ( I ~ o o )pp. Leaving aside those caused by siege, the famines or shortages I have been able to discover in the sources are the following: ( I ) a tradition of high prices under Basil I, cited in connection with the speculations of Xicephorus 11:Cedrenus, ed. Bonn, 11,372f.; (2) the great famine of 927, caused b y an unusually severe winter: Theophanes Cont., De Romano Lecapeno, c. xxvii (ed. Bonn, p. 417) ; (3) under Romanus I 1 in October, 960: Ps.-Symeon Magister, Chronicon, ed. Bonn (1838), p. 759; no cause given. (4) the famine of the summer of 968-sources discussed below-allegedly caused by the winds Cedrenus, ed. Bonn, 11,381, mentions a shortage that in May; (5) in the reign of John Tzimisces (970)~ had continued five years; no specific cause assigned. (6) famine of 1037 a t Constantinople, caused by a six months' dry spell: Cedrenus, ed. Bonn, 516. One has only to leaf through the lives of the saints to realize that local famines in the countryside were almost annual occurrences, but general shortages affecting Constantinople itself were rare indeed. For two reasons i t seems probable that even were the sources complete, the list would not be significantly extended. (a) The famine of gz7/gz8 remained a landmark, even in imperial legislation. See the novel of Kicephorus 11, dated 967, in Imperatorum post Iustinianum novellae constitutiones, Coll. 111, nov. xx, ed. Zacharia, Jus. Gr.-R., 111, pp. 296-299; cf. Ostrogorsky, Byzantine State, pp. 24zf., 253. (b) The Isaurians and Amorians were notoriously out of favor with their historians. Since famine could be-and frequently was-laid a t the emperor's door, would not ecclesiastical chroniclers have been quick t o seize upon any such evidence in their sources and attribute it t o the Iconoclast in power ? Even Theophanes had t o admit that prices were low under Constantine V, attributing them t o high taxes and the consequent need of the peasantry to sell in large quantities. Theophanes, ed. de Boor, I , 443. From Xicophorus, however, we learn that an alternative explanation of abundance was current in the city, Xicephorus, ed. de Boor, p. 76. 143 Ps.-Symeon, De Romano Const. Porphyr. f., c. iii (ed. Bonn, p. 759); Theophanes Cont., De Romano C . P . f., c. xiii (ed. Bonn, p. 479). 141

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 115 year throughout the east, and prices were particularly high and supplies crucially short in the Egypt of the 1 k h ~ h i d i d s . l ~ ~ But it would be difficult indeed to trace an aggravation of the famine of 968 to influences other than the expedition into Osrhoene and Syria, undertaken by Nicephorus Phocas, for the shortage had continued into the reign of that unpopular Emperor, and the chroniclers did not hesitate to accuse him of turning the situation to his own profit. Whatever justification there may be for their accusations -and there seems to be very little --Zonaras, Cedrenus, and even Liudprand of Cremona admit that the high prices were caused initially not by the Emperor's manipulations but by a preexisting di~equilibrium,l4~ and it is Liudprand, all unwittingly, who provides further evidence on the source of the imbalance : the impact of military demand. Fortunately that disgruntled writer visited the imperial city in 968, the very year in which Nicephorus prepared for his great campaign. Once again, it had been a year of generally bad harvests, and the Emperor probably knew well that whatever God's bounty might be, he would be forced to campaign in a wasted land. Had he ever read t h e ~ cerimo.itiis e he would have learned Constantine's advice to that effect.146Recent experience would have taught him the same lesson, for a shortage of supply had forced John Tzimisces to raise the siege of Massisa and even an ally of the arch-enemy, Saif ad Daulah, to break up his camp.147 Only in what was presumably the theme of Mesopotamia had the field mice spared the harvest of 968. So from that region Nicephorus gathered his grain, transporting it possibly to Caesarea where expeditionary forces were usually collected.l4* On the 19th of July, Nicephorus despatched his fleet and ordered Liudprand to wait upon him. After a foul meal on the twentieth, climaxed by perjury on the part of the Emperor, the Bishop of Cremona returned to the lodgings he found so miserable. The next sequence of events might best be given in his own words.149 "His conversation took place on Monday, the 20th of July, and for the next nine days I received no supplies from him at all. This, too, was at a time when the famine at Constantinople was so severe that three gold pieces were insufficient to provide one meal for my twenty-five attendants and our four Greek guards. On the fourth day of that week Nicephorus left Constantinople to march against the Assyrians." Could there be any clearer indication of the disruption that the departure of an expeditionary army might create in the equilibrium of 144 The cause was insufficient floods of the Sile over several years, with costs particularly high and shortages acute in 968: Histoire de Yahya ibn Said d'Antioche, ed. and trans. J. ICratchkovslry and A. A. Vasiliev, Patvologia Orientalis, XVIII (1924) pp. 812f. 14Wedrenus, ed. Bonn, IT, 372-374; Zonaras, Compendiz~mhistoriarum, XVI, 28, 1-13, ed. Bonn, I11 (1841-1897), pp. 513f.; Liudprandus, Relatio de legatione, cc, xxxiv, xliv (ed. Becker, MGH, PP. 193, 198f.I. De cerimoniis; I , App. (ed. Bonn, I, 486). '47 Yahya ibn Said, ed. Vasiliev, 793f (A.D. 964); Sicephorus himself had been forced to withdraw from Antioch for the same reason. 148 Liudprandus, Relatio de legatione, c. xliv (ed. Becker, ofi. cit., pp. 198f). Liudprand's observations a t this point confirm the impression of normally abundant supplies suggested by the evidence in note 142, above: "This year famine had wasted the land of the Argives to the point that two Pavian sextarii of corn could not be bought for one gold piece, and this where abundance usually holds sway." lbid., c. xxxiv (ed. Becker, op. cit., p. 193). 8"

116

J O H N L. T E A L L

civilian demand and supply? The advice Constantine VII had given his son was wise indeed: Upon setting out on an expedition, make sure that the city remains in a posture of defense during your absence, but be discreet in your preparations, lest a shortage of grain result .150 Nor is it surprising that the famine continued and was still raging in 970 under John Tzimisces,151for with the change to a pattern of predominantly offensive warfare, necessitating the siege and reduction of great cities, methods of supply had to be altered. The advice on matters of supply to be found in the treatise, De castrometatione, composed after 988, differs not only from the suggestions of Maurice and Leo, but from the system outlined in the De velitatione bellica as well. Siege warfare at the end of the tenth century could be supported only by continuous exports from the "land of the Romans" itself. Herein lies the essential change, and it seems advisable to translate in full the pertinent passage from the De ~astrometatione.~~2 "Those who wish to capture walled cities in war and thereby bring the enemy completely down should make use of frequent incursions against the fields of those cities by skirmishes and irregulars (those called corsairs by the westerners) and by the use of other forces of foot or horse. This they should do in order that the enemy may be easily overcome. . .as a result of not being able to work the land as he wishes, but being rather in complete distress. For unless the vines and the fruit-bearing trees have first been cut down, the harvest destroyed by fire, and the animals carried off, in order that the pressure of famine may make the enemy flee and take refuge elsewhere, then it will be very difficult to capture cities abundantly endowed, as they will be, with things to make them secure and with an ample force of fighting men. He who has not taken measures of this sort against the enemies, but seeks to capture offhand their armed strongholds, will mount an expedition against them in vain. "Just as it is necessary to starve the enemy, so abundance must be ensured to the army. If an expedition is to be undertaken against the land of the Saracens, while the provisions have been winnowed out by frequent devastations, and it is impossible in turn for the army to bring from its own fields by its own beasts more than twenty-four days supply of barley, while the cities which are to be attacked are well-inhabited and have supplies for a long time, it will be necessary to raise the siege unless necessities are supplied from the land of the Romans. "If in the land of the Saracens-so fertile and offering such an abundance of peasants-it is impossible in a short time to succeed and accomplish something according to plan unless the army is supplied from our own land, so much the more is this true in the land of the Bulgars where there is a De cerimoniis, I, App. (ed. Bonn, I , 450).

Cedrenus, ed. Bonn, 11, 381.

152 Scriptor Incertus, Liber de re militari, c. xxxi (ed. VBri, pp. 37f.). (Wote that the title De

castrometatione, sanctioned b y popular usage, does not correspond to the contents of the book). lS0

151

T H E GRAIN SUPPLY O F T H E BYZANTINE E M P I R E

117

lack of all things, particularly of barley. For assuredly if the supplies for the

army are not forthcoming from our own land, they will have to turn back

home without coming to grips with the enemy. What the greatest force and

power of the enemy could not do, the lack of supply will effect and it will

be necessary to raise the siege."

To sum up, until the great period of military endeavor signalized by this passage, Byzantine armies remained comparable in size to those organized under Maurice and in consequence somewhat smaller than the largest expeditionary forces Belisarius commanded. If, until the tenth century, the armies were small, they maintained both then and later the trend towards decentralization already noticeable under Justinian. In yet another way Byzantine military organization continued in a direction already defined during the sixth century, but unlike precedents offered by the fourth century from its great age of military ref0rm.1~~ In sharp contrast to the armies of Constantine, the medieval Byzantine armies were rurally based. Not until the tenth century, finally, did strategy consistently demand the export of grain to support foreign operations. The consequent strain is clearly visible, and it was fortunate indeed that the Empire's productive capacity had increased to a level capable of at least partially satisfying the demands made upon it. How this feat was accomplished is a question of production or supply, to which we may turn in conclusion. 2. The Supply of Grain A complete analysis of the agricultural economy of the Byzantine Empire would consider, first, the extensive variety of its soils and climates; second, the techniques employed to produce the harvests these factors would permit; and finally, the modes of agrarian living whereby men shaped the environment to their needs and adapted themselves to their environment. While such an analytical structure may be kept in mind, it cannot be attempted here. The object of this section of our study is far more limited. I t attempts to answer only the following questions, which are essential to evaluating the success of the agricultural adjustment from 619 to 1025: What were the new sources of supply? Had the impact of the new demand, altered and reduced from its counterpart in the later Roman Empire, been met by improvements in the extent and intensity of cultivation ? Some of its new sources of supply the Empire found outside its own political frontiers. At times and in varying degrees, Byzantine Constantinople depended upon the peoples north of the Bosphorus, chief among them the Bulgars of the Danube. Commercial provisions in the treaty of 716 and the proposed accord of 812 between Byzantium and the first Bulgarian Empire, the war of 894 with Symeon-originating as it did in a dispute over the conditions of trade-all these are evidence that merchants travelled between Constantinople 153 Zosimus, Historia nova, 11, 54, As$etti sociali, pp. 320-329.

ed. L.

Mendelssohn, (Leipzig, 1887), pp. g ~ f . ;cf. Mazzarino,

118

J O H N L. T E A L L

and the Black Sea ports of Anchialus and Mesembria. Among the products they chose to export it seems safe to include grain. If in the seventh and eighth centuries the Bulgars were not themselves capable agriculturalists, they could follow the practice of the Avars and draw upon the services of Slavic peoples subject to or associated with them. By exporting the products of the soil in which their land could be quite rich, the Bulgar aristocracy and later the church could maintain a balance of payments for liturgical objects as well as for those luxury goods they cherished so dearly. By the end of the tenth century, Byzantium depended so closely upon exports from and through the Bulgar kingdom that Basil 11's great victory at Cimbalongus, followed by the swift disintegration of Samuel's state and the consequent termination of the commercial blockade he apparently imposed, produced a new era of abundance of the "necessary things" at Con~tantinop1e.l~~ In contrast, Cherson seems to have been often, if not consistently, the center of a grain deficit area on the north shore of the Black Sea. In 654-655 the exiled Pope Martin complained bitterly that grain was often mentioned but never seen in the Crimea; it was only from the crews of Byzantine ships venturing in search of salt that he was able to buy his bread.155A fragmentary report which it is a temptation to write off as, at most, the chance record of a temporary famine received confirmation from Constantine Porphyrogenitus in the tenth ~ e n t u r y : "If l ~ ~the Chersonites do not journey to Romania and sell the wax and hides they get from the Pechenegs, they cannot live. If grain does not pass across from Aminsos, and from Paphlagonia and from the Boukellarioi, and the flanks of the Armeniakoi, the Chersonites cannot live." Byzantine literary sources have nothing to say about the agricultural and commercial situations of the other Greek cities of the Black Sea coast, and it is always possible that theirs may have been better. Nor do these same sources mention exports from the peoples east of Cherson. Rich though it was in agricultural products, the land of the Khazars seems to Arab and Persian geographers mention a trade have exported only i~ing1ass.l~~ in furs undertaken by the Bulgars of the Volga, and a similar export wherein the Russian Kaganate of the Kuban delta engaged, supplementing it-for Cherson and the Byzantine market -with deliveries of swords and slaves. A people without culfivated fields of their own, these Rus consumed the grain they levied as tribute from the Aso-Slavs of the Don and Azov regions.l58 After the formation of the Kievan state in the ninth century, Russian merchants from the north entered into direct commercial relations with 154 Sources: Theophanes, ed. de Boor, I, 497; Theophanes Cont., De Leone Basilii filio, c. ix (ed. Bonn, pp. 357 1.); Michael Attaliates, Historia, ed. Bonn (1853), p. 234. Discussion in I. Sakazov, Bulgarische TVivtschajtsgeschichte, Grundriss der slavischen Philologie, V (Berlin, 1929), pp. 49f., and for a contrasting view on agriculture: G. FCher, Les monuments de la culture protobulgare, Archaeologica Hungarica, V I I (Budapest, 1g31), passim. 15j Martinus, Epp. XVII, XVIII (Migne, P L , 87, cols. zozf.). lj6 Constantinus Porphyrogenitus, De administvando imperio, c. liii, ed. G. Moravcsik and R. J. H. Jenkins (Budapest, 194g), p. 286. 157 D. M. Dunlop, T h e Histwvy of the Jemish Khazars (Princeton, 19j4), pp. 224-234, esp. 228. 15* Ibn Khordadbefi, Khitab a1 ~ n a s n l i k ed. , and trans. de Goeje, BGA, VI, 115. See G. Vernadsky, Ancient Russia (New Haven, 1g43), pp. 261-307, esp. 282 ff.

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 119 Constantinople. Their commercial fleet, composed of the monoxyla or "singlestraker ships," sailed down the Dnieper from Kiev in June, bearing furs of castor, black fox, sable, and ermine, together with honey, wax, and slaves. To these forest products, the Russian Primary Chronicle, Constantine Porphyrogenitus, and Ibn Khordadbeh alike fail to add grain. I t has been suggested that grain constituted an important object of internal trade between north and south Russia, but the Kievan state did not export any to Byzantium before the twelfth century.159 In fact, there is reason to doubt that the supply from any of the regions to the north of Constantinople seriously challenged, in volume, the grain furnished the capital from other directions. If exports from South Russia cannot be proven to have existed before the twelfth century, the land of the Bulgars must often have failed its Byzantine markets. Krum relied not on trade but on booty to increase his wealth. His successors suffered famine and complained of over-crowding in the kingdom, praying the Empress Theodora to grant them land for living space.lGO Even in the eighth century the shipping lane past Abydus to the south had not lost its importance; in the tenth century its control was essential to domination of the capital. Thus in 742, when Constantine V besieged Artavasdus at Constantinople, the usurper sent out provision boats, laden with grain, which were captured by the imperial fleet based at Abydus.lG1At the time of their revolt in 987, Bardas Sclerus and Bardas Phocas captured ships, whose home ports were in Asia Minor, and used them to harass Mediterranean shipping lanes and to blockade the entrance to the Hellespont, hoping thereby to reduce Constantinople through starvation.lG2 Of the potential centers of export to the south, Syria seems to have won a reputation as at least an occasional granary for Constantinople. An eleventhcentury version of the tale of Abraham the Jewish moneylender and Theodore the navicularius has the latter trading in silks and grains at Syrian ports in the time of Heraclius.lG3A second reference to Syrian exports occurs in Zonaras' retelling of a tale concerning the Emperor Theophilus. According to the version found in Joseph Genesius and in the Scriptores Post Theophanewz, Theophilus saw from the palace a grain ship sailing around the city, apparently in passage from the south towards the ports along the Golden Horn. Upon discovering that its owner was none other than the Empress Theodora, Theophilus ordered the boat burned, exclaiming that he would not permit his wife 159 SO G. Vernadsky, Kievan Russia (New Haven, 1948), pp. 99-102, 116-121. Sources: De administrando impevio, c. ix, ed. Moravscik and Jenkins, pp. 59-63; Ibn Khordadbeh, loc. cit., note I 58, above; The Russian Prilnavy Chvonicle, trans. S. H. Cross, 2nd ed. (Cambridge, Mass., 1gj3), pp. 64-90. 160 Sakazov, Bulgarische Wirtschaftsgeschichte, p. 57. On Krum, see Scriptor incertus de Leone, ed. Bonn (1842), pp. 344f., and Theophanes, ed. de Boor, I, 484f. For relations under the Amorians: Ps.-Symeon Magister, De Michaele, c. xxv (ed. Bonn, p. 665) and Theophanes Cont., De Michaele Theophili filio, c. xv (ed.Bonn, p. 165). Theophanes, ed. de Boor, I, 419. 182 Leo Diaconus, Historia, X, 7, 9 (ed. Bonn, pp. 170, 173). 163 J. Starr and B. Nelson, "The Legend of t h e Divine Surety and t h e Jewish Moneylender," Annuaire de l'institut de philologie et d'histoire ovientales et slaves, V I I (1939-1944), p. 307.

J O H N L. T E A L L

to make of him -the Emperor -a shipmaster engaged in sordid trade.164This, in sum, is what the earlier historians have to say concerning the incident, and from their pages emerges an unsympathetic Theophilus, endowed with a gentleman's aristocratic contempt for mercantile activity. Zonaras, later in time, but generally careful of his sources, expands the tale in such a fashion as to emphasize Theophilus' concern for justice, a characteristic which even earlier hostile historians could not entirely ignore in their chronicles of the heretic's reign. In Zonaras, the Emperor becomes an apostle of "free trade"; by destroying the boat he puts an end to a threatened monopoly that could only deprive rightful merchants of their revenues. Zonaras sees fit to add one other important detail. The ship's cargo came specifically from Syria,165and, whatever the proportion of fact and myth mixed by Zonaras or his source, external evidence suggests that if either one or the other fabricated the Syrian origin of the ship's freight, he had reason to do so. Leo V was forced to issue an edict forbidding his subjects to trade with the peoples in Egypt and Syria.166 Upon other occasions the Byzantine people and even the state adopted an attitude far less strict. As Cecaumenus observed, the Byzantine general or merchant had no objection to commerce with the enemy save when t h e battle was hottest, while the anonymous author of the D e velitatione bellica urged the shrewd general to use, for intelligence purposes, merchants travelling across the Taurus frontier.16' Despite unanimous praise of the island for its fertility, much less can be said concerning the possibility of exports from Crete. The observations of John Cameniates may be cited as typical of what Byzantine historians and hagiographers, and Arab geographers as well, had to say in honor of Crete. Although they well knew that to land on Crete meant that they would be delivered i n t i the hands of an Arab slaver, Cameniates and his comrades, taken prisoner at the fall of Thessalonica in 904, were not entirely miserable as their ship approached shore, for whatever else might befall them, the wonderfully fertile island would at least offer bread enough to avert starvation. Unfortunately, neither Cameniates nor, seemingly, any other source has anything to say about the export of the island's abundant foodstuffs to Con~tantinop1e.l~~ We can note only that the loss of the island in 825 was no more felt than that of North Africa or Sicily, and thus consider the fall of Crete as simply another in a long series of catastrophes, ever limiting the consumers within the Empire to the resources of their own land. One by one the Mediterranean granaries, actual or potential, had passed under the control of Islam where other uses were often found for them. In the north a Krum or a Samuel might 164 Theophanes Cont., De Theophilo, c. iv (ed. Bonn, pp. 88f.); Genesius, ed. Bonn, pp. 75f. '65 Zonaras, XV, 25, 37-43 (ed. Bonn, 111, 355f.). '66 F. Dolger, Regesten, I (Munich, 1924)) no. 400, p. 49. 187 Cecaumenus, Strategicon, c. lxiv, ed. Vasilievsky and Jernstedt, p. 33; Ps.-Nicephorus Phocas, De velitatione bellica, c. 7 (ed. Bonn, p. 196). 16* Ioannes Cameniates, De excidio thessalonicensi, ed. Bonn (1838), p. 384. Compare Theophanes Cont., De Michaele Anzor., c. xxi (ed.Bonn, pp. 73f.), Vita Nicolai Studitae, Migne, PG, 105, col. 868; A1 Istakhri, trans. A. D. Mordtmann, Das Buch der Lander (Hamburg, 1845)) p. 43.

T H E G R A I N SUPPLY O F T H E B Y Z A N T I N E E M P I R E 121 withhold supply while the people of Cherson looked south to Asia Minor for life's neccessities. While even hostile frontiers were by no means formidable barriers to trade, would not Constantinople and the army have had to search for the bulk of predictable supply in Bithynia, Lydia, Thrace, Phrygia, or Macedonia? I t is George of Pisidia, the epic poet of Heraclius' Persian Wars, who provides the answer, terse but unmistakable, for the first crisis in 6 1 9 . l ~In ~ the midst of a lengthy encomium on the Emperor he paused to note: "If He who weighs all things had not through you driven off the furies from near to us, who-in so great a shortage of the things necessary to life-had persuaded the cities to share their sustenance?" When (if the Annales of Eutychius can be credited) the young Heraclius provisioned Constantinople from ~hessalonica, he apparently set a pattern for the following years.170 Above all, it is by tracing, with the aid of the Miracula Sancti Demetrii, the relations between thesetwo great cities that we learn how men of the seventh century organized and distributed their resources in grain so as to share their sustenance. Even the authors of this remarkable document, informative though they are on economic relationships, do not of course discuss such matters as independent objects of interest; only insofar as they are incidental to two sieges that Thessalonica withstood in the seventh century do details concerning sale and transport of grain find a place in the pages of the Miracula. The first siege, the Avar-Slav attack of 617 or 619, caught many of the inhabitants outside the walls, gathering their own harvest. Yet Thessalonica did not depend entirely upon its own hinterland; in the midst of the battle came grain. Since the fact of the siege was merchants from ~ o n s t a n t i n o ~bearing le not known at the capital, the merchants presumably had not been dispatched as an emergency measure.171 Particularly interesting is their reaction to the siege itself. Unlike the merchants who had abandoned Rodosto in Justinian's time, unlike the fearsome sailors who had shunned Thessalonica in 597, the captains and crews alike took up weapons and helped beat off the barbarians. There could be no better illustration of what is really meant by the "militarization" of the Byzantine Empire. Militarization lay not in so formal a matter as the union of civil and military power in the hands of the strategos, but rather in learning to live with risk, a lesson which the Byzantine people seem to have learnedrather early in the seventh century. For a talent like this the people of Thessalonica found good use during a second great siege, undertaken probably in 662 by Drogouvites, Rhynchinians, and Sagoudates. 'IjJell aware upon this occasion of the coming storm of Slavic peoples, the Emperor at Constantinople wrote to the chief men of the city, urging them to conserve grain in the storehouses. This they refused to do. Greedy for profit, they sold the grain at seven modii to the nomiswza, shipping it out in such haste that they did not pause even to pay the export duties. Georgius Pisides, Expeditio Persica, 111, 300, ed. Bonn (1836), p. 40. Eutychius: Annales, Migne, P G , 111, col. 1085. Miracula S . Demetrii, secs. 170, 177, 178 (Acta SS., Oct. IV: 167, 169, 170). Cf. ibid., secs. 66, 71 (Acta SS., Oct. IV, 128, 12gf.). la9

I7O

J O H N L. T E A L L

Fortunately, the Emperor himself saw to Thessalonica's needs sending boatloads of grain from Constantinople to help the besieged city resist her attackers. But the men he had chosen for this task proved unworthy. They sold the grain at advanced prices to the citizens, forcing the latter to send a deputation to purchase grain from the Velegezites, a friendly tribe of Slavs located near Demetrias in Thessaly. Luckily, the siege was shortly over. From the Emperor arrived not merely the 5,000 wzodii Thessalonica had requested, but 60,000 measures in all, while the deputation returned from Thessaly bearing provisions from the Slavs. With significant assistance from certain Slavic peoples settled in Greece, the cities-as George of Pisidia had indicated-shared their supplies in time of Ships and carts laden with grain often travelled in the opposite direction, from Macedonia and Thrace eastward to Constantinople; but before examining the evidence for such exports, we should note briefly the description of Thessalonica and its environment which John Cameniates provided not quite three centuries later. If we contrast this later description to that of the seventhcentury city of the Miracula, it is possible to trace agricultural and commercial development, matters that must often remain subject to hypothesis at Constantinople. The city, John noted, had recently expanded. Both the walls and the plains around them offered refuge to subjects of the Byzantine emperor whenever they had to flee from islands that were prey to the depredations of Arab pirates. The years of warfare were long past ; swords had become plowshares as the Slavs elected the ways of peace. To the east of the city one could see cultivated fields, interspersed with growing vines, while to the west lay another urban center, that of Berrhoea. About the walls of this second city lay villages, cloaking the countryside like a mantle. Some of them were subject to Byzantium's emperor; others paid tribute to the Bulgars. In times of peace a healthy commerce flourished between them, increased and multiplied by the great rivers flowing nearby. Up the Vardar ships from the sea bent their course, while on fast days the river supplied fish in abundance to provision the urban market. In one respect Thessalonica of the tenth century seems to have differed somewhat from its predecessor of the seventh. While the land outside the walls still lay open to agricultural exploitation, the city now maintained a fleet of ships to search out and import grain for the urban market. Although, as is manifest from details in the Miracula such as the collection of the export duties, a water-borne grain trade had by no means ceased to exist; urban growth had brought with it an increased demand which could be satisfied only by an organized system of imports.173 Turning back to the Miracula, and moving with its narrative along the ThracoMacedonian littoral, we find that lands not far from Thessalonica may have furnished grain in exchange to Constantinople. Once the siege of 662 had ended, the Slavs returned to their villages, abandoning their hopes for a home 172 A. Tougard, D e l'histoire profane d a m les actes grecs des Bollandistes (Paris, 1874), pp. 158, 164, 166, 187; Miracula S. Demetrii, sec. 193 ( A c t a SS., Oct. IV, I 75). 173 Cameniates, De excidio thessalonicensi, cc. iii-vi, lxi (ed. Bonn, pp. 490-496, 5 72).

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 123 in Thessalonica. Two tribes alone remained an object of concern to the emperors : the Strymonians and the Rhynchinians. The Miracula explains why Constans I1 and his successors found it necessary to mount expeditions against these people to the west of their ~apita1.l'~ "And once again assistance came from the intercession of Him who loves the city. For while the other Slavs in these parts were settling down and had laid down their arms, those from the Strymon and the Rhynchinus continually seized numbers of sailors returning from the transport of produce to the imperial city. And they returned with many craft to their homes, making prisoners of those from the islands and from the narrows of the sea, together with those in Parium and the Proconnesus; even did they make prisoners of those in the toll station together with the ships of the fleet." Thus grain came to Constantinople from the west. Years later, in 821 to be precise, the biographer of a chance voyager in the neighborhood of the Strymon provides further information on shipping in that region. Saint Gregory of Decapolis, turned back probably by force majeure from Constantinople, took ship to Chrystopolis on his return voyage to Thessalonica. Disembarking for reasons his biographer did not disclose, he pursued his journey by a great road, probably the Via Egnatia. In the words of the biographer, Ignatius: "From Enos by boat he went to Chrystopolis. From there, disembarking from his boat, he came to a certain river, on which he fell in with Slavic pirates, coursing along the banks of the river with their small craft, and plundering the boats they came upon."175 Evidently, according to the observations of the editor of the Life, the Slavs in the early ninth century were not quite the peaceful people Cameniates makes of them in the tenth. They seem to practice their piracy on a great river, probably to be identified with the Strymon. In 821, then, shipping must have moved inland from the sea, up the course of a river famous for the productive fields along its banks. Unfortunately Ignatius has little else to say of the sights Gregory may have seen in his journey from Constantinople to Thessalonica, and it is left to the Arab prisoner of war, Harun ibn Yahya, to mention the many villages and fields of grain to be seen as one made the twelve days' journey from east to west.176One of the most important ports in the region was the harbor of Rodosto. In the eleventh century the logothete of Michael VII, Nicephoritzes to his familiars, established at its gates a depot or foundax wherein he endeavored The significance of Rodosto, to corner all the grain the hinterland pr0d~ced.l'~ Miracula S . Demetrii, sec. 190 (Acta SS., Oct. IV, 174). Vita Gregorii Decapolitae, c. x (ed. Dvornik, pp. 5 4 f . , and c f . p. 31). Between t h e periods represented b y t h e Miracula o n t h e one hand (early and middle seventh century) and t h e l i f e o f Gregory o n t h e other (early n i n t h c e n t u r y ) , t h e cleisurae o f t h e Stryrnon had been organized b y Justinian I1 and garrisons installed there. Possibly exports f r o m t h e S t r y m o n are t o b e dated f r o m t h a t point. Constantinus Porphyrogenitus, De Thematibus, ed. A. Pertusi (Studi e Testi, 160) ( R o m e , ~ g j z )pp. , 88f. 176 A. A. Vasiliev, " H a r u n i b n Y a h y a and His Description o f Constantinople," Seminarium Kondakovianum, V (1932),p. 162. 177 Bratianu, Etudes byzantines, pp. 141-157. 175

124

J O H N L. T E A L L

however, began before the eleventh century. In the tenth, Gregory, a disciple of Saint Basil the Younger, paused at the shrine of Saint Stephen to pray for a safe voyage by land and sea as he journeyed to his proasteion to gather the season's harvest in the summer's heat.l78 Much of the economic adjustment after 619 can certainly be explained simply by the history of Rodosto itself. Abandoned and disused in Justinian's time, its port was used in the tenth to gain access to fields of century by at least one of constantinople9s grain he must have accounted himself fortunate to own. Yet if the Arabs at Cyzicus in the seventh century deprived Constantinople of Sicilian grain, they must also have kept from its markets supplies from the Strymon and possibly from Rodosto as well, thus forcing the city to gather what it could from the European suburbs and the provinces to the east. Of vital importance were the grain fields located in what had been known, under the later Roman Empire, as the European district of the diocese of Thrace. When the Avars fell upon the city in 626, they captured citizens who had ventured out beyond the Theodosian walls to the tenth milestone to gather the harvest they had planted. At the time of the siege of Thessalonica in 662, a bureaucrat from Constantinople owned a proasteion in Thrace, while the sources of the Khitab a1 Uyun report that Maslamah's attack in 717 deprived the citizens of gainful employment by land and sea.180 During those grim yearsagain according to the Khitab-the suburban fields could not support the massed army, but their condition seems gradually to have improved. When the army of ConstantineV took up its position in Thrace near the city walls, it deprived the partisans of Artavasdus, besieged within, of the necessities of life.181 Finally, the ninth-century source of the Khitab a1 Uyun has a glowing report to submit on agricultural conditions in the lands nearest the Theodosian walls :I82 "If an army went at the present time to A1 Kustantiniyya, when it was in need of provisions, and there was no importation of corn, their provender dealers would bring them more than they wanted, from the places nearest them.'' There seems little reason to doubt this source. Krum was able to gather rich booty from the Thracian jbroasteia; Leo V founded numerous cities in Thrace and Macedonia, and the rich Thracian proasteia of a certain Stephen were spared the effects of drought thanks to a miracle worked by a refugee saint.183 No less important were the lands across the Bosphorus and on the south V i t a Basilii Iunioris, c. xlii (Acta SS., Mar. 111, App., p. 28).

C ~ Y O RPasch., .

ed. Bonn, p. 717. lso Tougard, Actes grecs, p. 152; Khitab a1 Uyun, trans. Brooks, "Campaign of 716-718," JHS, p. 23. Nicephorus, ed. de Boor, p. 61. Khitab a1 Uyun, loc. cit., note 180, above. See the discussion of this passage in P. J. Alexander, T h e Patriarch Nicephorus of Constantinople: Ecclesiastical Policy and Image Worship in the Byzantine Empire (Oxford, 1958), pp. 123f., who suggests that Leo I11 or Constantine V may have been responsible for the abundance of the early ninth century which contrasts so strongly with the dearth of grain in 718. "Perhaps the mass transplantations of colonists to Thrace under the Isaurian dynasty were prompted by the desire to recolonize this region and render it capable of supplying the city with grain." This hypothesis seems all the more reasonable in view of the similar measures undertaken, under strikingly comparable circumstances, by llehmed 11. See above, note 62. 183 Ps.-Symeon Magister, De Leone, c. viii (ed. Bonn, p. 612) ; Theophanes Cont., De Leone, c. xix (ed. Bonn, p. 30); Acta Davidis, Symeonis, et Georgii Mitylenae, c. xxi, ed. in Analecta Bollandiana, XVIII (1899)~P. 236. 179

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 125 shore of the Sea of Marmora. The biographer of Michael Maleinus reported Bithynia to be abundant in flowing water and in all good things. We need not, however, wait until the tenth century to find evidence of the importance of its land to the subsistence of Constantinople.ls4When in 641 Valentinus and his army plundered within the city's walls vineyards owned by many of its citizens, economic interests dictated that an early peace be concluded between the opposing forces.185 I n the ninth century, the patron of Saint Basil the Younger owned a vineyard near Chalcedon wherein he wished to bury the Saint's relics, while Leo the Wise assigned the wine from imperial estates in Bithynia to the support of Saint Euthymius' rnonastery.lg6 The Bithynian lands of the Opsician theme, covered with the small villages St. Theodore encountered every day in his journey from Saccudio to Lampsacus, also produced grain in abundance. On the family properties at Boscytium, not far from Saccudio, Theodore and his uncle Plato hauled out the dung from the stables, irrigated the land, and broke new furrows with the plow.18' In the district of Prusa, the monastery of Saint Eustratius owned proasteia whence the monks transported by cart the grain they needed at home. ls8 Once the Arab armies had been driven off in 718, boats set out from Constantinople to Chalcedon and to the southern shore of the Sea of Marmora, bound to collect grain and other provisions for those who had survived the siege. In and 866, ~ ~ m b a t i u s ~ e o r g ePiganis, the strategos of the Opsician theme, laid waste at harvest time the country estates of Constantinople's powerful families.189 Many a monastery at Constantinople owned proasteia in the Opsician theme and elsewhere in Asia Minor. But the lives of the saints, abundant as they may be in tales woven about the journeys a steward of such an estate might make to the capital, fail to indicate whether or not the functionary brought with him grain to supply the mother house.lgO 1n-the life of Saint George of Amastris, the Pontic coast at the end of the ninth century is reported as the scene of vigorous activity in shipping, with merchants from Amastris found at Trebizond and with the harbors near the mouth of the Sangarius in constant use. The themes for which such harbors provided natural outlets exported their grain, as we have seen, to the Crimea; of any shipments to Constantinople there seems to be no evidence.lgl Among the ports situated on the southern and western coasts of Asia Minor, destined-for later importance when Italian merchants exported their cargoes of grain from there, Ephesus at least may have been a "city that shared its 184 Vita Michaelis Maleini, c. xv, ed. L. Petit, "La vie de saint Michel Maleinos," Revue de l'orient chre'tien, VII (1902), p. 560. 185 Nicephorus, ed. de Boor, p. 30. Is6 Vita Basilii Iunioris, c. lv (Acta SS., Mar. 111, App., p. 31); Vita Euthymii, IX, 16f., ed. de Boor (Berlin, 1888), p. 28. ls7Theodorus Studites, Ep. I, 3 (Plligne, PG, 99, col. g16f.); Vita Theodori Studitae, cc. vi, vii (Migne,

ibid., cols. 121-123).

Vita Eustvatii hegumeni, c. xxix (ed. Papadopoulos-Kerameus, Analecta, IV, p. 387). Theophanes, ed. de Boor, I, 397; Theophanes Cont., De Basilio, c. xix, (ed.Bonn, p. 240). lgO See, e.g., Vita S. Ioa?znicii, cc. xv, xxxiii, lviii (Acta SS., Nov. 11, pt. 1, 361, 392, ~22). lgl VitaGeorgii Amastvidos, c. xxvii, ed. V. G. Vasilievsky, Trudy, I11 (St. Petersburg and Leningrad, Is8 Is9

1908-1930)~PP. 42,

54f.

126

J O H N L. T E A L L

sustenance" with Constantinople. Even in the eighth century it held a fair, while, not far from it, the Menander valley had won a reputation among ninthcentury sources for its richness and fertility.192In the harbor of Ephesus, Saint Gregory of Decapolis found many ships, bound for Constantinople, but fearful of venturing into waters that pirates had rendered dangerous. Gregory persuaded the masters to set sail; further evidence that coastal shipping continued despite the risks.lg3For the support of a monastery founded near Ephesus by Saint Lazarus, monks in the eleventh century sought out grain in the "ports of Lydia," while two centuries earlier pious citizens of Smyrna with a boatload of grain alleviated a famine afflicting the monks and their charges on Lesbos.194 Finally, for the ports on the southern coast indications are too few to permit generalization. Exports from Attalia, situated in a populous plain according to a1 Istakhri or his sources, supported the army of Nicephorus I1 in Crete, but the ninth-century Life of Saint Anthony the Younger reveals no more than that the port was active in that century.lg5 But if the cities in the sixth century could not "share their sustenance," how could they have done so in the seventh? What adjustments of production or distribution permitted the Empire's heartland to satisfy in general a demand it never met successfully under Anastasius, Justinian, or Maurice ? Roads and bridges were maintained in a condition adequate to bear heavy, constant traffic. Good roads united Caesarea in Cappadocia with the market towns surrounding it, and communications between Berrhoea or Thessalonica and their hinterland seem to have been swift and easy. The Via Egnatia could still be used in the ninth century, lesser roads elsewhere were heavily frequented in the succeeding centuries. In his journeys about Greece, Nicon the Metanoite found land travel by no means difficult, while the voices of those passing on the highway nearby annoyed Lazarus at his monastery near Ephesus. His monks complained that these same travellers continually ate the beans planted regrettably close to the thoroughfare. In Galatia, Luke the Stylite grieved over the poor who lined the great road. The construction and maintenance of the highways was a corvke that burdened the taxpayer, and the peasants near Lazarus' mountain retreat were apparently skilled in such tasks, for, in their eagerness to be able to visit the holy man more often, they broke up rocks and built a fine road up to the recluse's cell.196 While such information from the lives of the saints demonstrates that transportation facilities were maintained in good condition, it has not been lg2 Theophanes,

ed. de Boor, I, 469; Cedrenus, ed. Bonn, 11, 199. Vita Gregorii Decapolitae, c. ix (ed. Dvornik, p. 53). 194 VitaLazavi in monte Galesio, c. xc (ActaSS.,Kov. 111, 536) ; ActaDavidis, c. xiii (ed. AnBoll, XVIII [18991, p. 225)lg5 A1 Istakhri, trans. Mordtmann, p. 43; compare the accounts from ibn Hauqal translated by b1. Canard in A. A. Vasiliev, Byzance et les Avabes, 11, pt. 2 (Brussels, 1935-), pp. 414-416; Vita Antonii iuniovis anachovetae, passim, ed. A. Papadopoulos-Kerameus, Pravoslvanyi Palestinskii Sbonzik, X I X (= fasc. 57), (1go7), pp. 186-216. lg6 Caesarea and Ephesus: Vita Lazari, cc. xi, xxvi, xxxiv, xxxvi (Acta SS., Kov. 111,512, 517, 520) ; Thessalonica: Cameniates, ed. Bonn, p. 500; Via Egnatia: Vita Gregorii Decapolitae, cc. xviiif. (ed. Dvornik, pp. 62f.); Greece: Vita Niconis, ed. Lambros, pp. 15gf., 176; Vita Lucae Stylitae, c. ix (ed. Delehaye, p. 204).

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 127 possible to discover any evidence of technological improvement that would have cheapened transport and have rendered the hinterland, in consequence, more accessible to far distant markets. References to the cartage of grain are frequently to be found in the saints' lives; mules were sometimes applied to the task as well as oxen, but whether loads were greater and costs cheaper than formerly we do not know.lg7 \Ve cannot prove either that a significant diminution of local demand was one of the factors in the economic adjustment, for that could have resulted only from an extensive and permanent reduction in the quantity of cities, together with a profound alteration in the character or quality of urban life. A recent study of Byzantine cities, it is true, hints at such developments, suggesting that industrial centers, as well as those lacking in commercial or political importance, disappeared, contracted, or turned to agriculture during the seventh and eighth centuries.198Even in Asia hlinor, the author contends, the decline of Ephesus and Pergamum and the disappearance of Heracleia lend substance to a comment made by the anonymous author of the "Regions of the \Vorld":1g9 "In the days of old, cities were numerous in Rum, but now they have become few. Most of the districts are prosperous and pleasant and have each an extremely strong fortress on account of the frequency of the raids. .. . To each village appertains a castle where in times of flight they may take shelter." While urban life was certainly threatened and possibly partly extinguished in some provinces, it is impossible either to accept without question the author's general conclusions from his evidence or, in turn, to deduce from his conclusions a lessening in the urban demand once represented by the provincial cities. Even in Greece and the Balkans, cities and the urban way of life subsisted in some regions, and elsewhere soon began again to flourish. Commercial activity of some proportions, as we have seen, distinguished Thessalonica in the seventh century, and even in 747 Constantine could recruit brick makers and potters from Thrace and Greece.20° Episcopal sees in Thrace, Haemimontus, Europe, and the Rhodopes reappeared as the Iconoclast emperors sought to convert the Slavs and protect their northern frontiers.201In Thrace and Macedonia Leo V founded cities, which may not have been dedicated exclusively to purposes of defense.202Late in the ninth century, perhaps early in the tenth, Saint Peter at Argos taught the children of his city, as well as strangers under his protection, the liberal arts and crafts that would be most useful to them if lg7 V i t a Eustratii, c. xxix (ed. Papadopoulos-Kerameus, Analects, p. 387) ; V i t a Lazari, c. ccix ( A c t a S S . , Kov. 111, 572); BrBhier, Civilisation byzantine, pp. 175f. lg8 A. P. Icazhdan, "Vizantiiskie goroda v. T'II-XI vv.," Sovetskaia Arkheologiia, X X I (19j4), pp. 164-188. For a different point of view, see Kirsten, "Byzantinische Stadt," pp. 19-34, who believes that the ancient polis was partly replaced in the middle ages b y the fortress settlement. lg9 T h e Hudud a1 A l u m : T h e Regions of the World, trans. V. Xinorsky, Gibb hlemorial Series, N. S., 11 (London, 1g37), p. 157. For flight to the castra, see Theophanes, ed. de Boor, I, 452; Ps.-Kicephorus Phocas, De velitatione bellica, cc. ii, xii (ed. Bonn, pp. 188, 215). Theophanes, ed. de Boor, I, 440. For craftsmen and builders in Bithynia in the early ninth century, see V i t a Ioannicii, cc. xx, xli, xlv ( A c t a S S . , Nov. 11, pt. I , 351f., 407f., 410). 201 li. Dvornik, Les Slaves, Byzance et Rome a u IXeme siBcle (Paris, 1926), pp. 74-99. 202 Genesius, ed. Bonn, p. 28.

J O H N L. T E A L L

they moved to another city, to the fields, or to a f i ~ o a s t e i o nAthens . ~ ~ ~ manifested signs of activity in the tenth century, while the life of Saint Nicon tells of Jewish merchants at Sparta and of specialized craftsmen engaged in the building t rades.204 In the themes of Asia Minor both the coastal cities and the unwalled market continued to exist, and in the eighth century, as we have town or KG)~OTTOAIS noted, Ephesus held its fair.205 Round about the major urban centers, the countryside continued to support unwalled market towns throughout the seventh and eighth centuries. In one of his novels, Heraclius mentions the emporia, and the parents of Saint Stephen the Younger are supposed, in the company of many others, to have fled Constantinople during the Iconoclast ~ . ~ ~ ~ and Caesarea were large persecutions, taking shelter in a K W U ~ T T O A I Nicaea and populous in the tenth century, the former containing an important colony of Jewish merchants.207 The Life of Saint George of Amastris, in particular, reflects an urban point of view. The chief men of Amastris, along with the clergy, visited the Saint and begged him to become their bishop, suggesting that it was his duty toward the city which had educated and sustained him.2O8 Some further points should be noted. The appearance in many provinces of the Empire of the fortress settlement with its reduced precincts is significant, but does not necessarily prove that population and demand diminished. Furthermore, the conditions that led men to abandon the extensive area of the ancient polis and to take refuge in the fortress would hardly have been conducive to the export of agricultural surpluses. I t was precisely those port cities, which, as we have seen, had once been in a position to challenge Constantinople for grain supplies that survived adversity and hard times. Thessalonica and Ephesus, in fact, maintained some degree of urban specialization, and consequently of need for agricultural products, even during the seventh and eighth centuries. In the absence of any indication of improved facilities for land transportation it is difficult to believe that the supposed disappearance of inland urban centers could have worked greatly to the benefit of Constant inople. Within the agricultural economy itself there seem to have been no technological improvements significant enough to be considered adjustments. Although it is possible that the horseshoe and the improved collar were used after the ninth century, the wither-yoked ox remained the foremost beast of labor.209Mules, it might be added, won, or maintained, considerable importance. With admiration for his hard work and humility, the biographer of Theodore 203 Vita Petri ep. Argivorum, c. xii, ed. A . Mai, Novae Patrum Bibliothecae, IX, iii (Rome, 1852-1905), 8. 204 For Sparta, Vita Niconis, ed. Lambros, pp. 163f., and on Athens, Kazhdan, op. cit., p. 177f. 205 See above, note 192. 206 Novellae constitutiones imperatorum Post Iusfinianum, Coll. I, nov. xxiv, ed. Zacharia, Jus. Gr.-R.,111,p. 42. Vita S. Stephani iunioris, ed. Migne, PG, 100, col. 1088.See also Ducange, Glossarium ad scriptores mediae et infirmae graecitatis (Lyons, 1688), S.V. 207 Theophanes Cont., De Constantino prophyrogenito, c. xlix (ed. Bonn, p. 464); Vita Constantini olim judaei, c. li (Acta SS., Nov. IV, 642); Vita Lazari, c. xxvii (Acfa SS., Nov. 111, 517). 2os Vita Georgii Amastridos, c. xv (ed.Vasilievsky, p. 26). 209 Lefebvre des Noettes, "Le systeme d'attelage du cheval et du boeuf & Byzance," Me'la~ges D iehl, I, (Paris, 1930), pp. 183-191; BrBhier, Civilisation byzantine, pp. 175f.

T H E G R A I N SUPPLY O F T H E B Y Z A N T I N E E M P I R E 129 the Studite tells how the youthful Saint bent his shoulders to heavy and repellent baskets of mule dung on his Bithynian estate.210In the region of Thessalonica, mules were used for the lighter tasks after the oxen had broken new land or fallow. Whether mule or ox, however, the animal never pulled behind him the wheeled, overturn plow. Work in the fields was conducted with the aid of the old aratrum, supplemented by a double-pronged hoe and a spade.212 In general arable lands failed to receive adequate manure. While the peasants of the Farmers' Law let herds into their parcels of land when all harvests had been gathered, they also made use of woods for pasturage.213The lack of rich alluvial plains in many of the provinces forced the peasant in some instances to continue the old practice of mountain pasturage. Leo the farmer, in the Life of Saint Paul of Latros, left his goats near the Saint's cave and went back down into the plain to cut his grain at harvest time.214 When Constantine VII complained that the sale of mountain land might injure one's neighbors he indirectly suggested that the use of such pasturage may have been widespread where Mediterranean conditions of soil obtained.215Thus in some regions the peasant had perforce to deprive his arable lands of manure, a practice all the more dangerous since the small farmer was generally unable to stock his lands properly. Lacking the financial resources and fodder to support a herd, he might beg from kindly saints a single ox when his own sickened and died. The herds of swine or smaller animals, so often guarded by small children, could never provide the manure that larger animals would have supplied in abundance.216 While it would not be difficult on the basis of such evidence to condemn Byzantine agriculture as backward and unprogressive, for two reasons criticism of this nature misses the point. Had the peasant of Western Asia Minor, the Thraco-Macedonian littoral, or Greece, ventured to use the northern overturn plow for his spring labors, he would quickly have found that the hot sun baked out the precious moisture the clods contained. A recent student of east European and Middle Eastern agriculture has suggested that manure may be of no great advantage under dry-farming conditions.217 Given Mediterranean soils and -

Vita Theodori Studitae, c. vii (Migne, P G , 99, col. 124). P. Iioukoules, Private Life of the Byzantines, V, p. 255. See Theophanes Cont., De Michaele Amor., c. iv (ed. Bonn, pp. 43f.);Vita Lazari, c. ccix (Acta SS., Nov. 111, 572). 212 BrChier, Civilisation byzantine, pp. 171f.For the seventh century see the tools mentioned in the Nbpo5 r ~ w p y l ~ bcc. ~ , xxii, lxii, ed. W. Ashburner, "The Farmer's Law," Journal of Hellenic Studies, XXX ( I ~ I O ) , pp. 100, 105. 213 "Farmer's Law," cc. xliv, lxxviii (ed.Ashburner, op. cit. pp. 103, 107). 214 Vita Pauli in monte Latro, c. xiii (ed.AnBoll, XI [1892], p. 45); Vita Lazari, c. ccxxxv (Acta SS., Xov. 111,. -581). , 215 NOV. constt. imperatorum post Iustinianum, Coll. 111,nov. vi (A.D. 947), ed. Zacharia, Jus Gr.-R., 111, 256. 216 Vita Ioannicii, c. ii (Acta SS., Xov. 11, 333); cf. Acta Davidis, c. iv (AnBoll, XVIII [18g8], p. 214); Vita Eustratii, c. xxv (ed.Papadopoulos-Kerameus, Analects, IV, p. 384); ibid., c. xiii, p. 377; Vita Philareti eleemosynarii, ed. Fourmy and Leroy, p. 119. 217 D. Warriner, Land and Poverty in the Middle East (London, 1948), p. 123; cf. C. E. Stevens, "Agriculture and Rural Life in the Later Roman Empire," Cambridge Economic History, I (Cambridge, 1911-), P 94. 210

211

J O H N L. T E A L L

climates, agricultural practice in the Byzantine Empire demanded the investment of a large amount of labor, together with the most careful use of the resources at its command. Although the dramatic struggle of the poor and the powerful has led us sometimes to forget it, the Byzantine rural classes-whether peasant or large landowner-never forgot that success depended upon work, tenacity, and knowledge. Articles in the Farmer's Law required that the land be carefully plowed, that vineyards be dug, dug again, dug yet again and fenced aboutI2ls and the owners of large estates, to judge from the composite picture drawn from the lives of the saints, never left such matters to chance or to the operation of law. Nicon was sent out by his father to examine the family estates; the parents of Luke the Stylite were wise in agricultural lore; the father of Luke the Younger, a very wealthy man, personally oversaw his estate; when fortune failed him, Philaretus turned his own hand to the plow.219Even an emperor, Constantine IX, brought soil from the mountains, transplanted trees, made sterile fields into fertile parks. According to Psellus, Constantine took a personal interest in all these operations, and the phrases the historian uses suggest that the Emperor was by no means ignorant of economic rationalization, selective breeding, labor-saving processes, and the suppression of fallow.220 Popular belief attributed similar knowledge to that fascinating man, Leo the Mathematician. Appointed Archbishop of Thessalonica, Leo arrived in the city at a time of famine and distress. By astronomical calculation he advised his people of the proper time to plant their seed. The good citizens were amazed as spring approached and the crops in the fields were plentiful.221 The Patriarch Photius, a fellow-member with Leo of the intellectual circle formed around Caesar Bardas, had read ancient agricultural theory and his comment on Vindanius Anatolius is particularly worth noting:222 "Read the work of Vindanius Anatolius.. .and as our own experience has shown us in many instances, it is useful for the cultivation of the land and agricultural works, perhaps the most useful of all treatises on the same subject." Careful as Byzantine agricultural practice always was and learned as it could become on occasion, farming for subsistance rather than for the market seems to have remained its objective. Cecaumenus, cautious no less in his farming than in his fighting, urged his son to equip his estate with mills and with workshops. Grow grain, vines, trees, and look after your flocks that you 218 For the practices, and for the crucial problems of water rights and ownership of trees, see "Farmer's Law," cc. xxxi, xxxii, lxxxiii, lxxxiv (ed. Ashburner, op. cit., pp. 102,107f.). For an interesting Marxian analysis of agricultural practice in the law-which sees, in contrast to the view expressed above, evidence of sparse occupation of the soil and extensive agricultural practice-cf. E. E. Lipshitz, "Vizantiiskoe krest'ianstvo i slavianskaia kolonizaciia," Vizantiiskii Sbornik (Moscow and Leningrad, 1945)) pp. 96-143 (in German translation: Byzanz und die Slaven [Weimar, 19511, PP. 32-105, esp. 53-57). 219 Vita Niconis, ed. Lambros, p. 134; Vita Lucae iunioris in Hellade, c. xi, ed. AnBoll, XI11 (1894), pp. 85f.; Vita Lucae Stylitae, c. v (ed. Delehaye, p. 200); cf. Laudatio Philothei Opsiciani, c. xvi (Migne, PG, 136, col. 156). 220 Michael Psellus, Chronographia, VI, I 73-1 75 (ed. Renauld, 11, pp. 56 ff .). 221 Theophanes Cont., De Michaele Theophili f., c. xxviii (ed. Bonn, p. 191). z22 Photius, Bibliotheca, cod. 163, ed. I. Bekker (Berlin, 1824)~ pp. 106f.

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 131 may live in self-sufficiency, he c o ~ n s e l l e d . ~The 2 ~ parents of Luke the Stylite were highly praised indeed when they were lauded for having lived selfsufficiently.a4 Judging from agronomic analogy, Byzantine agriculture probably sought its profits from the vine and fruit trees rather than from the cultivation of cereals. These were the forms of agricultural endeavor that necessitated the greatest outlay of work and that were to be undertaken only by specially trained men. l i e even read of a Symeon Ampelas who won both fame and fortune from properties devoted to the cultivation of the ~ i n e . 2 2 ~ To sum up then, if Byzantine agriculture met the demand for grain that Constantinople, the armies, and the lesser cities imposed, it did so because the total acreage of arable land was fully exploited, an achievement traceable ultimately to the augmentation of the labor force. From many different quarters came the new hands to till the fields. The waging of successful wars led to the acquisition of numerous slaves, and the slave seems to have won new importance in the labor force of the Byzantine e ~ t a t e . Liberal ~ 2 ~ policies of settlement and colonization increased the number of persons, both free and servile in status, available for work. In 762 Constantine V welcomed Slavic refugees from the Bulgar Empire, establishing them along the Artana in Bithynia. I t was a settlement more massive and probably more permanent than the military colony of "special people" that Justinian had founded in the Opsician theme.227Peoples from the eastern frontiers were brought into Thrace, and to judge from the gift that the future Leo I11 presented to Justinian 11, their life there must have been a prosperous one.228Money for land and grain for seed were given to the Saracen prisoner who consented to receive baptism. In turn, the father of the family who agreed to accept such a one as son-inlaw enjoyed three years' exemption from taxes on head and hearth.229Indelicate methods were sometimes employed as well, as the widow Athanasia learned when an imperial edict ordered her to accept a barbarian as husband.230 From the coasts and islands came refugees to swell the population of the inland Their reception was not always cordial, and an episode in the Life of Luke the Younger demonstrates how necessary and effective was Cecaumenus, Strategicon, cc. lxxxvii, lxxxviii (ed. Vasilievsky and Jernstedt, p. 36). Vita Lucae Stylitae, c. v (ed. Delehaye, p. 200). 225 Leo Diaconus, VII, I (ed. Bonn, p. I 13) ; a professional vintner in Vita Lazari, c. xix (Acta SS., NOV.111, 512). 226 See "Farmer's Law," cc. xlv-xlvii (ed. Ashburner, op. cit., p. 103); Cedrenus, ed. Bonn, 11, 130, 207, 341; Leo Grammaticus, ed. Bonn, p. 258; Vita Michaelis Maleini, c. xi (ed. Petit, ROC, VII [ ~ g o z ] p, p. 557f.); Vita Andreae Sali, sec. I (Acta SS., May VI, 5); Vita Pauli in monte Latro, c. iii (ed. AnBoll, XI [1892], p. 23); Vita Athanasiae Aegineticae, c. xvi (Acta SS., Aug. 111, 174), and the references above, notes 84, 85. 227 Theophanes, ed. de Boor, I, 364, 432; Nicephorus, ed. de Boor, p. 36. For contrasting views on the extent and significance of these settlements see P. Charanis, "The Slavic Element in Byzantine Asia Minor," Byzantion, XVIII (1948), pp. 69-83, and Ostrogorsky, Byzantine State, pp. 116f., 150. 228 Theophanes, ed. de Boor, I, 391, cf. 422, 364, 365, and for later settlements under Theophilus, Leo Grammaticus, pp. 215, 231 f. 229 Constantinus Porphyrogenitus, De cerimoniis, 11, 49 (ed. Bonn, I, 695). 230 Vita Athanasiae, c. ii (Acta SS., Aug. 111, 170). 231 See, e.g., Cameniates, ed. Bonn, p. 504; Acta Davidis, c. xviii (ed. AnBoll, XVIII [18gg], p. 232f.); Vita Pauli in monte Latro, cc. ii, viii (ed. AnBoll, XI [18gz], pp. zof., 33); Vita Athanasii ep. Methonensis, c. iv, ed. A. Rlai, Novae Patrunz Bibliothecae, IX, pt. 3, pp. 34f. 2s3

224

9*

132

J O H N L. T E A L L

the work of the imperial bureaucracy in creating a home for such persons.232 Luke's grandfather had come down to Castorium in Thessaly, and God, reports Luke's biographer, had been good to him, and his flocks and herds had prospered. Perhaps for this very reason the villagers greeted him with suspicion and hostility, and it was only when he was armed with the authority of an imperial edict that he could settle on the land with his fellow refugees. Prope;ties were divided up, concludes the biographer, and truly miraculous peace and happiness reigned from that time on. This we may doubt, but the episode reveals, better than any other, what migrations and settlements meant in human terms. An expanding population, resulting from a surplus of births over deaths, played as large a role in rural as it did in urban expansion. Certainly by the tenth century, and probably before, isolated freeholds (riypi8la) or large estates (7rpodtm~1a) developed out of nucleated villages. I t is the origin of these separate settlements that interests us particularly, and the Treatise o n Taxation fortunately offers precise and interesting information on their origin. A peasant often left many children at the time of his death, to some of whom he would have assigned his properties within the village; to some his properties outside. Those whose share lay outside very likely did not wish to remain far from their property, and so moved out to it, built a house, and improved the land. Others may have had many slaves or cattle, some may have felt crowded by evil neighbors; they, too, moved outside the village to improve their land and to enjoy the space of a larger undertaking.233Thus anonymous men from the villages joined monastic heroes such as Lazarus, Athanasius of Athos, and Dorotheus of Thrace. Trees were planted and precious streams of water diverted to irrigate them. The forests were cut back, water mills erected, boundaries carefully marked, and harvests grown and gathered with plow, hoe, and sickle.234 While we cannot estimate the extent of arable land, evidence drawn from hagiography (and confirmed in documents of a more official nature), seems to justify crediting Byzantine agriculture with turning back a frontier, attacking with determination and ingenuity the encroachments of wood and waste. Such a conclusion, to be sure, implies an estimate of the Byzantine rural achievement that differs from that usually offered. Byzantinists have concentrated intently on the struggle between the powerful and the poor, forgetting that this was a society of success as well as of failure. Without doubt there were many who But "went to the wall" as did Philaretus or the father of John Psi~haites.~35 there were others as well; had Byzantium not been as much a society of winners as of wasters, it would never have survived. -

Vita Lucae Helladis, c. iv (ed.AnBoll, XI11 [1894], pp. 83f.). F.Dolger, Beitrage zur Geschichte der byzantinischen Finanzverwaltungbesondersdes10und11,Jahrhunderts (Leipzig, 1927), p. 115. 234 Vita Lazari, c, xxxiv (Acta SS., Nov. 111, 520); Vita Lucae Helladis, Migne, PG, 111, col. 464; Vita Dorothei, c. x (Migne, PG, 120, col. 1064); ibid., c. xvi (Migne, PG, 120, col. 1072); Vita Athanasii Athonitae, cc. xxv, xxvi, edited in AnBoll, XXV (1906), pp. 35f., 49, and passim. Cf. Vita Theodori Studitae, c. vii (Migne, PG, 99, col. 124); Theodori Studitae, Oratio XII, 11 (Migne, P G , 99, col. 814). 235 Vita Ioannis Psichaitae, c, ii, ed. P. van den Ven, "La vie de s. Jean le Psichaite," Le Muse'on, N. S., 111 (1902), p. 105. 232

233

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 133

IV. CONCLUSION

The solution to the problem posed in this study has proven to have been one of interlocking changes within the categories of demand and supply. If the fifth century is taken as a point of departure, it is possible to delineate for that age a grain economy that had to be conducted under imperial auspices on a Mediterranean-wide basis for the benefit of the greater cities and the expeditionary armies. When coupled, during the sixth century, with the increasing strain of the military effort, urban demand threatened continually to exceed an agricultural production already hampered by certain serious deficiencies. Change came at the end of that century not through the application of deliberate remedies to weaknesses within the productive sector of the economy, but rather through a diminution and deconcentration of the two great demand factors. Considering certain conscious imperial policies, and, even more, the impact of disease, we have every reason to believe that the need for grain was far less in the Constantinople of Maurice than it had been for generations preceding his reign. Ironically, the very weakness of Maurice and the other emperors of the late sixth century helped to ensure survival after the crises of 619 and 641. The army that faced the Persians and the Arabs had learned to find what grain it needed in the regions nearest it. Supplies from Egypt, we may suspect, were rarely found on the Euphrates or Danube frontier even before that province was lost to the Empire. Since neither Sicily nor Africa could compensate in full measure during the seventh and eighth centuries for what had been lost with Egypt, it was fortunate that demand both at Constantinople and among the military remained at a low level. Resources were carefully husbanded; then and later the authors of handbooks of strategy were greatly concerned lest an excessive concentration of troops in one locality overwhelm agricultural resources that were to be shared among the cities. By the beginning of the ninth century, the worst hardships were past. The producing population became ascendant as forest and waste were cut back in the areas of Thrace and Eastern Macedonia, and in Western Asia Minor, where Constantinople had then to turn for its grain. Transportation facilities were adequate; the possibility of an increased local demand constituted no threat, and in fact it might actually have declined; in some quarters agricultural technique or experiment remained at a level that justifies its being compared with the best classical agronomic theory. Yet, in none of these respects do the sources permit us to speak with assurance of an adjustment, of a change contrasting clearly with the economy of the fifth or sixth century. If supply increased, its greater quantity was the result primarily of many more hands-free and servile alike-to work the fields. So numerous were the hands, so scant the resources, that a historian has recently described medieval Byzantine rural society as caught up continually in a struggle for the

134

J O H N L. T E A L L

land.236 As a consequence of these population pressures, the amount of fallow land may have diminished in contrast to its extent in the fifth and sixth centuries. Certainly, the epibole of the tenth century was, if not a totally different fiscal device, at least not at all the burden Procopius described.2S7 I t was well that the labor force had increased. The tenth century saw Constantinople reach a climax in its medieval growth, while the emperors adopted a military policy that strained the forces of production to the utmost. A new era was about to begin: one as troublesome for the grain economy as for many other sectors of Byzantine life. APPENDICES A. THE POPULATION O F CONSTANTINOPLE

The estimate given in the text above is based upon the following factors: I. A reconsideration of the definition of the 4,388 domus listed in the Notitia Urbis Constantinopolitanae. They do not seem to be-as A. AndrCadks, "De la population de Constantinople sous les empereurs byzantins," Metron, I, 2 (1920), pp. 68-120, and others considered themexclusively senatorial dwellings, but rather private detached houses. See, e.g., Codex Theodosianus VII, 8, 8 (400); XVI, 2, 22 (420). If such houses could contain members of the lower classes as well, then there is no reason to assume that the domus count indicates a senatorial class much larger than that of which contemporary Rome could boast. Nor is it necessary, in consequence, to conclude that the proletarian population must have been inflated to a comparable degree. For the Byzantine house, see Ph. Koukoules, l l ~ p i7Tjv pvlav-r~vTjvoi~iav,' E ~ r ~ ~ q p'E-ralp. ls But. XI1 (1936), p p 76-138. ZITOV~GV, 2. A calculation by planimeter of the city's area, using the maps in R. Janin, Constantinople ) ,A. M. Schneider, B y z a n z : Vorarbeitenjiir Topographic u n d Archaologie byzantine (Paris, ~ g j ~and der Stadt, Istanbuler Forschungen, VIII (Berlin, 1936) Constantinople within the Theodosian walls embraced 14.3 sq. km. (1,430 hectares), less than half the extent of Paris in 1815 and about the size of Rome within the Aurelian walls. Of the total, about 700 hectares lay within the thinly populated XIVth region. Judging from the domus count in the Notitia, 68% of the population (i. e., 3,522/4,388 domus) lived within the 375 ha. of the first ten regions. Assuming the total population to be 800,ooo (a conservative estimate for Rome made by A. von Gerkan, "Die Einwohnerzahl Roms in der Kaiserzeit," Mitteilungen des Deutschen Archaologischen Itzstituts: Rcmische Abteilung, LV [1g40], pp. 172-173), and applying the proportion, we would arrive at a population density of 1,37o/ha. in an area where already much of the land was devoted to public buildings and palaces. Among modern European cities, only the 88.5 ha. of the port area of Naples in 1881 approached this figure. So G. Pardi, "Napoli attraverso i secoli," Nuova revista storica, VIII (1g24), pp. 193-195 Unlike the tenements of Naples, the houses in Constantinople seem never to have exceeded four or five stories in height. 3. The XIVth region to the west was never heavily populated and thus formed a favorite location for monasteries. From R. Janin, L a gkographie ecclbiastique, pt. I, vol. 3: Les Lglises et les monastdres (Paris, 1gj3), the following list of locations of monasteries (IV to IX centuries) may be compiled: Lycus and Exokoinion, 15; Blachernae, j ; Dexiokratae, 9; Deuteron, 3; Psamathia, 10; Golden Horn, 6 ; Propontis, j ; eastern extremity, 6. 236 G. Rouillard, La vie rurale duns Z'Ernpire byzantin (Paris, 1g53), pp. 181-189. Demographic trends are also emphasized by Lemerle, "Histoire agraire," RHist, CCXIX, pp. 63ff., who finds early evidence of increasing land-hunger in the "Farmer's Law." 237 SO F. Dolger, BZ, XXXVI (1936),pp, 157-161, whose point of view is indicated by the title of an earlier study: "Das Fortbestehen der Epibole in mittel-und-spatbyz. Zeit," Studi Albevtoni, I1 (Padua, 1g34), pp. 3 ff. See Appendix D.

T H E G R A I N S U P P L Y O F T H E B Y Z A N T I N E E M P I R E 135 4. The crucial and continual water shortage would have constituted an effective outer limit t o population growth. See Procopius, Anecdota, xxvi, 23 (ed. Loeb, VI, pp. 308f.), and Joannes Malalas, Chronographia,ed. Bonn, p. 492 ; for discussion : R. Meyer, Byzantion-Konstavz,tinopolis-Istanbul: eine genetische Stadtgeographie, Akademie d. Wissenschaften, Wien: Denkschriften, Phil.-Hist. Klasse, 71, 3 (Vienna, 1943), p. 5.

In the absence of any comprehensive study of supply within the later Roman Empire-since Rostovtzeff's "Frumentun~," Pauly-Wissowa, Real-encyclopddie der classischen Altertums, VII, cols. 126-187-the following indications may be offered, based particularly upon the sources for the period immediately preceeding the loss of Egypt. Mention may be made of the survey by G. Richard, "Le problitme du bl4 & Byzance," Infarmation historique, XIX (19j7), p p 93ff., which, for the period under review here, is concerned chiefly with matters of fiscal policy. Egypt, as we have seen, was the major granary; although local disturbances may have rendered its deliveries less predictable under Tiberius, Maurice, Phocas and Heraclius (see R. H. Charles, ed. and trans., The Chronicle of John, Bishop of Nikiu [London, 19161, passim, esp. pp. 8, 15, 21, and the commentary of G. Rouillard, L'administration civile duns 1'Egypte byzantine, 2nd ed. [Paris, 19281, pp. 174f.), the direct statement of Nicephorus (ed. de Boor, p. IZ), leaves no doubt that the events of 619 were severely felt at Constantinople: "In addition t o these things, the bitterness of famine assailed the state. For Egypt did not deliver the rest (-rb Ao1~r6v)and the result was that the greater part of the emperor's grants of grain ceased." Yet, as Nicephorus' own phrasing suggests, and as the laws on coemptiones further indicate (C. Th. XIV, 16, I and 3, and Novellae imperatoruv~Post Iustinianum, Coll. I , nov. xii, 6, [ed. Zacharia, Jus Gr.-R., 111, zgf.]), Constantinople enjoyed secondary sources of supply as well. In the fourth century, Africa was bound to support the city and those eastern provinces where military demand might exceed local resources (C. Th. XIII, 9, 2). While the establishment of the Vandal kingdom undoubtedly interrupted these shipments, after Justinian's reconquest arrangments similar to those of the fourth century seem once again to have been in effect. When Heraclius commandeered African ships for his revolt against Phocas, the diversion of shipping caused a famine at Constantinople (Theophanes, ed. de Boor, I, 296; cf. Procopius, Anecdota, xxv, 7-10), I n fact, civilian as well as military supply was organized on a Mediterranean basis. To Sicily the citizens of Thessalonica applied in 597 (Miracula S. Demetrii, sec. 74 [Acta SS., Oct. IV, I ~ I ] ) as , did the patriarch of Alexandria in 615 (Vita Ioannis eleemosynarii, c. xiii, ed. H. Gelzer, Leontios von Neapolis Leben des Heiligen Johannes [Freiburg im/Br., 18931, p. 27.) To Rome in the fourth century, possibly to Constantinople in the seventh century, Macedonia upon occasion exported grain. (Symmachus, Ep. 111,55, e d 0 . Seeck, MGH, Auct. Antiq., VI, I, [Berlin, 18831, p. 88, and a curious detail connected with Heraclius' voyage from Africa to Constantinople: Eutychius, Annales, trans. R. Pococke, Migne, PG, 111, col. 1085.) For military supply on a Mediterranean basis, see above, pp. 92f. and refs. There are a few indications of export from regions of the eastern Empire closer to Constantinople. In Thrace, Peter Barsymes made coemptiones after the plague of 542, but, in Procopius' terms, they were not enough. (Anecdota,xxii, 17f.) The fields of Thrace adjacent to Constantinople were given over to agriculture in the sixth century, being inhabited by a peasant class to whom Belisarius turned for troops in j 58 (Agathias, V, 16 [ed. Dindorf, Hist. Gr. Min., 11, p. 374f.l). Whether they produced grain or, more probably, vegetables and wine we do not know. I t should be noted, too, that the reserves it was able to stock sometimes permitted the city itself to act as an emergency granary for its hinterland. In 375, when invasions forced the peasants to abandon their fields and seek refuge in castella, the monk John sought and obtained the necessary supply from "magnates" resident in Constantinople. (Callinicus, De vita Hypatii, ed. Seminarii philologorum Bonnensis sodales [Leipzig, 18951, pp. 64f). North of the Danube there is little evidence of agricultural activity in the earlier part of the period; upon occasion, in fact, the Empire seems to have supplied or exported grain rather than

136

J O H N L. TEALL

t o have imported the commodity. (See, e.g., for the Huns: E . A. Thompson, A History of Attila and the H u n s [Oxford, 19481, p. 137; for the Kutrigurs: Procopius, Bellum Gothicum, IV, xix, 16 [ed. Loeb, V, 248-2501.) Relationships changed with the appearance of the Slavs early in the sixth century and with their submission t o Avar power after the middle of the same century. The former were agricultural peoples, and the Strategikon of Maurice (XI, 5 [ed. Scheffer, p. 2821) emphasizes the quantities of supply they produced early in the seventh century and the facility with which it could be transported into the Empire: "Do not waste.. .the supplies to be found in the land. Take care to transport them into your own land by boats or animals. Since their rivers mingle with the Danube, transport by boat is easily effected." When Maurice ordered his commander Peter, in 602, t o lead his troops into winter quarters across the Danube, thus relieving the pressure on the Empire's supply, he might well have been acting on the basis of the information to be found in the Strategikon. (Theophylactus Simocatta, VIII, 6 [ed. de Boor, p. 2951.) Priscus had received supply from the Avar khagan when the siege a t Tomi had reduced hisown troops to starvation (Ibid., VII, 13, p. 267f.). On the agricultural capacities of the Slavs see F. Dvornik, T h e Slavs: Their Early History and Civilization (Boston, 1956), pp. 53ff., and in greater detail, L. Niederl6, Manuel de l'antiquite'slave, I1 (Paris, 1923-1gz6), pp. 184-210. According to Rostovtzeff, "Frumentum," col. 138, after its foundation Constantinople laid primary claim t o the slight amount of grain still produced on the north shore of the Black Sea in the hinterland of the Greek cities. Possibly it is t o exports from this region that Socrates referred when he noted (371) that the Euxine furnished the city with grain "to any extent it may require" (Historia ecclesiastica, IV, 16 [Migne, PG, 67, col. 5011). I n the latter half of the sixth century the eastern shore of the Black Sea seems, oddly enough, a likelier candidate than the northern littoral for the role of Constantinople's supplemental granary. Colchis, the land of the Lazi, witnessed a marked advance in material culture under Justinian which may be gauged by the contrasting monopolies instituted there early and late in the Emperor's reign. Prior t o 535 an important customs station had been erected a t Petra where John Tzibus exercised a monopoly over the import of bread, salt, and other commodities (Procopius, Bellum Persicum, 11, xv, 11 and 16 [ed. Loeb, I, 388, 4061). I t was, Procopius remarked, a land lacking in all good things, and we may assume that its initial importance t o the Empire was strategic. Justinian subsequently made notable attempts t o civilize the wild peoples and their tough land, constructing roads and churches, as well as instituting agricultural settlements. (Procopius, De aedificiis, 111, 6 [ed. Loeb, VII, 206-2121). After 556 local production was ample t o tempt John the African t o institute a monopoly not of imports, but of grain exports. (Agathias, IV, 22.) I n 562 it was customary for the king of the Lazi to support the Sunni with offerings of grain (Menander, frg. 3 [ed. de Boor, p. 1781). Regrettably, the sources do not permit similar generalizations for Asia Minor where potential resources were great and where Justinian's reconstruction of roads and harbors can be expected to have increased facilities for export. (Procopius, De aedificiis, IV, viii, 4-7 and ix, 17-21; V, ii, 6-13, iii, 4-6, and iii, 12-15 [ed. Loeb, VII, 284, 296, 322-324, 324-326, 3281.) From "all Asia" grain fleets are supposed t o have come in the early fourth century to satisfy the needs of Constantine's newly-founded capital (Eunapius, Vitae sophistae, p. 462, ed. and trans. W. C. Wright, Philostratus and Euna#ius [London and New York, 19221, p. 382), but the sources of the later period fail to accord equal prominence to shipments from the western coast. At the end of the sixth century, Evagrius described abundant fields of grain in the vicinity of Chalcedon (Hisioria ecclesiastica, 11, 3 [Migne, P G , 86 bis, cols. 21gzf.I) in terms which recall the stores laid up by Hypatius in his monastery of Rufinus, near the same city. (Callinicus, De V i t a Hypatii, ed. Sem. phil. Bonn. sodd., p. 104.) With Bithynia's supply, and that of Phrygia as well, Barsymes attempted t o supplement his coemptiones in Thrace, but the combined resources of all three regions could not suffice without those of Egypt. Thanks t o the fragmentary nature of the sources, the results of a survey of the several producing regions within or near the Empire are undeniably disappointing. That some of them exported grain to Constantinople cannot be denied. Procopius admitted that the coemptiones of Barsymes were not without precedent ; others after him continued the unpopular levies (Anecdota,

T H E GRAIN SUPPLY O F T H E BYZANTINE E M P I R E 137 xxiii, 14 [ed. Loeb, VI, 2721). When emergency dictated, almost any region in the Mediterranean world was liable to be called upon to make up a deficit; intermittently, at least, transports from Africa seem to have been regularly organized; Thrace and Bithynia were probably most often called upon to supplement Egypt's "happy transport." At the end of the sixth century the Slavs were winning new prominence as agricultural producers, but Egypt's offerings were still of crucial importance. C. SUPPLY FROM NORTH AFRICA AND SICILY

A survey of North Africa during the period of the Islamic conquests suggests that until the last quarter of the seventh century any possibilities of private supply from the province should not be discounted. During the initial stages of their attack, the Arabs desired chiefly men and booty won as cheaply as possible for the masters of Egypt and Syria. The Arab raids were sporadic and inconclusive in their effects on the grain-producing regions; after the attack of 647, for example, fifteen years of peace preceded the raids signalling the renewed offensive in 665. (See the survey in C. Diehl, L'Afrique byzantine [Paris, 18961, pp. 563-593, and the more recent, if briefer, prCcis of G. Marpis, L a berberie musuZmane [Paris, n.d.1). I t is difficult to believe that the land could have suffered critically; Arab historians describe the rich fields of grain in Ifriqya at the time of the final conquest (Lewis, NavaZ Power, p. 70) ; the surplus was adequate in quantity to support the foundation and growth of cities such as Kairouan, Tunis, etc. (Else Reitemeyer, Die Stiidtegriindungen der Araber in Islam [Munich, 19121); the resources of the province were considerable enough in the middle years of the century to make its exploitation seem desirable to Constans I1 (Diehl, L'Afrique byzantine, p. 569); seals of the commerciarii extend to the reign of Constantine IV (ibid., pp. 500-502). In the seventh and eighth centuries Sicily did not suffer complete catastrophe either. Although by the seventh century the island had become primarily the granary of Rome, on one occasion at least (between 602 and 610) the archontes of Thessalonica sent thither for grain to relieve a famine in their city. (See above, Appendix B.) As in the case of the North African conquests, Arab raids were sporadic, permitting a continuity of settlement in some parts of the island well into the ninth century. For the narrative, see the classic M. Amari, Storia dei mtuulvtani d i Sicilia, 2nd ed., I (Catania, 1936), passim, esp. pp. 194-209, 216,293-300; also the study of D. C. Dennett, "Pirenne and Muhammad," Sfieculum, XXIII (1948)) p p 165-190, esp 168-173. For a convenient summary of the Pirenne thesis, with criticisms, which reprints without documentation the work of Dennett, Lopez, Riising, and others, see T h e Pirenne Thesis: Analysis, Criticisnz, and Revision, ed. A. F . Havighurst (Boston, 1958). A recent interpretation, not always fair to Byzantium, is that of W. C. Bark, Origins of the Medieval World, Stanford Studies in History, Economics, and Political Science, XIV (Stanford, Calif., 1958)) esp. pp. 5-29. While Syracuse suffered badly, Palermo, Messina, Nota, Paterno, Modica, and Ragusa maintained themselves: B. Pace, "I barbari ed i bizantini in Sicilia," Archivio storico siciliano, XXXVI ( I ~ I I pp. ) , 35-45, and pertinent sections in the two studies of G. Pardi, "Storia demografica della citt&di Palermo," Nuova revista storica, I11 (1919), p. 190, and "Storia demografica di Messina," ibid., V (1921), pp. 28f. From the lives of the saints, we learn that ships plied between Catania and Constantinople in the eighth century ( V i t a Leonis catanensis, cc. viii, ix [Acta SS., Feb. 111, 2281) ; in the seventh century there were naz~cleroia t Syracuse ( V i t a s. Zosinzi, Acta S S . , Mar. 111,837) and an influential community of Jews; at the end of the seventh and early in the eighth century Sicilian ports had replaced those of Italy on the sea routes from Constantinople as the primary ports of call: Agnellus, Liber pontificalis ecclesiae ravennatis, secs. 131, 136, 145, edited by 0. Holder-Egger, MGH: S s . Rer. langobard. et ital. (Hannover, 1878), pp. 364, 367, 372f Constans I1 reorganized the Sicilian coinage and the newly-established mint at Syracuse produced and maintained a high level of gold coinage until the first part of the reign of Justinian 11;towards the end of the reign of Constantine V, gold coinage again became abundant in the Syracuse mint, presumably as part of the general administrative reform after 752: D. Ricotti-Prina, "La monetazione siciliana

J O H N L. T E A L L

nell' epoca bizantina," Numismatics, XVI ( ~ g j o )pp. , 26-61, and for evidence of the more active defense of the island in the latter part of the eighth century: ibn el Athir in Bibliotheca arabosicula, ed. M. Amari, I (Turin and Rome, 1880-1881), p. 363. The available evidence suggests, then, that Sicily maintained production, possessed ships for the cargo, and the men to sail the ships. Relations with Constantinople were maintained on an active basis until ca. 680, with interruptions until 752, and consistently thereafter until the final collapse of Byzantine power. However, it is only in references to the patrimonies of the Ravennate and Roman churches that specific information on the export of grain may be found. During the tenure of Maurus as archbishop of Ravenna (650-671) grain and other commodities were exported for the use of the church in Italy, and a money payment was offered the imperial government (Agnellus, Liber pontificalis, sec. 111 [ed. Holder-Egger, p. 3501). Although the papal patrimonies were more than once a source of contention between Rome and the Byzantine emperor, it is not clear whether they were valued a t Constantinople for the grain they produced or (as in the case of Ravenna) for the taxes to be realized from them. See Theophanes, ed. de Boor, I, 410, and Hadrianus I, Epist. 56, Migne, P L , 96, col. 1218. That the "confiscation" of the patrimonies in 731 may have entailed a loss of grain at Rome is suggested by certain rearrangements in the city's supply system undertaken during the eighth century. For its grain Rome turned from the Mediterranean to domuscultae or agricultural settlements in its own hinterland. See 0 . Bertolini, "Per la storia della diaconie romane," Archivio della societd romana d i storia $atria, LXX (1g47), pp. 1-228; R. Vieillard, Les origines de la Rome chre'tienne (Miicon, 1g40), p p 110-122; 0. Bertolini, R o w a d i fronte a Bizanzio e a i Langobardi (Bologna, 1g41), p p 506-512. Yet, even if some grain travelled to Constantinople, particularly after 752, it is doubtful that the quantity was at all considerable. The eventual loss of the province was treated almost with a curious indifference: Ps.-Symeon, Chronicon, ed. Bonn (1838), p. 622. D. THE EPIBOLE AND THE WASTE LAND PROBLEM

On the epibole, see in addition to the materials cited above in note 28: G. Ostrogorsky, "Das Steuersystem im byzantinischen Altertum und Mittelalter," Byzantion, VI (1g31), p p 229-240, and the same author's further defense of his thesis in "Agrarian Conditions in the Byzantine Empire in the Middle Ages," Cambridge Economic History, I, pp. 196-204, which he maintains also in Byzantine State, p. 121. "The late Roman system of the epibole had provided for the compulsory conveyance of waste land to the owner of cultivated acres and consequent liability for the appropriate tax on this additional land. . . .This had now [by the late seventh century] been altered, for the tax on the fallow land was laid on the neighboring farmers, who at the same time acquired the right to enjoy the use. . . .This new regulation making the community liable for the taxes. . . appears later on under the designation of allelengyon. The transference of the land becomes merely a consequence of the transference of the tax." With increasing reserves of manpower that could be depended upon for working the land, the state ceased to be interested primarily in ensuring cultivation. Its major concern after the seventh century was to maintain continuity in the payment of taxes. Amplifying Ostrogorsky's arguments, Danstrup, "State and Landed Property," pp. 242-245, points out that the later Roman Enlpire combined two institutions by no means necessarily interconnected: the attribution of vacant land (adiectio sterilium) and collective responsibility for taxation. They were to be found, in fact, existing in isolation from each other first in Ptolemaic Egypt and later in medieval Byzantium. Could the arguments of 0 . and D. be accepted, the epibole legislation, and the subsequent transmutation of the practice into the allelengyon, might be used to trace the density of rural settlement. In constrast to a thin rural settlement in the late Roman period, after the seventh century the lands of the Byzantine Empire were more fully exploited. (So Ostrogorsky, "Agrarian Conditions," op. cit., pp. 197, 202 f .) For two reasons this thesis seems to me to lack cogency, although I agree with Ostrogorsky (against Dolger) on the later interpretation of the epibole. I. His analysis of the legislation of the fourth, fifth, and sixth centuries considerably oversimplifies the problem. While C. J. XI, jg (58),

T H E GRAIN SUPPLY O F T H E BYZANTINE E M P I R E 139 and C. J. X, 3, 4 (383) support his conclusions, I cannot agree that in Nov. J. CXXVIII, 7 (545)the transfer of lands is still the primary motive. Further doubt is cast upon his conclusions by J. Karayannopulos, "Die kollektive Steuerverantwortung in der friihbyzantinischen Zeit," Vierteljahrschrift fiir Sozial- und Wirtschaftsge.schichte, XLIII (1956), pp. 289-322, analyzed by Lemerle, "Histoire agraire," RHist., CCXIX, p. 38, note 3. 2. A distinction should be made between masterless or abandoned land and marginal or inferior land t o the cultivation of which imperial policies might drive the peasant. Forced settlement of the latter seems implied in C. J. X, 3, 4, (383)and even more clearly in two laws dealing respectively with emphyteusis and inheritance: C. J. XI, j g (j8), 2 and C. Th. XI, I, 17. Thus, in and of itself, the epibole cannot be used as a conclusive index of settlement. Whenever the practice appears, it proves only that a discrepancy exists between demand and the resources available to satisfy i t ; we cannot always be certain whether the demand is for taxes or for agricultural produce. For evidence of increased settlement, then, we must depend upon the materials used above, possibly upon indications of increased rapacity and litigiousness, manifest in "la lutte pour la terre," described by G. Rouillard, L a vie rurale d a m 1'Empire byzantin (Paris, 1gj3), pp. 181-189. Perhaps the fairest and more accurate statement of the problem as it stands at the moment is that of K. M. Setton, "On the Importance of Land Tenure and Agrarian Taxation in the Byzantine Empire," American Journal of Philology, LXXIV (19j3), pp. 238-239: "During the early Byzantine period the government had forced the landowners to take over abandoned lands by holding them responsible, under the epibole as we have noted, for the tax on such lands. If the landowners had thus to pay the land tax, they had best take the land also, and, if possible, put it under cultivation to minimize or avoid loss.. . .During the middle Byzantine period the system of responsibility for the taxes on abandoned land continued to fall on the unfortunate neighbor of the peasant who had fled. . . .The surtax was now called allelengyon. The government, however, was now chiefly interested in taxing as a source of revenue those whose own property was near an abandoned property, although here, too, the taxpayer would inevitably seek to cultivate the new lands. . .and this had become an easier undertaking in the eighth and ninth centuries because of the increased availability of agricultural labor." I

http://www.jstor.org

LINKED CITATIONS - Page 1 of 2 -

You have printed the following article: The Grain Supply of the Byzantine Empire, 330-1025 John L. Teall Dumbarton Oaks Papers, Vol. 13. (1959), pp. 87-139. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C87%3ATGSOTB%3E2.0.CO%3B2-F

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 11

Conséquences et intérêt démographiques de la Peste noire de 1348 Yves Renouard Population (French Edition), 3e Année, No. 3. (Jul. - Sep., 1948), pp. 459-466. Stable URL: http://links.jstor.org/sici?sici=0032-4663%28194807%2F09%293%3A3%3C459%3ACEIDDL%3E2.0.CO%3B2-A 11

Plague and Economic Decline in England in the Later Middle Ages John Saltmarsh Cambridge Historical Journal, Vol. 7, No. 1. (1941), pp. 23-41. Stable URL: http://links.jstor.org/sici?sici=1474-6913%281941%297%3A1%3C23%3APAEDIE%3E2.0.CO%3B2-K 80

Arabic Lists of the Byzantine Themes E. W. Brooks The Journal of Hellenic Studies, Vol. 21. (1901), pp. 67-77. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281901%2921%3C67%3AALOTBT%3E2.0.CO%3B2-2 142

Byzantines and Arabs in the Time of the Early Abbasids E. W. Brooks The English Historical Review, Vol. 15, No. 60. (Oct., 1900), pp. 728-747. Stable URL: http://links.jstor.org/sici?sici=0013-8266%28190010%2915%3A60%3C728%3ABAAITT%3E2.0.CO%3B2-R

NOTE: The reference numbering from the original has been maintained in this citation list.

http://www.jstor.org

LINKED CITATIONS - Page 2 of 2 -

212

The Farmer's Law Walter Ashburner The Journal of Hellenic Studies, Vol. 30. (1910), pp. 85-108. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281910%2930%3C85%3ATFL%3E2.0.CO%3B2-V 213

The Farmer's Law Walter Ashburner The Journal of Hellenic Studies, Vol. 30. (1910), pp. 85-108. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281910%2930%3C85%3ATFL%3E2.0.CO%3B2-V

NOTE: The reference numbering from the original has been maintained in this citation list.

The Armenian Chronicle of the Constable Smpad or of the "Royal Historian" Sirarpie der Nersessian Dumbarton Oaks Papers, Vol. 13. (1959), pp. 141+143-168. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C141%3ATACOTC%3E2.0.CO%3B2-K Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:36:42 2008

T H E ARMENIAN CHRONICLE O F THE

CONSTABLE SMPAD OR O F

T H E ccROYALHISTORIANM

H E Chronicle written by the Constable Smpad (1208-1276)) brother of King Het'um of Cilicia, is one of the principal Armenian sources for the history of the Crusades1 The section of it which extends from the year 951 to 1162 follows mainly the History of Matthew of Edessa and of his continuator Gregory the Priest, occasionally including information derived from other sources; it is, on the other hand, particularly important for the history of the thirteenth century, since it was written by a man who had direct access to the official documents and who, moreover, played a major role in many of the events he related. An anonymous writer continued the Chronicle down to the year 1331. The two Armenian editions, one published in Moscow in 1856 by Osgan of Erivan, the other in Paris in 1859 by G. Chahnazarian, have long been out of print. In 1862 V. Langlois published a French translation of certain sections only;2 in 1869, the greater part of the text, beginning with the year 1092, was included by E. Dulaurier, together with a French translation, in the Recueil des historiens des C ~ o i s a d e s . ~ All of these publications were based on two manuscripts of fairly late date in the Library of Etchmiadzin and on copies of them made towards the middle of the nineteenth century. The new Armenian edition, recently published by Father Seropk Akelian, deserves, therefore, special a t t e n t i ~ nI.t~is based on a manuscript of the late thirteenth or early fourteenth century which had been bought in Constantinople in 1876 by Seropk Markar Alishan and given to his elder brother, Father Leonce Alishan, who deposited it in the Library of San Lazzaro (no. 1308).~There are several lacunae in this manuscript. The initial folios, relating the events of the years 951 to 974, are lost and the manuscript is also incomplete at the end, where it stops abruptly in the middle of an account of the events of the year 1272; the other missing parts occur after folios 11, 35, 45, and 156 (A.D.1023-1029, 1063-1064, 1070, 1230-1251). In order to present a continuous text, the editor has copied the missing parts from the Paris edition of Smpad's Chronicle and in one instance from the History of Matthew of Edessa, clearly indicating these additions through the use of smaller type.6 Unfortunately, this new publication is not a critical edition. There are frequent references in the footnotes to the corresponding passages of Matthew

'IIP

1 I n transliterating Armenian names I have followed the pronunciation of the Western Armenians for all names of the Cilician period, since by that time the change in the consonant sounds had taken

place.

Victor Langlois, Chronique de Sempad, Extraits (St. Petersburg, 1862). Documents armbniens, I, pp. 610-72. S. Akelian, Smpada Sbarabedi Darekirk' (Chronicle of the Constable Smpad). (Venice-San Lazzaro,

1956).

Ibid., pp. ix-x. "bid., pp. 1-11, 27,59, and 226-8 are taken from Smpad's Chronicle; pp. 71-4 are taken from the Histovy of Matthew of Edessa (Vagharshabad, 1898))pp. 194-7.

SIRARPIE DER NERSESSIAN

of Edessa and occasional references to those of the Chronicle of Samuel of Ani, but in rare instances only is the text compared with that of the earlier editions of Smpad's Chronicle. This is particularly regrettable because of the many important differences, and the reader who does not himself make this comparison remains unaware of the particular significance of the manuscript on which the new edition is based, and of the new information that it provides. Ever since the late nineteenth century, when L. Alishan published his important study on the Kingdom of Cilicia, translated into French a few years later,' it has been known that there existed in the Library of San-Lazzaro a manuscript that contained a more detailed text than did Smpad's Chronicle known through the editions of Moscow and Paris. Alishan referred to this as the work of the Royal historian, or as the Cilician Chronicle; he quoted numerous passages from it in his Sissouan and published several extracts in ~ of the Middle Ages his collection of texts entitled H a y a b a d o ~ m .Historians and of the Crusades, in particular Claude Cahen, have made frequent use of the passages known to them through the French translations of Alishan's works;9 and in 1948 the late Robert P. Blake read a paper at the Byzantine Congress in Paris entitled "La chronique royale de la Cilicie armknienne." Through the courtesy of the Abbot of San Lazzaro Blake had obtained a complete microfilm of this manuscript, but the rights to publish the text had been reserved for the Mekhitharist Congregation. I t is most surprising, therefore, that in this new edition, made by one of the Mekhitharist Fathers, there is no indication that the manuscript used is the very one referred to by Blake, in other words, Alishan's "Royal Chronicle." That this is the case is beyond doubt. One need only collate the quotations in the Armenian edition of Sissouan and the extracts included in Hayabadoum, with the corresponding passages in the new edition to be convinced t h a t they are taken from the same manuscript.1° I t is this identity that makes the new publication particularly valuable, for we now have the long-awaited text of the "Royal Chronicle'' and can compare it with the editions of SmpadJs Chronicle. Such a comparison shows clearly that the text of the Etchmiadzin manuscripts of the Chronicle, known through the Moscow and Paris editions, is an abbreviation of the text of the Venice manuscript. This abbreviation, which may be the work of the continuator of the Chronicle or of a later scribe, was Lkon le Magnifique,premier roi de Sissouan ou de I'Annknocilicie, Ccrit par le R. P Leonce M. Alishan, traduit par le R. P. George Bayan (Venice-San Lazzaro, 1888).LBonce Alishan, Sissouan ou 1'ArmknoCilicie (Venice-San Lazzaro, I 899). L. Alishan, Hayabadoum, (Venice-San Lazzaro, 1901))pp. 412, 436-41. C1. Cahen, "Quelques textes nCgligCs concernant les Turcomans de RQm au moment de l'invasion rnongole," Byzantion, 14 (1939), pp. 131-9; see especially 133-5. Idem, La Syrie du Nord d I'e'poque des Croisades et la principaute' franque dlAntioche (Paris, 1940)) passim. See his bilbiography, p. 99: "Aussi officiel et plus cironstancie [than Smpad's Chronicle] est I'Historien Royal de llextrCme fin du XIIIe sihcle qui n'est connu jusqu'ici que par les extraits d'Alishan dans 'LCon le Magnifique' et 'Sissouan,' extraits qui donnent l'impression d'un ouvrage considerable et rendent extrgmement desirable une Cdition." Steven Runciman, A History of the Crusades, I1 (Cambridge, 1952))pp. 483-4 and passim. lo Compare in particular Alishan, Hayabadoum, pp. 412, 436-41 with Akelian, op. cit., pp. 194, zoo, 203-4, 206--7, 208-11. I have also had the advantage of having at my disposal the photographs made for Robert P. Blake.

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 145 carried out in a very uneven manner. Sometimes the events of a single year are told in fairly great detail, closely following the text of the Venice manuscript, but in numerous instances the long accounts of the latter are reduced to a few sentences or even entirely omitted, so that concerning several successive years there is no information at all. The basic agreement between the two redactions becomes even more apparent when they are compared with the History of Matthew of Edessa, for in several instances they include events that are not recorded by the latter.ll There are, however, a few minor additions in the text of the Etchmiadzin manuscripts,l2 and some of these must be considered more closely in order to determine whether the Venice manuscript contains the work of an anonymous writer, as Alishan was inclined to think,l3 or whether it is a more faithful copy of Smpad's Chronicle. Since the Venice manuscript is defective a t the beginning and at the end, it does not give the name of the author which might either have accompanied the title of the work or have been mentioned in the colophon. Throughout the text the third person is used in all the references to Smpad. In the Etchmiadzin manuscripts, on the other hand, there are five distinct references to Smpad as the author. The first of these specific references is included in the praise of King Leon I, where it is stated that whenever this King heard of a man who was wise, learned, or expert in warfare, he attracted him to his court, "and I, the Constable Smpad, author of this history, am one of these rnen.'j14 Under the year 1208 we read: "The Constable Smpad, author of this book, was born in this year."15 I n relating the Seljuk invasion of 1246 the author writes: "The father of the King and I, the Constable Smpad, entered Tarsus."16 Under the years 1248 and Ia jo the brief entries read : "I, the Constable Smpad went to the Tartars. And I came back to my brother, King Het'um."l7 Are these sentences part of the original redaction or are they later additions ? The last three entries occur in a part of the text where there is a serious lacuna in the Venice manuscript. Ten folios following folio 156 are now lost and the missing text must have been at least five times as long as that preserved in the Etchmiadzin manuscripts. This is particularly regrettable since the lost narrative may have given more detailed information about Smpad's mission to the Mongol court than do his remarks in his letter to his brother-in-law, Henry I of Cyprus.ls The date of Smpad's departure on this mission, as given l1 For instance, the events of the years I 129 to I 13j ; Akelian, ofi, cit., pp. I 59-60 ; Recueil des historiens des Croisades, Doc. arm., I , pp. 615-6. l2 For instance, after the account of the murder of the sons of Mandale and the return of T'oros to Vahga, the Etchmiadzin manuscripts give the following sentence which occurs neither in Matthew of Edessa ( R H C , Doc. arm. I , p. 100) nor in the Venice manuscript (Akelian, op. cit., p. 137): "He established the inhabitants of the fortress on the banks of the river Baradis, in the locality now called Gragga ( R H C , Doc. a m . , I, p. 613). l3 Alishan, Hayabadoum, pp. I I 1-2. '"HC, Doc. arm., I , p. 634. l5 Ibid., p. 643. l6 Ibid., p. 650. l7 Ibid., p. 6j1. Is William of Nangis, "Vie de saint Louis," Recueil des historiens de la Fraqzce, X X , pp. 361-3; L. Alishan, Assises d'dntioche (Venice, 1876), pp. 90-1.

146

SIRARPIE D E R NERSESSIAN

in the Etchmiadzin manuscripts, is not correct. The Armenian year 697 began in January 1248; Smpad's letter was written from Samarkand on February 6, 1248, when he had been gone for eight months, and as P. Pelliot has shown, he must have set out on his journey the previous year, that is in 696 = A.D.1 2 4 7 . ~ ~ This error in the date could not have been made by Smpad himself and must have originated in a later copy, as did so many other erroneous dates in the Etchmiadzin manuscripts. I n regard to the first autobiographic mention, it should be recalled that Smpad h a s only eleven years old when King Leon I died and it does not seem very likely that he would have referred to himself as one of the "men" who attracted the King's attention because of their wisdom, knowledge, or proficiency in warfare. Even the use of the first person in relating the Seljuk invasion of 1246 bears the stamp of an interpolation, for in the preceding paragraphs the third person had been used twice in relating the invasion of the previous year.20 l f w e now turn to the content of the two redactions, we observe that in the Etchmiadzin manuscript the most summary treatment occurs in the latter part of the text, which extends from 1248 to 1275. Several years are entirely omitted, and the entries for the others consist of brief mentions, the longest being of only four ~entences.2~ In the Venice manuscript on the other hand, the account of the events of the years 1252 to 1272 (following the lacuna after folio 156 mentioned above) covers forty-seven pages.22I t is very difficult to suppose that a man who was interested in writing the history of his country would have so drastically curtailed the account of the period with which he was most familiar, and in which he had played a leading role. We must, therefore, conclude that Smpad is the author of the "Royal Chronicle" and that his text has been faithfully preserved in the Venice manuscript, while the other redaction, known through the Etchmiadzin manuscripts, is a faulty, abridged copy, with a few additions of a later date. Smpad's Chronicle, as we now see it, was a homogeneous composition. The first part is primarily a summary of the History of Matthew of Edessa, as mentioned above. Very few facts have been omitted; but Smpad has condensed the verbose descriptions of his source. He has, moreover, eliminated the long speeches as well as the frequent references to the Scriptures. He has greatly abridged, too, the passages relating to Matthew's native city of Edessa,23but it is evident that he also used other sources. For instance, in speaking of the arrival of the Crusaders, he recalls the predication of Peter the Hermit, not mentioned by Matthew of Edessa, and instead of listing the names of the principal -

19 P. Pelliot, "Les Mongols et la PapautC," Revue de 1'0rient chre'tielz, XXIV (1924), pp. 326-7 and XXVIII (1931-1932) pp. 18-9. In the Chronicle of Het'um (Hayton) of Gorigos, which has now been published for the first time, the date of Smpad's journey is given as 1246. See V. A. Akopian, Malzr jamanakagroztt'younner, XIII-XVIIIc, I1 (Erevan, 1956), p. 67. 20 RHC, Doc. arm., I, p. 649.

21 22

Ibid., pp. 651-3. Folios 157-7gV.Akelian, op. cit., pp. 229-54.

23 Note, for instance, the omission of the long letter which the Catholicos Gregory is said t o have written t o the clergy of Edessa: RHC, Doc. arm. I, pp. 63-7; Akelian, op. cit., p. 121.

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 147 Crusaders he writes: "their complete history and the names of the princes are recorded by the Frank historian^."^^ In several other instances also, Smpad's account, though shorter, contains specific information lacking in Matthew of Ede~sa.~~ Smpad follows less closely the text of Gregory the Priest, the continuator of Matthew of Edessa. The order of events is not always the same and Smpad is often better informed than Gregory. The Chronicle is particularly valuable for the period from 1163 to 1273, and since this section has been so badly preserved in the later redaction, I propose to offer, for the convenience of historians of the Middle Ages who are not familiar with the Armenian language, a comparison of the two redactions and a translation of those passages which are not restricted to the local history of Cilicia. But before beginning this comparison, I shall translate the account of Manuel's entry into Antioch in 1159, which does not occur in the later redaction, and which also differs in many respects from the account given by Gregory the Priest.26 Smpad first speaks of the arrival of Manuel in Cilicia, tells how the Emperor forgave T'oros, thanks to the good offices of the King of Jerusalem and other, Latin princes, and adds: "Then the King of Jerusalem, the Prince of Antioch, the sebastos T'oros and the Frkres thought of saving the Christians and, setting out in great numbers, they came and encamped before Antioch. But, although the King of Jerusalem and all the leaders exhorted the Emperor of the Greeks to hasten to the salvation of the Christians, and he agreed to do this and gave them his promise, yet he did not do so truthfully and sincerely. For he intended to enter Antioch not for a usefulpurpose but for debauchery and lust for women; he had thought of marrying one of the daughters of Bohemond, Lord of Antioch, and he wanted to go to see her and find out whether he liked her, but he did not betray his thoughts to anyone. And during these days he gave very handsome presents to Baldwin, the King of Jerusalem, who was a very tall man and most pious; he crowned him with the royal diadem and gave him costly garments; he prepared a royal palace which he filled with all kinds of gold and silver vessels, and furniture, as is the custom, and presented that to him.27He also offered costly gifts to the latter's princes, and one of these princes, named Philip, spoke words worthy of remembrance. When the Emperor sent him three talents of gold and costly garments, he arose, thanked and praised the King, and said to the messengers: 'Tell your King that we have not come to you for 24 Akelian, op. cit., p. 100. Here the name of Peter the Hermit is given as Peter-Sarkis, but in the manuscript itself Peter is added in the margin, as a correction of the name Sarkis, written by mistake in the text. 25 There are also instances where Smpad's text is more detailed; see, for instance, his account of the negotiations between an armorer called George and Bohemond in 1099, before the former opened the city of Antioch to the Crusaders (Akelian, op. cit., pp. 106-8). 2"kelian, op. cit., pp. 179-83. For Gregory the Priest see RHC, Doc. arm. I , pp. 188-93. 2 7 William of Tyre (XVIII.24) also refers to the costly gifts presented by Manuel to Baldwin and to his followers.

SIRARPIE DER NERSESSIAN

treasures and garments, but for the salvation of the Christians. If you have this in mind and accomplish it, we shall place at your service our own selves, all our soldiers and all our belongings, and wherever fighting occurs you will see what brave soldiers you have. Give your gold to your poor. But if you should not do what we request and what you have promised, that is, the salvation of the Christians, then your gold is of no use to us.' Then they began to make the following arrangements for entering the city. They decorated the gates of the city and all the ramparts, they sent the royal banner to be raised on the fort, they placed the soldiers and their leaders to man the gates of the city and the streets, they established head-men for the streets and filled the entire city with soldiers. Then they sounded the trumpets, and the Emperor wearing imperial garments and, on his head, a crown adorned with precious stones which shone like the glittering stars, came forth on a horse harnessed in gold. On the Emperor's right and left the soldiers walked with measured tread. Before him rode the King of Jerusalem crowned with a diadem and seated on a horse, while the Lord of Antioch, to show his humility, preceded him on foot. And with such splendid pomp did Manuel, the Emperor of the Greeks, and Baldwin, the King of Jerusalem, enter the city of Antioch. The Emperor then proceeded to the center of the city and came to worship at the holy church and apostolic throne of Saint Peter; then he returned to his h0use.~8When the great Emir of Aleppo, Nur ed-Din, son of Zengi, heard of this assembly of Christian Kings, he was dismayed; in his fear he equipped all his fortresses, prepared for battle, placed commanders in all places and conveyed his goods and plunder to the other side of the Euphrates river.29Then, after a few days, the Emperor sent an ambassador to Nur ed-Din; he wrote him a missive demanding all the land of the province of Antioch, Edessa, and its surrounding area which Nur ed-Din had wrested from the Christians, and also requested the return of the men of Christian nations who were held captive in the prisons. When the great Emir of Aleppo saw the ambassador and read the missive of the Emperor, he ceased to be troubled and afraid, for being a shrewd and wise man he realized how limited was the strength of the army; for this demand had not been made with sword and arrows, but with paper and ink. He answered, therefore, that he would not comply with the order and would never do these things that were asked of him, but that if he [the Emperor] was able, let him obtain them by force. Upon hearing this, the Emperor summoned everyone to advise him as to what he should answer. Then the King of Jerusalem, the Lord of Antioch, and all the princes fell at the feet 28 The triumphal entry of Manuel into Antioch is related by William of Tyre (XVIII.z5), by Nicetas Choniates (111.3) and especially by John Cinnamus (1V.21). See also F. Chalandon, Jean 11 ComnBne et Manuel I ComnBne (Paris, 1912), pp. 450-2. 29 The joint expedition of the Byzantines and the Franks, which advanced only as far as Balana, mentioned by Gregory the Priest (RHC, Doc. arm., I, p. 1891, is only hinted a t by Smpad, but he knew that Nur ed-Din had prepared to withstand the allied attack. For an account of these events see also Abu Shama, "Book of the Two Gardens," RHC, Doc. or., IV, pp. 104-5.

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 149 of Manuel, Emperor of the Greeks, and said, '0 King, do not turn our great joy to sadness, for the enemies of Christ are in consternation because of our union, and if you should not fight them, but establish friendship with them, they will eradicate the name of the Christians from the face of the earth, they will despise and have no esteem for the Christian nations, and we shall become the target of their mockery.' But he [Manuel] alleged important reasons, saying: 'News has come from my capital and because of this I must hurry back,' and with false words he fabricated justifications for his return. Then all were greatly saddened and urgently pleaded with him to march on Aleppo, were it only for three days, and then, if he so wished, to establish friendship with its people. But Manuel did not listen to their pleas and was not willing to help the Christians; so he sent a friendly word to Nur ed-Din. When the infidels heard this, they were amazed by this unhoped-for occurrence, for all had expected to be destroyed, and now they saw that without bloodshed and war they were saved from ruin, and in their joy they could not believe that this was true and did not know what answer to give. They thought that the legates were spies sent to them, until they ascertained the truth from them, and when they had learned the truth, they sent to him [Manuel] many treasures and costly presents, fine horses and mules, and fifty of the Christian prisoners.30Then the good Emperor of the Greeks, Manuel, who had come like a powerful eagle, returned like a weak fox, and with his numerous cavalry he departed like a fugitive and reached the land of the Sultan Kilij Arslan. Then the Turcomans, coming out of the border region,31 fell on the rearguard and slaughtered 12,000 men. Because of this there was great enmity between the Emperor and the Sultan, and T'oros was left in peace." I shall now pass on to the comparison between the two redactions of Smpad's Chronicle and list below the events lacking in the later one. The initial figures give the pages of the new edition, those in brackets refer to the pages in the first volume of the Documents armkniens in RHC. Pp. 188-9 (p. 623), A.D. 1167. Further information on the Catholicos Nerses the Gracious and remarks on the embassy of the Byzantine legate Theorianos. Pp. 190-1 (p. 625), A.D. 1173. Information on Nerses' successor, his nephew Gregory. Mention of the death of Nur ed-Din and the accession of his son Malik as-Salih. P. 192 (p. 625), A.D. 1175. Rupen seizes Msis, as well as Adana and Tarsus. P. 194 (p. 628), A.D. 1187. Praise of the Baron Leon.

Pp. 194-200, A.D. 1187. Long account of the battle of Hattin and the fall

30 According to Abu Shama it was Manuel who sent gifts to Nur ed-Din (op. cit., p. 105). The small number of prisoners, mentioned by Smpad, must refer to the leaders; Cinnamus speaks of 6,000 men (IV.21), while Gregory the Priest writes that ~ o , o o oprisoners were freed, which seems a gross exaggeration (op. cit., p. 190). 31 Smpad uses here the Turkish word up, which means "extremity" or "tip," and was applied by the Seljuks to the border lands especially. I am indebted to Mr. H. Berberian for this explanation.

SIRARPIE DER NERSESSIAN

of Jerusalem, entirely omitted in the later redaction, except for a short entry under the year 1188 : "The sultan Saladin took Jerusalem from the Christians" (p. 629). Following is a translation of the original passage : "At this time [A.D. 11871 the lord of Aleppo, Damascus, and Egypt was Husep', son of Ayub, called Salaheddin. There were two brothers, from the province of Dvin, the sons of a Kurdish peasant called Ayub; the name of the first son was Husep', that of the other was Hedl. They had come from their country for the purpose of drinking wine and had become the vassals of Nur ed-Din, the Lord of Aleppo, who, taking pity on them, occasionally gave them alms, and they served him faithfully. As their fortunes rose day by day they acquired some wealth and whatever they could lay hands upon [they spent] in eating and drinking with every one.32 Because of this they were liked by all and came to rule over many countries. Salaheddin became a very wealthy, arrogant, and belligerent man and began to threaten the Christians; his power increased from day to day, and, through his wisdom and shrewdness, he overthrew and utterly annihilated the authority of the Christians throughout his domains. In this same year Salaheddin marched against the King of Jerusalem ; the King, the Franks of the coastland, the Count of Tripoli, and the Frkres who wear garments marked with the cross joined forces against Salaheddin. The soldiers of the Franks were encamped on a hill, and the infidels suffered from lack of water. Then the apostate Count of Tripoli sent word to Salaheddin saying: 'What will you promise to give me if I should move the Christian armies and make them set camp in a place without water, so that you and your army may encamp by the water ?'33 And he [Saladin] promised him many treasures, confirming it in writing. Then the infidel Count set about advising the King and the chiefs and said: 'It is not good for us to remain here; come, let us leave this place and encamp on the hills where the wings of our army can be supported by our fortresses.' And he made everyone believe his wicked words, and when the Christians left that place, the Sultan encamped by the water, so the Christians could no longer reach the water, and were in great distress and uncertainty, and could find no solution to their predi~ament.3~ Then, in their despair they accepted death and rushed into battle. And when their front line was in battle 32 This account of Saladin's early life is purely imaginary. The continuator of Samuel of Ani also thought that Saladin's family had humble origins and that his father and uncle worked as wood carriers in Tekrit, ( R H C , Doc. arm., I , p. 453). For Saladin's origins see V. Minorsky, Studies in Caucasian History, (London, 1953)~ pp. 107-57. 33 Smpad, like other Armenian and Syrian writers, echoes the accusations of treachery made against Raymond of Tripoli by several Western sources, but he alone mentions explicitly the message supposedly sent to Saladin. For a discussion of Raymond's so-called treachery see in particular Marshall W. Baldwin, Raymond 111of Tripolis and the Fall of Jerusalem, (Princeton, 1936)~ pp. 156-60. 34 For a discussion of the battle of Hattin see Baldwin, ibid., pp. 151-5 and Runciman, op. cit., 11, pp. 486-91. Smpad seems to have known that the Franks did not camp a t Hattin during the night, but moved up to the hill the following day, as has been shown by a recently discovered account: Jean Richard, "An Account of the Battle of Hattin referring to the Frankish Mercenaries in Oriental Moslem States," Speculum, 27 (1952) pp. 168-77.

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 151 formation, the infidel Count of Tripoli and his men deserted the Christians, which brought about the destruction of the Christians. The latter, being resigned to death, joined battle and, when the fighting had lasted a long time, they were defeated, for their men and beasts were disabled and weakened by thirst. There was great heat and a parching wind, and the hand of the Muslims was strengthened and they slew everyone. Then the King, who, together with a few soldiers was entrenched on the summit of a hill, begged to be taken to the Sultan, and the latter sent men and had him brought to him. When the King arrived, the Sultan went forth to meet him, bowed down, kissed, and embraced him; then, taking him by the hand, he made him enter his tent and sit on a cushion. And he sat humbly before him and said: 'Holy King, you are a thousand times welcome to the house of your brother. Be not saddened, for such is the fate of warfare, sometimes defeat and sometimes victory. You are a just and faithful King, true to your oath, and your righteousness pleases me. Because of this not one hair of your head shall fall; because of you I shall show mercy to many, and because of my love for you, I shall set many free.'35 While he was saying this, they brought before him Raynald, Prince of Tripoli.36When the King saw him he arose; the Sultan also arose out of respect for the King, and he said to Raynald, Prince of Tripoli, who had betrayed the King, '0 infidel! I do not rise because of you, but because of your King.' And the Prince said: 'And I do not thank you, but my King.' And the King asked for some water, and the Sultan ordered it to be brought in a gold cup, mixed with rose-water and snow; and taking it, the Sultan, like a taster, sipped the water then presented it to the King; the King, taking it, drank half and gave it to the Prince of Tripoli, and he also drank of it.37And the Sultan said to the Prince: 'I am not giving this drink to you, but to your King.' And the Prince said to the Sultan: 'And I do not thank you, but my King.' And the Sultan said to the Prince: 'Unfaithful one, how many times have you sworn to me and pledged your faith, and have not kept your promise, but have enslaved and killed so many men; you seized my treasures on the route of Damascus and also, forgetting your oath, caused much blood to be shed at Sersim. What answer can you give now?'38 And the Prince, answering, said to Sultan 35 These words of welcome do not occur in any of the other sources, though there may be an indirect reference to them in one of the versions of the Estoire d'Eracles in which we read that Saladin felt greatly honored to have such rich prisoners in his power. (RHC, Hist. occ., 11, p. 67). 36 There is a strange confusion here between Raymond and Reynald of Chatillon, for although the name is correct (Ernaghd = Reynald, not Eremund = Raymond), he is designated as Count of Tripoli. In Samuel of Ani's Chronography we find the same confusion (RHC, Doc. arm., I p. 457). 37 One of the versions of the Estoire dJEracles also mentions the gold cup and the fact that the Sultan drank first before presenting the cup to the King (RHC, Doc. occ., 11, 67, third text). In all the other accounts the King drinks first and then passes the cup to Reynald. 38 Saladin's reproaches are more specific than in the Western texts (RHC, Doc. occ., 11, pp. 67-9; Chronique d'Emoul, ed. Mas Latrie [Paris, 18711, pp. 172-4), and are closer to the accounts of the Arab writers, for Smpad also refers to the treacherous capture of the caravan by Reynald on the way to Damascus (RHC, Doc. or., I, pp. 55-6, 675-6, 687; IV, 258-9, 275-6; Beha ed-Din, Life of Saladin, PPTS [London, 18971, p. 115; Kemal ad-Din, in Revue de I'Orient latin, IV [1896], pp. 179-81). Bar Hebraeus says that the nobles urged Saladin to kill Reynald: The Chronography of Gregory

152

S I R A R P I E DER NERSESSIAN

Salaheddin: 'Do not bark so. Do what you wish. I have already exacted the price of my blood from the Muslims for forty years, and henceforth I care nothing if I die.'39 Then the Sultan signalled to the servants who seized him [the Prince] by the feet and hands and [threw him] before the Sultan; and he, drawing- his steel sword, struck him on the loins and the servants killed him immediately. Seeing this, the King was greatly saddened, and the Sultan said: 'Do not grieve for the loss of one who was unfaithful to you.' Then they brought the Frkres together with their Master and made them stand before him; and the Sultan said to the Master: '0 venerable head of your brethren, though you slew many of our soldiers, I love you because of your bravery. Now, abjure your faith and embrace our religion, and I shall give you presents and confer honors on you, and I shall set you all, and especially you, above my greatest men.' The Master answered and said: '0 great Sultan, willingly will I comply with your wish, but if you will so command, let me speak to my brethren and persuade them all to obey you.' And he granted him permission and said: 'Whoever does my will shall live, and whoever does not listen shall die by the sword.' Then the Master assembled everyone and said: '0 brethren, behold the days of the salvation of our souls have come, by which we shall inherit heaven. I beseech you, remain united and steadfast in the love of Christ ; let us today mingle our blood with His redeeming blood; be not afraid of those who kill the body, but of those who subjugate the soul and the body, and let not passing greatness deceive us.' And he spoke many other words from the Holy Scriptures, enjoining them to die in their faith. And he returned to the Sultan and said: 'Some are obedient to your orders, but some are opposed; now, command that they be brought before you.' And when they came, he [the Sultan] began to question them, one by one, but they could not be swayed, a n d so he commanded that they all be killed. Then he said to the Master. 'And you, why do you consider accepting our faith?' And the Master, filling his mouth, spat in the Sultan's face so that being infuriated he [the Sultan] might kill him at once, and he should thus join his spiritual brothers. And he said to the Sultan: 'I exhorted them all to accept death, that they might earn the celestial life, so why should I do your will ?' And the Sultan ordered that he also should be killed; and when they killed him, light descended on them from heaven [and shone] for three-days, to the confusion and shame of the infidels and to the glory of the believers.40 Abd'l Faraj . . . Commonly known as B a r Hebraeus . . . trans. from the Syriac by E. A. \V. Budge, I (Oxford, 1g32), p. 324. I have not been able t o identify the locality called Sersim. Reynald intercepted the caravan as it was passing near Shobek, that is, near the ICrak of Rlontreal (Beha ed-Din, ibid., p. I I ~ )but , it is not absolutely clear in the Armenian text whether the bloodshed a t Sersim refers t o the capture of the caravan or t o another event. 39 I n the same version of the Estoire dlEracles mentioned above, Reynald also gives a truculent answer t o Saladin, though the actual words are different: R H C , Doc. occ., 11, pp. 68-9. 40 The Master of the Frbres must be Gerard of Ridfort, the Grand Rlaster of the Temple who, as we know from other sources, was one of the principal prisoners. Smpad's account, entirely favorable t o Gerard of Ridfort, differs from that of the other sources. He is mistaken in thinking that Gerard was killed by Saladin, for we know that he was sent t o Damascus with the other prisoners, was released

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 153 And when all this had happened in this manner, the Sultan freed the King, giving him many presents, and also those who were with him. And he commanded that each inhabitant of Jerusalem give one Egyptian tahegan per head, and that, taking from his house whatever he himself could carry he depart in peace, while those who wished to remain were to give yearly a red tahegan per head. Many remained, and many arose and departed. And Salaheddin took Jerusalem and its provinces, and gradually he seized all the coast land and the province of Antioch, and all the Christians trembled through fear of him. This same year, a certain Turcoman, called Resdom, gathered innumerable Turcomans and entered the land of the Cilicians, and he boasted of suppressing the very name of the Christians. He advanced as far as Sis and encamped before the city of Ravin, and the face of the earth was covered with this infinite multitude. Then Leon, strengthened by God, marched against them with thirty men and slew their leader Resdom, and they all turned and fled. Pursuing them as far as Sarvantik'ar Leon slaughtered them; and it is said that two warriors were seen to descend from the citadel of Sis and to slaughter them, and they were believed to be Saint George and Saint T h e ~ d o r e . " ~ ~ Pp. 200-1 (pp. 629-30). After mentioning the death of the Constable Baldwin in 1188 and the subsequent capture of Bragana by Leon, the Venice manuscript adds that Leon killed the Master of the Citadel, the Emir Dip'li, who had murdered Baldwin. Mention of the fall of Jerusalem, erroneously dated in the year 1188 in the earlier publications, is omitted. The marriages of Leon's two nieces, Alice and Philippa, and his own marriage to a niece of Sibylla, the wife of Bohemond of Antioch, are dated 1188 instead of 1189. Events of later years, such as the death of the Catholicos and the election of his successor, are also placed under the year 1189 in the earlier publications, while in the Venice manuscript (pp. 204-5) they occur in their proper place, namely 1193. The account of the events of the intermediary years, 1189 to 1193, has been entirely omitted in the later redaction. I t reads as follows : shortly thereafter, and died in the battle before Acre, or, according to some sources, mas captured and killed b y Saladin. ( E m o u l , p. 253; R H C , Doc. occ., 11, p. 130; Doc. or., IV, p. 425; Ambroise, TIze Crusade of Richavd L i o n Hearted, trans. by Merton J. Hubert [New York, 19411, p. 143 and note 51). Only I
.. .

154

SIRARPIE DER NERSESSIAN

Pp. 201-4. "In theyear 1189 theKing of the Germans set forth with numerous soldiers and came to Constantinople ; and he marched on Iconium, seized it, and slew the soldiers of Kilij Arslan. And Kilij Arslan gave him thirty great men as hostages and ~oo,oootahegans and established friendship with him. And the Emperor came as far as Seleucia and, as it was summer and the heat was very great, the King went to swim in the river, but, overpowered by the onrush of the waters, for he was an old man, he was drowned. I t is said that he had been foretold that he would die by water ; that was why he had come by the long land route. As for his son, he went as far as Acre and died there after six months; then his soldiers scattered and departed. "And in the year 1191 the King of the Franks came by ship to Acre with numerous soldiers and laid siege; Salaheddin also came, for the city belonged to him, and he encamped opposite them. The Frank soldiers dug three trenches round about themselves, they fortified their position, and scattered iron caltrops all around. The city was in great straits and the Sultan could not come to the help of the citizens. At that time the King of the English came to Cyprus and took it from the Greeks; he captured the Duke Comank [Isaac Ducas Comnenus] and brought him to Acre. The two Kings, joining forces, fought bravely against the Sultan and the citizens. Then the Sultan sent word to the Kings and said: 'Take back your city and sell the men for their weight in silver and gold.' They answered and said: 'Out of respect for you, it would be right for us to do so, but because we have sworn before the Holy Sepulchre of Christ to kill everyone by the sword, we cannot perjure ourselves.' And so they took the city and killed 36,000 men there; and Salaheddin fled.42 "In the year 1192 there was a severe famine, and it is impossible to describe in writing the greatness of the suffering; so many men died that the dead could not be buried, and the living envied those who had already died. When springtime came, they ate the grass like sheep, and because of this unwonted food they fell and died. This same year Kilij Arslan, the Sultan of Iconium died, and Salaheddin began to harass great Antioch with a view to capturing it. But the astrologers told him, 'You cannot conquer it,' and the Sultan desisted from his plan. Antioch suffered from famine, for no food could be brought into the city for fear of the Sultan. And the citizens said to the Prince: 'Behold, we die of hunger, what should we do?' And the Prince said: 'Wait for me for fifteen days and then I shall give you an answer.' Then the Prince arose and took with him five horsemen and went to Salaheddin while he was still encamped opposite Acre. He arrived at the door of the pavilion and said to the doorkeepers: 'Tell the Sultan the Prince of Antioch is here and wishes to see him.' When the Sultan heard this he immediately came out to him, and taking him by the hand, made him enter the pavilion and bade him sit down. And he [the Prince] said: 'I have a request; if you should grant it 4 W n the siege and capitulation of Acre see Runciman, Hist. of Crusades, I11 (1954)~pp. 47-51.

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 155 I shall sit down.' And he said: 'What you request is granted. Speak.' And the Prince said: 'I beg you to grant me Antioch.' And the Sultan said: 'It is granted, and moreover I shall give to you and your city sufficient food for three years.' And they became friends and he returned to Antioch, and the city was filled with abundant food.43 "In the year 1193when the Prince returned from [his visit to] Salaheddin he spoke to his wife of his intention of seizing Leon, and she said: 'Do not do such a disloyal deed for he is my son-in-law, and he always comes willingly to your service and helps you in your wars.' But he did not desist from his evil plans and sent an invitation to him. And Leon went forth to Baghras. And the wife of the Prince sent word in secret to Leon. And Leon invited the Prince to come to him at Baghras so that he might entertain them there and then they would proceed together to Antioch. And they came willingly ; Leon went out to meet them, escorted them to Baghras, and there seized the Prince and, taking him, imprisoned him in the citadel of Sis, and kept a careful watch over him.44 "This same year the Sultan Salaheddin sent word to Leon bidding him to cede the land of Cilicia after which he would be allowed to depart safely. Leon was in doubt as to what he should do, but putting his trust in God he said to the messengers: 'Tell the Sultan I have no land to give you, but should you come to my land I shall let you taste of the double-edged sword, as I did to your coreligionary Resdom.' Upon hearing this the Sultan growled like a lion, and made ready his army and set out towards Cilicia to annihilate the believers in Christ; he had come as far as the Black River when he fell ill and died there. And his son called Melek' Dahr succeeded him."45 There are no significant differences between the two redactions for the events which took place in Cilicia during the years 1193 (after the death of Saladin) 43 I have found no other reference to the famine in Antioch. According to the Arab sources and to Bar Hebraeus (op. cit., pp. 340-1) Bohemond of Antioch visited Saladin on October 30th a t Beirut, not a t Acre. Saladin "entertained him hospitably, granting him the territory of el-'Amk-corn lands, the crop bringing in an annual return of fifteen thousand gold pieces (din2rs) annually" (Beha ed-Din The Life of Saladin, p. 395. See also Abu Shama, RHC. Doc. or. V, pp. 89, 91). Bohemond's request, as reported by Smpad ("I beg you to grant me Antioch") should be interpreted as referring to the plain of el-'Amk, known in ancient times as the Plain of Antioch. Saladin had acquired this territory "in the year 584 (A.D. 1188-1189) a t the time of the conquest of the coast land," Beha ed-Din, op. cit., P. 38. 44 The new edition gives the correct date, 1193, for the guet-apens of Baghras while in the later redaction this is related under the year I I g j , along with the arrival of Henry of Champagne (RHC, Doc. arm., I, p. 631). I n the Venice manuscript the arrival of Henry of Champagne and the negotiations for Bohemond's release are told separately, in their proper places (op. cit., pp. 206-7). The guet-apens is mentioned before the death of Saladin, but Saladin died on March 4, 1193 (Beha ed-Din, op. cit., p. 406) and Bohemond was still in Antioch in September of that year (Cahen, op. cit., p. 583 and note 4). 45 We have no other information about Saladin's intention to invade Cilicia. After his meeting with Bohemond a t Beirut, he went to Damascus on November 4, 1192 (Beha ed-Din, op. cit., p. 396; Bar Hebraeus, op. cit., p. 341), and spent the winter months there, being too ill t o undertake any campaign. The Black River is probably the Kara Su which joins the Orontes in its lower course; a Muslim expeditionary force may have approached that region, but it could not have been directed by Saladin, who died in Damascus. Melek' Dahr is llelik az-Zahir, Lord of Aleppo.

156

SIRARPIE DER NERSESSIAN

to 1200.46 For the year 1201 there is no entry in the later copies, while in the Venice manuscript we find the following passage: P . 212. "In the year 1201 the Sultan Rukn ad-Din went East with many

soldiers and, not by warfare but by peaceful means, captured the city of

Theodosiopolis, which is the city of G a r i r ~Then, . ~ ~ arrogantly,

he proceeded

to Mjngerd, and while he was attacking that fortress the Georgian soldiers

arrived and routed him.48 They seized the Lord of Erzinjan, Shah Vahram,

and returned to their country with rich booty.49The Sultan appointed his

brother, Tughril Shah, Lord of the city of Garin5QHe was a kind man and

throughout the days of his life he remained on friendly terms with King

Leon, and he greatly loved the Christians. His brother, the Sultan, return-

ed to his country."

There are some differences between the two redactions in the account of the events of the year 1203. The Venice manuscript mentions the intervention of the catholicos Grigoris on behalf of his nephew Het'um, which is omitted in the later copies, but does not give as many details about the destruction of the church vessels carried out by order of the catholicos H ~ h a n n e s . ~ ~ Under the year 1204 it records the death of Rukn ad-Din and the accession of his son Soliman Shah (p. 213). The entries for the year 1205 are the same in both redactions, but the events of the years 1206 to 1216 are either omitted or very briefly mentioned in the later redaction.

Pp. 214-5 (pp. 642-3). "In the year 1206 Bohemond, Prince of Antioch,

died, and his son, the one-eyed Count of Tripoli succeeded him.52 Leon

4 6 See Alishan, Lion le Magnifique, pp. 173-7, who, in his account of the coronation of Leon I, reproduces the passage in which the bishop and barons of Armenia are listed (In Akelian, op. cit., pp. 208-10). 47 Smpad does not speak of the raid into Cilicia by Rukn ad-Din which is known through other sources (Cahen, op. cit. p. 601). According t o Abu'l Feda, Rukn ad-Din took Theodosiopolis (= Erzerum) from >lohammad ibn Saltak (RHC, Hist. ov., I, p. 79). 48 This victory is told in great detail in the Georgian Annals. Histoive de la Ge'ovgie depuis l'aqztiquzie' jusqu'au X I X e sidcle, traduite du GCorgien par by M. Brosset, I (St. Petersburg, 184g),pp. 456-63. 4"he Georgian Annals do not give the name of the Lord of Erzinjan, but we find his full name, Melik Fakhr Eddin Bahram Shah, in the Seljuk Nameh (Ch. Schefer, Quelques chapitres de l'abvege' du Seldjouq Nameh [Paris, 18891, pp. 39-40). 50 Rukn ad-Din appears to have first named Aha Eddin Saliqy as governor and later deposed him in favor of his brother, Tughril Shah (Schefer, ibid., p. 40). 51 I n the Etchmiadzin manuscripts, after the enumeration of the precious vessels and other treasures destroyed by the Catholicos, we read: "And do not blame us for writing this, for we know i t not b y hearsay, but we saw and touched with our hands these objects while they were being destroyed" (Doc. avnz., I , p. 641). Alishan had already remarked that this sentence could not have been written by Smpad, since he was only a few years old when the destruction took place and he was not a t Hromgla. I t should, therefore, be considered as a later addition, like those mentioned above (p. 146). There are other signs that this passage has been altered b y the addition of information derived from a source hostile to the Catholicos, for the introduction of several words into the preceding paragraph, relating to the person of the Catholicos, results in obvious contradictions. Hohannes is said to be "humble in heart and perfidious, simple in his appearance, but indifferent t o spiritual matters. He greatly loved virtuous people." The corresponding passage in the Venice manuscript reads: "he was humble in heart, simple in his person and zealous in spiritual matters. He greatly loved virtuous people (P. 213). 5 2 The correct date of Bohemond 111's death is 1201 (Cahen, La Syrie du Nova p. 594). Het'um (Hayton) of Gorigos has the correct date (Akopian, op. cit., 11, p. 61).

. . ."

A R M E N I A N C H R O N I C L E O F T H E C O N S T A B L E S M P A D 157 sent a messenger to him and showed him the treaty which the Count's father had made with the King whereby the son of Bohemond's eldest son was to succeed him, of which we spoke previously, but he [the Count of Tripoli] rejected the treaty and did not recognize [his nephew's] rights. Then the King sent to the Patriarch of Antioch and showed him the same document, and the Patriarch testified to the King's rightful claim, but the Count would not agree. Then the Patriarch excommunicated him and ordered that no bells should be rung in Antioch, no Mass be celebrated, no dead be buried, but he [the Count] refused to obey. He even dared to lay hands on the Patriarch and to put him in prison, where he was tormented by hunger and thirst. He sent [men] to him saying, 'Testify on my behalf, that I am the rightful master of Antioch, and you will be set free and live.' The Patriarch never agreed to do this, and he died in prison of hunger and thirst without uttering a lie. From this time on there was bitter enmity between the King and the Prince. "In the year 1207 the Counts of Venice and of Flanders came to Constantinople and took the city from the Greeks; they massacred and put to flight everyone, and from then on the Greeks ceased to rule in Con~tantinople.~~ "This same year, under some pretext, Leon seized the Sebastos Heri and his sons Constance Camardias and Joscelin and Baldwin; he chained and imprisoned them. Heri was the son-in-law of Hohannes, the Catholicos of the Armenians, and this caused serious dissension between King Leon and the Catholicos Hohannes. Their hatred for one another grew to such an extent that the King, the prelates, and the barons of Cilicia deposed the Catholicos and in his place appointed David, Archbishop of Msis and Abbot of Ark'agaghin as Catholicos of the Armenians. And Hohannes remained at Hromgla in retirement, and through his courage and wisdom he was able to circumvent the King's numerous secret or overt intrigues. Meanwhile, the Sultan Khosrov shah [Kaikhaus I], son of Kilij Arslan, reigned over the land of Rum, and began to threaten King Leon; induced and encouraged by Der Hohannes, he marched with many soldiers on Pertous, captured it, and seized its master Krikor, son of Leon; and from then on Pertous ceased to belong to the Armenians. And this happened in the year 1208." In referring to the establishment of the Empire of Nicaea, Smpad mentions how the Seljuk Sultan was killed by the soldiers of Lascaris in 1209 (p. 216). He then relates that Leon appointed Raymond-Rupen heir to the throne, and how he blinded and imprisoned Kork, the illegitimate son of Mleh.54 He also speaks of the invasions into the territory of Antioch, where Leon destroyed villages and cut down trees and vineyards (p. 217). I n 1210 Leon sent Het'um of Lampron to the Pope and to the German Emperor and obtained a crown for 5 T h i s incorrect date is repeated in the later redaction. BHC, Doc. avm., I, p. 642.

b4 Parts of this section are quoted by Alishan, Lio7z le Magnifique, pp. 278-9.

158

SIRARPIE D E R NERSESSIAN

Raymond-Rupen; he married Sybil, the sister of the King of Cyprus while another sister was married to the young Rupen (p. 217). I n 1211 after the death of the Catholicos David, there was a reconciliation between the King and the former Catholicos Hohannes, and the sons of Heri, Joscelin and Baldwin, were released. P. 217. "This same year [IZII] the Lord of Garin, Tughril shah, prompted by Leon, marched with a large army on the city of Cesarea; Leon also came with a largeforce and the two together made war against Kaikhaus, Tughril's nephew, but after tarrying several days without being able to accomplish anything, they returned each one to his country." For the years 1211-1213 Smpad mentions briefly the death of Zak'arit, Lord of Ani, the complete reconciliation between the Catholicos and the King, and the marriage of Leon's daughter Rita to John of Brienne at Acre (pp. 218-9). The description of Leon's entry into Antioch in 1216, which is recounted in some detail, the passage relating to the selection of Leon's daughter as heiress to the throne in lieu of Raymond-Rupen, and the account of the siege of Damietta have been quoted by ~ l i s h a n - a n dneed not be repeated here.55 The Seljuk wars are related more fully in the later redaction15'jbut in almost every other instance the Venice manuscript records events which are either omitted or greatly abridged. Pp. 222-3, A.D.1219. Account of King Leon's death and burial. Pp. 223-5, A.D.1220. Revolt of some Armenian and Greek barons against the bailiff C ~ n s t a n t i n e .Death ~ ~ of the Catholicos Hohannes and election of Gost andin (Constantine) of Partzrpert . P. 225 (pp. 647-8)) A.D.1222. Marriage of Zabel, daughter of King Leon, to Philip of Antioch.58 Imprisonment and death of Philip. P. 225 (p. 648)) A.D.1226. Marriage of Zabel to Het'um, son of the bailiff C~nstantine.~~ P. 225, A.D.1229. The German Emperor seizes J e r u ~ a l e m . ~ ~ P. 229, A.D.1252. Death of Queen Zabel. Pp. 229-30, A.D. 1253. King Het'um goes to the court of Mangu Khan.61 Het'um's daughter, Sibyl, marries Bohemond of Antioch. 55 Alishan, op. cit., pp. 257-8, 286-9; Akelian, op. cit., pp. 219-20. 56 Akelian, p. 221-2; RHC, DOC. arm., I, p. 644-5. See the translation of excerpts from this passage in Alishan, Sissouan, p. 314. The stipulations set by the Armenians before the marriage are mentioned in the later redaction and not in the Venice manuscript, but in the latter we read that Philip "began to send to Antioch, little by little, the riches collected by King Leon and his predecessors." This statement agrees with the remarks of the historians Kirakos and Vardan (RHC, Doc. arm., I, pp. 428, 442-3). s9 The account in the later redaction is again more detailed. I n the Venice manuscript the story of Zabel's flight to Seleucia is omitted and one can understand that Smpad preferred not to dwell on Zabel's reluctance to marry his brother. This section in the Etchmiadzin manuscripts shows obvious signs of modifications and, to some extent, of confusions. The birth of Het'um's and Zabel's eldest son is recorded under the same year 1226, instead of 1236, and likewise under the same year we find the account of the Seljuk invasion of Cilicia of 1245. (Doc.arm., I, p. 649). so Ten folios are missing from this point down t o the year 1252. See above p. 143. Smpad's account, though shorter, is in substantial agreement with that of Kirakos (ed. Tiflis [~gog],pp. 350-7).The only addition contained in Smpad is the statement that Mangu Khan appointed two of his servants to accompany Het'um: "Margadea, who was one-eyed, and Batcho." 5'

58

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 159

P. 230, A.D. 1254, Raids by a certain Turcoman named Islam-Beg.62 P. 230, A.D. 1255. Death of Prince Geoffrey of Cilicia, who had been made a eunuch. Pp. 231-2, A.D. 1256. Het'um returns from the Mongol court in the month of September, and makes successful raids into the sultanate of Riim.63 On November 15 Het'um's son Leon is knighted in the presence of a large assembly of Armenian and Latin princes and barons.

Pp. 232-3. "In the year 1258, the Nation of the Archers set out and marched on the city of Babylon [Bagdad] with their generals and captains, having as their supreme commander the Khan Hulagu, and they surrounded and harassed the city. The inhabitants of the city were very numerous, all the Muslim nations being assembled there for two reasons: first, because the year before they had heard that the Khan would march against them and had made all the preparations for war; second, because the Caliph who resided in Cairo was no longer there. At the time of King Baldwin who resided in Jerusalem and molested Egypt, the Sultan of Aleppo came and killed the Caliph, and from then on the impure see of their pontiff was transferred to Bagdad. That is why they were assembled there by order of their leader. Then the inhabitants of the city marched against the Nation of the Archers, and they killed many of their soldiers and routed them, and after pursuing them for a while they returned to their city. But they [the Mongols] again marched against the city, and the Caliph sent to Khan Hulagu, saying: 'Take your soldiers and depart from us, lest when I display the mantle of the prophet, you be utterly lost.64But if there be among you men who follow our religion, I shall spare them.' When the Khan heard these words he [and his companions] mocked the Caliph; they spat as if they were spitting in the Caliph's face, and Hulagu, raising his voice, said: 'With the help of God and the counsel of Jenghiz Khan we shall destroy by the edge of the sword all of you who boast in the prophet.' After these words, the two armies marched against one another; the Nation of the Archers attacked and raising a loud cry they struck at the enemy with great force, routed them completely, and, entering the city after them, massacred so many men and women that for many days the wide Euphrates river which ran through the city was tainted with blood. They continued to slaughter until they were exhausted; then they appointed chiefs and tax collectors and returned to their country with rich booty." P. 233, A.D. 1258. Het'um's younger brother, Leon, dies on May 30 in the city of Adana, just before setting out for Cyprus. Raid of the Turcoman Sarum.65 Alishan, Sissouan, p. 377. C1. Cahen, "Les Turcomans de Rfim," Byzantion, 14 (1939), p. 133.

Quoted in part by Alishan, ibid., p. 187.

64 The reference to the prophet's mantle or standard also occurs in Kirakos, op. cit., p. 365, but

Hulagu is said merely to have answered, "God knows what He will do." Alishan, Sissouan, p. 377; Cahen, op. cit., p. 133. 62

s3

160

SIRARPIE DER NERSESSIAN

P. 234, A.D. 1259. Het'um goes to Tripoli to make peace between the Prince of Antioch and the Count of Tripoli. On the day of the feast of Pentecost, Het'um's brother Baldwin is consecrated bishop at Tarsus, and called John (Hohannes) ; on the same day Het'um's son, T'oros, is knighted. Pp. 234, A.D. 1259. "But before this a baron by the name of Oshin, of Greek nationality, captured the fort of Muntas. Hearing this, the Sultan of RGm, Rukneddin, assembled a large army, appointed a chief, and sent it to besiege the aforementioned fort of Muntas. Then a man was sent forth from the fort, and coming to King Het'um, told him that numerous Christians were gathered there and that the Sultan had marched against them."66 Pp. 235-7. "In the year 1260 Khan Hulagu and his mighty host advanced like a spring torrent, seizing the forts of the infidels wherever they passed, some peacefully, some by fighting. Thus he advanced as far as Aleppo, surrounded the city, and sent word to King Het'um to join him; and the latter immediately came to the Khan with his army, and the Khan received him joyfully. And the victorious Khan harassed Aleppo with numerous engines, and in seven days he opened a way for the invaders in spite of the great width of the ramparts and the depth of the Then they all raised such a loud cry that the earth shook and trembled at the sound; entering the city they marched on the citadel and, drawing their swords, mercilessly slaughtered the nation of the Ismaelites. However, they did not sin against the faithful by killing them, but only plundered their belongings. No one can tell how many were killed. Taking their booty and captives they proceeded to Damascus, and subjugated the towns, villages, and fortresses, all the way to Jerusalem. Over that region the Khan appointed as governor and commander a man called Kitbuqa, and the brave and powerful Khan, taking his soldiers and his son, Abaga, went East to his own country. However, the commander Kitbuqa did not follow the instructions of the Khan who had appointed him to remain at that place, but collecting his soldiers and 500 men from Cilicia whom he had asked the King to send him, set out with the intention of entering Egypt.68When the Egyptian spies saw this and immediately notified their people, an army was called up and went forth to meet the enemy some four days' journey from their place, at a locality called Brr.69The two The remainder of this account, and the successes of the Armenians under the leadership of Constable Smpad himself, are quoted by Alishan, ibid., p. 191.I n the later redaction this long story is reduced to the statement that Het'um took the fort of Muntas (RHC, Doc. arm., I, p. 651). 6' The seven days' siege agrees with the account of Rashid ad-Din (Histoire des Mongols de la Perse, ed. Quatremkre [Paris, 18361, p. 335) and of Abu'l Feda (RHC, Doc. or., I, p. 140).Hayton in La Flor des estoires de la tevre dlOrient, says that the siege lasted for nine days and that the citadel resisted for eleven more days (RHC, Doc. arm., 11, pp. 171 and 302). 68 Smpad and the author of the History of the Nation of the Archers are the only ones who state that Kitbuqa initiated the Mongol-Egyptian war. History of the Nation of the Archers (the Mongols) by Grigor of Akanc'. Edited with an English Translation and Xotes by Robert P. Blake and Richard N. Frye (Cambridge, Mass., 1954), p. 349 (81). s9 All the Muslim sources say that the battle took place a t Ain Jalud, that is, the Springs of Goliath, between Nablus and Beisan, and the same locality is designated by Hayton under the varying forms

A R M E N I A N C H R O N I C L E O F T H E C O N S T A B L E SMPAD 161 armies encamped half a day's distance from one another. When the sun arose, they clashed and fought violently, but because of the great heat and the sickness among their horses the Nation of the Archers was defeated and fled. The Egyptians pursued them and few of them escaped; Kitbuqa was killed on the battlefield, while his wife and sons were taken captive to Egypt. The remnants of the army went to Hulagu Khan and related what had happened, and he, growling like a lion, promised to go and seek revenge for the blood of his soldiers. "This same year Het'um, King of the Armenians, assembled his soldiers and, having set forth as for an excursion, he passed between Cappadocia and Iconium, going out to meet the Nation of the Archers at Gangra, in Galatia, which was on the borders of Smyrna and [the land] of Lascaris, for [the Mongols] had previously asked him to join them. However, when the King made preparation for his departure, [the Mongols] had already gone beyond that place, but because of his fear of them, he did not interrupt his journey, and, though facing death, he proceeded with a few men, as we have said. But nothing was achieved, for the men against whom they went had heard that they were coming and had fortified themselves in the caverns and grottoes of the rocks; and they [the Armenians] all returned empty-handed to their homes. And one of the barons of King Het'um, of Greek nationality, called Vasil Geronents, died on the way, and they brought his body to Cilicia, and buried him there, in the sepulchre of his ancestors."

P. 237, A.D. 1261. Het'ums daughter Rita, marries Constantine, the son of the Lord of Sarvantik'ar. Geoffrey of Sarvantik'ar dies, leaving three sons, Constantine, Smpad, and Oshin. There is no entry for the year 1262. Pp. 237-8, A.D. 1263. Constantine, father of King Het'um, dies on Sunday, February 24. Pp. 238-40. Long account of the struggle with the Karamanides which took place before the death of Constantine. This entire passage, except for the words of advice addressed by Constantine to his son, King Het'um, has been quoted by Alishan and commented on by C1. Cahen.'O Pp. 241-2. "In the year 1263, Het'um, King of the Armenians, went on a friendly visit to Antioch to see the city, and he took with him the venerable doctor and Archbishop of Anazarba, Der Hagop, also some priests and deacons, and many golden and silver treasures from the treasury of his father Constantine, in order to distribute them among the poor and to offer them to the sanctuaries as a memorial to his soul. When the King entered the city, he was joyfully greeted; walking through the of Aimeloc or Haymaloth (RHC, Doc. arm., 111, pp. 175 and 305). Kirakos and Vardan say that the battle took place a t the foot of Mount Thabor (Kirakos, [Tiflis, 19091, p. 373; Vardan [Venice, 18621, p. 152). The place called Brr by Smpad is probably Zer'in which is only a short distance west of Ain Ain Djalut). Jalud (Encyclopedia of Islam, S.V. 70 Alishan, Sissouan, pp. 344-6; Cahen, in Byzantion, 14 (1g3g), pp. 133-5. II

SIRARPIE DER NERSESSIAN

city, he visited Saint Paul and Peter and other churches, and offered them and made his presents; he also went to the monastery called Djebik' father Constantine a fellow-member of their brotherhood; he, the King himself, became their brother and he gave them many gifts, by testament, so that they should come to his country each year to collect them. He remained in Antioch for some time and then returned to his country, Cilicia. "This same year, in the month of June, King Het'um went east to Hulagu Khan, because every year the people of Cappadocia molested the Cilicians who lived at the foot of the mountains. And because he was very fond of the King, Khan Hulagu sent equitable Mongol judges after himj71 who journeyed with the King as far as Heraclea; the Sultan of RGm Rukneddin also arrived and they remained there for a few days, swore a pact of friendship, and made peace. And the King and the Sultan became as father and son, and each one returned to his country. This same year Kyra Mary, Countess of Jaffa and sister of King Het'um, who had come to see her father, Constantine, because of her sorrow, reached the end of her lifetime and died at the fort of Lampron; she was buried in the holy monastery of Sgevra. She left two sons and three daughters." Pp. 242-4. ''In the year 1264 King Het'um gathered together an army, and with many foot soldiers from among the common people, he invaded the province of Aleppo and the small towns of Martmsrin, Srmin and also they won some booty and a few servants; it was there F ~ g h a . 'There ~ also that King Het'um had a narrow escape. Taking with him two eunuchs from among his servants and one of his barons, Constantine Ablhasnants, he left his soldiers and entered that town, walking carelessly without cuirass or helmet. Suddenly they met twenty infidels, dressed in armor, who were on their way to the citadel, which was in the middle of the town, in order to protect those who had fled and gathered there. When they came face to face, the King did not know what to do; the infidels did not know who he was, and one of them pounced to strike the King with his sword, but the eunuch Joscelyn intercepted the stroke. Once more the man raised his sword to strike the King and the Baron Constantine diverted the stroke, and, drawing the King to one side, he faced the man. Theinfidels went on and entered the citadel, and thus the King escaped alive; he

[?I

71

The word used in the text is arghoutchik'. According to the Malkhasiants Dictionary (Erevan,

1944) this word is derived from the Tatar yargufi which means "judge" and serves to designate the Mongol judges. I n the Chronicle ascribed to King Het'um 11, under the year 1263 we read: "King

Het'um and Sultan Rukneddin and the arghoutickik' assembled a t Heraclea for the purpose of dividing the forts between them." (Akopian, M a n r jamanakagrout'younner, X V I I - X V I I I c , I [1g51], p. 82. Akopian gives good reasons for ascribing this Chronicle to Het'um I 1 rather than to Het'um [Hayton] of Gorigos, pp. 65-73.) See Pelliot's comments on the tribal name Argun and his suggestion that this name may be the Argons mentioned by Marco Polo (P. Pelliot, "Une ville musulmane dans la Chine de Nord sous les Mongols," Journal asiatique, CCXI [1g27], p. 265, note 3).Marco Polo says that these "Argons," who sprang from two different races, were handsomer and more capable than the other natives and consequently acquired considerable authority among the Mongols (ed. Yule-Cordier, I [New York, 19031, p. 284). 72 These are the towns of Ma'arra Mipin (in Latin: Megaret Basrin), Sermin and Fou'a, southwest of Aleppo: Cahen, L a Syrie d u Nord, p. 156.

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 163 returned to his land with rich booty and there was great rejoicing. This same year Het'um again assembled his soldiers and marched on the fort of Aintab, but returned to his land without having been able to make any headway against the fort. After a short time, during the winter, King Het'um thought of returning against the same fort of Aintab, but when he reached Bourdj ar-Rasas he had to stop, for the sun was darkened by a cloud and did not appear for five days. Because of the violence of the wind and the rain, they could not leave their tents, and the servants and common footsoldiers suffered greatly. They, therefore, thought of returning, saying, 'God does not want us to go farther.' There was among them a Frank named Martin, a surgeon honored by the King, and he spoke to the King and nobles who were gathered at their deliberations, some saying, 'Let us turn back,' and others saying, 'Let us not turn back.' [Martin said] '0 King and princes, spend this night outside your tents, and then discuss whether to turn back or to stay.' And the words he spoke convinced them; many of them praised him and they returned peacefully. This same year Het'um again called an assembly for he wanted to plunder the district of Aleppo, but it being winter he could not put his plan into effect. In the year 1264-1265, Khan Hulagu sent one of his generals, together with a large army, to the impregnable fort of Bir which was in the hands of the infidels.73The general, whose name was Durba, raising bastions, sorely harassed the fort and demolished the walls of the ramparts and the tower. Durba then sent to ask King Het'um to join him, and the King, having mustered his soldiers and his relatives at the fortress called T'il Hamdun, celebrated there the feast of Epiphany. Then, after a few days' Het'um assembled his soldiers and went out to meet Durba, and proceeded as far as a place called Pampgtzor; he sent zoo cavalry men ahead to Durba and was himself to follow them. At that time came the welcome news that a son was born to Leon, baron of the Armenians, at Msis, and this was in January. . . . And who can describe the joy of the King, of the nobles, and of the people at this news? And many, even of those who were least worthy as well as some of the eunuchs, were honored by being knighted. Then news came that Durba had left Bir because the Sultan of Egypt had arrived against him; hearing this, the King returned to his home." Pp. 244-5. This same year, at Easter, Prince Leon's son is baptized and Smpad's two sons, Het'um and Vasil the Tatar, are knighted. The Sultan of Egypt, Baibars, comes as far as the outskirts of Antioch, at the Black River, but hearing that the Armenians had assembled a large army, he withdraws without entering Cilicia. P. 245 (p. 651). Oshin, Lord of Gorigos, brother of the King, dies on December 26 at Tarsus and is buried at Sis. 73 Baron C. dlOhsson, Histoire des Mongols, I11 (The Hague and Amsterdam, 1834), pp. 404-6. 11*

164

SIRARPIE DER NERSESSIAN

Pp. 245-7. "In the year 1266, the Sultan of Egypt once more assembled an army and, advancing as far as the fortresses of the Frkres who wear the garments marked with the cross, he captured Arsuf, Safeth, and other forts.74 Then turning away from Cilicia, he came to Damascus and remained there for some days. And after receiving the ambassadors of King Het'um, who had come to persuade him to make peace, he sent his own ambassadors; for the Sultan wanted to make peace, but demanded from the King fortresses and other localities which bordered his own lands. The King did not comply with these demands for two reasons; first, because of his fear of the Archers [Mongols], lest they should say that King Het'um was an ally of the Sultan of Egypt, and had given him the places and fortresses freed by them; and second, although the Sultan's demand was not great -being but a ruined place called Sheh75-and he said of it, 'Give me that locality so that I might make a market-place for us and for you,' the King would not cede it so as not to come under the authority of the Sultan. For he had been a victorious and famous King for many years, while the Sultan had been the servant of a vile servant and had of late become so powerful that everyone feared him. The King several times sent his noblemen to the Sultan with most handsome presents, so as to win his friendship, but the Sultan persisted in his demand for the abovementioned -place. Assembling his army, the Sultan came as far as Aleppo, appointed as supreme commander of his army one of his noblemen called Smlmot', and as his seconds, Alfi and the Sultan of Aleppo, and he sent them to Cilicia against the army of King Het'um, and he himself remained where he was.76They arrived at the place called Nicopolis, at the foot of the Black Mountain, and encamped there. The soldiers of King Het'um were devided into three groups; one was with the King who had gone to ask the Nation of the Archers to help his army; the second was at the place called the Gate [Portella], and the third had been sent against the infidels at the place called Mari, on Monday, October 23, and had encamped there.77 74 Other sources date the capture of Arsuf in February-April 1265: Cahen, op. cit., p. 714. The Chronicle ascribed to Het'um I1 (Akopian, op. cit., I, p. 83) mentions the capture of Arsuf from the Hospitallers in 1264, and the loss of the forts of Safed and Arassous by the Templars in 1266. Het'um (Hayton) of Gorigos gives the correct dates: Arsuf was captured in 1265 and Safed on July 22, 1266 (ibid.,11, pp. 72-3). 75 Sheh is the locality known as Chih al-Hadid or Shih al-Hadid (Alishan, op. cit., p. 512; Cahen, op. cit., p. 137). Grigor of Agner reports that after the disaster suffered by the Cilician army, the nobles reproached the King for not ceding this locality. They said, "What use is Sheh to us, that for the sake of Sheh you lost your sons?" (History of the Nation of Archers, p. 364 [g6]). The geographic name has been misunderstood in the English translation and interpreted as "inhabited locality," as if it were derived from the word "shen" (ibid.,p. 391, note 65). Smpad's text, already quoted by Alishan, leaves no doubt that Sheh designates the border town of Shih al-Hadid. 76 Smlmot' = Izz eddin Igan surnamed Semm el-maut (the deadly poison), who died in 1276-1277 (cf. Makrizi, op. cit., I. 2, p. 146). He is also mentioned by Vardan (Venice, 1862, pp. 163-4) and in the Chronicle of Het'um I1 (op. cit., p. 83; see, also RHC, Doc. arm., I, p. 487). Abu'l Feda says that Baibars placed his army under the command of El-Malek el-Mansur, Prince of Hamah (RHC, Doc. or., I, p. 157). d'Ohsson (op. cit., 111, p. 421) referring to Kuwa'ri, mentions both Semm el-maut (Yzz-ad-din Aigan) and the Prince of Hamah, and adds the name of the Mameluk emir Qalawun (Seif ud-din-Calavoun). 77 Mari = Phylae Amanides, the passage from Syria to Cilicia (RHC, Doc. arm., I, pp. XXVIXXVIII and 827), is also mentioned in the History of the Nation of the Archers, p. 357 (89).

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE SMPAD 165 Early on Tuesday morning the infidels arrived at the place where the Christian army stood, and for a short while they remained facing one another; then, through the wiles of Satan, the Christian soldiers turned to flight without fighting, but the sons of the King, Leon, Baron of the Armenians, and his brother T'oros, left the soldiers, turned back, and advanced to fight against the infidels. T'oros was killed during this battle, while the Baron Leon was taken prisoner together with Vasil the Tatar, the son of Smpad, general-in-chief of the Armenians, as well as other men among whom were a certain Djilard [Gerard?] and Adom." Pp. 247-50, A.D. 1266-1268. The author gives a fairly detailed description of the destruction of Sis and of the terrible slaughter of the population; he then relates the protracted transactions between the King and the Sultan for the release of the Baron Leon and the other prisoners. In return the Sultan asks the Armenians to obtain the release of his favorite, Sonqor al-Ashkar, whom the Mongols held captive.78Het'um himself goes to the court of Abaga in 1267, but is told that the prisoner cannot be found. The following year Leon, the son of the Constable Smpad, is sent to the Mongols, succeeds in finding Sonqor al-Ashkar, and returns with him to Sis. Messengers bring the news to the Sultan.

P. 250, A.D. 1268. "Then the Sultan set forth with all his soldiers and marched on Tripoli, and caused great damage to the city. From there he marched for five days without stopping day or night, and suddenly descended on the famous city of Antioch; he took it after four days, on Saturday, May 6.79No one can describe the carnage that took place, or the number of those who were taken captive, or the amount of treasure carried away to Egypt. But few Armenians were killed, for those who were from Cilicia were allowed by the Sultan to return thither. The commander of Antioch also saved himself with his family, and came to Cilicia; it was said that it was through his advice that the city had been betrayed into the hands of the Sultan, but only the Lord knows the 78 The imprisonment and release of Leon are related in great detail in the History of the Nation of the Archers, pp. 356-70 [88-1021. 79 The exact date of the fall of Antioch differs in our various sources; the most reliable document is the letter written by Baibars to Bohemond of Tripoli (translated in Michaud-Reinaud, BibliothBque des Croisades, IV [Paris, 18291, pp. 507-11). Baibars says that he left Tripoli on a Wednesday, Shaban 24; he arrived before Antioch at the beginning of the great Ramadan, and captured the city on Saturday, Ramadan 4. These dates correspond to Wednesday, May g and Saturday, May 18. (H. G. Cattenoz, Tables de concordance des Bres chre'tienne et he'girienne [Rabat, 19541). Smpad is correct in stating that the city was talren on a Saturday, but the date-May 6-is wrong, not only because it does not agree with Baibar's letter and with Makrizi, who gives the same dates (ed. Quatrembre, I, 2, pp. 52-53)? but because May 6 was a Friday not a Saturday. To the sources listed by Runciman, op. cit., p. 326, and Cahen, op. cit., p. 716, we can now add the exact information given by Het'um (Hayton) of Gorigos: Antioch fell on May 17, 1268 (Alropian, OF. cit., 11, p. 74). so The Commander of Antioch was the Constable Simon Mansel. He was captured at the head of a small force, but was allowed to return to negotiate for the capitulation of the city (Michaud-Reinaud, op. cit., pp. 508-9; Makrizi, op. cit., p. 52, note 63).

166

SIRARPIE DER NERSESSIAN

Pp. 250-1 (p. 652), A.D. 1268. Final arrangements for the release of Leon who returns to Cilicia. Appointment of the Catholicos Hagop on February 12, 1268. Pp. 251-2, A.D. 1269. Severe earthquake. Leon, Baron of the Armenians, goes to visit Khan Abaga and returns with many presents.81 Smpad's son Het'um dies on July I5 and his other son Vasil on September 29. King Het'um dies on Tuesday, October 29, at s u n ~ e t . ~ 2 P. 252, A.D. 1271. Leon is anointed King on January 6 in the church of Saint Sophia at Tarsus in the presence of a large assembly.

P. 253. "This same year the Sultan of Egypt Baibars Bunduktari set out again to invade Cilicia. And King Leon sent ambassadors to him and made him return to Egypt, while he himself went east, to the Khan Abaga, and the Khan offered him 20,000 men to take back with him for the protection of his land, and he [the Khan] promised to come a few months later. And King Leon, taking a few of these men, returned to his country. This same year a Frank King called King Edward [Eredouard] came by boat with 2,000 men to Acre and waited there for his companions, the other kings." P. 253-4. In October of the same year, a son is born to Leon. A.D. 1272. Heavy snowfall on January 6th. Death of Giiragos vartabed. A Muslim woman called Maria plots to poison the King and is apprehended. Leon builds the fortress called Gadariats at the foot of the Taurus mountain. Leon's son is baptized in the presence of the Syrian Catholicos Ignatius, and called T'oros after his uncle. P. 254. "This same year [ I Z ~one ] of the servants of King Edward, who had crossed the seas and had come to Acre, one day entered the King's chamber as he was seated alone, clad in a thin garment, and made all the servants go out, and, coming close to the King as if he had a secret to tell, he drew his knife and stabbed him in the chest. The King lept up and put out his right hand to the sword, but the man stabbed him again. . . ."83 8 1 Bar Hebraeus, op. cit., p. 448 dates Leon's visit in the month of July. For other sources see N. Akinian in Handes Amsorya, XLII (1948), pp. 269-79. 82 The correct date is 1270. Several contemporary notices in Armenian manuscripts record the King's death on Tuesday, October 29, 1270 (see Akinian, op. cit., XLVI [1g52], pp. 467-72). Makrizi, however, states that Het'um died in A.H. 669, on the 21st day of Rebi I ; this corresponds to Kovember 7, 1270 (op. cit., p. 84). Quatrembre adds (Makrizi, op. cit., note 104) that according to Nova'iri Baibars received a letter from Leon on the 27th day of Rebi I saying that his father Het'um had died on the 28th of Teschrin I, which corresponds to the 21st of Rebi I. 83 The details given by Smpad, namely, that Edward was dressed in a thin garment, agree with the account of the Templier de T y r , Gestes des Chiprois, ed. G. Reynaud, [Geneva, 18871, p. 201), where it is said that Edward was "vestu soulement en chemise et braie." See also R. Rohricht, "La Croisade du Prince Edouard dlAngleterre," Archives de Z'Orient latin, I (Paris, 1881), pp. 625 -7; MichaudReinaud, Bibliothbque des Croisades, IV, p. 530. Het'um (Hayton) of Gorigos records in his Chronicle that on May 18, 1272 one of the Assassins attempted to kill PrinceEdward, and that on September 13th Edward sailed back (Akopian, op. cit., 11, p. 76).

A R M E N I A N C H R O N I C L E O F T H E CONSTABLE S M P A D 167 The narrative stops abrupty at this point. One folio of the last quire is missing and there was, most probably, an additional quire that has disappeared. The Etchmiadzin manuscripts have an entry for the year 1274, mentioning Baibar's raid as far as Gorigos. Smpad's original text may have continued until 1274 or, at the latest, 1275, since he died in 1276. The excerpts translated above give a sufficient idea of the significance of the new edition and of the importance of Smpad's Chronicle. Instead of brief notations, with numerous gaps, we now have a continous and fairly detailed account, the work of a cultured man interested not only in the history of his fatherland, but also in that of the neighboring countries. As Constable of the kingdom and brother of the King he had access to the royal archives and other official documents, and his knowledge of French enabled him to consult contemporary Western sources. Smpad's broad interests, his use of earlier Armenian writings as well as of Western documents were already evinced by his other works. In 1265 he revised the Law Book, composed in Armenia in 1184 by Mkhit'ar Gosh, which he summarized, adapted to the usages of his own time, and rewrote in mediaeval Before Armenian so as to make it more intelligible to his ~ontemporaries.8~ this he had translated parts of the Assizes of Antioch for use by the law courts of Cilicia where many usages and customs of the feudal West had been adopted. The original French text is now lost and this mediaeval Code is known to us only through this Armenian translation, which was done with the greatest care from a reliable In his preface Smpad informs us that the model was sent to him by his close relative Simon, Constable of Antioch, who, in turn, had received it from his father, the late Constable Sire Mansel. "When I completed the translation," adds Smpad, "I sent it again to the Court of Antioch, so that they might confront it [with the original]; and they confirmed in writing and by their testimony that it is correct and agrees [with the original] word for word."86 Smpad was also interested in ancient Greek and Byzantine philosophical and theological writings which had been translated into Armenian. A brief note written by him, and preserved in a copy of later date, informs us that he had summarized (or written a commentary on) the Dialectics of John Damascene.87He owned the manuscript of the Scholia of Cyril of Alexandria and the works of Dionysius the Areopagite, copied in 1175 by Nerses of Lampron,ss 84 Josef Karst, Annenisches Rechtsbuch, I . Semfiadscher Kodex aus dem 13. Jahrhundert oder mittelarmenisches Rechtsbuch (Strasbourg, 1095). A new edition of this Law Book has recently been published by A. G. Galstian, together with a Russian translation ( S w b a t Sfiarafiet. Sudetnik [Erevan, 19581). According to Galstian, this Law Book is not a translation or a revision of the text of Mkhit'ar Gosh, but an original composition based on earlier sources. The similarities between the two works are to be explained by the use, independently of one another, of the same sources. 86 Alishan, Assises dlAntioche (Venice I 876). Ibid., pp. 2-3. Alishan, Sissouan, p. 74. s8 F. N. Fink and L. Gjandschezian, Verzeichnis der armenischen Handschriften der koniglichen Universitatsbibliothek z u Tiibingen (Tiibingen, 1go7), p. I I I.

168

SIRARPIE DER NERSESSIAN

and he commissioned a copy of the Categories of Aristotle and the Treatise on the Nature of Man attributed to Gregory of Nyssa, to which he added a fairly long colophon in verse.89Two other manuscripts of his personal library have survived: a Lectionary to which he again wrote a long versified colophon,gO and a Gospel book illustrated with full-page m i n i a t ~ r e s . ~ l Catholicos Garegin I Hovsep'ian, Colofihons of Manuscripts (in Armenian) (Antilias, 1951)) cols. 951-6. The manuscript, begun in 1239, was completed in 1244. Vatican, Borgianus armenus 61, A.D. 1268: E. Tisserant, Codices arnzefzi Bibliothecae Vaticanae (Rome, 1gz7), pp. 91, 102. Alishan, Sissouan, p. 74. 91 Erevan, Matenadara?~,no. 21117644: G. Hovsep'ian, op. cit., cols. 957-8. For the inscription on one of the churches erected by Smpad, see Alishan, op. cit., pp. 75-6.

Cornuti: A Teutonic Contingent in the Service of Constantine the Great and Its Decisive Role in the Battle at the Milvian Bridge. With a Discussion of Bronze Statuettes of Constantine the Great Andrew Alföldi; Marvin C. Ross Dumbarton Oaks Papers, Vol. 13. (1959), pp. 169+171-183. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C169%3ACATCIT%3E2.0.CO%3B2-2 Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:37:05 2008

CORNUTI:

A T E U T O N I C C O N T I N G E N T I N T H E SERVICE

O F CONSTANTINE T H E G R E A T

AND ITS DECISIVE ROLE I N T H E BATTLE

AT T H E MILVIAN BRIDGE

With a Discussion of

BRONZE STATUETTES O F CONSTANTINE

THE GREAT

by

MARVIN C. ROSS

H E Art Museum of Princeton University has recently acquired a bronze statuette of an early Byzantine Emperor (figs. I, 2) which, because of its unusual historical importance, deserves a fuller discussion than is usually devoted to objects of such small artistic merit.1 I t is a seated figure, only 0.125 m. high, crudely cast in bronze. A survey, by Mr. Marvin Ross, of all the known specimens of this class of object will be found directly following this paper; thus we need not here go into technical details. I t will suffice to note that the Princeton statuette appeared for the first time in the famous it was mentioned by Frieda Schottmiiller,3 Gr6au c ~ l l e c t i o n .Subsequently ~ who identified it correctly, as we shall see, as the Emperor Constantine I , whereas F. 0. Waag6,4 the next scholar to discuss it, left its attribution to a specific ruler open. This is all that has been written on this object save for the recent paper by E. Schaffran5(see postscript, p. 179). The form of the monogram of Christ on the shield (fig. 2) cannot be earlier than the period of ~heodosiusI , and the globe surmounted by a cross (which is now broken off, though its base is still visible) gives the same terminus Post quem. The exaggerated round eyes remind us of the Carmagnola head and other related imperial portraits; so the actual date of the object must be sought in the fifth or even in the sixth century. The crudeness of the cast and the deformation of the figure by incompetent copying prevent us from using the criteria of stylistic analysis. ~ ; t here are-other features which point to a much earlier date for the prototype of our statuette. The bejewelled diadem and the rectangular precious stone in its center go back to Constantine who is the earliest possible candidate for this portrait. c he Jupiter-like costume with the naked torso and the mantle (draped over the left shoulder and covering the legs down to the ankles) is, on the other hand, hardly conceivable for any Christian emperor after Constantine. Even in his case, I know of only one such representation in the guise of the supreme god, a parallel which is particularly striking because it is found on an official portrait, circulated after the victory over Licinius, when the profession of the Christian creed stood in the foreground of court propaganda. This is a bronze medallion struck in Rome, the best specimen of which, in the Museo Archeologico of

9Rp

For the photographs of this object and the permission to publish it, my best thanks are due to Professor Ernest De Wald and Miss Frances Jones of Princeton. The Constantine medallion in Florence is reproduced by the courtesy of Professor G. Caputo; it was selected from a large series of photographs taken in 1957 and 1958 in Italian museums, with the help of the American Philosophical Society of Philadelphia. The section on primitive secret societies is part of my research sponsored by the WennerGren Foundation of New York. Other acknowledgements will be found in the notes below, but I wish to record here my indebtedness to Marvin C. Ross for his valuable collaboration. Collection J. GrCau, Catalogue des bronzes antiques et des objets d'art d u moyen-&ge et de la renaissance (Paris, HBtel Drouot, June 1-9,1885), p. 64, no. 319. Frieda Schottmiiller, "Bronzestatuetten und Gerate," Bibliothek der Kunst- und Antiquitatensammler, 12 (1918), p. 39, fig. 23. F. 0. Waag6, A J A , 39 (1g35), p. 81 and fig. 2. ti E. Schaffran in Rivista d i Archeologia cristiana, 32 (1956; published in 1958), pp. 243-249. This paper was brought to my attention by Professor E. Kitzinger.

A N D R E W ALF(IjLD1 Florence, is reproduced in figure 3. Here, as I remarked elsewhere,6 Constantine as the Supreme God, the deus praesens of the heathen, receives sovereignty over the Orient from his eldest son who, following the example of Alexander the Great, had conquered this region like a new Dionysos. The panther, sacred animal of Dionysos, makes its proskynesis before the new Jupiter. Thus our statuette must be an image of Constantine. If any doubt should remain on this point, our discussion of
The shield on which Constantine's left hand rests shows two incised emblems: a chrismon of the cruciform type and below it a crescent-shaped device with an oblique hatching suggesting the texture of a fleece, and terminating in two confronted goats' heads. Almost a quarter of a century ago, I discovered8 the same shield-badge (figs. 5 , 6) on a relief of one of the pedestals of the Arch of Constantine in Rome,g and I explained it as the emblem of a unit of the Late Roman army called the cornuti, the "horned ones," i.e. the he-goats. The alternative suggestion, proposed by F. Altheim,l0 may be discarded. I t needs no discussion here, because our explanation has proved to be correctl1 and has been confirmed by the new publication and elaborate study of the Arch of Constantine by H. P. L'Orange, who has demonstrated the role of the cornuti in the famous battle near the pons Milvius on September 28, 312. A. Alfoldi, JRS, 37 (1947), p. 15.

Cf. A. Alfoldi, "L'evoluzione spirituale del principato," in Corvina 3, series I (1952), p. 32 seq.

A. Alfoldi, "Ein spatromisches Schildzeichen keltischer oder germanischer Herkunft," Germania, I9 (1935), PP. 324-328. Ibid., pl. 45. Cf. H. P. L'Orange and A. v. Gerkan, Der spatantike Bildschrnuck des Konstantinsbogens, Studien zur spatantiken Kunstgeschichte, 10, (Berlin, 1939)~PIS. 28 and 32 (hereafter quoted as L'Orange). Cf. the reproductions of the drawings of the Notitia dignitaturn with similar shield emblems on pl. 46 of my paper quoted in note 8. lo F. Altheim in his "Runen als Schildzeichen," Klio, 31 (1938),p. 55 seq., and in his Literatur und Gesellschaft im ausgehenden Altertum,~(1948)~p. 231 seq. derives such emblems from the runes. He attaches no importance to the goat-heads on the shields of the cornuti on the Arch of Constantine, although the drawings of the Notitia clearly show similar shield-emblems. Now the Princeton statuettegives us definite proof that my explanation of the carvings on the Arch had been correct. The wolf-heads of the shield-emblem on Valentinian's silver patera at Geneva (R. Delbrueck, Spatantike Kaiserportrats, [Berlin, 19331, pl. 79 and p. 179 seq.) show that other military units had paired animal heads on similar crescent-shaped devices. The horned helmets discussed by L'Orange, and again by me in this paper, have not been noticed by Altheim; they explain the peculiar shape of these emblems which R. Egger ("Der Grabstein von Cekantevo," Schriften der Balkankommission, Antiquarische Abteilung, vol. 11, 2 [Vienna, 19501, p. 16) interprets as the crescent of the moon. l1 H. P. L'Orange, pp. 43 and 124. 8

CORNUTI

L'Orange contends that the officer with the shield of the cornuti also has horns on his helmet. I think, rather, that the two curved objects emerging from the crown of the helmet are feathers. But much more important than this detail is that L'Orange has been able to show the presence of warriors with horned helmets beside soldiers in regular uniform in the battle scenes of the Arch12 (fig. 4). In the attack on Verona these cornuti appear as spearmen with the battling Emperor himself. Their commander wears the garb of a Roman officer.13 Far ahead of the main body of the attacking soldiers another high ranking officer with a horned helmet is advancing toward the enemy.14 A member of the cornuti is also conspicuous in the battle scene by the river, in this instance fighting with his sword.15 In the relief that shows Constantine accompanied by the goddesses Roma and Victoria, with the god of the River Tiber gazing at them, a cornutus immediately follows the divine escort of the ruler.16 The accomplishments of this regiment could not have been more highly exalted. In the representation of the victorious entry of Constantine into the City, the "horned ones" are also present.17 They are marching with the "regulars" and the Moorish spear-throwers.18 The glorification of the cornuti is, indeed, even greater than L'Orange had thought. Here I must register some slight disagreement with his views because I can no longer follow his suggestion,lgbased on my previous paper, that the horned helmet was not the exclusive emblem of one regiment. In late imperial times Roman helmets did not have horns. And although we know from the Notitia dignitaturn that several variant forms of the shield-emblem of the cornuti were used by other military units, there is only one such emblem on the Arch of Constantine, namely that of the cornuti. Furthermore, no other unit of the victor's army was singled out in this fashion by the sculptors of the Arch. L'Orange's contention20that the presence of two Rhine legions, the I Minervia and the XXX Ulfiia, was also indicated by the animal badges on the helmets and shields of their soldiers cannot be sustained. The symbol of Neptune, known to have been the emblem of the legio X I Claudia and of the XXX Ulfiia, decorates the hexagonal shields of a tropaion; so these belong to the vanquished and not to the victorious army. Moreover, the trident with two dolphins was so popular an ornament that it may have been used here without a precise connotation. L'Orange's observation21 that the officer's helmet on pedestalrelief no. 6 has a ram's head, the badge of the I Minervia, deserves more serious consideration, but I do not find it convincing. If the ram's head had been meant to record the achievements of this legion and had not been merely an Ibid., p. 43. Ibid., p. 63. l4 Ibid. l5 Ibid., p. 68 seq. ' 6 Ibid., p. 66 seq. l7 Ibzd., PP. 72-73. l8 Ibid., p. 76 seq. l9 Ibid., p. 43. Ibid., pp. 43 and la

l3

a Ibid., p. 117.

I 10.

174

ANDREW ALFOLDI

ornamental motif, the heroic deeds of the I Minervia would have been represented in the battle-scene reliefs as were those of the cornuti. We may, therefore, conclude that the heroes of the great battle were the Moorish cavalry and the cornuti alone. The dramatic turning point in the reign of Constantine was inseparably connected with them. The Emperor and the Senate gratefully acknowledged this on the Arch, and the early Byzantine statuette with the cornuti shield asserts it unmistakably once more. Ammianus attests that the Teutonic war cry, the barditus,22 was introduced think into the Roman army of the fourth century by the valorous c o r n ~ t iI. ~ ~ we can go a step further. The adoption by the Roman army of the barbarian shield-emblem of the cornuti seems to stem from Constantine's gratitude towards these all-important auxiliaries: he not only allowed them to use the barditus as a battle cry, and, as we shall see, to keep their barbarian shielddesign, helmet, and standard, but he used these same emblems for the other Teutonic formations incorporated into the cadre of his new army and recruited chiefly from areas beyond the frontiers of the Empire.

We can gather more details about the cornuti in the century following Con~tantine.2~ They were originally an infantry unit, as the Arch of Constantine clearly dernon~trates.2~ and Ammianus Marcellinus also knows them as in the Notitia dignitatam they appear as aaxilia palatina, i.e. among the elite of the infantry regiments. Like so many other Constantinian formations in the West, they were split into the cornuti seniores (occ. V [14] 158 = VII g) and the cornuti iuniores (occ. V [24] 169 = VII 18), while in the East they appear undivided, together with the bracchiati, another contigent of barbarian origin which was coupled with them as an operational unit, perhaps already under the first Christian emperor (or. VI [g] 50). This multiplication of the corvtuti units and their distribution between East and West reflect the organizational measures of Valentinian and Theodosius I, a subject that does not concern us here.27The cornuti were further expanded by the introduction of a mounted detachment of the same name which has the rank of vexillationes palatinae in the Notitia. As in the case of the infantry cornuti, there was a division into seniores and iuniores in the West, while in the East they remained undivided, and were again combined with the bracchiati.28 Tacitus, Germania, 3. Ammianus, XVI, 12, 43. Mommsen, Gesammelte Schriften, 6, 240. 24 For t h e relevant facts see R. Grosse, Romische Militargeschichte von Gallienus bis zum Beginn der byzantinischen Themenverfassung (Berlin, 1920), p. 38 seq. 25 L'Orange, p. 42. 26 Ammianus, XXXI, 8, 9. R. Grosse, of. cit., p. 39. S7 The detailed history of these regiments will soon be published in a comprehensive survey of the late Roman and early Byzantine army by my pupil Dietrich Hoffmann of Basel. 28 Equites cornuti seniores, occ. VI [6], 48 = V I I 162 (or 168); numerus cornutorum seniorum, CIL, VI, 32963. 28 ; Equites cornuti, Not. or. VI [g], 50; Equites bracchiati seniores, occ. VI [4], 45 = VII, 161; Equites bracchiati iuniores, occ. VI [46] = VII 170; Equites bracchiati, or. IV, 29. Cf. CIL, V, 8740. 22 s3

8760.

CORNUTI

175

As R. G r o ~ s ehas ~ ~pointed out, this new type of infantry, unknown before 312,included in the fourth century a certain percentage of provincial Roman elements along with the ever-growing numbers of Germans from beyond the Limites: mixti cum arctois Germanis Galli, as Ammianus30 describes them. I n the reign of Julian, Valentinianus, the future Emperor, a native of Pannonia, was commander of a batallion of the c o r n ~ t iBut . ~ ~even before that the commander of the cornuti had been a Teutonic officer of renown, B a i n ~ b a u d e s . ~ ~ In spite of such fluctuations, our remarks concerning the shield-emblems, name,33 and battle cry of the cornuti clearly indicate that these were free Teutonic warriors, at least when they first appeared in the battle of the Milvian Bridge. Their constant partners in warfare,34 the bracchiati, who shared their Teutonic barditus, had the same origin. Their name derived from the barbarian armrings which they ~ o r e ~ ~ - c e r t a i nwith l y their national costume and weapons rather than with the Roman military uniform. We are, therefore, entitled to seek analogies for the organization of the "horned warriors" among the social and military institutions of the free Teutons. The curious shield emblem of the cornuti, which also appears in finds made in German territory not subject to the Romans, may give us the key to this complex problem.

Excavations in Teutonic territories have brought to light half a dozen representations of warriors wearing on their helmets the same device with confronted animal heads as the cornuti. These are bronze ornamental reliefs of about A.D. 600 which seem to have been disseminated from Scandinavian workshops to what is now Germany.36I t is sufficient for our present purpose to recall the two most characteristic examples: the bronze core for repousse work from Torslunda, Sweden3' (fig. 11) and the fragment of a decorative pattern on the royal helmet of Sutton Hoo (fig. 7).38The occurrence of animal Op. cit., p. 39.

Ammianus, XXV, 6, 13.

31 Chron. Pasch. (54g6 Dindorf): Oircrhav~1~1uvb5 ~ptpoGvos -r6-re i j v -r&y~crros Kopvohwv o h w

Asyo~dvqvow~8pc;Cf. Theophan. a. 5855 ( 5 1 de ~ Boor). 3a Ammianus, XVI, 12, 63. 33 Ammianus, XX, 4, 13, says that the petulantes were Gauls. This probably refers to their country of origin, but does not imply that they were Roman citizens. 34 Ammianus, XVI, 12, 43. 35 Lydus, De mag. I, 46: p p a x 6 ~ 0 j1- r o ~CrpylAAiyspo~cysA~oqbpo~. 36 J. Werner, Acta Archaeol., 21 (1g50),p. 79. These have recently been discussed by R. L. S. BruceMitford, "The Sutton Hoo Ship Burial: Recent Theories," Proc. Sugolk Institute of Archaeology and Natural History, 25 (1949)~p. 50 seq. and pl. 10. Cf. Bruce-Mitford, in R. H. Hodgkin, A History of the Anglo-Saxons, 3rd ed., 2 (Oxford, 1g52), p. 720 and pl. 86. K. Hauck, "Herrschaftszeichen eines Wodanistischen IConigtums," Jahrbuch fur frankische Landesforschung, 14 (1g54), p. g seq. (hereafter quoted as Hauck). 37 Often reproduced, e.g. in 0. Almgren, Hallristningar och kultbruk ( K u n g l , Vitterhets Historie och Antikvitets Akaderniens Handlingar, 25 [Stockholm, 19271'1, figs. 7 and 45a. 0. Montelius, Kulturgeschichte Schwedens (Leipzig, 1906), figs. 167 and 217. Statens Historiska M u s e u m : Tiotusen By i Sverige (Stoclrholm, 1g45), fig. 185. 38 H. Maryon, in Antiquity, 21 (1g47), p. 139. Bruce-Mitford, op. cit., pl. go. Hauck, p. 40, fig. 6. Our figure 7 incorporates the results of the latest restoration work done at the British Museum, kindly communicated to me by Mr. R. L. S. Bruce-Mitford. 29

30

176

ANDREW ALFOLDI

heads on helmet horns can be traced back to the La Tkne culture of the Celts. The pair of horns on the Celtic "Torrs Chamfrein" in Edinburgh (fig. g) with duck-like heads on their extremities may be associated with this group. The "chamfrein" has recently been analyzed by R. J. C. Atkinson and Stuart Pigg0tt,3~who conclude that the horns were made during the earliest phase of the Celtic occupation of Britain and did not originally belong to the bronze head-piece to which they were attached in modern times. In the opinion of these scholars the horns were either mounted on a helmet or were the terminals of two long drinking horns. The reconstruction of the second alternative, which the authors consider more likely than the first (pl. 85 of their study),is not, however, very convincing. Such a curved finial would have been awkward for the drinker and, to my mind, inconsistent with Celtic style.40The supposition that this pair of horns belonged to a helmet is more plausible. Such helmets are known from the La Tkne culture; but instead of adducing the well-known horned helmets from Gaul, it will be sufficient to point to the famous silver cauldron from Gundestrup in Copenhagen. In the cavalcade represented on this object the first horseman has a bird on his helmet, the next a boar, the third stag horns, and the last goat horns (fig. 10). The helmets surmounted by boars on another Torslunda plaque (fig. 12) may be of the same Celtic origin, and recall the variety of zoomorphic helmets worn by the Gundestrup riders. These Celtic parallels may show the provenance of the crescent-shaped horns of the cornuti. In the late Scandinavian Bronze Age (fig. 8)41we already find horned helmets provided with eyes and nose, evidently an imitation in metal of animal masks worn by warriors in prehistoric times.42 I should like to add a few words about the possible meaning of these primeval thinks that the Torslunda plaque (fig. 11) represents practices. K. Wodan's fight with the Fenris wolf. This ingenious guess is not, however, convincing. One eye of the cornutus is admittedly missing, but this does not necessarily make him lliodan the one-eyed; and since on the Sutton Hoo helmet (fig. 7) there are two such figures, they cannot both stand for liodan. I t is important to note that the Sutton Hoo cornuti are not a mechanical reduplication, but form an integrated couple. Nevertheless, Hauck is surely right in postulating a mythical prototype for these scenes, the exact definition of which must be left to competent scholars. Such a mythical prototype need not have refer red to a prominent divinity, ut could simply have represented the legendary ancestor of a tribe. This may be confirmed by the fact, stressed by Hauck, that the cornutus of the helmet of Old Uppsala was a member of the Swedish royal house, and that the tombs of Valsgarde on which the cornuti once more occur, belonged to some important clan of the nobility.@ Since the 39 Archaeologia, 96 (1955), pp. 197-235. 40 For the normal shapes of drinking horns, cf. Archaeologia, 96 (1955)) pls. 64-69. 41 Cf. R. Djupedal-H. C. Broholm, Aarb. f. Nord. Oldk. och Hist. (1g52), p. 5 seq. R. Halbert, Arsberattelse Lund (1954/55), p. 221, fig. 3. 42 Cf. e.g. K. hleuli, Schweizer Masken (Ziirich, 1948)) p. 104. W. Krause, Gott. gel. Nachr. (1948), p. I O I seq. 43 Hauck, pp. 40 seq. and 46 seq. 44

Ibid., p. 45.

CORNUTI

pair of cornuti on the Sutton Hoo helmet are derived from the same Swedish iconographical repertoire, it is possible that this helmet was not a looted object, but the property of a prominent member of the ruling clan of East Anglia, which was of Scandinavian origin. The mythical prototype of the cornuti in the service of Constantine the Great must be sought in the same milieu that produced the Scandinavian representations of cornuti. I t has, I believe, been overlooked that the horns on the helmets of the cornuti on the Arch of Constantine (fig. 4) are different from the usual type of prehistoric helmet, on which the horns are placed over the temples. In this case the horns, instead of being laterally disposed, grow from the center of the rim above the forehead of the warrior. This is because they are goats' horns. Thus the cornuti of the Roman army had goat helmets. Their shield emblems are a secondary feature, perhaps only the Roman adaptation of a symbol that had originated as a standard or as a headgear. Since in the social and religious life of primitive nations both the standard and the headgear denoted the sacred origins of a tribe or a clan, it is not surprising that the cornuti should have identified themselves by such meaningful symbols.45 A curious feature of the Scandinavian cornuti is that they are shown dancing. This is not a display of the sword dance, the chief amusement of Teutonic tribes.46 On the Torslunda plaque the warrior with the wolf's mask (fig. 11)) who draws his sword for a fatal duel with the cornutus, is not dancing like his opponent, and this contrast suggests that the latter is performing the tripudium, the wild dance that preceded battle and produced the desired emotional frenzy, wzixta fietulantiae rabies.47This wild explosion of brutal instincts was noted by the Romans among all the barbarians of the West; the Alpine tribes,48 the S p a n i a r d ~ and , ~ ~ the Thracians50 practiced it, as well as the Celts and the the tripudiztm was not a chaotic outbreak, but a wellT e ~ t o n s Sometimes .~~ disciplined collective performance, like the war dance of the Ambrones before the battle of Aquae S e ~ t i a e . ~ ~ The comical appearance of such a dance was, of course, completely lacking in the early stages of cultural evolution. The cult-dance of the he-goats was a highly serious act among the early Greeks.53 Horned or animal-shaped 45 Cf. also ibid., p. 34. 4 T a c i t u s ,Gevmania, 24, Cf. K. Meschke, Schwerttanz und Schwerttanzspiel im germanischen Kulturhveis (Leipzig and Gerlin, 1g31), p. 162.Hauck, p. 42 seq. 47 Sax0 Gramm. V I, ii. G. DumCzil, Mythes et dieux des Gevmains (Paris, 1g3g), p. 87 seq. 48 Liv. XXI, 42, 3. 49 Liv. XXIII, 26, 9. Tac. Annales, IV, 27, 4. 5l Tac. Hist., j, 17. 52 Plut. Mar. 19, 4 (vol. 3, 258 Ziegl.): o i r ~ &T&TOIS 0388 p a ~ l d . 1 8 ~~ ~E1P ~ ~ E V 8p6po1~, O I 068' &vapepov

"

&AaAaypbv ~ ~ V T E S&Aha , K ~ O ~ O V T EP&pq S T& bnha ~ a oiw v a h h 6 ~ ~ ~T~~ 0 V T1E,S &pa T ~ V&ijv E T ~ ~ Y Y O V T O vohA&IS ~pooqyopiav,'Ayppwva~, K T ~ . 53 I have in mind the famous passage of Herodotus, j, 67: T& TE 83 &Aha 01 ZIKV~VIOI d ~ i p a vT ~ V 'A8pqmov ~ a 83 i ~ p b ~h s r & ~ aa h 0 0 ~ p a y ~ ~ oxopoicr~ icn dykpa~pov,~ b pkv v A16vvoov 06 T I ~ ~ V T E S , T ~ V 8kXA8pqmov. Cf. K. Ziegler, RE, 6 A, cols. 1899 seq., esp. col. 1917.Fr. Brommer, Satyroi (Wiirzburg, 19371, p p 6 seq., 10 seq., 13. On Dionysos as a goat, see M. V. de Visser, Die nicht menschengestaltigen G6tter der Griechen (Leiden, 1903)) pp. 47, 208 seq. M. P. Nilsson, Geschichte der griechischen Religion, I (PvIunich, 1941), p. 218. Idem, Opuscula selecta, I (1951))p. 61 seq. K. Th. Preuss, in Vortrage Bibl. Warburg (1927/28), p. I seq. E. Maass, Gott. gel. Anz. (188g), p. 803 seq. (phyle of t h e aiyl~opais). I2

178

ANDREW ALFOLDI

helmets suggest that even the Greeks once used in battle the symbols of their zoomorphic ancestors. Similarly if the aias ("goats") of the Rig-veda were better known, they would reveal themselves, I think, as warriors with goatmasks. However, these few remarks must suffice to show that the Teutonic cornuti did not stand alone in the prehistory of the Indo-Europeans. A dancing human figure with goat horns on his head (discovered on a prehistoric rock carving at Tanum, north of Goteborg) was identified by H. Giintert as the god Thor.54 The ritual dance of the Longobards around the head of a goat that was sacrificed to their divinity55 may also be significant in this connection. 0. Hofler has stated that the "he-goat" names in the Sagas refer to warrior groups, an explanation which he hopes to substantiate in a future publication.56 Survivals of such warrior groups have been discovered by anthropologists in the traditional revenant gatherings of masked village youths around New Years' Day. These youths are called "Haberer" (from the "Habergais")5' and have a counterpart in a youth organization in Zurich named "he-goats" and in other similar groups.58 This is the background from which Constantine's cornuti emerged, a warrior group of the kind we find even today among the savages of many regions. The addition to the Roman army of such a contingent, untouched by Roman civilization but, after the victory over Maxentius, exalted above all the legions, gives a vivid illustration of Constantine's revolution. The glorification of the cornuti on Constantine's Arch and on his statues-the Princeton weight is a crude derivative of one of the latter -illustrates the sudden elevation of these barbarian mercenaries in the hierarchy of the Roman army. The bracchiati and other similar contingents soon followed. The Princeton statuette is important not only because it confirms the role played by the cornuti in the battle of the Milvian Bridge; it is also an additional testimonium of Constantine's conversion. In a famous passage of De mortibus ~ersecutorum,Lactantius states that the Emperor, admonished by the divinity The goat-skins of female divinities who were originally leaders of warrior-groups, such as the aigis of Athena or the goat-skin of Juno Sospita, are also pertinent to this problem. Cf. K. KerCnyi, in EranosJahrbuch, 17 (1g4g), p. 65. 54 H. Giintert, Der arische Weltkonig und Heiland (Halle, 1923)~ p. 45 seq. 55 This has been discussed by bliillenhof and many scholars subsequently; cf. e.g., R. Wolfram, Schwerttanz u n d Mannerbund (Kassel, 1938), 186 seq. 5 6 0. Hofler, Kultische Geheimbiinde der Germanen, I (Frankfurt, 1934)~ note 41.

57 R. Wolfram, op. cit., p. 233 seq.

58 E. Hoffmann-Krayer, "Knabenschaften und Volksjustiz in der Schweiz," Schweizerisches Archiv

f u r Volkskunde, 8 (1905)~p. 177. Cf. also P. Thielscher, in Forschungen u n d Fortschritte, 28 (1954)~ p. 374 seq. I n a letter dated August 20, 1949, Prof. 0. Hofler gave me the following additional details: "Professor Burgstaller, Linz, schreibt mir, daf3 das Haberfeldtreiben im Innviertel 'Bockshauteln' heil3t. Das ist gewiB wichtig, denn der Name des Haberfeldtreibens-jener oberbayrischen Form des von kultmaskentragenden 'Haberern' ausgeiibten damonischen Riigegerichtes (das ich im 2 . Band meiner Kultischen Geheimbiinde als eine Funktion typischer Kultmannschaften hoffe erweisen zu konnen)-birgt in dem Glied 'Haber' wohl sicher das mit caper, alt-nord. hafr, identische Wort fiir 'Bock'. Ferner verweist Professor Burgstaller auf L. Siess, Sagen aus dem oberen Miihlviertel, 4 (1898), p. gff., wo eine Felsenhohle, genannt 'Bockstube', erwahnt wird, in der sich nach der Sage menschliche 'Bocke' aus der Ortschaft Neundling bei Altenfelden aufgehalten haben sollen. Habergeissmasken gibt es bei vielen hiesigen Burschenschaften." He also mentions Swedish youth groups with the name "he-goats," descendants of our cornuti.

STATUETTES O F CONSTANTINE T H E GREAT

179

in a dream, ordered the initials of Christ to be painted on the shields of his soldiers. Our bronze figure shows this very monogram on the shield of the cornuti, thus providing unmistakable confirmation of Lactantius' story. I t is very likely that this feature of the shield derives from the monumental prototype of our figure. The cornuti won the battle not by their valor alone, but also because of the protection of the wonder-working sign, the symbol of their Emperor's heavenly patron.

Postscript. This note had been written when I received the latest issue of the a paper by E. Schaffran on the Rivista di archeologia c ~ i s t i a n acontaining ~~ same statuette, which he had seen in a private collection in Vienna. Schaffran has not followed the lead of Robert Zahn, who had attributed a similar statuette from Pergamon to Constantine I.60He has failed to realize that the crudeness of this object is due to its being a derivative work, separated from what must have been a famous prototype by a long series of increasingly debased molds, and he attributes it to one of the Longobard rulers between A.D.583 and 626. This obliges him to consider the Jupiter costume of the Emperor as a toga, possibly with a tight-fitting tunica over the chest. The stylistic degeneration is interpreted as Germanic "rhythm" (p. 247 seq.). Schaffran rightly recognizes the Teutonic character of the shield emblem and considers it an apotropaic symbol coupled with the monogram of Christ (p. 248), but he does not perceive its relationship to the same emblems in the Notitia and on the Arch of Constantine. Consequently, though we must be grateful for the publication, I have nothing to add to my own interpretation.

B R O N Z E STATUETTES O F CONSTANTINE T H E G R E A T

A

GROUP of early Byzantine bronze weights in the shape of a seated emperor began to be published as early as 1885. These little statuettes have: been identified as Constantine the Great by Zahnl and Schottmiiller,2 and as Valentinian by Chittenden and Seltmaq3 while Protasov4 left the identification open. I t has remained for Professor Alfoldi to establish beyond question that these bronze weights derived from an original representing Constantine the Great. 59 Cf. supra, note 5. 6o

R. Zahn, "Zwei neue Schnellwagen im Antiquarium," Amtliche Berichte (Berliner Museum),

xxxv (1913), pp. 1-10.

R.. Zahn, in Amtliche Berichte aus den koniglichen Museen, 35 (1913), p. 9. F. Schottmiiller, Bronze Statuetten und Gerate, 2nd ed. (Berlin, 1921), p. 46, fig. 26. J. Chittenden and Charles Seltman, Greek A r t (London, 1946), no. 380, pl. 105. N. D. Protasov, "Sirijiskie reministsentsii v pamiatnikakh khudozhestvennoj promyshlennosti Khersonesa," Institut arkheologii i iskusstvoznanija, Trudy sektsii arkheologii, I1 (1928), pp. 100-117. 12*

180

MARVIN C. R O S S

Five of these little bronze statuettes are known in the museums of Europe and America. The first one to be published was the specimen now at Princeton which was illustrated in 1885 in the Grkau sale catalogue of bronze^,^ and ascribed to the sixth century without any attempt to identify the subject. Later it passed into the Kieslinger Collection in Vienna, and is said to have been brought to this country by a Viennese refugee. The best known of these statuettes is in the Staatliche Museen in Berlin (fig. 13) and is of particular interest in that it is complete with steelyard and weighing pan, so that one can easily see how such weights were used.6 This weight, said to have been found at Pergamon, was first published by R. Zahn and figured in the Exhibition of Late Antique Art from the Mediterranean Area held in Berlin in 1939.' In 1928 two more examples, excavated in Russia, were published by N. D. Protasov, who pointed out their similarity to the sculpture of the Arch of Constantine, but made no attempt to identify the subject. One of these8 (fig. 14) was found in the village of Perlevka of the Voronezhskaia province and is now in the State Historical Museum in Moscow. The other (fig. IS),' which is of better quality, was discovered at Kherson and is now in the local museum. The fifth of these weights, formerly in the collection of Professor A. B. Cook of Cambridge, England, and shown in the Exhibition of Greek Art held in London in 1946, is now in the Dumbarton Oaks Collection (fig. 16). Although all of these bronzes obviously represent the same emperor and derive from the same original, they may be divided into two groups. In the first group, consisting of the specimens in Berlin, Washington, and Kherson, the emperor holds a globe without a cross, while in the second group represented by the statuettes in Moscow (fig. 14) and Princeton (figs. I , 2) the emperor holds a globe surmounted (or once surmounted) by a cross. The shields, in so far as can be told from their present condition, have various decorations or none at all. This variation indicates that the bronze casters did not understand the significance of the shield-emblem of the Princeton statuette, which in this respect is surely closest to the lost original, as demonstrated by Professor Alfoldi. Curiously enough, India has given us irrefutable proof that small bronze statuettes of the early Byzantine period derived, at least in certain instances, from actual statues. Professor Benjamin Rowland has published a photograph of a bronze statuette of a seated St. Peter, holding a key and blessing with his right hand, that had been found at Charsada in the Northwest Frontier Province of India (fig. 17)."1° This object which, according to Rowland, must 5 Collection J. GrCau. Catalogue des bronzes antiques et des objets d'art d u m o y e n 6gige et de l a renaissance (Paris, H8tel Drouot, June 1-9, 1885), pp. 64 and 65, no. 319. 8 For a discussion of the use of such a scale see D. K. Hill, "When Romans went Shopping," Archaeology, V (March 1955), pp. 51-55. 7 H. Schlunk, K u n s t der Spatantike im Mittelmeerraum (Berlin, 1939), p. 50, pl. 42. Protasov, o$. cit., pl. VIII, figs. 3 and 4. O Ibid., pl. VIII, figs. I and 2 . lo B. Rowland, Jr., "St. Peter in Gandhara. An Early Christian Statuette in India," Gazette des b e a w arts, XXIII (1943), pp. 65-70.

STATUETTES O F CONSTANTINE T H E GREAT

181

have been imported into India in the fifth century, bears a remarkable resemblance to the cult statue of St. Peter that is familiar to all visitors to St. Peter's in Rome and which is generally regarded as a work of the early Byzantine period. This is adequate demonstration that such statuettes were at times copied from major monuments. The same has been suggested for the little bronze figure of a seated saintll found at Trier, although the original from which it was copied is not now known. The above examples support Professor Alfoldi's contention that the Princeton bronze reproduces a larger statue which must have been set up soon after the Battle of the Milvian Bridge in A . D . 312, when the role of the cornuti and the meaning of their "regimental badge" were still known to all. A stylistic comparison of the bronze weights with the carvings of the Arch of Constantine, set up in A.D. 315,'~ seems to confirm this conclusion. I wish to call attention especially to the relief representing the statues of Marcus Aurelius and Hadrian in the Roman Forum13 (fig. 18) and to the now headless figure of Constantine14 (fig. 19). Here we see the same squat posture, the same heavy folds of drapery incised with deep lines, even the same type of shoes as those we find in the bronze weights, although the latter must have lost through repeated re-casting certain traits of the model from which they were copied. In other words, the bronze statuettes appear to reflect the style of imperial statuary that was in official use in Rome ca. A.D. 310. As I have said, the Moscow statuette has a globe surmounted by a cross and the Princeton one also shows traces of what was presumably a cross. The earliest instances of the globus cruciger in imperial coinage belong to the first half of the fifth century (coins of Theodosius 11, Valentinian 111, etc.).15 I t may be argued, therefore, that the statuettes without the cross, namely, those of Berlin, Kherson, and Washington, represent the earlier group, and that the cross was added at a later date, when the globus cruciger became a customary imperial attribute. On the other hand, it is possible, at least theoretically, that the cross may have been a feature of the prototype, and may have been subsequently omitted because it presented difficulty in casting. The Princeton statuette, which by virtue of its shield-emblem appears to be the most authentic, might be claimed to afford an argument in favor of this latter alternative. However, since there is no evidence that the globus cruciger was, in fact, used F. Behn, "Friihchristliche Bronzestatuette aus StraBburg," Germania (1g34),pp. 284-286, pl. 31. See H. P. L'Orange and A. Von Gerkan, Der Spatantike Bildschmuck des Konstantinsbogens (Berlin, 1939). l3 For excellent illustrations, see A. Giuliano, Arco d i Costantino (Milan, ~ g g j )figs. , 34, 40, 42 and 43. l4 Guiliano, op. cit., fig. 44. l5 J. Tolstoy, Monnaies byzantines, I (1912), pl. 5, Theodosius 11; H. Cohen, Description historique des monnaies frappkes sous 1'Empire romain communSment appeleks midailles romaines, VIII (Paris, 1892), Licinia Eudoxia, p. 218 and Eudoxia and Valentinian, pp. 218-219. For personifications with globus crz~ciger,see J. M . C. Toynbee, "Roma and Constantinopolis in Late-Antique Art from 312 to 365," Journal of R o m a n Studies, XXXVII (1g47), pp. 135-144; p. 141 (reverse of coin of Nepotianus with personification of Roma holding a globe surmounted by a Christian symbol, nearly a century before such globes appear in the hands of emperors) ; idem, " R o m a and Constantinopolis in Late-Antique Art from 36j to Justin 11," Studies Presented to David M . Robinson, I1 (St. Louis, 1953), pp. 261277, esp. p. 269 and footnote 85. l1

l2

182

M A R V I N C. ROSS

at so early a period, one may well hesitate to accept it as a feature of the Constantinian prototype. In searching for a monumental prototype, one is naturally led to think of Constantine's statue set up by vote of the Senate immediately after the battle of 312. This famous statue, however, is said by Eusebius to have held a spear or standard in the form of a cross,l6 and it is difficult to imagine that the prototype of our statuettes had such an attribute. The provenance of our statuettes appears in most cases to be eastern. The same may also be said of bronze weights in the form of an emperor's or empress' bust, most of which have been found in the eastern regions of the Empire (Constantinople, Asia Minor, Syria, Cyprus, etc.). This observation does not militate against the Roman origin of the seated-emperor type, since the casts may have been in the first instance conveyed from Rome to Constantinople, and subsequently disseminated from the latter center to the eastern provinces of the Empire. The respect accorded to imperial portraits, whether they were made of precious or base material,17 naturally meant that such weights were regarded as reliable standards bearing the stamp of government authority. The dissemination of these weights from the capital to the provinces may also have served to popularize the official art of the fourth century. The same purpose was, of course, achieved on a larger scale by coins and medallions, but one should not disregard the role played by small articles of mass-produced sculpture. A similar observation has already been made by Dr. C. C. Vermeule with regard to small terracotta reliefs modelled after the zoccoli of Constantine's Arch,18while Dr. Doro Levi has used a small bronze bust, probably of Fausta, in the Arles Museum to account for the "advanced" style of a fourth-century bust, from Chichester (now in the Princeton Art Museum), which is made of local chalk.lg Like the cheap terracotta replicas and small imperial effigies, the weights, too, must have served as carriers of the artistic trends introduced in the fourth century. As for a date when these bronze statuettes were made, there is very little basis for ascribing them to any particular period, since they are of rough workmanship and doubtless were mass produced for commercial purposes. Imperialbusts intended to be used as weights appear to have been particularlypopupp.

Is See H. GrCgoire, "La statue de Constantin et le signe de la croix," L'antiquite'classique, I (1g32), I 35-143, esp. 138ff. ; J. GagC, "La victoire imperiale dans 1'Empire chrCtien," Revue d'histoire et de phi-

p.

losophie religieuse, XI11 (1933), pp. 370-400, esp. p. 385ff.; A. Alfoldi, The Conversion of Constantine and Pagan Rome (Oxford, 1948), p. 42ff. with further bibliography; C. Cecchelli, "La statua di Constantino col 'salutare segno' della croce," Actes du VIe congrks international des e'tudes byzantines, Paris, 1948, I1 (Paris, 1g51), pp. 85-88. B. Stephanides, "Constantin le grand et le culte des empereurs," 'E.rra-rqp~s ' E - r a ~ p s i aBulav-r~vBv ~ X~rou6Bv8 , (1931), pp. 214-226, esp. pp. 217-218 (see criticism by F. Doger in BZ [1g32], p. 441). l7 See K. M. Setton, Christzan Attitude towards the Emperor in th.e Fourth Century (New York, 1941)~ pp. 196ff. Is C. C. Vermeule, "Roman Coins and Sculpture: Six Small Studies of the Reliefs from a Triumphal Arch," Numismatic Circular (July 1g53), pp. 297-300; idem, "Herakles Crowning Himself: New Greek Statuary Types and their Place in Hellenistic and Roman Art," Journal of Hellenic Studies, LXXVII (1957)j P P 283-299, esP. P. 290, note 27. l9 Doro Levi, "A Chalk Bust from Chichester," Record of the Museum of Historic A r t , Princeton University, 111, no. 2 (1g44), pp. 9-19.

S T A T U E T T E S O F CONSTANTINE T H E GREAT

183

lar in the fifth centuryI2Obut continued in use as late as the reign of Phocas (602-610) .21 I t was probably during this period, i.e. from the fifth to the early seventh century that the statuettes of Const antine were made.22

\IASHINGTON, D.C., JANUARY 1959

MARVINC. ROSS

See R. Delbrueck, Spatantike Kaiserportrats (Berlin, 1933)~pp. 229-231 and pl. ~ z z f f . See the weight in the British Museum illustrated by 0. M. Dalton, Catalogue of E a r l y Christian Antiquities . . . of the British M u s e u m (London, I ~ O I )p., 98, no. 485. 22 This dating is based in part on the fact that no dated fourth-century monument or coin with an emperor holding the globe surmounted with a cross is known. However, the principle seems to have been accepted in the time of Constantine the Great, as seen on his silver medallion, struck in A.D. 317, showing the Emperor holding a scepter with a globe a t the top surmounted by a cross (see A. Alfoldi, "Das Kreuzszepter Konstantins des Grossen," Schweizev Miinzblattev, IV [1g53], pp. 81-86). The question has recently been raised by Phyllis Williams Lehman, "Theodosius or Justinian? A Renaissance Drawing of a Byzantine Rider," T h e A r t Bulletin, XLI (1959), pp. 38-57, esp. P. 44. Two other small bronze statuettes of Constantine the Great are known in the literature on the subject. One, an equestrian statuette, was found a t Altinum in Italy in the nineteenth century, and was acquired by the Museum in Vienna. Dr. No11 has recently made a study of it and believes i t to have been made in the "Solar Period" of the Emperor, i.e. circa A.D. 310-325 (R. Noll, "Der Reiter von Altinum," Jahveshefte des osterreichischen Archaologischen Institutes, XLIII [1g56-19581, pp. 113-119). The tube-like rendering of the drapery and the general roughness of the workmanship recall the bronze weights. Another bronze statuette, a standing figure of the Emperor, found in Jutland in 1730 and now in the Copenhagen Museum, was identified as Constantine the Great by M. B. Mackprang ("Eine in Jutland vor 200 Jahren gefundene Kaiserstatuette," Acta Archaeologia, I X [1g38], pp. 135-I~I),who thought it possibly a replica of the famous statue on the porphyry column in Constantinople. These bronze statuettes doubtless reflect the great esteem in which the Emperor was held during his lifetime and long after his death, and, considering his popularity, it was natural for replicas ofhis likeness to have had broad public appeal. 20

21

Ol~vcrse

Rcve~-sc

3. Florence, Museo Archeologico. Bronze Medallion of Constantine I

4. Rome, Arch of Constantine. Detail of Warrior's Head

Rome, Arch of Constantine. Detail of Relief on Pedestal

7. London, British Museum. Royal Helmet from Sutton Hoo, detail

8. Copenhagen, National Museum. Scandinavian Helmet

9. Edinburgh, National Museum of Antiquities. Torrs Chamfrein

11.

12.

Stockholm, Statens Historiska Museen. Torslunda Plaque

Berlin, Staatliche Museen. Scale, with bronze Statuette of Constantine'

14. Moscow, State Historical Museum. Bronze Weight 2s

15. Kherson, Museum. Bronze Weight

16. Washington, D.C., Dumbarton Oaks Collection. Bronze Weight

17. Statuette of St. Peter. Location unknown

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: Cornuti: A Teutonic Contingent in the Service of Constantine the Great and Its Decisive Role in the Battle at the Milvian Bridge. With a Discussion of Bronze Statuettes of Constantine the Great Andrew Alföldi; Marvin C. Ross Dumbarton Oaks Papers, Vol. 13. (1959), pp. 169+171-183. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C169%3ACATCIT%3E2.0.CO%3B2-2

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 15

Roma and Constantinopolis in Late-Antique Art from 312 to 365 J. M. C. Toynbee The Journal of Roman Studies, Vol. 37, Parts 1 and 2. (1947), pp. 135-144. Stable URL: http://links.jstor.org/sici?sici=0075-4358%281947%2937%3C135%3ARACILA%3E2.0.CO%3B2-4 18

Herakles Crowning Himself: New Greek Statuary Types and Their Place in Hellenistic and Roman Art C. C. Vermeule The Journal of Hellenic Studies, Vol. 77, Part 2. (1957), pp. 283-299. Stable URL: http://links.jstor.org/sici?sici=0075-4269%281957%2977%3C283%3AHCHNGS%3E2.0.CO%3B2-K 22

Theodosius or Justinian? A Renaissance Drawing of a Byzantine Rider Phyllis Williams Lehmann The Art Bulletin, Vol. 41, No. 1. (Mar., 1959), pp. 39-57. Stable URL: http://links.jstor.org/sici?sici=0004-3079%28195903%2941%3A1%3C39%3ATOJARD%3E2.0.CO%3B2-%23

NOTE: The reference numbering from the original has been maintained in this citation list.

Fourth Preliminary Report on the Restoration of the Frescoes in the Kariye Camii at Istanbul by the Byzantine Institute, 1957-1958 Paul A. Underwood Dumbarton Oaks Papers, Vol. 13. (1959), pp. 185+187-212. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C185%3AFPROTR%3E2.0.CO%3B2-O Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:37:22 2008

F O U R T H PRELIMINARY R E P O R T O N T H E RESTORATION O F T H E FRESCOES I N T H E KARIYE CAM11 A T ISTANBUL BY T H E BYZANTINE INSTITUTE, 1957-I 958

N July of 1958 the restoration of the paintings in the Parecclesion of the Kariye Camii at Istanbul was completed. This Fourth Preliminary Report, dealing with the results of the seasons of 1957 and 1958 is, theref ore, the last of the series to be published in consecutive volumes of Dumbarton Oaks Pafiers.l As has been the case since 1952, the technical work of restoration was again mainly in the hands of Mr. Carroll Wales and Mr. Constantine Causis. Their skillful and persevering labors, now fulfilled, are deserving of highest praise and grateful acknowledgement. The general direction of the field work of the Byzantine Institute, which included a number of other projects in Istanbul, was again shared by the author with Mr. Laurence Majewski, Deputy Director, and Mr. E. J. W. Hawkins, Assistant Director. The fruits of their labors and those of the other members of the staff during this time are visible in other parts of the Kariye Camii, in the Fetiye Camii and the Zeyrek Camii. Students and amateurs of Byzantine art will also wish to join the Byzantine Institute in expressing gratitude to all private benefactors and to the Bollingen Foundation and the Samuel H. Kress Foundation for their sustained support which has made possible the uncovering and preservation of the paintings in this extraordinary monument of Byzantine art (figs. I and 2). To the government of the Turkish Republic and its officers the Byzantine Institute herewith expresses its deep appreciation for permitting and actively encouraging its work in the Kariye Camii and a number of other public monuments of Istanbul. The author wishes to express his personal thanks to Bay Feridun Dirimtekin, Director of the Ayasofya and Kariye Camii Museums, for his unfailing courtesies and to the members of the Commission for the Preservation of Ancient Monuments of Istanbul and the officers of the Directorate General of Museums in the Ministry of National Education who have so greatly facilitated the work. I t is now possible for the first time to present a list of all subjects depicted in the Parecclesion together with diagramatic plans indicating the position of each of the paintings in the building. The plans and the list of subjects, with cross-references to earlier reports, will be found immediately before the illustrations at the end of this report. Plan A gives the position of each of the scenes and medallions in the vaults and lunettes above the cornice (numbers 201-~42)~while plan B indicates the locations of the standing figures and medallions of saints which compose a frieze upon the walls below the cornice (numbers 243-269). This fourth report deals with the latter group, but excludes the figures in 1 The three earlier Reports appeared in volumes 9-10 (1956)) 11 (1957)) and 12 (1958). A final publication of the Kariye Camii in several volumes is expected to follow shortly. It will include completely illustrated catalogs of the mosaics and frescoes, a volume of studies in the representational arts, and will end with a historical and architectural survey of the building which will also present the results of archaeological investigations made during the years of work in the Kariye Camii.

P A U L A. U N D E R W O O D the bema which were published in the Second Preliminary Report.2 Outside the bema the frieze of figures consists mainly of martyrs, a majority of whom are military saints represented in full armor. I t can be readily seen in plan B that only the south wall could exhibit the ideal layout for the series of saints. In this wall each of the two bays contained an arcosolium placed in the center between the piers (fig. 3). At the sides of these two arcosolia, and again on the piers, full-length military saints were depicted while two bust portraits of saints, in medallions, were placed in the spandrels above the arcosolia, as can be seen in Figure 3 which represents the "ideal" composition of the individual units of the walls. Elsewhere in the Parecclesion this arrangement was either unattainable or was subsequently altered. In all probability the south wall of the western bay was originally composed in the same manner, but the later insertion of the sculptured ornament above and around the arcosolium of Michael Tornikes3 resulted in the destruction of the two medallions that had probably existed above the arch and of parts of the two military saints at the sides (figs. 12 and 13). In the whole of the north wall this ideal arrangement could only be approximated because the presence of two doors on that side of the chapel made it necessary to place the two arcosolia off center of the bays thus destroying the rhythmic composition of the wall and reducing the amount of wall surface on which paintings could be placed. There is among the military saints, as Delehaye has remarked,4 a kind of &at-major. First in rank are the military saints far excellence to whom the the Greeks gave the name oi a y ~ op&p-rvp~s ~ oi o-rpa-rqh&-ral. Among these, Pseudo Codinus5 lists Demetrius, Procopius, the two Theodores, and George. St. George, however, is recognized as chief among the military saints and is often termed ~po-rraloq6pos. Delehaye further remarks that Saint Mercurius should be added to these because he was only slightly less popular than the other^.^ I t is evident that the painters of the Kariye Camii arranged their military saints with an eye to protocol. The very group that is named above occupies most of the south wall, beginning at the bema, where it is arrayed in this order: George, Demetrius, Theodore the tiro, Theodore Stratelates, Mercurius, and P r ~ c o p i u sThe . ~ more usual arrangement would have been to 2 Dumbartan Oaks Papers, 11 (Cambridge, Mass., 1g57), pp. 211-220, figs. 44-51. See P. A. Underwood, "Notes on the Work of the Byzantine Institute in Istanbul: 1955-1956," Dumbarton Oaks Papers, 12 (Cambridge, Mass., 1958), pp. 271-274. 4 H. Delehaye, Les ZLgendes grecques des saints militaires (Paris, ~ g o g )pp. , 2ff. De officiis, ed. Bonn, p. 48. 0p. cit., p. 9. I t is interesting to note that this is the very same group that fills the soffits of the three great arches surrounding the naos a t Hosios Lucas, north, south, and west (but excluding the medallions a t the tops of the arches), where they are likewise represented as standing figures in full armor, in almost the same order of precedence and in the same pairings, as in the Parecclesion of the Kariye Camii. At Hosios Lucas, George and Demetrius are pendant figures in the western sides of the north and south arches. The two Theodores are pendants in the eastern sides of the same arches, the tiro being given precedence in the north arch over the "general" in the south arch. Mercurius and Procopius are paired in the great western arch, Mercurius in the north side, Procopius in the south. These important positions in the scheme of church decoration give the military saints a degree of "priority" over all other saints and martyrs represented in the church. For layout, see R. W. Schulz and S. H. Barnsley, The Monastery of Saint Luke of Stiris, in Phocis, etc. (London, I ~ O I ) , pl. 34, or, the plan given in E. Diez and 0. Demus, Byzantine Mosaics in Greece (Cambridge, Mass., 1931).A t the Protaton,

T H E F R E S C O E S I N T H E K A R I Y E CAM11

189

rank the saints, according to importance, as pendant figures on both sides of the chapel, as was done at Hosios Lucas and elsewhere. In such an arrangement, St. George would have been placed on the north wall immediately west of the bema, next to the figure of Christ that in all probability once filled the wall of the bema on the north side. St. Demetrius would then have been placed in the corresponding position on the south side. That this was not done here is due to the fact that no space was available on the north wall close to the bema and that the wall on that side was so greatly cut up by doors and niches. In presenting the individual figures we begin, then, with those on the south wall immediately west of the bema and continue in a clock-wise direction around the circuit of the walls (numbers 250 to 269 on plan). 250. ST. GEORGE (Figures 4 and 5) Between the bema and the arcosolium in the south wall of the eastern bay stands the figure of St. George attired in full armor.* The identification is certain for the inscription is clearly legible though in places the impasto had been scraped and required some strengthening. Placed, as he is, nearest the sanctuary, and immediately beside the Eleousa, he is given the position of honor at the head of all the saints on the walls of the chapel. The accepted type of St. George, which is followed in this fresco, is that of a beardless youth with a rather round face, often with a low brow, and with a bushy head of hair which is short at the back of the head since it always disappears from view at the level of the ears. As in this case, the hair is nearly always treated as a mass of small tight curls which often have such heavy highlights as to suggest their being blond.9 Mt. Athos (G. Millet, Monuments de Z'Athos [Paris, 19271, pls. 48 and 49), very nearly the same protocol is observed, although the set of military saints is enlarged to eight. Sts. George and Demetrius are paired, as pendants, and given the positions of honor on the eastern faces of the transept walls near the corners where they turn into the nave (George on the north, Demetrius on the south). The two Theodores come next as they stand side by side on the north wall of the nave between the north transept and the arch to the west, the tiro having precedence by being placed to the east of the "general." Their pendants on the south wall of the nave are Mercurius and Artemius, who are thus paired, while the pendant figures of Procopius and Eustathius form a pair a t the western ends of the north and south walls, respectively. 8 Dimensions: height of panel, including red borders, 2.28 m.; minimum width of panel, including red borders, between the face of the pier of the bema and the left jamb of the arcosolium, 1.09 m.; height of figure, from toe a t left to top of head, 1.85 m.; diameter of nimbus, .383 m.; height of green zone, .52 m. The type is quite distinctive and remarkably consistent throughout the history of Byzantine art. Compare the fresco with the mosaic version of St. George in the outer narthex a t the Kariye Camii, p. 136, or, with his representations illustrated in color in A. Grabar, Byzantine Painting (Geneva, 1953)~ a t Hosios Lucas (one in mosaic in the nave, another in fresco in the lower church), Cefalh, Monreale, Cappella Palatina a t Palermo, Martorana a t Palermo, Protaton on Mt. Athos, Grabanica, Zica, etc., in all of which his features more or less uniformly present the same type. The color of the hair may be important as a feature of the type. I n the mosaic a t the Kariye Camii the hair is executed in brown glasses with so much yellow stone for the lights as to make him appear blond. I n the fresco version, however, the hair now appears brown, but the surface paint in the area of the hair has suffered much loss. The most striking similarity, as to type of face, hair, and even coloring, exists between the mosaic of St. George in the Kariye Camii and the figure of a saint who is usually identified as Demetrius (although no inscription accompanies the figure) on thewestern face of the northern pier of the chancel in the church of St. Demetrius a t Salonika (Grabar, op. cit., illustration facing the title page), often dated in the seventh century. As E. Kitzinger has observed ("Byzantine Art in the Period between Justinian and Iconoclasm," Berichte z u m XI. Internationalen Byzantinisten-Kongress, IV, I [Munich, 19581, p. 25),

P A U L A. U N D E R W O O D St. George stands nearly frontally, but with a slight turn of the body and head toward the right.10 With his right hand he holds a long spear very nearly vertically, the point of the spear actually touching the red border at the top.ll His left arm supports his shield and with his left hand he holds his sheathed sword in a composition which somewhat expands the figure on the right-hand side to convey the impression of filling the spandrel of the arch. His left arm and the drapery that covers it are thrust through the arm straps that are attached to the inner face of his circular shield. Since the costumes of the military saints are often very similar, much of the following detailed description of St. George also applies to others, and subsequent descriptions will, therefore, be more summary. The saint wears violet colored pantaloons which are ornamented just below the knees by narrow bands of yellow. His feet and ankles are bound with gray winding strips which are tied at the tops. Over the pantaloons he wears a short warrior's tunic of blue. The skirt of the tunic, visible below his armor, contains a considerable amount of blue pigment which is generally used rather sparingly in the paintings of the Parecclesion. The darks of the skirt are actually the blueblack of the background with a few darker black lines indicating the drapery folds. The sleeves of the tunic are similarly painted and have reddish yellow cuffs. His armor consists of the type of cuirass of bronze to which two valances, or overlapping skirtings, are attached at the bottom to protect the thighs and hips, and at the armholes to protect the upper arms. The breast plate of the cuirass (thorax) is the most highly decorative of any worn by the military the latter figure is utterly different from the other representations of Demetrius in the same church (which are remarkably consistent) or the usual type accorded Demetrius elsewhere. He is so similar to the type of St. George that in my opinion the Salonika mosaic should be identified as one of the earliest of the representations of St. George that has come down to us. (In the color plates of Grabar, op. cit., contrast the figure in question on the page opposite the title page, with the typical version of St. Demetrius a t Salonika on p. 50. For illustrations of St. Demetrius in other mosaics in his church a t Salonika, see G. and M. Soteriou, The Basilica of St. Demetrius at Thessalonika (in Greek), 2 (Athens, 1952)~pls. 61, 62, 67, 72, 77. For further comments on the type of Demetrius, see infra, p. 192 and note 18). Kitzinger's suggestion, loc. cit., that this mosaic, on the western face of the southern pier a t the chancel, represents St. Bacchus (because i t is pendant to the mosaic of St. Sergius, Bacchus' twin) is difficult to sustain. Not only should Bacchus have his attribute, the maniakion, as Sergius has in his mosaic (Soteriou, 09. cit., pl. 65a), but he should be represented with long locks of hair that fall to the shoulders behind the ears as is again the case with his "twin" Sergius in the mosaic. I t is true, however, that in the pre-iconoclastic encaustic icon from Mt. Sinai, now in Kiev (0.Wulff and M. Alpatoff, Denkmaler der Ikonenmalerei in kunst-geschichtlicher Folge [Hellerau bei Dresden, 19251, p. 13, fig. 3) the two saints have short curly hair and, as types, closely resemble St. George. However, as usual, they wear the maniakion. (see infra p. 206 and note 55). The striking similarity of type that exists between the mosaic here suggested to be that of St. George, a t Salonika, and the saint who stands a t the right in the encaustic icon representing the enthroned Madonna with two standing saints a t the Monastery of St. Catherine, Mt. Sinai (G. and M. Soteriou, Icones du Mont Sinai', I [Athens, 19561, pl. qff.) would seem t o identify the latter as another pre-iconoclastic representation of St. George. His pendant, a t the left, probably represents St. Theodore Tiro (see infra, note 20). See also the later icons of St. George a t Mt. Sinai (ibid., figs. 52, 62, 167, 169, 177). 10 I t will be seen that the standing figures on the south wall of the eastern bay and the pier between it and the western bay (figs. 4, 8, and 10) all turn toward the right with increasing emphasis, and that in the western bay the first figure (fig. 12) stands frontally, followed by figures who turn toward the left (figs. 13 and 14). 11 Evident in several of these figures is the artist's desire to fill the spaces available to him and even a t times to extend beyond the confines of his borders (figs. 10 and 13).

T H E F R E S C O E S I N T H E K A R I Y E CAM11

191

saints. From a boss in the form of a rosette in the center there radiates a kind of pinwheel design in S curves. For the middle tones the bronze material of the breast plate is represented by an olive green color. The same color is used for the darks, but in so dark a hue that the effect is that of a deep brown, almost black. The highlighting, which defines the design, is painted in yellow. The two valances around the hips and those protecting the upper arms, which probably represent series of vertically suspended strips of leather, are painted in red, orange-yellow, and yellow, with white for highlights. The same colors are used for the wide collar which laps over the cuirass at the shoulders around the neck. The skirts and armlets seem to attach to metal (iron) girdles, painted gray, which in turn are attached to the cuirass itself. The costume is completed by a red mantle thrown loosely over his right shoulder and left arm and covering the back of the figure. The shaft of his spear is painted reddish brown, the head in grays. The scabbard of his sword is dark blue-violet as is the belt that twines around it. The metal tips of the scabbard and the belt and the two bands on the scabbard, near the hilt, are yellow. The hilt of the sword itself is reddish brown. The inner face of the shield is painted in yellows and browns in a pattern of scales, while the small portion of the outer face of the shield, visible along the right side, is in blues and blacks with white highlights. 251, 252. ST.FLORUS (Figure 6) and ST.LAURUS (Figure 7). The two saints chosen to be represented in the medallions12 above the arcosolium in the south wall of the east bay were the twin brothers Florus and Laurus. According to their notices in the synaxaria, they were natives of Byzantium where they were employed as sculptors' apprentices. They were martyred in Illyricum, and their relics brought to Constantinople in the reign of Constantine the Great. Almost exact counterparts of these medallionportraits of the two saints are to be found, in mosaic, in the outer narthex of the Kariye Camii in the soffit of the western arch of the second bay. There they are also paired with one another and bear facial characteristics that are very similar to the frescoed versions. The two saints turn their heads very slightly toward one another and both doubtless held small crosses in their right hands, as Florus does and as is usually the case in the extensive series of martyr medallions in the Kariye Camii, both in mosaic and in fresco. Florus holds his left hand vertically, palm outward, beside the cross. The backgrounds of the two medallions are composed of three concentric zones which darken in value toward the center.13Thebackground of the medallion of Florus is rendered in three values of the same pink color that was used in 12 Dimensions: diameter of the medallion of Florus, including border, .51 m.; diameter of halo of Florus, 265 m. The diameter of the medallion of Laurus was originally the same as that of Florus, but owing to the fracture in the wall, which passes through the medallion, and the displacement of .08 m. caused by the earthquake, the present horizontal width is .59 m. The diameter of the halo of Laurus is .285 m. l 3 The two outer zones measure about .06 m. in width.

192

P A U L A. U N D E R W O O D

those of Melchizedek (no. 239)14 and the Archangel Michael (no. 242),15 while that of Laurus is painted in three values of yellow. The outer garment of St. Florus is worn over the shoulder and breast at the right. I t is painted in earth reds with highlights of pink and is ornamented with gold brocades at the neck and shoulder. Beneath this is a blue garment that is visible over the shoulder at the left. Most of this area is the blue-black of the general backgrounds, but the impression of a blue garment is conveyed by a few strokes of light blue pigment. This garment seems to have borne a tablion on the breast which is partly concealed by the outer garment. The tablion is painted in yellows and browns. Only one of the garments of St. Laurus is at all preserved or visible in the two widely separated fragments. These are reddish brown. The fragment at the left retains a small part of a gold brocaded ornament, possibly part of a tablion. 253. ST. DEMETRIUS (Figures 8 and g) The pendant figure to St. George, on the right-hand side of the arcosolium in the south wall of the east bay, is St. Demetrius.16 Although most of the impasto of his inscription had been scraped away there can be no doubt as to the identification of the figure for parts of the letters mu and eta are well preserved and some of the other letters can be deciphered on close inspection. The Saint (fig. 8) stands nearly frontally, but with his body turned slightly to the right while the head turns a little, and is inclined, to the left. His full weight rests upon his right leg, and his left leg is flexed. His right hand, extended far to the left, holds the shaft of his spear which rests on the ground near his right foot." The diagonal line of the Spear and the extendedright arm thus help to fill the wide space available in the spandrel of the arcosolium and bring the figure into compositional relationship with St. George and the medallions above the arch. His left hand rests lightly on the rim of his shield which, at his left, hangs suspended from his right shoulder by a strap. Behind the shield the sheathed sword is held in the crook of his left arm. The face (fig. g) is that of a beardless youth and in general conforms to the type adopted for St. Demetrius.lg I t is one of the best preserved heads in the "Second Prelim. Report," op. cit., pp. 184f., and fig. 14. "First Prelim. Report," op. cit., pp. 279-281, and fig. 84. la Dimensions: height of panel, including red borders, 2.25 m.; minimum width of panel, including borders, between the right jamb of the arcosolium and the pier a t the right, 1.10 m.; height of figure, from toe at left to top of head, 1.805 m.; diameter of nimbus, .385 m.; height of green zone, .51 m. l7 A posture which is very similar to that of the same saint a t the Protaton (Millet, 09. cit., pl. 48, I ) . l8 I t is more difficult to determine the characteristics of Demetrius than those of George (see above, p. 189 and note 9). He is definitely a youth and beardless, tends to be tall and slender, with a fairly long neck and face, as in our fresco and again in his seventh-century mosaics, such as the one on the eastern face of the south pier a t the chancel in his church a t Salonika (Soteriou, op. cit., pl. 67). His hair is short, sometimes rather closely cropped, as a t Salonika, but never as bushy as that of St. George. While it may be wavy, it is often rather straight, and in the works of the better masters is never curly. I n his images a t Salonika, which can be assumed to be among the earliest and most nearly "canonical" examples, the hair is dark. One can see these characteristics rather dryly expressed in the figure of Demetrius a t Cefalh and note how he differs from St. George who stands beside him. Other good examples are found among the icons of Mt. Sinai (Soteriou, op. cit., figs. 47, 70), in the 14

15

T H E FRESCOES I N T H E K A R I Y E C A M I I

193

entire series of frescoes. The hair is brown with yellow lights. The shading of the face, especially at the left, and the shadow below the chin are terre verte. Brown is merged into the terre verte in the modelling of the chin and the right side of the face. The flesh tones in the full lights are warm yellows brightened by pink at the cheeks, the upper part of the forehead, and in the right side of the neck. Fine, short, diagonal strokes of white mark the highest lights, especially on the left side of the face, neck, and eyes. The lips are pink with outlining in reddish brown. His pantaloons are yellow with narrow ornamental bands of blue below the knees. His feet and ankles are bound with spiral wrappings of gray which are tied at the top. Over these he wears a short red tunic with tight sleeves and narrow yellow cuffs. The lining of the tunic, visible at the lower left, is pink. Over the tunic, and suspended from the cuirass, are three skirtings which form part of his armor and protect the thighs. Similar protective elements cover the upper arms where sleevelets are attached at the arm holes of the cuirass. These flexible elements of the armor are painted yellow in the middle lights and brown in the shadow areas. The edges of the strips that compose the skirting are outlined at the ends in white lines, while the narrow strips that alternate with wider ones are edged with white and blue. The general tone of the cuirass is a greenish yellowish gray which seems to have been applied over red. On this a pattern of small diagonal scales was painted in thin white paint with heavy accents for the highlights in white impasto. White and gray dots of heavy impasto are applied on a number of the individual scales. The bottom edge of the cuirass, the narrow edging of the joint along the left side, a collar around the neck, an elaborate flange around the upper arm, and a kind of girdle across the middle -all are painted in two values of gray with white lights. A short green mantle that hangs down the back from his left shoulder completes the costume. Only the inner side and right edge of his circular shield are exposed to view. The inner face of the shield is painted in grays and has a rectangular ornament in the center between the two semicircular arm straps. These and the strap from which the shield hangs are painted black. The rim of the shield is yellow. The sword, with a gray and white hilt, is sheathed in its red scabbard which is reinforced by two yellow bands.

254. ST. THEODORE TIRO (Figures 10 and 11) St. Theodore the tiro (fig. 10) occupies the face of the southern pier between the two bays of the Parecclesion.19 The Saint stands in a rather exProtaton (Millet, op. cit., pl. 48, I ) , and in the church of St. Demetrius a t Ochrid (V. R. Petkovi6, La peinture serbe du moyen dge, I1 [Belgrade, 19341, pl. 124), the latter being particularly close to the version a t the Kariye Camii. For information regarding the military saints in the Sicilian mosaics, which I have used in this report, I am indebted to my colleague, Professor Kitzinger, who has kindly permitted me to consult his file of photographs. Ig Dimensions: height of panel, including red borders, 2.25 m.; width, including borders, 1.11j m.; height of figure, from toe of right foot to top of head, 1.905 m.; diameter of nimbus, .39 m.; height of green zone, 485 m.

194

P A U L A. U N D E R W O O D

aggerated swaying posture which one associates with Gothic art of the fourteenth century, with his full weight balanced on the left side and his head turned strongly toward the right. The rising curve along the left side of the body carries toward the right through the arm at the left, continues through the shoulders, and returns through the loop formed by the arm at the right. The compositional direction toward the right is further emphasized by the strong diagonal lines of the sash across the breast, the spear which he holds in his right hand, and the shield which hangs behind his shoulder at the right. Both shield and spear break out of the confines of the right-hand border of the panel. The purpose of this composition can be observed if it is viewed in conjunction with the figure of St. Procopius (fig. 14) on the pier to the right, for the two Saints act as compositional pendants to one another and serve to bracket the composition of the wall between them by means of their strong counter movements. While the condition of the figure is good, the head (fig. 11) has suffered some losses of surface paint. More serious is an area of lost plaster in the left side of the face, about 8 cm. long, which extends vertically from the hairline through the right eye, parallel to the nose. The Saint is represented with a rather long narrow face, a pointed beard of moderate length, and a moustache.20 In his costume St. Theodore differs in some respects from all the others, notably in the sleeveless outer tunic worn over his armor and in the type of shield that he carries. His pantaloons are brown and his spiral leggings are gray. The short tunic with tight sleeves, worn under the armor, is red with gold brocade cuffs. The garment is visible also in the shoulders, and in the gaps between the protective sleevelets and the cuirass, where very little of the cuirass itself is visible. The lower part of the skirting of flexible armor can be 20 The type for Theodore seems to have been formulated in pre-iconoclastic times before a distinction came to be made between two Theodores. (Delehaye, op. cit., p. 15ff., discusses the evidence of an early but widespread cult of the Saint and suggests that originally only the tiro existed in legend whereas the "general" seems to have made a n appearance by the ninth century a t the latest. This explains why the pre-iconoclastic images so seldom give the surname in their inscriptions.) For an early pre-iconoclastic image of Theodore that conforms to the type of the tiro see the Coptic textile a t the William Hayes Fogg Art Museum (Cambridge, Mass.), accession no. 1939.112, published by SV. R. Tyler in the museum's Bulletin, IX, I (November 1g3g), p. zff., figs. 1-3, 5a and b. He is characterized by a long face with a dark, moderately long, pointed beard and a moustache. The hair of the head is short and curly. Essentially the same type appears again in the apse mosaic a t SS. Cosmas and Damian Rome (G. XIatthiae, SS. Cosma e Damiano e S. Teodoro [Rome, 19481, pls. 3 and 9) and in S. Theodore, Rome (ibid., pl. 38). The figure at the left in the encaustic icon of Mt. Sinai cited above, note 9, is one in which the type is perfectly expressed and for that reason it was suggested that the figure should be regarded as that of Theodore. Another icon of Theodore at Mt. Sinai, of the sixth or seventh century, this time inscribed, is of the same general type. (Soteriou, Sinai, fig. 15). After Iconoclasm, and after two Theodores came to be recognized, the tiro, in general, conforms to this type, as in the mosaic a t St. Demetrius in Salonika representing the Virgin in deesis to the left of a saint whom Soteriou, however (St. Demetrius, I, p. 195, and 11, pl. 66), regards as the "general." Still other examples of the type are found at Hosios Lucas (two mosaics, one in the north arch of the nave, the other above the window in the east wall to the south of the apse), at the Martorana and in the Cappella Palatina in Palermo. Usually, one of the points of distinction between the Theodores is that the tiro's beard, while it may be wavy, comes to a single point, as in our fresco and the examples cited above, whereas the "general" has a double beard, in two lobes of curly hair (infra). Sometimes, however, the tiro was given a forked, but pointed, beard as at the Protaton and Chilandari (Millet, op. cit., pls. 48, 3 and 74, 3). Other, more correct but late, examples of the tiro are: the Lavra (ibid., pl. 139, 4), and Pe6 (Petkovid, op. cit., 11, pls. gz and 105).

T H E F R E S C O E S I N T H E K A R I Y E CAM11 seen between the two tunics. The protective sleevelets that cover the upper arm and the shirting below are painted, as usual, in yellows and browns with white highlights. The sleevelets are ornamented with scales and coiling motifs. The cuirass can be seen mainly in a small area of the shoulder at the left. I t is painted in dark brown while the collar around the opening for the neck is yellow. The sleeveless outer tunic is painted yellow. His mantle, which is of a beautiful blue-violet color, is tied tightly, like a sash, across his breast and knotted below his right arm. Below and to the right, the mantle is blown out in a complicated train of flying drapery. The Saint carries a curved triangular shield whose lower point is concealed behind the figure.21The inner face is painted white and has a decorative border in yellow. I t is suspended from his shoulder by a black strap which attaches to the black arm-supports on the inner face. With his left hand he holds his sheathed sword by its belt, which is also black. The scabbard is red while the hilt of the sword itself is reddish brown and yellow. (Figure 12) To the left of the arcosolium in the south wall of the western bay, partially destroyed by the insertion of the sculptured decoration of the tomb of Michael Tornikes, is the figure of the second military saint who bears the name Theodore ~ ~ inscription giving his name is preserved in the left side of the (fig. 1 2 ) .The panel but his surname, which once existed to the right of the head, is now lost. He can definitely be identified, however, as Theodore Stratelates, the almost inseparable companion of Theodore the tiro in church dec0ration.~3The head of the Saint, too, identifies him as the "general," for the type most frequently adopted in later Byzantine art depicts him, as he is here, with bushy hair and a beard of moderate length which, in the best examples, is rendered as a double beard with two separated lobes at the bottom.Z4 The Saint stands frontally posed, with his weight evenly distributed on both feet. He holds his spear at a slight angle before him with his right hand. His shield and sword were doubtless at the far right in the areas now destroyed. He wears violet pantaloons with a decorative band of yellow around each calf. The foot at the right, partially destroyed, shows that he wore a kind of soft 2 5 5 . ST. THEODORE STRATELATES

21 Pointed shields are often found in representations of St. Theodore Stratelates, as a t the Lavra, Catholicon, Mt. Athos (G. Millet, op. cit., pl. 139, 4); twice in the churches a t Pe6 (Petkovid, op. cit., 11, pls. 92 and 105); a t PsaCa (ibid., pl. 171). Shields of this type appear also in the hands of soldiers in narrative scenes, as a t Lesnovo (ibid., pl. 156). 22 Dimensions: height, including borders, 2.245 m.; original breadth a t bottom, including borders 1.08 m.; height of figure, from bottom of his loosed boot to top of head, 1.995 m.; diameter of halo .38 m.; height of green zone, .49 m. 23 I n later centuries the two seem t o have achieved equal prominence in church decoration and are usually paired with one another as they are in the Kariye Camii. See, for example, Hosios Lucas (supra, note 7) the Protaton, &It. Athos (Millet, op. cit., pls. 48, 3, 49, 3 ) ; the Lavra (ibid., pl. 139, 4); Chilandari (ibid., pl. 74, 3 ) ; or several of the churches a t Kastoria (Pelekanides, Kastoria [Athens, 19531, passim), in all of which the two Theodores stand side by side. *4 See most of the examples cited in the preceding note, t o which add the fourteenth-century mosaic icon of the Hermitage, Leningrad (V. Lazareff, History of Byzantine Painting, I1 [Moscow, 19481, fig. 289) ; Dionysiou (Millet, op. cit., pl. 203, 2) ; Pe6 (Petkovi6, 09. cit., 11, pl. 92)) and the church of the Anargyroi a t Kastoria (Pelekanides, op. cit., pl. zra).

P A U L A. U N D E R W O O D boot or buskin which rises about one third of the way to the knee. These were red with two bands of yellow ornament around them. The buskin of the right foot has been removed and the Saint seems to stand upon it with his bare right foot.25 Here, one can see the slit in the bottom of the leg of the pantaloon and the bindings that wrap around the ankle. The origins of the phenomenon of the "one shoe off and one shoe on" are shrouded in ancient lore, and its meaning probably depended upon the category of the person who wore only one shoe or the nature of the occasion which called for this detail of c0stume.2~With specific reference to warriors, however, it appears to have become an attribute of the hero who is engaged in a perilous undertaking, although the ancients sometimes ascribed practical advantages to fighting with one bare The short tunic with long tight sleeves, worn under the armor, is red with yellow cuffs at the wrists. There are two skirtings or valances attached to the bottom of the cuirass. The lower skirting is in grays of three values and the upper in yellows and browns with some accents of very dark red. The cuirass is painted in the same colors that were used in suggesting the bronze material of St. George's cuirass, that is, a general tone of olive green, with dark brown and some black for the main lines of drawing, yellows in three values for the design, and a white dot in heavy impasto on each scale of the design. The main lines of the design of the cuirass make parallel, diagonal, curving lines from the upper right to the lower left. The girdle at the bottom of the cuirass, to which the skirtings are attached, and the two diagonal reinforcements of the cuirass which meet in a lozenge at the center, are painted in grays, as usual. An area of loss had occurred in the central lozenge which is filled with new plaster 25 A related phenomenon will be observed in the painting of Artemius or Nicetas (no. 264, fig. 25 infra) where the toe of the right buskin is torn open to reveal his bare toes. Similar instances of warrior saints with one bare foot (nzoaosandalos),or with toes of one or both feet exposed, occur elsewhere in Byzantine art, as in the Catholicon a t the Lavra, Mt. Athos, where George, Demetrius, and Theodore Stratelates have bare toes while Theodore tiro has a bare right foot (Millet, op. cit., pl. I39,3 and I39,4). I n these instances the Saints wear spiral leggings wound around their feet and ankles. Similarly, the Archangel Michael, when represented in full armor, sometimes has one bare foot, as a t the Patriarchal church a t Pe6 (Petkovi6, op. cit., I, pl. 78b). 26 I am indebted to Dr. VC7illiam A. S. Dale of the Toronto Art Gallery for references which he gathered in preparing a forthcoming study of the iconography of certain western medieval ivories. J. G. Frazer, The Golden Bough, Part 11, Taboo and the Perils of the S O ?(London, ~ 1g14), pp. 311f., associates the custom of having one foot bare with certain magical religious ceremonies, such as rites of purification, or with the liberation from the magical effects of knots, or as a sign of consecration or devotion on the part of those who find themselves in hazardous circumstances, especially fighting men. 27 Frazer, 106.cit., cites, for example, Thucydides' account of an attack by the Plataeans against the enemy's walls (111, 22). The Plataeans, says Thucydides, "not only were lightly armed, but also had only the left foot sandalled," and he explains this on the grounds that this was to prevent slipping in the mud. The Scholiast on Pindar's Pythian Odes (IV, 133) remarks that the Aetolians were shod on only one foot "because they were so warlike." Frazer also refers to Virgil's description (Aen.VI1, 689ff.)of the militia of ancient Latium, to Jason in his quest of the golden fleece, and to Perseus when he undertook to cut off the Gorgon's head, in all of which the heroes were shod on only one foot. In Christian art the representation of military saints with one shoe off is certain to have had closer connections with the classical use of the motif, as an attribute of the hero engaged on a dangerous mission (even if in origin it had magical significance as a protective talisman to ward off an evil fate), than i t has to the allegorical uses of the same motif in the sixteenth century in the West, as they are represented in the drawings of Urs Graf and others that are discussed by W. Weisbach, "Ein Fuss beschuht, der andere nackt," Bemerkungen zu einigen Handzeichnungen des Urs Graf, Zeitschrift fur schweizerische Archaeologie und Kunstgeschichte, IV (1g42), pp. 108-122.

T H E F R E S C O E S I N T H E K A R I Y E CAM11

197

and given a gray tone. A loose cloak or mantle hangs down the back of the figure. This is meant to convey a blue color, but is actually the blue-black of the background with the lines of the folds in deep black. After the marble reliefs were placed over the face of the arcosolium, some Byzantine repairs and repainting were carried out in the preserved parts of the figure. They occur at the right in areas of the skirting, half way up the cuirass, and in the right side of the head. A vertical strip of new plaster about .25 m. wide, extending from the bottom of the sculpture to the floor, represents a repair made by the staff of the Institute where original plaster had been lost. In addition, repairs were made by the Institute in a small area of the hair to right of center, and in two smaller areas, one in the hair at the left and another near the tip of the beard.

256. ST. MERCURIUS (Figure 13) To the right of the tomb of Michael Tornikes is the figure of St. Merc~rius.~8 The painting suffered damage, similar to that of Theodore Stratelates, when the reliefs were set in place. The Saint stands frontally posed, but with his head turned to the left while his glance is directed toward the right. If he carried a spear at all, which is doubtful, nothing of it now remains. His right hand grasps the hilt of his partially drawn sword and with his left he holds the scabbard. The circular shield, its inner face and rim at the right exposed to view, hangs suspended behind his left shoulder by a strap slung over his right shoulder. In spite of the damaged condition of the surface paint in the area of the head, the features are clearly preserved. He is depicted in the type that was usually adopted for Mercurius, that is, as a youth with a sparse beard and moustache.29The treatment of the head is a particularly good example of a phenomenon frequently found in early Palaeologan art in which, as it were, a profile view of the hair of the head is combined with a three-quarter view of the face-a mode of representation that results in very pronounced bulges in the forehead and in the back of the head. The legs are clothed in yellow pantaloons, decorated with green bands above and below the knees, and in gray spiral leggings. His short tunic with long sleeves is of blue-violet. Below the cuirass are three of the usual overlapping skirtings painted in yellows and browns with white highlights. These are covered at the left by what appears to be another skirting attached to the 28 Dimensions: height of panel, including borders, 2.22 m.; original breadth a t bottom, including borders, 1.08 m.; height of figure from toe to top of head, I. 855 m.; diameter of halo, .39 m.; height of green zone, .435 m. 29 The famous "Painters' Manual" of the monk and painter Dionysius of Fourna ('Ep~qvaia~ i j s Loypapl~ijsT ~ X V ~ ed. S , A. Papadopoulos-Kerameos [St. Petersburg, 19091, p. 157) characterizes Mercurius as vies drpxlyivqs-a youth who is beginning to grow a beard. One of the most literal interpretations of such a characterization is to be seen in the half-length figure of Mercurius in the Cappella Palatina at Palermo. There the head is covered by a mass of short curls and the beard is limited to a narrow fringe outlining the chin and cheeks from ear to ear. A faint moustache is indicated, but otherwise the face is beardless. A similarly sparse beard is given the Saint in his medallion at the Martorana. Other examples of the type are also found a t Hosios Lucas, lower church; the Protaton (Millet, op. cit., pl. 4g,1); Graranica (Petkovi6, op. cit., I, pl. 61a); Ped (ibid.,pl. gob); PsaEa (ibid., 11,pl. 170).

P A U L A. U N D E R W O O D bottom of the cuirass. This seems to be a solid, though pliable, piece of armor covered with small scales. I t is painted in the same colors as the skirtings to the right which are composed of separate vertically suspended strips. Over the cuirass, around the neck, is a wide collar which is treated like the skirtings below the cuirass. The cuirass itself is painted in dark olive green to represent bronze and is ornamented with rays, foliate designs and scales in yellow with occasional impasto dots of yellow. The girdles of gray metal around the bottom of the cuirass and at its arm holes are supplemented by a wide girdle around the middle which is also gray. The latter is ornamented at the center with a small bust figure with a pentagonal nimbus and by other small decorations. Drawn diagonally across the breast, and knotted at the center, is a red mantle which falls down the back of the figure to become visible at the lower left and right. The inner side of the shield is painted gray and is ornamented with a scale pattern. At the far right a little of the exterior face is indicated in yellow with outline drawing in reddish brown. The partially drawn sword is gray and the scabbard is brown with reinforcing bands and a fleur-de-lis in gray. The belt, which is coiled around the scabbard, is a dark blue-violet.

257.

ST. PROCOPIUS (Figures 14 and 15) The wide pier at the western end of the south wall of the Parecclesion provides space for a panel containing two figures (fig. 14). The first of these, at the left, depicts St. P r o c o p i ~ in s ~a~particularly striking attitude, with feet widely spaced, body turned somewhat to the left, and head and glance directed more sharply in the same direction. The full weight of his massive and very solidly constructed figure rests upon his left leg. With his upraised right hand he wields a naked dagger. His left hand, considerably extended toward the right, holds the pointed end of his sheathed sword as it-rises diagonally toward the left to lean against his shoulder and pass behind his head. His circular shield, whose exterior face is exposed in three-quarter view, stands vertically on its edge on the ground to the right. The three letters of the first line in the inscription which gives his name are largely preserved. While the second row of letters is effaced, the final omicro.n and part of the terminal sigma are preserved, thus fully identifying the figure. Even without the inscription the identity of the Saint would be apparent from the very pronounced type that was accorded to St. Procopius in Byzantine art (fig. 15). Characteristic of him are the round and beardless face of the youth, coupled with the treatment of the wavy, but not curly, hair which passes behind the ears, and falls almost to the sh0ulders.3~ 30 Dimensions: height of panel, including red borders, 2.22 m.; total width of panel, including borders (width of pier), 1.725 m.; height of figure of Procopius, from toe of foot a t right to top of head, 1.905 m.; diameter of halo, .38 m.; height of geen zone, .50 m. 31 I n these respects compare figure I j with representations of the Saint at Hosios Lucas (his mosaic in the western arch of the nave and his medallion, in fresco, in the lower church) ; the Cappella Palatina and the Martorana in Palermo; Mt. Athos: Protaton (Millet, op. cit., pl. 53, 2 ) , Chilandari (ibid., pl. 74, 4), St. Paul, chapel of St. George (ibid., pl. 192, I ) , Dionysiou (ibid., pl. 2 0 2 , 2); the Serbian , (ibid., pl. 195); or a t Kastoria: Church of the examples: Karan (Petkovib, op. cit., 11,pl. I I ~ )Pavlica Anargyroi (Pelekanides, op. cit., pl. 23b), St. Nicholas (ibid., pl. 58b), St. Athanasius (ibid., pl. 154a).

T H E F R E S C O E S I N T H E K A R I Y E CAM11 The faces of St. Procopius and those of the other two figures in this corner of the Parecclesion (St. Sabbas Stratelates, fig. 16, and an unknown saint, fig. 18) are the most perfectly preserved examples of facial representation in the entire chapel and, together with the face of St. Demetrius, form a true basis for judging the quality and style of the paintings in the Kariye Camii as they were in their original state. The skillful strokes of the brush are everywhere apparent in this head. The face itself is modelled with terre verte, shaded with brown, and has the main flesh tones of yellow brightened with red on the forehead, cheeks, and neck. The play of light is skillfully suggested by the short strokes of white that parallel the directions of the brush work with which the features were constructed. In this instance the hair, which so often suffered most from the effects of the covering materials, has survived in relatively good condition. The hair is rendered in browns with yellow lights. The Saint wears blue-violet pantaloons and, on his feet, white buskins that seem to be made of a soft pliable material, possibly leather. They fit rather loosely and slip down at the ankles. They are supported from the back by white thongs which are crossed and knotted below the knees. Over the pantaloons he wears a short blue tunic which is mostly the blue-black of the background, with drawing in black and with blue pigment very sparingly used. The sleeves of the tunic are of the same colors and have reddish yellow cuffs. Over the tunic the warrior wears what appears to be a short jerkin with short sleeves, possibly of leather or chain mail covered with scales. The skirt of this garment, which is slit up the left side, is visible below the cuirass and its short sleeves can be seen to emerge through the armholes of the cuirass. This armored garment is painted in a reddish yellow color with yellow scales. Rendered in the same colors and with the same scale pattern is a second armored garment which seems also to be of the same materials. I t is sleeveless and very short and is worn over what appears to be the cuirass. The latter is painted in gray and blue and is, therefore, similar in color to the under tunic. I t is ornamented around the waist, the neck, and the armholes by a decorative fringe of gray. Above the fringe at the waist the cuirass is bordered by three narrow bands, the lower one gray, the central one red, and the upper one yellow. The costume is completed by a red mantle which is bound around the breast as a sash, rises across the back of the shoulders, and wraps around his left arm where it falls in a cascade of drapery. The empty scabbard for the dagger is bluish violet with reddish yellow ornaments. I t is worn high up near his left shoulder where it hangs from a dark red strap that passes over his right shoulder and around his neck. The drawn dagger is white and gray. The scabbard for the sword is also bluish violet and its belt, which coils around it, is dark red. The shield is divided into concentric zones. The outer zone is gray and is decorated with ornamental motifs. The next, and widest, zone is a light reddish violet and is unadorned except for a few dark, scale-like, forms in the lower part. The inner zone of gray is subdivided into several lesser bands and at the center of the shield is a boss in the form of a human face, seen in near profile.

P A U L A. U N D E R W O O D

258. ST. SABAS STRATELATES (Figures 14 and 16) The last figure on the south wall is identified by the inscription as St. Sabas Stratelate~.3~ Although he is a military saint, he is not represented in warrior's costume, but in the tunic and chlamys of a courtier.33 He stands frontally posed, but with head (fig. 16) turned slightly to his left. His chlamys is thrown back over his right shoulder, thus exposing the tunic on that side. With his right hand he holds before him the small white cross that denotes the martyr. His left hand is concealed beneath the chlamys. Although parts of the costume have suffered loss of surface paint, the head is almost perfectly preserved and is, indeed, one of the finest specimens of portraiture to have survived in the Kariye Camii. His is a long thin face, with hollow cheeks and lofty brow. The dark brown hair is rather sparse at the top of the head, but falls in wavy curls behind the ears almost to the shoulders. The beard, of moderate length, terminates in two curly locks. The tunic is red and was decorated at the hem, collar, shoulders, and cuffs with gold brocades studded with red and green gems and with pearls, the latter represented by white dots of impasto. While the red of the garment is well preserved, the yellows of the brocades and much of the paint of the gems that lay over it, have flaked. The green chlamys, which is fastened at the breast, is not well preserved. I t bore a yellow tablion which is traceable at the vertical edge near the hand. Roughly incised lines scratched in the plaster define its general contours. The right foot has been entirely destroyed. As can be seen from the left foot, he wore red buskins with a yellow band of ornament at the ankles. UNKNOWN SAINT (Figures 17 and 18) The western end of the Parecclesion is separated from the outer narthex by two anta walls and two columns which support a very open tympanum. The southern anta wall34 had lost its frescoed plaster along its right edge throughout its height in an area that widened to the full width of the anta at the bottom of the painted panel containing the figure of a saint. New plaster was applied over the area of loss. The field of the panel was here given a dark gray tone and the lost parts of the borders were reconstituted in red. The figure painted on this anta wall is that of a martyr whose identity

259. AN

32 Dimensions: height of figure, 1.89 m.; diameter of halo, .38 m. Cf. note 30 for dimensions of the panel. The inscription to the right of the head preserves in its second line the letters tau, eta, lambda which suffice to identify the particular Sabas. A Gothic officer whose feast is celebrated on April 24, he is said by the Synaxarium Constantinopolitanurn (Propylaeum ad Acta Sanctorum, Novernbris, col. 627,10) to have been martyred in Rome. I n this source he is referred to as St. Sabas Stratelates (not to be confused with the later martyr Sabas Gothus). This Sabas is rather infrequently represented in extant works of art, but see his image a t Studenica (Petkovi6, op. cit., 11, pl. 3). He was depicted also a t Graranica, Ravanica, and Kaleni6 (ibid., pp. 30, 59, 63). 33 I n certain monuments that present a group of military saints some are depicted in military, others in court, costumes. For example, the two inner faces of the wings of the Harbaville Triptych contain a series of eight military saints in two zones. The four in the upper zone are armed, while those below wear the tunic and chlamys. Cf. A. Goldschmidt and K. Weitzmann, Die byzantinischen Elfenbeinskulpturen, I1 (Berlin, 1931), no. 33. 34 Average width, including red border at left and new plaster along right edge, .625 m. ; approximate height of panel when made to align with adjacent panel a t left, 2.21 m.; diameter of halo, .385 m.

T H E F R E S C O E S I N T H E K A R I Y E CAM11

201

cannot be definitely established. The entire inscription recording the name seems to have been confined to the upper left corner. With the exception of the 6 Q~os, the rest of the inscription has very nearly disappeared. However, in the third line, below and slightly to the right of the alpha of the second line, are traces of what appears to have been the letter nu. From its position in the line, it seems rather unlikely that this was the initial letter. To the right of the nu is the ghost of a more or less vertical stroke. In the fourth line, slightly to the left of center of the nu are parts of two strokes, a vertical one at the left and a diagonal one at the right, which could be construed as parts of the letter nu, or of a kappa. To the right, the next letter was curved, suggesting the left part of an omicron. In suggesting a possible identification of the figure, these data must be taken into account together with the type of the head, the probability that the painting represents a martyr (since he doubtless held a small cross in his right hand), and the fact that the figure is not attired in military garb. On these grounds, but still very much in the realm of speculation, one might conjecture that the figure represents the martyr Akindynus. A very striking similarity exists between the features of our Saint (fig. 18) and those of St. Akindynus in the Protaton of Mt. Ath0s.3~Especially characteristic of the two images are the arched eyebrows, the prominent cheek bones, the extent to which the beard covers the face, the moderate length of the beard, and the treatment of the hair.36 The surviving fragments of letters in the inscription could be construed to fit the name of Akindynus. The Saint wore a rather short tunic whose color is indeterminate, for the paint of this garment is now so thin that only gray and black paint survive. I t is likely that a blue garment was intended and that all blue pigment, which is used so sparingly in the frescoes of the Parecclesion, has disappeared. I t was decorated with gold brocade hem, cuffs, and collar, the latter being very wide. The area of the cuff and hand has been largely effaced and little more than underpainting now remains. Above the hand -is an irregular and extensive area of lost plaster (now newly filled) which probably contained the small white cross carried by all martyrs in the mosaics and frescoes of the Kariye Camii save those who wear the arms and armor of the warrior. At the lower left, below the hem of the tunic, is a small area of the right leg where it can be seen that the Saint wore green pantaloons. Over the shoulders, but open down the front, he wore a red mantle with yellow tablia at both edges. His costume closely resembles some of those worn by the full-length figures in mosaic in the transverse arches of the outer narthex at the Kariye Camii. In particular, the pantaloons worn under short tunics, and the mantle worn over both shoulders and open down the front, are features that are paralleled in the figure of St. Andronicus.3' Millet, op. cit., pl. 54, I. hTot all representations of Akindynus, however, display these traits, but the frequency with which the features of saints a t the Protaton resemble those a t the Kariye Camii indicates that the painters in both places had a common body of understanding in the representation of types. 37 Th. I. Schmit, Kakhrze-diami (Sofia, 1906), pl. LXI, no. 128. 35

36

P A U L A. U N D E R W O O D Except for the serious loss of paint in the beard and moustache, the head, and especially the face, are in a remarkably fresh state of preservation. The colors are unusually bright in the face. The red of the lips, which seem to "float" in an otherwise effaced area, is untouched. In places the hair is also extremely well preserved. This was painted in browns and still retains its very light blue highlights in contrast to the white strokes used for the facial highlights. 260. ST. DAVID OF THESSALONIKE (Figures 19 and 20) On the narrow eastern face of the northern anta wall is the image of the dendrite saint, David of T h e s ~ a l o n i k e .A~ ~contemporary of the Emperor Justinian, David was a famous miracle working hermit of Thessalonike where, according to the menaea, he spent part of his ascetic life sitting in an almond tree "like a singing bird."Z9 Representations of this Saint are very rare and never, to my knowledge, in quite the form in which he is depicted at the Kariye Camii40 he trunk of the tree is presented in a rather columnar form until, at the top, the branches and foliage assume the shape of a nest (did the painters think of the comparison to a "bird" ?), or basket, from which rise the torso and head of the Saint. The analogy to the manner of depicting the stylite saints (cf. fig. 30, for example), is obvious: the tree trunk substitutes for the column shaft, and the foliage resembles the capital. The Saint himself is portrayed as a very old and emaciated figure with both hands held palm outward before him in an attitude of prayer. He wears a scapular, the attribute of the monk, over his tunic. Visible over the scapular are the black cords and loops to which he could attach his mantle which, however, he does not wear. The base of the tree is gnarled and so widely spread that part of it is painted on the adjacent wall to the right. Near the top of the trunk, on the left, are two stumps of branches. The surface paint is rather effaced and what is now visible is mostly a dark olive green with yellow highlights. At a height of 1.20 m. from the bottom of the panel the rather flat bottom of the nest of leaves begins. This too widens out to such an extent as to encroach on the adjoining wall to the right. The leaves of the tree are of two values of light green with white highlights, the spaces between the leaves are black, and the stems are of dark olive green with yellow highlights. The Saint's loosely fitting tunic is light red-violet highlighted with white. the scapular is dark green with drawing in black. 38 Dimensions: height of panel, including red borders, 2.20 m.; width of anta wall, including red border a t left, .50 m.; diameter of halo, .305 m.; height of green zone, .585 m. The inscription was found to be clearly legible and has been somewhat strengthened by in-painting. 3 Cf. A. A. Vasiliev, "Life of David of Thessalonica," Traditio, IV (1946), pp. 115-147. 40 Vasiliev, op. cit., pp. 143-147, who collected the representations of the Saint, did not, of course, know the painting in the Kariye Camii. The examples he cites in which David is represented in a tree are late works, mostly Russian, of the sixteenth to nineteenth centuries, and in none of them is he shown in a tree which forms a nest-like enclosure such as we have a t the Kariye Camii. The earlier examples, namely, a t the Church of the Anargyroi in Kastoria (A. K. Orlandos, T h pvlav-r~vh~vqwaia - r f ~ Kamopias [Athens, 19381, p. 50, fig. 33) and a t the Protaton (Millet, op. cit., pl. 45, I), depict him as standing, without his tree, and wearing an extremely long beard. The painting in the Kariye Camii is, therefore, by far the earliest example of its type that has comedown to us, and in several respects is unique.

T H E F R E S C O E S I N T H E K A R I Y E CAM11 The head (fig. 20) is that of an emaciated ascetic with long thin features, hollow cheeks, and prominent bony structure. The very sunken eyes are overhung by bushy brows and the eyes themselves are hardly visible. The flesh, in its present condition, appears rather yellow, suggesting a sallow complexion. The hair and beard are gray and yellow with some lights in white.

261.ST. EUSTATHIUS (Figure 21) In the south wall of the Parecclesion, the left half of the western pier is penetrated by an arched opening which gives some light and ventilation to an interior chamber. There is space on this pier for only one standing figure: that of St. E ~ s t a t h i u sAs . ~ ~a pagan, before his conversion, the Saint was named Eustathius Plakidas,42 and under this name he is sometimes represented in the m o n ~ r n e n t s .Judging ~~ from the frequency with which St. Eustathius is represented in art, he was, after the group of six who outrank all others,'@one of the most popular of the military saints. The "Painter's Manual" describes him as having a half gray, rounded beard,45 and so is he represented in our fresco.46 More precisely, he is usually depicted with wavy, partly gray hair that falls in locks behind the ears almost to the shoulders, a short, but heavy, beard that is rounded in shape below the chin.47 In the fresco the warrior stands frontally with head turned slightly to the left. With his left hand he holds the scabbard of his sword while his right hand holds the hilt. Through the crook of his right arm a long spear passes downward, diagonally behind his right hip and leg, to rest on the ground between his feet, and upward, in front of his right upper arm, to the very edge of the border at the top of the panel. Another, shorter, spear whose surface paint is very nearly lost, rises vertically from the crook of his arm to pass in front of his shoulder. The lower end of the shaft emerges from behind the warrior's mantle. What is now visible of this spear seems to be the underpaint of green. I t is not possible to dertermine whether it represents a first rendering, later painted out, or whether the Saint actually carried two spears, one of which has flaked. He wears pantaloons which are covered with small scales drawn and highlighted in yellow. The feet and ankles are bound with gray spiral leggings. The short tunic with long tight sleeves is understood to be a blue garment, 4 1 Dimensions: height of panel, including red borders, 2.205 m.; width, between opening at left and edge of pier at right, .88 m.; height of figure, from toe at right to top of head, 1.865m.; diameter of halo, .385 m.; height of green zone, .62 m. 4 2 See the Synaxarium Constantinopolitanurn for Sept. 20 (op. cit., col. 59,25ff.). 43 See, for example, the inscription above his medallion at the Martorana in Palermo, or that above his full-length portrait as a warrior a t Detani (V. Petkovi6 and Dj. BoSkovi6, Dec'ani [Belgrade, 19411, Album, pl. 158).Cf. also the "Painter's Manual" of Dionysius of Fourna (op. cit., p. 158) where his name is given in full. 44 Supra, p. 188 and note 7. 45 LOC. cit., p. 158. 46 See also his medallions a t the Cappella Palatina and the hlartorana in Palermo. His full-length portraits in warrior's costume at the Protaton (Millet, op. cit., pl. 52, 2 ) and at DeCani (loc. cit.) are especially similar to that of the Kariye Camii. He is also represented at Studenica, &Ca, Staro Nagoricino (where he is called Plakidas), Gratanica, Psata, etc. 47 In seeking the required roundness, some painters gave him great moustaches that curve upward, as at Chilandari and Dionysiou, Mt. Athos (Millet, op. cit., pls. 74, 4 and 202, 2).

P A U L A. U N D E R W O O D though little if any blue pigment now remains. Over the tunic is a still shorter jerkin which is also covered with scales. Its color, however, is greenish brown with drawing in dark brown and scales highlighted in white. Over this is worn a single skirting of vertical pieces of armor suspended from the cuirass. The vertical pieces are painted in greenish brown, brown, and white, as are the individual pieces of armor that form the sleevelets to protect the upper arm and the collar that appears over the shoulder at the left under the mantle. The cuirass itself, like the jerkin and the pantaloons, is covered with scales. The colors are those that were used in the jerkin. Finally, covering the back and the right side of the breast is a red mantle. The scabbard is brown and the belt which twines around it is blue-violet. 262, 263. ST.SAMONAS (Figure 22) and ST.GURIAS (Figures 23 and 24) As can be seen on the plan, the northern wall of the western bay is very largely occupied by an arcosolium and the door of the passageway that leads from the Parecclesion to the nave of the church. I t is very doubtful that fulllength figures existed on the wall of this bay even before the sculptured canopy was applied around the face of the arcosolium. Two medallions, however, were probably painted in the spandrels above the arcosolium before the sculptures were put in place, though the evidence has been destroyed. The only other wall space on which figures could be painted in this bay is that above the semicircular head of the doorway, where the half-length figures of St. Samonas (no. 262, fig. 22) and St. Gurias (no. 263, figs. 23 and 24) are to be found.48The painting of St. Samonas, at the left, is still intact, including the inscription. The figure of St. Gurias was partially cut off at the right on the insertion of the sculptures, and his name, which fell on that side, was partly destroyed and partly painted out when the Byzantine workmen, after inserting the reliefs, painted a red border along the broken edge of the plaster, just to the left of the sculptures. When much of this secondary red border was removed in a search for the inscription, part of the initial letter of the name came to light, namely, the vertical stroke and a bit of the horizontal stroke of the letter gamma. This evidence tends to confirm the identification with Gurias, for he and Samonas are almost as inseparable companions in art as they are in the synaxis of November 15 where their feast commemorates their common martyrdom at E d e ~ s aThe . ~ ~identification of Gurias is confirmed by comparing his features in the fresco with those of his mosaic portrait in the outer narthex of the Kariye Camii which are remarkably similar. St. Samonas (fig. 22) bears the small white cross in his right hand before him. He wears a long blue tunic which is actually executed in cool grays, black Breadth of present wall surface from pier a t left to marble sculpture a t right, 1.635 m.; diameter of haloes, .39 m. 4O Cf. the Synaxarium Constantinopolitanum for November 15 (op. cit., col. z25,z), the synaxis of Sts. Gurias, Samonas, and Abibus, the latter also of Edessa. The three Saints, in that order, are represented in the mosaics of the fourth bay of the outer narthex of the Kariye Camii where the space in the arch soffits permitted only three medallions. Cf. Schmit, op. cit., pl. 65, nos. 136-138, prior to cleaning. Sts. Gurias and Samonas are paired also in the southern arch of the Cappella Palatina, while a t GraEanica they are pendants on the two western piers (Petkovi6, op. cit., pp. zg,30).

T H E F R E S C O E S I N T H E K A R I Y E CAM11

205

lines for the folds, and white highlights. The garment has yellowish red brocades at the hem, cuffs, and collar. These are studded with green and dark red gems and with pearls. The buskins (not illustrated in Figure 22, which cuts off the lower part of the figure to the left of the door) are of the same colors as the brocades. Over the tunic he wears the chlamys, buttoned at his right shoulder, but thrown back on that side to expose part of the tunic. The chlamys is of an unusual reddish brown color which is unique in the paintings of the Parecclesion unless it is the same as that used in the "shot" colors of the garment of the woman at the far right in the scene of Christ raising the daughter of Jairus,sO or in one or two of the marbleized panels in the dado below. The chlamys is decorated with a yellowish red tablion which is studded with green gems and with pearls. The garment also has a narrow edging of yellowish red. St. Gurias (fig. 23) extends his cross in his right hand toward his companion at the left. The directions of the drapery folds on the right side, where part of the figure was cut off, indicate that on this side too his hand and forearm were extended. He wears a red tunic with tight sleeves which is decorated with His chlamys, which he wears in the brocades similar to those of St. same manner as does his companion, is green. Like the chlamys of Samonas, this garment is also adorned with gold brocaded tablion and edging.

amo on as:

264. ST. ARTEMIUS or ST. NICETAS (Figures 25 and 26) The warrior Saint who occupies the pier in the north wall between the two bays of the Parecclesion51(fig. 25) cannot be definitely identified owing to the complete disappearance of the inscription, but his head (fig. 26), which bears the features of Christ, reveals him to have been either St. Artemius or St. Nicetas, both of whom were customarily depicted in the image of Christ in Byzantine art a t least as early as the twelfth century.52 A rather u n u s u ~ liconographic detail is the treatment of the Saint's right foot. Although this area of the painting has suffered considerable loss of NO. 203, "First Prelim. Report," op. cit., p. 277. Dimensions: height of panel, including red borders, 2.265 m.; width of panel with red borders, below marble sculpture, 1.09. m.; height of figure, 1.94 m.; diameter of halo, .385 m.; height of green zone, .535 m. 62 The "Painters' Manual," 09. cit., p. 157, states that Artemius was to be represented Zlpoto~TOG X p ~ m o G-rb ~1605.Concerning Nicetas, who follows immediately after Artemius, it is said, ~ a ai h & Zlpotos ah+. These directions are more than a literary tradition, for Byzantine monuments contain many examples of this practice and often pair the two "Christ-like" Saints. At the Cappella Palatina, Artemius seems to be given precedence over Nicetas by being placed in the north arch with or near eight other military saints. Nicetas is paired with the seldom represented Andrew Stratelates a t one side of the western arch as pendants to the two western military Saints John and Paul. Both Artemius and Nicetas closely resemble the Christ type that was used a t Palermo. I n the Martorana Artemius (but not Nicetas) is again represented in the north arch along with his military brethren and here too he resembles Christ. At the Protaton, in a painting that very closely resembles our fresco (Millet, op. cit., pl. 49, I), Artemius (but not Nicetas) was represented among the military saints. Artemius was also represented in: the chapel of St. George in the monastery of St. Paul, Mt. Athos, (ibid., pl. 192, I ) , Zira, southern chapel (Petkovid, 09. cit., 11, p. 21), and a t Sisojevac (ibid., p. 61). Artemius and Nicetas appear together a t Studenica (ibid., pl. 5), Graranica (ibid., pl. 66), Ravanica (ibid., p. 60), the church of the Panaghia of the Archon Apostolake a t Kastoria (Pelekanides, 09. cit., pl. 242, a), in all of which they are placed side by side. At Vatopedi, Mt. Athos (Millet, op. cit., pl. 83, I ) , they are placed as pendant figures a t each side of the apse of the north transept, Artemius a t the left in the position of greater honor. Both are dressed as martyrs, and both are portrayed in Christ's image. For Nicetas see also Derani (V. Petkovid and Dj. BoSkovid, op. cit., pl. 151). 60

51

206

P A U L A. U N D E R W O O D

surface paint, and part of the foot has been entirely lost, yet the essential details of the footwear are clearly evident. I t can be seen that he wears the same type of soft boot, or buskin, that was worn by Procopius (fig. 14) and Theodore Stratelates (fig. 12). Whereas the latter is depicted as having removed his right boot, and thus stands with a bare right foot,53 the toe of the boot of St. Artemius or St. Nicetas is represented as having been cut or torn open, thus exposing the toes of his right foot. The Saint stands with his full weight upon his left leg and with his right leg bent at the knee. In his right hand he holds the shaft of his spear which rises diagonally to the upper right corner of the panel. His sword is girded a t his waist and his left arm is inserted through the supports in the reverse side of his circular shield which is turned to expose the front face in a three-quarter view. The painted surface of this panel is now rather thin, especially in the areas of yellows and browns, and the original coloring is, therefore, sometimes difficult to determine. The figure wears what seem to have been yellowish brown pantaloons. The buskins are light reddish violet. The short tunic, visible in the skirt, right sleeve, and in an oval area below the cuirass, is red with ornamented cuffs at the wrists where only a yellow tone now remains. The armored skirting around the thighs, which in this instance seems not to have been attached to the bottom of the cuirass, is almost entirely effaced, but one can see faint traces of a design which is similar to that used in most of the other figures. Above the skirting is an ornamented gray band like that of St. Procopius (fig. 14). Some details of the design on the cuirass can be observed in the upper part where it is painted in yellow over red. Around his chest the cuirass is reinforced by a wide band of gray, representing metal, on which are traces of an embossed bust framed in a lozenge and flanked by foliate ornaments. He also wears a highly ornamented collar. Tied tightly across his chest is a redviolet mantle which, after wrapping itself around his left upper arm, behind the shield, reappears below in folds of drapery. The sheathed sword is long and curved, and, like the belt from which it hangs, was painted black. The shield is ornamented with three concentric zones: the outer one gray and ornamented, the second one a light red-violet originally ornamented with scales in a darker value of the color, and the inner zone, now yellow, but subdivided by parallel lines of dark red. 265, 266. ST. BACCHUS (Figure 27) and ST.SERGIUS (Figure 28) Like the north wall of the West Bay, that of the East Bay has relatively little surface available for the zone of saints owing to the fact that in it were constructed an arcosolium at the left and a door at the right. In the spandrels above the arcosolium are two medallions which depict the "twin" * counterparts to those over the military Saints, Bacchus and S e r g i ~ s , ~the Supra, p. 196 and notes 25-27, for comments on this iconographic detail. The order in which they are referred to in the synaxaria, in the vocables of their churches, and the usual order of rank in their joint images is Sergius, first, and then Bacchus. I n their positions in the Parecclesion Sergius is given priority by being placed closer to the sanctuary. From the standpoint of protocol the saints on the north wall should be counted from right to left. 53

T H E FRESCOES I N THE KARIYE C A M I I arcosolium in the opposite side of the chapel. Their inscriptions are almost perfectly preserved, but their identification would be certain in any case because they are given their traditional attribute-the maniakion, or necklace, which, with few exceptions, was accorded only to them.55 The medallions5'j consist of busts set in backgrounds of three zones, each progressively lighter in color value toward the outer edge. The color of the background for St. Bacchus is light green, and that for St. Sergius is pink. The types adopted in the portrayal of the two Saints are indistinguishable: both are beardless, with brown hair, lighted with yellow, that reaches to the shoulders. In both instances large parts of the face were totally destroyed and in those areas new plaster was applied and toned a neutral color. Each held a white cross before him in the right hand. The undergarment of St. Bacchus is red and seems to have been decorated with brocade at the collar. His mantle, buttoned at the center, under his chin, is blue-violet and bordered at the edges by a narrow reddish yellow band and ornamented at each side by a tablion of brocade which was studded with pearls and possibly also with colored gems. The garments of St. Sergius are even less well preserved. His mantle was buttoned at the left and the left side was thrown back over the shoulder to expose the undergarment in a small area on that side. The mantle was green and had at least one tablion of brocade. The undergarment was dark gray, representing blue, and was adorned with a yellow brocade collar and with other ornaments at the shoulder. The maniakia are painted in yellow to represent gold.

267. AN UNKNOWN MILITARY SAINT (Figure 29) The only full-length figure in the north wall of the eastern bay is partially preserved on the narrow piece of wall between the arcosolium and the door.57 Owing to the complete loss of plaster in a wide area that runs diagonally between the two arches, the middle parts of the figure no longer exist5$and the painted surface of the surviving parts is so poorly preserved that much of the 55 That this is the correct terminology in this instance for the necklace as an insigne of military rank (in classical usage more frequently called the streptos) is evident from the account of the "lives" of Sergius and Bacchus in the Synaxarium Constantinopolitanum (09.cit., col. 115) where they are said to have been deprived, on their martyrdom, of their badges of office, the maniakia and zonai (belts). The artistic tradition wherein they wear necklaces is certainly pre-iconoclastic. See the encaustic icon of Kiev and the mosaic of St. Sergius in the church of St. Demetrius, Salonika (supra, note 9). For later examples, see Daphni (Diez and Demus, op. cit., figs., 68, 69); the figure of Bacchus a t MileSevo (A. Frolow, La peinture du moyen dge en Yougoslavie, I [Paris, 19541, pl. 75, 4); the Protaton, Mt. Athos (Millet, op. cit., pls. 40, 41); Lavra, Catholicon (ibid.,pl. 137, 3 ) ; or, the mosaic medallion in the soffit of the western arch of the fifth bay of the outer narthex a t the Kariye Camii, an almost exact duplicate of the fresco portraits in the Parecclesion. I n the Serbian examples a t Staro NagoriEino and Lesnovo (Petkovid, op. cit., 11, pls. 54 and 162) they do not wear the maniakion.. A possible instance of another saint wearing the necklace is found in an icon of Mt. Sinai where the central figure between Procopius and Nestor is identified by Soteriou (op. cit., fig. 47) as Demetrius. I n his text volume (p. 64), however, he admits that the inscription is largely effaced. 56 Dimensions: diameter, .51 m.; haloes, .30 m. 5 7 Dimensions: height, including red borders, 2.265 m.; width a t bottom, including borders, .62 m.; height of figure, 1.82 m.; diameter of halo, .375 m.; height of green zone, .625 m. 58 Other losses of plaster occur in the head and feet.

P A U L A. U N D E R W O O D detail is lost. I t can be seen, however, that the Saint was clad in full armor and that he was portrayed as a young and beardless warrior. He turns his head toward the left and holds his naked sword perfectly vertically in his right hand, his right arm doubled back and extended horizontally from the shoulder. The narrowness of the wall space and the close proximity of medallions at each side seem to have dictated this pose with its emphasis upon the vertical. The position occupied by this unknown Saint is one of great importance since he stands at the head of the rank of full-length figures of saints on the northern side of the chapel and is nearest to the lost figure of Christ that doubtless filled the northern wall of the bema.59 The Saint wore blue-violet pantaloons and his legs and feet were bound in gray leggings tied at the top. His short tunic was blue-black (representing blue), as can be observed in his right sleeve whose cuff was olive green and yellow. The olive green cuirass was reinforced by a gray metal band across the chest. The sleevelets, as can be seen on his right upper arm, were decorated with scale patterns. The mantle, which survives only at the left and over both shoulders, was red. 268. A FRAGMENT OF A MEDALLION OF AN UNKNOWN SAINT (Figure 29) The relieving arch of the door leading into the Prothesis of the Parecclesion (formerly the diaconicon of the comnene church) so nearly fills the remaining space on the north wall of the eastern bay that there was room for only two medallions in the spandrels above. A fragment of one of these still remains, but there is no means of establishing the identity of the Saint, for no part of the bust now survives and only the left side of the halo and the left quadrant of the background of the medallion are preserved. The background was painted in three zones of green, like that of St. Bacchus. 269. A FRAGMENT OF AN UNKNOWN STYLITE SAINT (Figure 30) I n the western soffit and jamb of the arch leading to the prothesis of the Parecclesion is the painting of a stylite SaintG0which is complete save for the head of the Saint himself and his inscription. The column, which consists of the upper part of a column base drawn as though seen from above, a short shaft,

"

In a sense he was, therefore, pendant to St. George, and in more normal conditions occupies the position where that chief of all military saints should have been placed. Among the more popular of the warrior saints there is one who is conspicuously absent from the set in the Parecclesion, namely, St. Nestor, who is often placed in the company of his more celebrated brethren. A t Cefalh, for example, he is placed in the south wall of the chancel along with Sts. Theodore, George, and Demetrius; in the Cappella Palatina, with Theodore, Demetrius, and Mercurius; in the Martorana he is in the northern arch with six other more prominent military saints; at Hosios Lucas he is one of seven military saints in the main arches surrounding the naos. A t DeEani (V. Petkovid and Dj. BoSkovid, op. cit. pl. 78,2), in a fresco that depicts Nestor in a type comparable to our fresco, he is the companion of Demetrius. Here, Nestor is beardless, and has wavy hair which passes behind the ears and almost reaches his shoulders, a type that conforms also to that in the Martorana. However, in the other examples mentioned above, the outstanding characteristic is that the hair is rendered in stringy locks with a ragged outline. The head of the Saint being discussed here is so poorly preserved that the type remains very uncertain and his identification as Nestor cannot be definitely established. Dimensions: height of fragment, including red border beneath, 1.47 m. on the curved surface; width of reveal, .52 m.; width of column shaft, .185 m.; height of green zone, .535 m.

T H E F R E S C O E S I N THE K A R I Y E C A M 1 1 and a capital, is surmounted by a cauldron-shaped vessel from which the bust of the Saint emerges. The Saint is shown frontally, holding a small white cross before him in his right hand, and with his left hand, palm outward, held vertically. The base is painted in cool but dark grays. The shaft appears to have been gray also, but in quite a light value. A wide and very light gray highlight runs down the right side and a narrower brownish gray stripe along the left edge provides a shaded area of modelling. At the base of the shaft is a vertical band, in its center 18 cm. high, which is painted in the cool blue-gray color of the base. At the top of the shaft is a pinkish horizontal band on which the sculptured capital, of the same pink color, is placed. The vessel in which the Saint is placed is grayish green, perhaps terre verte. The Saint is clad in a blue-violet mantle which is buttoned at the breast. Below, the edges of the mantle are slightly parted to reveal a blue-black undergarment. Of the head, only the lower part of the beard survives. This is painted in browns and yellows, indicating that he was an old man. FRAGMENTARY FRESCOES IN T H E DOMES O F T H E PROTHESES OF THE CHURCH AND THE PARECCLESION

The fourteenth-century protheses of the church and the Parecclesion were reconstructions of Theodore Metochites built upon the sites of the earlier prothesis and diaconicon of the comnene church, although not necessarily in the same forms. The Prothesis of the church, to the north of the main apse with which it communicates,61 is covered by a windowless melon dome with eight flutes. In the course of repairs to the masonry of the dome, fragmentary figures of angels were discovered in the consecutive flutes numbered 6 , 7, 8, and 1.'j2 No trace of a medallion is apparent in the summit of this dome, and it is certain that its only decoration consisted of eight figures of angels. These are now so very fragmentary, and their paint is so effaced, that no illustration is provided here. The Prothesis of the Parecclesion63 is covered by a ribbed dome (fig. 31) in eight segments, which is raised upon a high drum. In the latter, three windows in the eastern side still remain and a fourth, on the south side, had been filled. The eight ribs rise from the cornice, through the drum, to a central medallion at the summit. In six of these segments of the upper zone, above the level of the windows, fragmentary paintings of apostles still exist. The decoration of The door between the two was filled by a cupboard during the period of the building's use as a mosque, but this was removed by the Byzantine Institute. 6 z Flute no. I is that on the eastern axis; the others are numbered counting from it in a clockwise direction. 63 I n its north side, a portion of the eleventh-century wall, which is an integral part of the south wall of the bema of the main church, is still preserved. This is penetrated by an arch that had been filled with masonry in the fourteenth century. On removing the fill, fragments of fresco decoration of the eleventh century, consiting of a double armed cross accompanied by the inscription 'I(qooii)~I X(p1m6)~/ N i 1 Ka, were discovered in the soffit. The existence of this door indicates that in the eleventh century the diaconicon occupied the site, and that communication between it and the bema of the main church was provided by the arch. I4

210

P A U L A. U N D E R W O O D

the ribs consisted of meander patterns and foliate designs in alternate ribs. Of the apostles, St. Peter was represented in the eastern segment of the upper zone. Assigning to him the numeral one, and counting the segments in clockwise direction, the fragments of apostle figures in the upper zone are arranged as follows: I. Peter (inscribed), now preserved in little more than the cartoon sketch; 2 . Paul, identified by the remains of his head with the characteristic brow and black hair; 3. Andrew, preserved only in the top of the head, but accompanied by part of his inscribed name; 4. an unknown apostle whose lower extremities are all that remain (upper left, in figure 31) ; 7. an unknown apostle; 8. John the Theologian, identified only by the inscription. Not counting the blocked window to the south, four spaces were available in the lower zone for the accommodation of the other apostles. Of these only the figure in the lower zone of the fourth segment is at all preserved. Even here little more than the underpainting now remains, but he can be described as being turned toward the right, with his right hand extended before him. I n the medallion at the center there was a rather small bust of Christ whose halo very nearly filled the available space. The halo was yellow, and at the left the drawing of one of the arms of the cross is visible. At the bottom of the medallion, i.e. to the eastern side, the shoulders spread out, and what appears to be the cartoon drawing of the hair falls over the shoulder at the right. St. Peter, in the first segment to the east, wore a yellow outer garment, as he always does in his representations at the Kariye Camii. Very little of the surface paint now remains. St. Paul, in the second segment, is turned toward the right to face toward St. Peter who was nearly frontally posed. Paul's tunic is bluish gray, with a bluish black clavzzs, while his outer garment is reddish violet. The apostle in the fourth segment, upper zone, is preserved from the knees downward. He wore a blue tunic and a yellow outer garment. The apostle in segment number seven wore a green outer garment. St. John (number eight) carried a book and wore a red-violet outer garment.

KEY TO PAINTINGS

A. The Paintings in the Vaults, above the Cornice

B. The Paintings on the Walls, below the Cornice

Key to the Paintings of the Parecclesion

T H E F R E S C O E S I N T H E K A R I Y E CAM11 K E Y TO T H E P A I N T I N G S O F T H E P A R E C C L E S I O N s 4

The Paintings in the Vaults and Lunettes above the Cor~ice(Plan A) 201. The Anastasis; First Report, pp. 264-275, figs. 63-75 202. Christ Raising the Widow's Son; First Report, pp. 278, 279, figs. 81-83 203. Christ Raising the Daughter of Jairus; First Report, pp. 275-278, figs. 76-80 204. The Second Coming of Christ; Third Report, pp. 241-254, figs. 1-19 205. The Land and Sea Giving up their Dead; Third Report, pp. 259, 260, figs. 25 and 29 206. The Angel Carrying Lazarus to Abraham; Third Report, pp. 257, 258, figs. 26 and 30 207. Abraham with Lazarus and other Souls in Paradise; Third Report, p. 258, figs. 23 and 27 208. The Rich Man Suffering the Torments of Hell; Third Report, p. 259, figs. 24 and 28 209. The Torments of the Damned; Third Report, pp. 254-257, figs. 20-22,36,38 210. The Entry of the Elect to Paradise; Third Report, pp. 260-263, figs. 31-35 211. The Virgin and Child, medallion; First Report, pp. 281, 282, figs. 86-89 212-223. Twelve Angels of the Lord; First Report, pp. 282-285, figs. 86,go-IOI 224. John Damascene as Hymnographer; Second Report, pp. 175-179, figs. I, 5, g 225. Cosmas the Poet ; Second Report, pp. 179, 180, figs. 2, 7, 10 226. Joseph the Poet; Second Report, pp. 180-182, figs, 3, 6, 11 227. Theophanes as Hymnographer; Second Report, pp. 182-184, figs. 4, 8, 12 228. Jacob Dreaming, Jacob's Ladder, and Jacob Wrestling with the Angel, Second Report, pp. 194-198, figs. 27, 29, 30, 40, 41 229. &losesand the Burning Bush, two incidents; Second Report, pp. 198-200, figs. 28, 29, 31, 42 230. Moses and the Burning Bush, third incident ; Second Report, pp. zoo, 201, figs. 32, 33, 43 231. Bearing the Ark of the Covenant to Solomon's Temple; Second Report, pp. 188, 189; Third Report, pp. 263-265, figs. 37-39 232. Bearing the Sacred Vessels to Solomon's Temple; Second Report, p. 189, figs. 20, 21 233. The King and all Israel before the Ark; Second Report, pp. 189-192, figs. 22, 24, 25 234. Placing the Ark in the Holy of Holies; Second Report, pp. 192-194, figs. 23, 24, 26 235. Isaiah Prophesying; The Angel Slaying the Assyrians before Jerusalem; Second Report, pp. 201-205, text fig. A, and figs. 34-37 6* The positions of all subjects in the Parecclesion can be found by referring, by number, to Plan A or B. References are given to the Preliminary Reports in which the various paintings were described and illustrated. "First Report" = Dumbarton Oaks Papers, 9-10 (Cambridge, Mass., 1956); "Second Report" = ibid., 1 1 (1957); "Third Report" = ibid., 12 (1958).

I4*

P A U L A. U N D E R W O O D 236. Aaron and his Sons before the Altar; Seco~dReport, pp. 205-209, text fig. B, and figs. 38, 39 237. Fragment of a Figure in West Tympanum (David or Solomon?); Second Report, pp. 186-188, fig. 19 238. Souls of the Righteous in the Hand of God (Fragment); Second Report, pp. 185-188, figs. 17, 18 239. Melchizedek the Righteous (medallion); Second Report, pp. 184, 185, figs. 13, I 4 240. Christ (medallion); Second Report, pp. 184-185, figs. 15, 16 241. Christ (medallion) ; Fourth Report, figs. I, 2 242. Archangel Michael (medallion);First Report, pp. 279-281, fig. 84 The Paintings on the Walls below the Cornice (Plan B) 243. Unknown Church Father (Fragment) ; Second Report, pp. 211-216, fig. 46, left 244. St. Athanasius; Second Report, pp. 211-216, fig. 46, right 245. St. John Chrysostom; Second Report, pp. 211-216, fig. 47 246. St. Basil; Second Refiort, pp. 211-216, figs. 45, 48 247. St. Gregory the Theologian; Second Report, pp. 211-216, figs. 45, 49 left 248. St. Cyril of Alexandria; Secofid Report, pp. 211-216, figs. 45, 49 right 249. Virgin and Child (Eleousa) ; Second Report, pp. 217-220, figs. 45, 50, 51 250. St. George; Fourth Report, pp. I ~ ~ ~ figs.I 3,~4,I5 , 251. St. Florus (medallion) ; Fourth Re+ort, pp. 191, 192, figs. 3, 6 252. St. Laurus (medallion);Fourth Report, pp. 191, 192, figs. 3, 7 253. St. Demetrius; Fourth Report, pp. 192, 193, figs. 3, 8, g 254. St. Theodore Tiro; Fourth Report, pp. 193-195, figs. 3, 10, 11 255. St. Theodore Stratelates; Fourth Report, pp. 195-197, fig. 12 256. St. Mercurius; Fourth Report, pp. 197, 198, fig. 13 257. St. Procopius; Fourth Report, pp. 198, 199, figs. 14, 15 258. St. Sabas Stratelates; Fourth Report, p. 200, figs. 14, 16 259. Unknown Saint; Fourth Report, pp. 200-202, figs. 17, 18 260. St. David of Thessalonike; Fourth Report, pp. 202, 203, figs. 19, 20 261. St. Eustathius; Fourth Report, pp. 203, 204, fig. 21 262. St. Samonas; Fourth Report, pp. 204, 205, fig. 22 263. St. Gurias; Fourth Report, pp. 204, 205, figs. z3,24 264. St. Artemius or St. Nicetas; Fourth Report, pp. 205, 206, figs. 25, 26 265. St. Bacchus; Fourth Report, pp. 206, 207, fig. 27 266. St. Sergius; Fourth Report, pp. 206, 207, fig. 28 267. Unknown Military Saint; Fourth Report, pp. 207, 208, fig. 29 268. Fragment of a Medallion of an Unknown Saint ;Fourth Report, p. 208 fig. 29 269. Unknown Stylite Saint; Fourth Report, pp. 208, 209, fig. 30

1. Kariye Camii. The Parecclesion. General view looking East

2. The Parecclesion. View into the Dome from the Southcast

16. Detail, Head of St. Sabas Stratelates (258)

31. Fragmentary Paintings in the Dome of the Prothesis of the Parecclesion

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: Fourth Preliminary Report on the Restoration of the Frescoes in the Kariye Camii at Istanbul by the Byzantine Institute, 1957-1958 Paul A. Underwood Dumbarton Oaks Papers, Vol. 13. (1959), pp. 185+187-212. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C185%3AFPROTR%3E2.0.CO%3B2-O

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 3

Notes on the Work of the Byzantine Institute in Istanbul: 1955-1956 Paul A. Underwood Dumbarton Oaks Papers, Vol. 12. (1958), pp. 269-287. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281958%2912%3C269%3ANOTWOT%3E2.0.CO%3B2-Y 20

In Paper Covers Edgar Dale; T. C. Holy; R. W. Tyler; C. C. McCracken Educational Research Bulletin, Vol. 9, No. 1. (Jan. 8, 1930), pp. 24-25. Stable URL: http://links.jstor.org/sici?sici=1555-4023%2819300108%299%3A1%3C24%3AIPC%3E2.0.CO%3B2-2

NOTE: The reference numbering from the original has been maintained in this citation list.

Notes on the Work of the Byzantine Institute in Istanbul: 1957 Paul A. Underwood Dumbarton Oaks Papers, Vol. 13. (1959), pp. 215-228. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C215%3ANOTWOT%3E2.0.CO%3B2-C Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:38:48 2008

NOTES ON TEE WORK OF THE BYZANTINE INSTITUTE IN ISTANBUL: 1957

HE preceding issue of this series of notes on the work of the Byzantine Institute1 contained some observations on a number of sepulchral monuments in the Parecclesion and the narthexes of the Kariye Camii in the early stages of their cleaning or discovery. At that time the uncovering of three arcosolia in the outer narthex had barely begun and only one of them was sufficiently opened up to permit comment and illustration. While it is still too early to present a detailed study of all the wall tombs, it is felt that a preliminary report should be made regarding the three in the outer narthex in order to supplement, and on certain points to correct, the information in the previous report concerning the tomb in the fifth bay,= and to present an account, for the first time, of the newly uncovered tombs in the fourth and second bays of the outer narthex. I t has always been evident that the early fourteenth-century faqade (the west wall of the outer narthex) of the Church of the Chora was of a very open type with six tall arches that extended almost to the floor (fig. 9). In the interior, the arches penetrated into the tympana above in which scenes of the early life of Christ were depicted in mosaic (fig. 10). When the church was converted to Moslem use all these openings, with the exception of the door, were filled with cut stone and rubble leaving only the small pointed windows, high up in the fills, to illuminate the interior. In the course of restoration work in 1956 it became evident, however, that the exterior

T

Paul A. Underwood, "Notes on the Work of the Byzantine Institute in Istanbul: 1955 and 1956," Dumbarton Oaks Papers, 12 (Cambridge, Mass., 1958), pp. 269-287. Ibid., pp. 279-282, figs. 11-15 and text illustration, p. 281. I n some respects the present report supersedes the earlier account of this tomb, but the reader is referred to it for details concerning the paintings in the soffit of the arch and for illustrations which are not repeated here. 1

sides of three of these great windows (those in bays two, four, and five, counting from the north end) had been filled in late Byzantine times by thin curtain walls of cut stone so placed that half their thickness projected on the exterior plane of the faqade (fig. 9). These walls are so unequal in thickness, and their stones are of such differing dimensions, as to suggest that they were constructed at different times. The curtain walls converted the three windows into great niches (figs. I, 5, and 7), or arcosolia, within which sarcophagi, long since removed, were placed and portraits and appropriate sacred images painted on the back walls and, in one case, in the jambs and soffit of the arch as well. In the Turkish alterations the upper parts of these late Byzantine fills, which formed the backs of the arcosolia, were dismantled to a level of about 2.80 m. above the floor, the small windows placed upon these truncated sections of the Byzantine curtain walls, and new walls built flush with the interior planes of the openings, thus concealing the existence of the arcosolia and their paintings (fig. 10). With these alterations, of course, the paintings in the upper parts of the back walls of the niches were destroyed. The two windows of bays one and six, which had not been made into niches, were filled by the Turks at the same time, and in the same way, that the tombs were filled. THE EARLY

FOURTEENTH-CENTURY

\VINDOWS OF THE FACADE

After opening the windows of bays one and six, which had never been transformed into arcosolia, some aspects of their original treatment became evident, and it is to be presumed that the windows of the other bays, before they were converted into arcosolia, had been treated in the same fashion. All window openings were quite uniform in dimensions. From their sills, which are about 22 cm. above

216

P A U L A. U N D E R W O O D

floor level, to the mosaic surface in the soffits of their arches (figs. 5, 7), they measure between 4.37 and 4.44 m. The widths of their openings from masonry to masonry (their revetments have disappeared) vary between 1.79 and 1.84 m. The thicknesses of their jambs, with the exception of bay six, vary between .88 and .89 m. In bay six this is reduced to .74 m. I t is evident from the presence of metal cramps in the masonry of the reveals that the marble revetments of the walls of the narthex turned in, to line the jambs of the windows. In the second bay small fragments of revetment, still in situ, are wedged between the back wall of the arcosolium and the masonry of the jambs (fig. 8), thus enabling one to determine that in the northern jamb of this window the revetments extended into the reveal to a depth of .635 m. Evidence in the south jamb of the window in bay four indicates a return there of .66 m. In all probability these figures represent an average return in all windows with the exception of that in bay six where the wall is thinner. Balustrades, about 1.20 m. high from the finish level of the sills, had filled the lower parts of the window openings. Evidences of these exist in bays one and six where slots were found to have been let into the lower jambs (fig. 11). I t is likely that similar slots exist also in the other three windows, but these are now inaccessible. The balustrades were capped by moulded copings; a fragment of one was found in the fill of bay six and it exactly fitted an imprint in the south jamb of that window immediately above the slot. While no very clear evidence now exists regarding the form of enclosure, if any, above the balustrades it seems likely that the openings would have been closed by means of grilles or window frames of some kind. If so, these would have been placed against the exterior edges of the marble revetments in the reveals. THE ARCOSOLIUM O F B A Y

5

(Figures 1-4) Some tentative comments on the arcosolium of the fifth bay, and some illustrations of its paintings, were published in the preceding issue of these notes devoted to the season of

1956. This arcosolium3 is unique among the three in the outer narthex in that the paintings were not restricted to the back wall of the niche, but were placed also in the soffit of the arch and in the jambs between the top of the sarcophagus and the cornice (fig. I). The original mosaic ornament in the soffit of the window was concealed behind an archivolt of cut stone voussoirs which rests upon the overhang of the cornice where it turned in to the window. The face of this archivolt and its intrados were plastered and thus provided with surfaces for painting. Below the cornice, the marble revetments in the jambs of the window were removed and plaster support for paintings was applied directly to the masonry. Thus, above the sarcophagus (which rose to a height of about 1.45 m. above the sill), all the surfaces within the arcosolium, with the possible exception of the tympanum of the arch, were painted. While the paintings in the soffit and the jambs are still relatively complete, though in varying states of preservation, only the lower part of the painting of the back wall of the niche still survives. The latter exists t o a height of about one meter above the presumed level of the top of the sarcophagus. The face of the arch was painted in a geometric design. The soffit, however, contains a bust figure of the orant Virgin and Christ child (Blachernitissa) within a cusped mandorla which half obscures four seraphim who peer out from behind. In the haunches of the soffit, flanking the Virgin, are two medallions containing bust figures of Sts. Cosmas of Mayuma (left)and John Damascene (right).4When the Turkish fill in the upper part of the arcosolium was removed, some evidence came to light, just outside the paintings in the soffit, suggesting that a semicircular window may have existed within the arch. If this was the case, the back wall of the arcosolium terminated at cornice level. The panel in the back of the niche (fig. 2 ) measures 1.57 -. m. in width within its borders. The inner border varies between 2.5 and 3 cm. Dimensions: height from sill to soffit of arch, 4.16 m.; height from sill to spring line of arch (top of cornice), 3.41 m. ; width between plaster surfaces in jambs, 1.77 m.; height from sill to top of fragment of painting on back wall, 2.46 m . 4 Ibid., figs. 12 and 13 for illustrations.

W O R K O F THE BYZANTINE INSTITUTE in width, andconsists of a narrow wave pattern of white painted upon a red ground such as was used throughout the arcosolium in framing its various parts, as well as in the tomb of Demetriuss in the inner narthex where it is used in the same way. Outside this frame is a wider border of red, about 5.5 cm. in width at the right side, and 4 cm. at the left. Finally, in the angles between the back wall and the jambs the wave border is repeated, this time as a frame for the paintings in the jambs. At the bottom of the panel, below the wave border, the red band is from 5 to 5.7 cm. in width and below this is a narrow white band, 2.5 cm. wide, which turns downward near the jambs and becomes the border of a horizontal panel, painted yellow, which formed a transition between the sarcophagus below and the portraits above.6 The plaster support for the paintings in the back and jambs was of poor quality. I t was very thin and seems to have been applied in only one coat. Owing to the rough and uneven surface ot the masonry, especially in the fill at the back, the plaster barely covers the more projecting stones and as a result the surfaces are unusually bumpy. There was little protection from seepage of moisture and the paintings have suffered greatly both in decomposition of the plaster and in flaking of surface paint. In many places cleavages had developed between the plaster and the masonry, and these have caused numerous, but relatively small, areas of total loss. The remnant of the portrait panel in the back of the niche (fig. 2) now measures .65 m. in height above its wave border at the bottom. This is a little more than two-fifths of its original height if it is considered to have been the same height as the panels in the jambs (i.e. 1.57 m.). Contrary to impressions published in the earlier report on this tomb,7 the back of the niche contained not two, but at least three, and probably six figures. Three figures stood in the foreground, and parts of the drapery of what seem to have been three other figures are to be seen at a higher level Ibid., pp. 276-278, fig. g. This lower panel seems to have been intended as an imitation of a yellow marble. It corresponds to a similar horizontal panel in the same position in the arcosolium of bay 4 (see infra). Cf. note I.

behind and between those in front, one at the upper left corner, another between the first and second figures from the left, and the third between the second and third figures. In-so-far as it is preserved and visible, the background was green, painted over black, with no evidence of setting or ground plane. The figures in the front rank appear to be those of an adult woman flanked by two children. The child at the left is dressed in a long caftan slit down the center and bound at the waist by a girdle. Owing to the flaking of the paint on this figure, especially in a zone around the edges, one can observe several of the steps taken by the painter in executing this and, perhaps, others of the portraits. Apparently he first put in the black of the background, but in so doing considerably overlapped the boundaries of the figure, a phenomenon frequently observed in other paintings in the Kariye Camii. He then applied green over the complete area intended for the garment. Much of this green paint and its overpainting has come away where it lay upon the black, thus lezving a zone that is now largely black around the edges of the garment. This zone is about g cm. wide at the bottom and about 5 cm. at the sides. Over the green the entire garment seems next to have been painted a bluish gray, possibly smalt. Foliate and floral motifs, representing the pattern on the material of the dress, were then painted over the gray. These consist of large leaf forms in yellow on which are traces of gilding. In the centers of the golden leaves are vermilion dots highlighted with white. Spaced at fairly regular intervals are small vermilion flowers composed of six roundish petals which radiate from a white dot. At the top of the fragment one can see the white girdle, with twisted folds, that marks the waistline of the figure. A white kerchief with fringed ends hangs from the girdle to left of center. I t seems likely that when the painter had completed this and other figures, he then applied the green paint that covers the black of the background. A second child, somewhat larger than the first, was portrayed at the far right, and placed so low in the panel that the lower border seems to cut off the lower part of the skirt leaving no space for the feet. The garment tapers to the waistline of the child at

P A U L A. U N D E R W O O D the top of the fragment. Here, to the left of the garment, there is a small piece of drapery, of the same color and pattern as the rest of the costume, which probably represents the lower edge of the right sleeve, suggesting that the garment had wide sleeves and that the right arm of the figure was extended toward the left. At the far right is another, similar, loop of drapery which is probably a bit of the left sleeve of the garment. Either the garment was first painted black, or, as seems to have been the case in the child at the left, it was painted over the black of the background. At any rate, a light yellowish brown was applied over black and on this a complicated series of motifs were painted to form the pattern of the material. A running vine motif in vertical bands which are spaced at regular intervals is painted in a heavy green paint. Interspersed in the coils of the vine are heart forms in white lines which are surrounded by single rows of red dots painted over black. Between the vertical vine patterns there are vertical rows of crowns alternating with small beasts. Both crowns and beasts are yellow highlighted with white. The crowns are drawn in perspective as seen from below and have three sharp prongs on top and two streamers below. Bluish vines appear here and there in these rows. In the lower left corner of the garment the letter omega is inscribed in white and beneath it what seems to be a ligatured gamma and eta. On the background, immediately to the left, is the merest trace of what appears to have been another letter. In this connection, it should be recorded that another inscription, of six lines, had been painted in the lower left corner of the panel. Regrettably, this inscription is no longer legible. Before discussing the figure of the woman in the center, between the children, it is convenient to make note of the fragments of painted drapery which at first glance might be regarded as pertaining to short outer garments, such as cloaks or mantles, worn by the two figures a t the left in the front row. This, however, seems not to be the case. The draperies in question are those mentioned above.s That these are not outer garments of any of the figures in the foreground can be deduced both from the fact that they do not P.

217.

envelop any of them and from other evidence to be given below. The drapery to the left of the child at the left is mainly painted in browns and yellows. The small brown triangle that projects to the left, below the drapery, appears to be the point of the right foot of a figure whose weight rests upon the left leg (invisible behind the child). The right foot seems to push the hem of the skirt out toward the left and the folds of the drapery accord with this posture. The piece of drapery found to the right of the same child has a strong vertical line in black which appears to mark the right-hand edge of a piece of fur trimming that comes down to a point. To the right of this line there is a wide hem of yellow, and above it the garment is green with yellow ornaments suggesting a vine motif. To the left of the fur trimming is a triangular area of reddish brown paint on which is part of a guilloche pattern in yellow. This probably represents the outer garment, while the fur probably served as its trimming, or possibly as a lining. If that is the case, the inner garment was green with a golden hem. The piece of drapery to the right of the central figure again has a wide yellow hem at the bottom and above it, and to the right is a vertical band of yellow, possibly the edging of the left side of an outer cape or mantle. To the left of this edging the garment is a dark greenish gray on which is a gridwork of yellow stripes. These three isolated pieces of drapery do not accord with one another in color or design and the directions of their hems are such as to make it unlikely that they pertain in any way to the figures of the front rank. The most important figure, and one which supplies evidence in the identification of the occupant of the tomb, is that of a woman who stands slightly to right of center between the two children and at a somewhat higher level than they, as though on a plane immediately behind them. She is attired in a vermilion caftan, open down the front, which bears a pattern of golden monograms within medallions and small heraldic birds with outspread wings. A sketch giving details of the monograms, and a more detailed description of the costume were published in the "Notes: 19 j 5-19 j6."9 As was noted there, three monograms are used in the pattern, O

Op. cit., p. 281.

W O R K O F T H E BYZANTINE INSTITUTE each recurring several times. These monograms can now be more definitely identified than they were in the earlier publication, and form the basis for identifying the personage who wears the costume. While they are linked together in horizontal and vertical rows, the monograms are to be read in the vertical sense. The one which appears most frequently, and in each of the four vertical rows, is easily identified as that of the P a l a e o l ~ g o i .In ~ ~the first and third rows, counting from the left, this monogram alternates with one based upon the letter alpha which is to be identified as that of the powerful family who signed themselves ~ a l a e o l o ~ Asanaioi, oi the dkscendants of the union, i n the year 1278, of Eirene Palaeologina, daughter of the Emperor Michael VIII Palaeologos, and the ex-Tsar of Bulgaria, Ivan I11 Asan, which produced four sons and two daughters.ll The best evidence for interpreting this monogram as one of the versibns used by the Asanaioi12 is its close resemblance to that which appears on the "shroud" of a still later descendant of these families, Maria Mangop (d. 1476), second wife of Stephen the Great, Voivode of Moldavia, whom Gabriel Millet has identified as the Maria Asanina Palaeologina who is recorded in the dedicatory inscription on an icon offered by that princess to the Monastery of Grigoriou on Mt. Athos.13 The monogram on lo For references to other examples, see "Notes: 1955-1956," op. cit., p. 281, note 30. See also the Poutna "shroud" of Maria Mangop referred to infra, note 13, and the embroidered fifteenth-century book cover at Grottaferrata (A. hlufioz, L'art byzantin a l'exposition de Grottaferrata [Rome, 19061, p. 142). l1 Many of their descendants regarded themselves as members of the dynastic family and figureprominently in Th. Papadopulos' genealogy of the Palaeologoi (Versuch einer Genealogie der Palaiologen [Speyer, 19381, nos. 44-46, 18, 21, etc.). l2 Another version appears on a costume in the arcosolium of the fourth bay, infra, p. 225, and fig. 6. l3 The "shroud," or embroidered silk cover for the tomb of the princess, is preserved in the treasury of the monastery at Poutna in Roumania. I t is best reproduced by 0. Tafrali, L e trLsor byzantin et roumain d u monasthre de Poulna (Paris, 1g25), pls. 43-45. The "shroud," of red silk embroidered in threads of gold, depicts the dead princess in ceremonial robes, crown, and pendants, within an arch reminiscent

219

the "shroud" is composed of an alpha of the type with a long horizontal bar at the top and a pendant cross-bar shaped like a Y. Attached to the left side of the alpha is a reversed sigma. These details are common also to the version painted on the costume here being discussed. But, whereas the Poutna "shroud" has a letter attached to the right of the alpha that can be viewed either as an eta or a nu, or both, and thus possibly gave the complete spelling of the name ACANHC, the examples on our fresco, with one exception, place a second sigma on the right. Nevertheless, it should be read as ACA[NH]C with the nu and eta omitted, but giving the two sigmas of the name. The one exception, the central monogram in the third row from the left, seems to have a nu attached at the right, but presents it in reverse, just as the sigma is reversed. This identification probably carries with it similar monograms on the costumes of two other women, one in the southeastern arcosolium of the Parecclesion which can still be very dimly discerned on close inspection and which was imperfectly sketched by A. Riidell, and the other in the arcosolium of the fourth bay (i.lzfra).14Those tombs, it would appear, contained the portraits of still others of the Palaeologoi Asanaioi for their costumes also bear the Palaeologan monogram. The second row adds a third monogram to the series of family names of the personage portrayed in the fresco and repeats all three in sequence. The best preserved example of of a wall tomb. The Palaeologan monogram appears twice among the embroidered ornaments of the arch. The whole is framed by a dedicatory inscription. Two of the corners are adorned by the double-headed eagle in medallions, and the other two by medallions containing the Asan and Palaeologan monograms (pls. 44, 45). G. Millet and H. des Ylouses, Broderies religieuses de style byzantin (Bibl. de 1'Ecole des Hautes etudes, Sciences Religieuses, 55) (Paris, 1g47), pp. 78-81, give the evidence for the identification of the Asan monogram and correct the error of Tafrali in reading it as "Marias." The inscription on the icon of Grigoriou is published by G. Millet, H. Pargoire, and L. Petit, Recueil des inscriptions , 5I I , chre'tiennes d u M o n t A thos (Paris, ~ g o q )no. p. 175. For genealogical information regarding Maria Mangop, see Papadopulos, op. cit., nos. 127, 127a. l4 Cf. "Notes: 1955-1956," op. cit., p. 274 and note 12; infra, p. 225, and fig. 6.

P A U L A. U N D E R W O O D this third monogram is found at the lower right corner of the costume, in the fourth row, where it seems to have alternated with the Palaeologan monogram. This monogram is arranged in cruciform fashion. At the top is the letter rho, and below it an alpha; at the right is a circular letter, probably an omicrolz, and at the left alambda-The monogramshould be regarded as that of the ' P A O ~ Afamily who, like the Asans, were allied to the Palaeologoi. There is also a marriage connection between this family and the Cantacuzenoi and, most interestingly, with the family of Theodore Metochites, ktetor of the Monastery of the Chora, thus bringing the Metochitai into blood relationship with these families, including the dynastic family. These relationships seem to stem from the marriage, about 1261, of the Protovestiarius John Raoul to Theodora Palaeologina,16 daughter of John Cantacuzenos. Their daughter Eirene was married, about 1290, to Constantine Palaeologos Porphyrogennitos,16 third son of the Emperor Michael VIII. In March of 1291 Constantine was cast into prison, for plotting against his brother Andronicus 11, and there spent the rest of his life. Before his death, Constantine took the monk's habit under the name of Athanasius and on his death, May 5, 1306, he was buried outside the Monastery of Lips. The date of the death of his wife Eirene Raoulaina Palaeologina is unknown, but it is recorded that she was buried in the Monastery of the Chora.17After the death of Constantine, the Emperor Andronicus I1 opened his court to his nephew, John, the only child of the marriage of Eirene and Constantine, and elevated the young man to the rank of Panhypersebastos, and ultimately to that of Caesar.ls Andronicus also arranged, about 1307, for John's marriage to Eirene, the daughter of Theodore Metochites, ktetor of the Monastery of the Chora. The relationship between the Asans and the Palaeologoi has been mentioned above. I t is evident also that the Raoul and Asan families were interrelated, for we find that among the relatives of Theodora Palaeologina, a niece of the Emperor Michael VIII, and Papadopulos, op. cit., no. 34. no. 3 7 . l7 See ilzfra, p. 221 and note 21. ls Papadopulos, op. cit., no. 38. l5

l a Idem,

foundress of the Constantinopolitan monastery of Our Lady of Good Hope, a Manuel Comnenus Raoul Asan, Grand Primikirios, is portrayed on one of the miniatures in the typicon of that monastery,lVwhere the ini ~ scription identifies him as yayrrpbs ~ iKTTJTOpiav.

If we turn now to the paintings in the jambs of the arcosolium, it would appear that we are dealing with a tomb of a person who was related in some way to Theodore Metochites as well as to the other illustrious families mentioned above. The painting in the jamb at the right (fig. 4) presents the figure of a nun (1.47 m. in height) standing within a framed panel (.58 x 1.57 m.) between the lid of the sarcophagus and the cornice above. She wears a tightly fitting headdress of black which covers all but the oval face and falls over the shoulders. The face now retains little of its flesh color, perhaps owing to loss incurred when the head was obscured by yellow paint. In the center of the face is a Byzantine repair and restoration that was carefully executed and still retains more of its paint and detail of drawing than the original parts of the face. The repair includes all of the eye at the left, most of the eye at the right, including the eyebrow, and all but the tip of the nose. The mouth, which is original, still retains some drawing and some of the red of the lips. Below the strip of lost surface paint that extends across the torso, the painted surface is relatively well preserved. At the two sides hang the folds of her mantle which is drawn in black, but painted in a dark reddish brown. The long dress, worn beneath the mantle, reaches the feet and is painted a light yellowish brown somewhat like a dark raw sienna. Over this, the drawing is in the reddish brown that was used in the mantle for the drawing. This color is also used for modelling the folds. The lower parts of the sleeves of the tunic emerge from behind the vertical edges of the open mantle at a point just below the strip of lost paint. The feet, which barely show below the dress, are painted black. Facing the nun, in the jamb at the left, is l9 Fol. 5 of Ms. gr. 35 of Lincoln College, Oxford, now in the Bodleian Library. Cf. H. Delehaye, Deux typica byzantim de 1'LPoque des Paleologues (Brussels, 1921), p. 13.

W O R K O F THE BYZANTINE INSTITUTE the standing portrait of a monk (1.51 m. in height) in a framed panel that measures .57 x 1.58 m. within its borders (fig. 3). The most serious losses of paint are in the upper parts of the figure, especially in an irregular strip at about the same elevation as the corresponding area of loss in the figure of the nun. In this figure, also, the area about the eyes was damaged. The monk wears a beard of moderate length and a black hood that covers the head and shoulders. The colors of the tunic and mantle are the same as those in the corresponding figure opposite. A black strip, which may represent part of the scapular hangs below the waist. These two portraits were accompanied by inscriptions. That for the nun (text figure)

probable that the nun Athanasia, who, according to the inscription, seems to have stood in some relationship to a ktetor, was also the personage whose family names are revealed in the monograms of the Raoul, Asan, and Palaeologan families in view of the fact that within the genealogies of these families an Eirene Raoulaina Palaeologina was the mother-in-law of the daughter of Theodore Metochites, the latter being referred to, in all probability, as the ktetor in the inscription of Athanasia. The hypothesis that this is the tomb of Eirene Raoulaina Palaeologina seems to be confirmed by a letter of one Demetrius Raoul Kabakes to his son, which was published by Leo Allatius.21 The writer, who claims descent

Inscription accompanying Portrait of the Nun Athanasia was inscribed in four short lines on each side of the head. On the left, where it is quite well preserved, the inscription reads: 'E~o~pfi&I [fi] 8oiihn TOG ~ ( E o ) G ['A]9avaufi[a]. I t continues at the right, where it is only partially legible : pova[x?] . . . . . . . . .p[o]G TOG. .. . .KT j ~ w [pos] ... .20 Regrettably, the inscription of the monk, which seems also to have been composed in four lines on each side of the head, is almost completely effaced. Nothing at all can be seen at the left, and at the right only a few letters in three of the lines are still to be seen. However, the third line contains .. . ~ a 1~ ~ [ f i ~?]. w p I t is likely that in accordance with common practice the monk and the nun Athanasia would also have been found among the personages portrayed in their court vestments in the back of the niche. In that case it is most 20 "Here sleeps the servant of God, Athanasia of founder.. ." the nun

...

...

.

from the Raoul and Metochites families, recounts what he had heard from his elders regarding some of his more illustrious ancestors. The pertinent passage, for our purposes, is this: "[Theodore Metochites] begot five sons and one daughter, and for [the latter] he obtained as his son-in-law the Sebastocrator [John Palaeologos] son of Constantine Palaeologos Porphyrogennitos, son of the Emperor Michael, who [i.e. Constantine], being born to royal rank, rightly sought the succession to the throne.. . . His wife was buried in our family's monastery of Christ in Chora, near the tomb of the ktetor and [that of ?]her husband. Her tomb conI n Roberti Creightoni apparatum, versionem, et notas ad Historiam ConciliiFlorentini scriptam a Silvestro Syropulo . . . exercitationes (Rome, 1674)~ pp. 616-621.I have not seen this very rare publication, but the document is republished ? T 6 p o ~A', by K. Sathas, M ~ a a l w v l ~BlPAiotWl~q, B u l a n l v h ' A v ~ K ~ o(Venice, T~ 1872),pp. p~$'-pAa'.

P A U L A. U N D E R W O O D tained the following inscription: 'The wife of Constantine Palaeologos, Porphyrogennitos, son of the most high Emperor of the Romaioi, Eirene Raoulaina Palaeologina.' "22 Admittedly, the document as a whole contains numerous errors and confusions, especially those of mistaken identities. Yet what has been quoted here seems to be historically correct, and on such a point as the place of burial of Eirene Raoulaina Palaeologina there is no reason to question its accuracy. Moreover, the document and the epigraphic evidence, especially the monograms, of the tomb itself seem to substantiate one another and make it virtually certain that the tomb in bay five of the outer narthex is indeed that of Eirene Raoulaina Palaeologina. From the point of view of chronology this is perfectly possible. While the date of Eirene's birth is not known exactly, it could not have occurred before 1 2 6 2 ~and ~ not much later than 1272, for in 1290 she was married to Constantine Palaeologos. Theodore Metochites' rebuilding of the church of the Chora must have been completed before 1320 and the windows of the outer narthex would probably not have come into demand as places for the construction of tombs until about would thus have been 1325 to 1 3 3 0 . ~Eirene ~ not much younger than 55 nor much older than 65 when the construction of tombs in the 22 . . . EITo~KE0 6 P ~ET' a h b v I T ~ V T E Wio\i5 K U ~8vycrripa wiav, ~ i 5c v Efipov y a ~ p p b v~ b voepamo~vpoOKwv~ p h o p a vuibv , wkv ~oOllopcpwpoy~vv~~ow mav~ivowTOOIlaAatoh6yow, WIOO66 TOO Pauthiw~ KWPOO M t x d h , 8s E Y E V V ~ ~ VET& ~ ~ V T ~ PS a u l A ~ i aT ~~ V &$iav, 65 ~ a Ei Z ~ ~ T E Iy~y&iw~Tilv 6ia60xjv T ~ S paoiheia~61~aiws. . . . 'H 68 y v v j &oO E-tacpq Ev T+ fiwCjv K ~ TOO I yivou5 fiwCjv wovau~fip~ov T ~ V X p ~ m b v~ j Xcjpav v E y y b ~TOOpvqy~iowTOOKT$~Opo5 ~ a ovwPiow i &(s. E ~ X68 E Ixtypacpjv 6 ~ & c p o ~

afiTq5 T O I ~ V ~ E .

'H orjlwyo~TOO llop~wpoy~vvfi~ow, KUPOOKWVmav-rivow TOOllahaioh6yow ~ a uioO i TOG OyqhoT ~ O P WaotABw~~ C j v 'Pwyaiwv' Hpfivq Paorjhatva 4 llaha~ohoyivq. 23 Her mother, Theodora Palaeologina Cantacuzene (Papadopulos, no. 34) was widowed in 1261 when her first husband, George Muzalon, died. Thereafter she married the Protovestiarius John Raoul, Eirene's father. 24 The tombs in the mortuary chapel (the Parecclesion) would doubtless have been assigned, but not necessarily occupied, before the windows of the outer narthex would have come into use as places of burial.

windows of the narthex began. A date of ca. 1330 should, therefore, be assigned to the paintings in the arcosolium of the fifth bay. In the light of the relationship between Eirene Raoulaina Palaeologina and Theodore Metochites, some of the lacunae in the inscription to the right of the nun Athanasia (who should be regarded as Eirene herself in the habit of a nun), might be filled as follows: ('ABavaufia) wovaxfi wq(~fi)pTOO yayi~poGr o c

... .

. . . .25

KT~~TWPO~

The questions arise as to the identity of the monk in the jamb at the left, and as to the large number of figures in the portrait panel in the back of the niche. As was the case in the Tornikes monument,26 where Tornikes and his wife are represented as monk and nun in the jambs, the figure in the left jamb should be regarded as that of Eirene's husband in the habit of a monk. I t is known that Constantine Porphyrogennitos became a monk before his death in 1306,~'and that he took the name of Athanasius. However, it is recorded that he was buried in or near the church of the Monastery of Constantine Lips.28 I t is not certain, therefore, that he was actually buried (or reburied) with his wife at the Chora although the text of Demetrius Kabakes suggests that he was. In that case the body would have been brought there from the monastery of Lips. I t is hardly possible, on the other hand, that all six personages represented in the portrait panel would actually have been buried in one rather small tomb, and it seems likely that the painting should be regarded as a group portrait of the deceased with members of her family.29 "[Athanasia], the nun, the mother of the son-in-law of . . . founder.. . ." Some faint traces of letters, or parts of letters, which are not positively identifiable fit into such a reconstruction of this part of the inscription. 26 Cf. "Notes: 1955-1956," op. cit., figs. 5-7, and pp. 2 73f. 27 Papadopulos, op. cit., no. 37. 28 Ibzd., p. 23. Pachymeres ed. Bonn, 11, pp. 424f., describes his funeral. 29 No comparable example of such a large number of portraits in a tomb is known to me in Byzantine art, but one is reminded of the extensive series of portraits of parents, husbands, sons, daughters, granddaughters, and in-laws (twenty portraits in all), on ten folios of the typicon of the foundress of the monastery of Our Lady of Good Hope, Theodora Palaeologina (cf. supra, p. 220, and note 19). In the south-

WORK O F THE B Y Z A N T I N E I N S T I T U T E THE ARCOSOLIUM OF BAY

4

(Figures 5 and 6) I n converting the window of the fourth bay into a wall tomb,30 the ornamental mosaics in the soffit of the arch, above the cornice, were left undisturbed and exposed (fig. 5). The undecorated band of setting bed between the mosaic ornament and the masonry fill in the arch was removed and a band of fresco ornament, consisting of a folded ribbon pattern in green and orange-yellow, was substituted to provide a transition between the mosaic and the fresco in the tympanum of the arch.31 The inner face of the Byzantine fill, which formed the back of the niche, was placed on a line with the termination of the marble revetments in the reveals, i.e., a t a distance of .66 m. from the interior face of the wall of the narthex. I n places one can see where the mortar of the masonry of the Byzantine fill accommodated itself around the back edge of the revetments, thus proving that the marble revetments, a t least down t o the top of the sarcophagus, had been left in situ when the arcosolium was constructed. With the exception of the ornamental border placed in the soffit of the arch the paintings in this arcosolium were, therefore, restricted t o the back of the niche. The evidence is that, unlike the other two arcosolia of the outer narthex, the wall and its painting in this tomb continued upward to fill the arch completely, leaving no window in the top. Moreover, had this not been the case there would have been no place in the eastern arcosolium of the Parecclesion of the Kariye Camii there are four portraits, possibly of three personages, one of whom may have been portrayed twice (see "Notes: 1955-1956, "ofi. cit., fig. 10). In the tomb of the fourth bay of the outer narthex (infra, p. 223, and fig.6), a family of three are represented. In these instances it is not a t all certain that all persons who are portrayed were actually placed in the tombs. 30 Dimensions: height from sill to soffit of arch, 4.37 m.; height from sill to spring line of arch (top of cornice), 3.38 m.; width between masonry jambs, I .79 m.; height from sill to top of fragment of painting, 2.38 m. 31 A detail of the fresco ornament was reproduced in "Notes: 1955-1956,'' op. cit., fig. 15, where a fragment of the painted plaster is shown turning onto the tympanum of the niche.

223

tomb for the painting of images of the sacred persons which must, therefore, have been placed in the tympanum above the group of portraits. The paintings began immediately above the top of the sarcophagus a t a level of about 1.21 m. above the sill, and from that level they are relatively well preserved to a further height of 1.17 m., or approximately half the original vertical dimension of the panel of portrait figures including its lower borders. Beneath the zone of standing figures (fig. 6) is a long horizontal panel, about 22 cm. high, painted a dark violet color, seemingly in imitation of porphyry. This is framed by a white band about 3 cm. wide, and by an outer border of red that extended a t the sides t o the marble revetments in the jambs and below t o the lid of the sarcophagus. The zone of figures was originally 1.675 m. in width, including its borders. The field of the painting itself, within the white line that formed an inner frame, measures 1.595 m. in width. The painting presented the portraits of three persons, probably a family consisting of a man, wife, and child. No attempt is made t o represent a ground plane or a setting of any kind, and as far as the painting is preserved the background is uniformly green. This color appears below the figures, a t the two sides, and between the man a t the left and the child in the center. Like the portrait panels in the southeast arcosolium of the P a r e c c l e s i ~ nand ~ ~ in the tomb in the fifth bay (supra), the mode of presentation is the traditional one in which those portrayed are aligned more or less frontally against a neutral background with no attempt t o depict a scene. They are in a class with such portraits as are found, for example, in Ms. gr. 35 of Lincoln College, Oxford,33where the figures, usually in pairs, are depicted against gold backgrounds, and small bust figures of the Virgin and Child, or Christ alone, are placed in the center above the heads of the fulllength figures; or with numerous examples of imperial portraits such as those collected and published by Lampros.34 The man, a t the left, stands isolated from Ibid., fig. 10. Cf. supra, p. 220 and note 19. 34 S. P. Lampros, A E O K ~ M p u~l a v r l v Q v &oKpcrr6pwv (Athens, 1930). 32 33

224

P A U L A. U N D E R W O O D

the other two figures who are grouped at the right, and thus occupies almost half the width of the panel. While his garments are represented as seen from the front, the feet are drawn in profile. From this one would infer that his head, at least, would have been turned to some extent toward the right. He wears a long heavy outer garment, or coat, which is open down the front. It has long sleeves that hang empty at the sides, indicating that it was worn like a mantle, hanging free from the shoulders. The tubular ends of the two sleeves can be seen at his sides.35 This garment is black with an allover pattern of alternately large and small interlacing medallions. The large ones contain the Palaeologan monogram, the small ones a quatrefoil motif. Large cusped spaces are left in the pattern of interlacing medallions, and these are filled by the heraldic doubleheaded eagles, the Byzantine imperial arms. All the ornaments are executed in gold leaf outlined with black lines. The lining of the garment is exposed to view in three places: in the vertical fold at the left where it has been turned back, in the back of the garment at the bottom, as though seen from below, and in the openings of the sleeves. In all cases the lining is painted in violet colors which are graded from light violet at the edges to darker values within. The inner garment was probably a long tunic with tight sleeves, open down the front and held together at the waist by a girdle, such as one finds in the men's costumes in the series of portraits in the Lincoln College miniatures mentioned ab0ve.~6This tunic is vermilion with a large pattern in gold consisting of a series of ogive curves which form segments of a circle surrounding a large and solid leaf-like motif in the center.37 The 35 A very good representation of a garment of this type is to be found on the "shroud" of Maria Asanina Palaeologina, now preserved in the Roumanian monastery of Poutna, mentioned supra, p. 219 and note 13. The Princess is dressed in a long coat which had extremely long empty sleeves that reach the feet. At the shoulders are slashed openings through the upper part of the sleeve through which her arms protrude. 36 Cf. note 19. 37 The pattern in gold is in general analogous to those appearing on the men's costumes in the Lincoln College manuscript, but still more

vertical and horizontal stripes, which seem to mark the edges of the garment, are painted a dark violet. Much prominence is given to the feet, both in size and in the descriptive manner of representation in profile. They seem to be clad in a type of buskin, and are at present the color of yellow ochre.38 Between the man at the left and the woman at the right stands a child who wears a heavy caftan which seems to have been parted down the front. I t is held together at the waist by a girdle from which hangs a kerchief. The field of the caftan is dark red. On this is an over-all diaper pattern in gold leaf outlined in black. The gold of the diaper is shaped into pointed ovals, and where these cross there is a small circle of gold. The girdle is knotted in the center and is made of two strands, each composed of multicolored stripes of dark red, vermilion, yellow, and white. On the white stripes are small red ornaments, while on the yellow and vermilion stripes the ornaments are green. The kerchief is white with light gray folds. I t is bordered at the ends in gold and has golden fringes. On the gold border are small scroll designs painted in black lines. The lining of the caftan, seen at the bottom as though viewed from below, is vermilion. On the feet, which had originally been painted much larger, but were reduced in size by repainting, the child wears white shoes bearing a design in red lines. Although the left side of the figure of the woman is obscured by the child, it is evident that she wears a dress made of a material with a foliate pattern, and over it a cloak with beautifully executed monograms. A broad strip of the dress is visible immediately to the right of the child. I t is of a black material that is slightly lighter in value than the black in the lightest parts of the outer garment of the man. This long garment, which almost covers the feet, has a wide golden hem at the bottom, about 12 cm. wide, bordered by two black lines at each side. The golden field of the hem closely comparable to that on the coat on the Poutna "shroud" described by Tafrali ( o f . cit., p. 5 2 ) as having been of red silk with embroideries in gold thread. 38 Since, under the Turks, the area of the feet had been daubed over with yellow paint, it is not clear what the exact color was originally.

W O R K O F T H E BYZANTINE I N S T I T U T E is decorated by three rectangles in red, one of which is surrounded by four small triangles, drawn in black lines, while above and below the other two there are small circles, also drawn in black lines, within which flecks of white paint still adhere. The over-all pattern on the black dress consists of leaf forms arranged in oval shapes, and in the center of each oval is a large red dot surrounded by a white circle. The leaves are rather sketchy and are painted in white and in yellow. Appearing below the golden hem are the points of the shoes which are also of gold. The vermilion and gold outer garment of the woman appears at the far right. I t would seem to have been a kind of mantle, similar in form to that worn by the woman in the southeast arcosolium of the Parec~lesion.~~ Like it, this mantle is covered with large monograms contained within circles. At the bottom is a golden hem, between g and 10 crns. in width, which is bordered at top and bottom by two black lines. Within the field of the hem rectangular gems and circular pearls are represented. The vertical edging, also of gold, and about 4 crns. in width, is interrupted at intervals by blue and red rectangles, and between them pearls are spaced in two rows. The vermilion field of the garment is mainly decorated by rows of great circles, each containing a monogram. The circles and the monograms are in gold leaf outlined in black. The spaces between the medallions are filled by blue ornaments consisting of four fleursde-lis forming a cross. The row of circles at the left contains three monograms and a small part of a fourth above. Each of the three is different, and they are repeated, apparently in the same order, in the vertical row at the right, where they are arranged so that the monograms in downward diagonal rows, from left to right, are identical. The first of these, at the top left, is partially cut off at the left by the golden edging of the garment. Its left part, however, is supplied by its counterpart, the second monogram in the row at the right. I t is arranged in cruciform fashion and is composed of the letters delta (at the top), rho (below),mu (left), and kappa (right). I t is probable that the letters should be read in that order. If so, they supply the a9

I5

Cf. "Notes: 19j j-1956," oP. cit., fig. 10.

name A ( E ) ~ ~ ( O ) K ( U ~ The T ~ ~ ) s. econd monogram in the row at the left is based upon a large alpha of the type used in one of the monograms in the tomb of Eirene Raoulaina Palaeologina in the fifth bay. In the present instance, however, what appears to be a minuscule sigmu is attached to the alplza at the left and a rather large nu at the right. This would seem to be a variant form of the monogram of the Asan family. The third monogram is the familiar one of the Palaeologoi executed in precisely the same form as that on the costume of the man at the left. There is very little documentary material relating to the Dermokaites family and their relationship to the ruling house of Constantinople. A document of August 1400, which is incorporated in the Acta of the Patriarchrecords a Theodora ate of Con~tantinople,~~ Palaeologina Dermokaitissa who was, by then, a widow. One gathers that she had at least one child (a daughter) who had married Nicholas Branas by whom she had a son who was still an infant in 1400. By that date the daughter had also died. As Papadopulos points out, it is not possible to determine from this single source whether Theodora's husband was by birth a Palaeologan or a member of the Dermoka'ites family who married a Palaeologina. Papadopulos assumes that a strategos Demetrius Palaeologos Dermoka'ites (who flourished ca. 1440) was a son of Theodora, but for this there is no evidence in the sources. THE ARCOSOLIUM O F BAY 2

(Figures 7 and 8) The arcosolium of the second bay (fig. 7)41 differed in several important respects from all others in the Kariye Camii, notably in the iconography and style of its painting. Its back wall, as seems to have been the case also in the tomb in the fifth bay (cf. supra), extended only to the level of the cornice. The arch, above the cornice, was filled by a semicircular window, but, unlike the arcosolium of the fifth bay, there were no paintings in 4O Miklosich and Miiller, Acta et difilonzata graeca medii aeui, 11,p. 420; Papadopulos, op. cit.,

nos. 140, 141. 41 Height from sill t o soffit of arch, 4.44 m.; height from sill t o top of cornice, 3.46 m.; width between masonry jambs, 1.79 m.

226

P A U L A. U N D E R W O O D

the soffit of the arch or in the jambs, and the only place available for portraits and images of the sacred persons was the panel in the back wall, between the sarcophagus and the cornice, where an enthroned Madonna and Christ Child are depicted with the deceased in a common setting. he early fourteenth-century mosaic ornaments of the original window were left intact and exposed when the window was converted to use as a sepulchral monument. As in the tomb of the fourth bay, a narrow band of fresco ornament was added along the outer edge of the mosaics in the soffit. This was done to extend the surface of ornament to the window frame. When the Turkish window and its surrounding Turkish masonry, above the stump of late Byzantine wall, were removed it was found that the outer edge of the band of fresco ornament terminated neatly along a groove in the mortar (fig. 12) which preserves the trace of the window frame that once filled the arch. The band of fresco ornament is very much like that used in the same place in the arch of the arcosolium of the fourth bay: a motif of folded ribbon in orange-yellows and in greens upon a field of black. In this instance, however, the ribbon was not trimmed with white lines, and the white dots in the black triangles at each side, which appear in the ribbon of the fourth bay, were omitted here. That there could have been no paintings in the jambs of the niche is evident from the fact that the painting on the back wall abuts neatly at each side against the broken remains of marble slabs that once covered the masonry of the jambs of the fourteenth-century window (fig. 8). The remnants of these revetments are found only as far down as the line of the top of the sarcophagus. On the right side of the painting, the marble fragments penetrate beyond the painted surface to a depth of 6 cm. This means that the inner face of the masonry of the back wall overlapped the revetments by about 4 cm. The wall at the back of the niche rose vertically to a height of about 3.46 m. (the top of the cornice) above the sill. I t is preserved to a maximum height of 2.62 m. At a level of 1.14 m. above the sill is the bottom of a horizontal course of roughly shaped stones which project about 8 cm. and vary in thick-

ness between 8 and 10 cm. On these there is a sloping bed of mortar containing pieces of marble, stone, and brick which builds up to a level of 1.305 m. From this level, the painted panel now survives to a maximum vertical dimension of 1.315 m. The top center of the fragmentary painting was cut away to form a rectangular slot, about 20 cm. in height, to accommodate the sill of the Turkish window. The original width of the painting between the revetments of the jambs is 1.703 m. In preparation for the paintings, the back wall above the sarcophagus was first given a thin coat of hard and impermeable pink mortar, containing quantities of pulverized terra cotta, similar to mortars used in Byzantine masonry construction. This is the only instance of the use of mortar of this kind, beneath the lime plaster support for paintings, that has been found at the Kariye Camii. In the lower corners of the painting two areas of this mortar can be seen where the lime plaster for the paintings has come away. The total thickness of mortar and plaster averages about 2 cm., the plaster itself accounting for about 1.5 cm. of the thickness. The only painting in this arcosolium was the rectangular panel (fig. 8) in which the representations of the sacred persons and the deceased are combined in a, scene of devotion entirely unlike the paintings in any other sepulchral monument in the Kariye Camii.42 At the left, the enthroned Madonna and Child are placed in angular perspective to face right toward the standing figure who turns toward them in three-quarters view. No distinction of scale is made between the figures who are here depicted in such intimate relationship. The panel is framed at each side by the usualnarrowwhite line andared border, 3.5 cm. wide. The same red border, but considerably wider (10.5 cm.), passes beneath the panel. Across the bottom, however, the narrow white line has been omitted and instead the front edge of the marble floor (4 cm. thick) of the scene above is represented as resting directly, as it were, upon the lower frame of the picture. 42 Cf. figs. 2 and 6, and "Notes: 1955-1956," figs. 5, 9, and 10. In the other tombs the two categories of portraits and holy images are kept quite separate, usually in different parts of the niche, and no psychological interplay exists between them.

W O R K O F THE BYZANTINE INSTITUTE The Virgin is clad in a red-violet outer garment which almost completely covers her blue tunic. The latter is visible in a small area on the wrist and a t the lower left where it touches the top of the footstool and casts its shadow upon it. At the edge of the redviolet outer garment, where it rises to pass over the extended arm, is a narrow edging of golden yellow. Very narrow yellow lines were also used to mark some of the sharp edges of the drapery folds in the lower left which fall over the front of the throne. Otherwise, the highlighting on the folds of the garment are in light gray or white. The shoes of the Virgin are red. The Christ Child, who seems to have been seated on his mother's left knee with his left leg drawn up, was clad entirely in golden raiment. The garment covers all that now remains of the figure with the exception of the feet. Part of a train of drapery can be seen at the right as it falls behind the Virgin's left knee. The effect of cloth of gold is achieved with yellow paint on which there is a considerable amount of hatching in thin light yellow lines. The Child wears sandals whose thongs are now only faintly visible. The wooden throne and footstool are yellow in three values: darkest on the righthand receding planes, slightly lighter on the left, and lightest on the top surface of the stool. The drawing is executed in a dark value of the same color. The cushion on the seat of the throne is red. At the right is the sombre but richly attired figure of the deceased, probably a woman if one can judge from the costume and the smallness of the feet. Clad entirely in black, she wears a long plain mantle which hangs at the sides and exposes the front of her richly patterned dress that appears to be of a heavy damask silk material. The mantle is painted in a very deep and rather glossy black pigment which appears not to be the usual charcoal used in the earlv fourteenth-centurv frescoes of the Parecclesion. The same pigment seems to have been used also in the delineation of the pattern on the dress. The field of this garment, however, is a duller black of lighter value, probably charcoal. The pattern consists of a series of ogive arches framed above by straight lines which form points On which are foliate finials. Interspersed are what seem to

227

be clusters of fruit and foliage. At the very top of the fragment the dress breaks off at a point that possibly marks the bottom of the sleeve of the outer garment. At the upper left, the black of the patterned dress is interrupted by a girdle of twisted materials of white and green. On the white there appears a small scale pattern in blue and on the green another pattern in red and yellow. In the absence of inscriptions, monograms, or historical sources that might throw light on the identity of the person for whom the sepulchral monument was constructed, its dating becomes difficult. There can be no doubt that it represents the last of the series of tombs in the walls of the outer narthex and, for that matter, in the entire church. I n excavating along the exterior of the west faqade, it was found that the curtain walls of all three arcosolia rest at a common level upon the top of the early fourteenth-century foundation wall which runs level throughout its length. But, whereas the lowest course of stones of the masonry fill in the arcosolia of bays four and five were above exterior grade level when they were laid, grade level had risen considerably by the time the wall of this arcosolium (bay two) was constructed. This is evident from the fact that the exterior sides of the cut stones of the lowest course in this tomb had not been dressed and were extremely roughly shaped, whereas the lowest courses in the other tombs were of carefully dressed stone, and must have been exposed to view at the time they were laid. The style of the painting, moreover, confirms the archaeological evidence that this tomb is considerably later in date than all others at the Kariye Camii. The painter appears to have combined some mannerisms that are basically Byzantine with technical devices and stylistic features that are characteristic of Italian Renaissance painting of the It is sufficient here merely fifteenth ~entury.4~ 43 Cf. P. A. Underwood, "Palaeologan Narrative Style and an Italianate Fresco of the Fifteenth Century in the Kariye Djami,w Studies in the History of A r t , Dedicated to W i l l i a m E . S u i d a on h i s Eightieth Birthday (London, 19591, pp. 1-9. There the contrast of 'style between this painting and all others in the Kariye Camii is more fully discussed and i t is suggested that the painting is a work of the second quarter of the fifteenth century by the hand of a

P A U L A. U N D E R W O O D t o point out that the spatial organization of the painting is no longer Byzantine, but that the achievements of Renaissance perspective have intervened in a considerable degree. The figures and furniture take their places optically upon a horizontal ground plane consisting of a marble floor whose front edge is brought forward to the picture plane where it rests upon the frame of the picture and becomes simultaneously a part of the scene above and of the frame itself. While the footstool is not drawn in true perspective, perhaps because of insufficient knowledge of its principles, its manner of drawing implies that Renaissance perspective had already been achieved. The orthogonals no longer diverge in depth, but Byzantine painter who had somehow acquired considerable knowledge of Italian Renaissance painting.

either converge to vanishing points or are drawn parallel to one another in the manner of isometric drawing. In certain passages of the drapery of the Virgin, especially in the folds between the knees and the feet, we observe a modification of Byzantine formulae, but folds like those a t the lower left, which stand forth sharply in deep relief, are not to be found in the repertory of Byzantine art. They reflect, rather, the sculptural qualities of western painting of the fifteenth century, as do the numerous instances of cast shadows which are to be observed here and there. Finally, the manner in which the pattern on the dress of the deceased is broken up to correspond to the individual folds represents a departure from the flat stencil-like application of pattern to drapery which is usual in Byzantine painting.

Inscription accompanying Portrait of the Nun Athanasia

1. Kariye Camii, Outer Narthex. Arcosolium of Bay 5, Tomb of Eirene Raoulaina Palaeologina

5. Arcosolium of Bay 4

7. Arcosolium of Bay 2

9. Arcosolia of Bays 4 and 5. Exterior

11. Window of Bay 6. Slot in Jamb for Balustrade

10. Window of Bay 6. R11 partly removed

12. Arcosolium of Bay 2. Trace of Window Frame in Arch Soffit

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: Notes on the Work of the Byzantine Institute in Istanbul: 1957 Paul A. Underwood Dumbarton Oaks Papers, Vol. 13. (1959), pp. 215-228. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C215%3ANOTWOT%3E2.0.CO%3B2-C

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 1

Notes on the Work of the Byzantine Institute in Istanbul: 1955-1956 Paul A. Underwood Dumbarton Oaks Papers, Vol. 12. (1958), pp. 269-287. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281958%2912%3C269%3ANOTWOT%3E2.0.CO%3B2-Y

Notes on the Work of the Byzantine Institute in Istanbul: 1955-1956 Paul A. Underwood Dumbarton Oaks Papers, Vol. 12. (1958), pp. 269-287. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281958%2912%3C269%3ANOTWOT%3E2.0.CO%3B2-Y 4

Notes on the Work of the Byzantine Institute in Istanbul: 1955-1956 Paul A. Underwood Dumbarton Oaks Papers, Vol. 12. (1958), pp. 269-287. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281958%2912%3C269%3ANOTWOT%3E2.0.CO%3B2-Y 5

Notes on the Work of the Byzantine Institute in Istanbul: 1955-1956 Paul A. Underwood Dumbarton Oaks Papers, Vol. 12. (1958), pp. 269-287. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281958%2912%3C269%3ANOTWOT%3E2.0.CO%3B2-Y

NOTE: The reference numbering from the original has been maintained in this citation list.

The Mother of God, "Stabbed with a Knife" George P. Galavaris Dumbarton Oaks Papers, Vol. 13. (1959), pp. 229-233. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C229%3ATMOG%22W%3E2.0.CO%3B2-0 Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:40:27 2008

THE MOTHER OF GOD, "STABBED WITH A KNIFE"

T

HIS paper is concerned with four unpublished Byzantine lead seals belonging to the Dumbarton Oaks Collection. The first, which is in excellent condition, is 2.6 cm. in diameter, 4 mm. thick, and weighs 14.8 grams (fig. I). I t shows on the obverse a figure of the Mother of God standing frontally. She holds in front of her breast a medallion containing a bust of Christ. Her face is very well preserved, and surrounded by a halo. The figure of the Mother of God is not accompanied by any of the familiar epithets. Instead, there is the following legend in two columns :

M-P e-8 H M A X A I I P WOEICA M ( f i ~ q ) pB(E)oG 4 ~axa1pw0eioa The Mother of God, "Stabbed with a knife." On the reverse is this invocation in five lines: $B&R0/ HBEIT WCW / A S A W ~ ~ E T W/AIAP KON W KAI/XAPTOOY/AAKI. Q(EOT~)KE pofieel T+ 6oirhq nd-rpq 81m6vq ~ a xap-roqirhm~. i

"Mother of God help thy slave, Peter deacon and chartophylax." The second seal (fig. 2), formerly in the collection of Mr. Howland Shaw, is an exact duplicate of the first, but is not as well preserved. I t is 2.4 cm. in diameter, 4 mm. thick and weighs 12.9 grams. The third seal (2.4 cm. in diameter, 3 mm., thick, and weighing 7.7 grams) presents the same type of the Virgin on the obverse and the same epithet with a slight difference in the arrangement of the letters (fig. 3 ) :

the first two seals is also of a very common type.2 Furthermore, all three legends make it clear that these seals belonged to the same person. The fourth seal (fig. 4) (2.3 cm. in diameter, I mm. thick and weighing 6 grams) has the same type of the Virgin and inscription on the obverse: ( M ~ P )078) HMA(X)AI/PWQEICA

The reverse, however, reads : (Q) ERO H (@)/ E I ~ N A I A/ K O N W K A I / X A P T O O Y / AAKITHCM/E(TA)A(HC) EKKA(H)/CIAC. (@EOT~)KE poj0e1 K w v o ~ a v r i v q61m6vq ~ a xap-roqirhm~ l M~(y&)h(q~)'E K K ~ ( ~ ) "Mother u ~ ~ . of God help

Constantine, deacon and chartophylax of the Great Church." Palaeographically the inscriptions of the first three seals can be matched by dated seals of the eleventh and twelfth centuries (especiallythe letters E, W, C, A).3 Stylistically the elegant stance and good proportions of the figure of the Virgin on the first two seals call to mind certain eleventh-century coins, for example those of Constantine X Ducas.4 An eleventh or twelfth-century date seems, therefore, very probable. The fourth seal cannot be dated on stylistic grounds because of its poor state of preservation. But the lettering of the reverse (note especially the M) suggests the thirteenth century.

Movusiov (Athens, 1g17), pp. 421ff. Cf. V. Laurent, "Bulletin de sigillographie byzantine, 1930," Byzantion, VI (1g31), pp. 81jff., idem, MP 88 "Les bulles m6triques dans la sigillographie HMAXAlPO/OEICA byzantine," 'EAA~vIKC~, IV-VIII (1931-1935), passim. The invocation on the reverse reads : IlETPO N/ G. Schlumberger, Sigillographie de 1'Empire l l A N A T N E / XAPTO@YA(A)/KANCKE/llOIC. byzantin (Paris, 1884)~pp. g ~ f f .Cf. Laurent, llC~pov, llavayve, xap~oqbhmavmdrro15 "Most Byzantion, VI (1931)~p. 78j. 3 N. Lichaeev, Istorileskoe znatenie italopure, protect Peter the chartophylax." This gvec'eskoj ikonopisi. Izobvaienija Bogomateri is a metrical inscription, such as is often (St. Petersburg, I ~ I I )pl. , IV, 16. G. Schlumfound on Byzantine sea1s.l The invocation on berger, "Sceaux byzantins inkdits," Revue des K. Konstantopoulos, 6vtav~1aUts~ o h v ~ 6 6 - e'tudes grecques, IV (1891), p. 124, no. 54. 4 Lichaeev, ibid., fig. 76. povhha TOG Iv 'A0fiva1~ 'EBWIKOG Nop~opcrr~~oG

G E O R G E P. G A L A V A R I S discussion, being of much later date, are consequently of some interest. As to the epithet Maxalpw&ioa, "stabbed with a knife," this is unique, as far as published material goes, and herein lies the major importance of the seals. The epithets given to the Mother of God are a valuable source of information concerning Byzantine religious life, history, and folklore. To give a comprehensive list of them would scarcely be possible here. However two large categories can be distinguished: I . Theological epithets, mainly derived from hymnography, such as ' O 6 q y j ~ p 1 a ,'Enim~yls, XBpa TOG ' A x w p j ~ o u e, t ~2.. Popular ~ epithets, which sometimes describe a particular iconography of the Virgin, such as rhwoqnhoijaa, 5 N. Kondakov, Ikonografia Bogomateri, I1 (St. Petersburg, 1915)~pp. 66ff. For a brief r a h m ~ o ~ p o q o i j o a ,e tc., or are related to discussion of the Blachernitissa type see miracle-working icons as, for example, l l o p ~ a C. Cecchelli, Muter Christi, I (Rome, 1946), i-rlooa, T p ~ ~ ~ p o G'oAa v, ~ ~ q w v j - r The ~ ~ a epithet .~ pp. 216ff. (to be used with caution; cf. A. on our seals is a new addition to this second Grabar's review in Cahiers arche'ologiques, 8 119561, pp. 259ff.). M. Vloberg, "Les types category.1° iconographiques de la mbre de Dieu dans l'art Inasmuch as sacred representations on seals byzantin," Maria; e'tudes sur la Sainte Vierge, I1 were usually modelled after famous icons,ll (Paris, 19jz), pp. 413ff. G. Soteriou, X p ~ o - r ~ a v ~ ~ f i our problem is to identify a miraculous image ~ a B i u ~ a v ~ i v f~it ~ o v o y p a q [ a@~ohoyia, , X XVII to which this epithet applied. (19 56))p. 11. E. Coche de la Ferte, "Decors en ceramique byzantine au Musee du Louvre," The numerous legends relating to miracuCahiers arche'ologiques, I X (1957), pp. 192ff. lous icons often refer to acts of agression or The suggestion that the iconographic type iconoclasm that call forth a response from the of the Virgin holding a medallion is a variant supernatural force embodied in the icons,l2 and of the Blachernitissa was first proposed by P. Dethier: see Schlumberger, Sigillographie, in such cases the icons may talk,l3 suffer,l4 be

If the legends on the reverse of these seals are of a common form, the same cannot be said of the iconographic type of the Virgin. According to Kondakov, this type of Virgin, holding a medallion, is called the Nikopoios, and he distinguishes this from the Blachernitissa in which Mary appears with hands raised either as a half figure or in full length.5 I t seems, however, that the Virgin of these seals is merely a variant of the Blachernitissa type showing Mary as a full length standing figure holding the medallion.6 There are few known examples of this variant on seals, and hitherto all of them have been of the sixth and seventh centuries.' The seals under

p. 37, note I, and more recently V. Grumel, "Le 'miracle habituel' de Notre-Dame des Blachernes & Constantinople," Echos d'Orient, X X X (1931)~pp. 129-146, esp. 144-145. I t seems best to reserve the term N~kopoiosor Kyriotissa for the Virgin holding the Child in front of her, one hand supporting His leg, the other resting on His shoulder: see A. XyngoV MTTEV&K~, poulos, Kcrrdihoyo5 -rBv E ~ K ~ V WMoumiov (Athens, 1936), p. 8, pl. 17h. Cf. 0. Wulff, Die Koinzesiskirche in Nicaa und ihre Mosaiken (Strassburg, 1903), p. 256ff., esp. p. 260. See examples in Konstantopoulos, op. cit., p. 78, no. 275. Schlumberger, Sigillographie, pp. 418, 420. P. Delatte, Le culte de la Sainte Vierge en Afrique d'aprks les monunzents arch& ologiques (Paris, 1907), pp. 84-127. Cf. G. Schlumberger, "La Vierge, le Christ, les saints sur les sceaux byzantin de Xe, XIe, XIIe sikcles," Memoires de la Socie'te'des Antiquaires de France, ge skrie, IV (1883), p. 2. To these examples we may add one more from the unpublished collection of seals of the Dumbarton Oaks Research Library and Collection: the seal (inv. no. 3130) that belonged to the well-known Bardas family and dates probably from the twelfth century.

'

For epithets of the Virgin in hymnography i v -rq 6yvoypa(~iq see S. Eustratiades, 'H OEOTOKO~ (Paris, 1930). Soteriou, p. 13. For popular epithets see P. Koukoules, 'Eriee-rk T I V T~ ~ @EOT~KOU, S ' Hy~poh6y1ov M ~ y a h q s'EhhaGos, X (1932), pp. 431,444. Cf. J. Themelis, llspi TBVErwvuyiwv ~ i j snavayias, Actes du I I I e congrhs international d'e'tudes byzantines (Athens, 1930 [1932]), pp. 311-314; Timothy of Jerusalem, At Inwvuyia~ TGS llavayias, Nia ZIBV, XLIV-LI (1952-1956), passim. lo For epithets of the Virgin appearing on seals see Schlumberger, Me'moires, pp. I j ff. l1 Cf. Hodegetria, Hagiosoritissa, Atheniotissa etc. Kitzinger in Dumbarton Oaks Papers, 8 (19541, PP. 83-149, esp. pp. 87ff. l 3 Cf. the stories of speaking icons told in Mt. Athos, G. Smyrnakes, "Ay~ov"Opo5 (Athens, 1go3), passim. l4 Cf. the weeping icons, Itine'raires russes en Orient, trans. by B. de Khitrowo (Geneva, 1899), pp. 89, 226. Ebersolt, Sanctuaires de Byzance (Paris, 1921), p. 22.

MOTHER O F

GOD,

"STABBED WITH A K N I F E "

wounded,15 or shed blood.l6 Usually those who strike the icons are unbelievers. John of Damascus, quoting from the writings of Athanasius Sinaita, speaks of a Saracen who shot an arrow a t a representation of St. Theodore, whereupon blood flowed from the figure of the Saint.l7 There is a famous story of an icon of Christ a t Beirut which was similarly wounded by unbelievers and which shed blood and w;ter.ls I t is among such icons that we should seek the prototype for our "stabbed" Mother of God.19 I n the outer narthex of the parecclesion of St. Demetrius of Alexandria in the monastery of Vatopedi on Mt. Athos, there is a n icon of the Virgin called 'Eupay~bvr), or the "slaughtered Virgin." It was struck not b y an unbeliever, but by a deacon or sacristan (the story varies) in a fit of anger, because his appointed task of cleaning the candlesticks in front of it every morning made him late for meals. To his horror blood gushed from the wound. The story may be found in many books about Mt. Ath0s,~0and although it has no connection with the seals under discussion here, the type of the "slaughtered Virgin" does suggest an analogy. The fine execution of these seals indicates that they were probably made in Constantinople, and therefore it is reasonable t o assume that they copy a Constantinopolitan icon. Many such miraculous images in the l5 See Pseudo-Athanasius, Quaestiones, PG, 28, 621. 16 Cf. the story of the Virgin's icon at Alexandria contained in a homily attributed to Theophilus, Patriarch of Alexandria: W. H. Worrell, The Coptic Manuscripts in the Freer Collection (NewYork, 1923),pp. 37off. Detailedinformation on these icons, the dating of the relevant texts, and general information on the magic properties of images are to be found in Kitzinger, 09. cit., P P 96ff. l7 De imaginibus, orat. I I , PG. 94, 1393. Is Pseudo-A thanasius, PG, 28, 796ff. ; Kitzinger, loc. cit. l9 I t is interesting to note that in western art, as early as the Gothic period, there are representations of the Virgin stabbing herself. See La Vierge duns l'art franyais, ed. Art et Style (Paris, 1g50), pl. 38. Cf. E. M%le,L'art religieux de la fin du moyen dge en France (Paris, 1g31), pp. 122ff. 20 Smyrnakes, p. 434. G. Soteriou, " A y ~ o v "Opo5 (Athens, n. d.), p. 144. R. Dawkins, The Monks ofAthos (London, 1936)~p. 361.

231

capital of the Empire were shown t o pilgrims.21 The scribe Alexander who visited Constantinople in 1393 saw one in the monastery of Theotokos Peribleptos, founded by Romanus I11 Argyrus (1028-1034). 22 "The image of the Holy Virgin," he writes, "was transfixed by a Jew .. . and the blood that issued from it can be seen t o the present day."23 More detailed is the account of Anthony, Archbishop of Novgorod who, on his visit t o Constantinople in 1200, saw a "stabbed" icon in the church of St. Sophia. He says: "We kissed the image of the Most Holy Virgin holding Christ. A Jew struck this Christ in the neck with a knife and blood flowed out. . "24 The same story is told by an anonymous English pilgrim who visited Constantinople afew years before the Latin conquest, and it is found also in Nicolaus Thingeyrensis' catalogue of the relics that existedin Constantinople in 1157.~~ The first of these two texts reads: in ips0 loco [in the church of St. Sophia, a t the place of the "Samaritan" well126 in angulo est imago sancte sanctarum Dei genitricis Marie, que portavit in ulnis suis Dominum nostrum Ihesum Christum, quem percussit quidam Judeus cum cultello in guture et continuo exivit sanguis el aqua.27 According t o Nicolaus

.

21 See anonymous Russian pilgrim, Itin. russes, p. 229. Ebersolt, op. cit., p. 22. 22 R. Janin, La gkographie ecclksiatique de 1'Empire byzantin. Le sikge de Constantinople, 111, Les Lglises et les monastkres (Paris, 1953)~ p. 229. 23 Itin. russes, p. 163. a4 Ibid., p. 87. 25 I am gratful to Dr. Cyril Mango of Dumbarton Oaks for bringing these two texts to my attention. 26 The "Samaritan" well was in a chapel called the Holy Well ("Ay~ov@piup) attached to St. Sophia. For the location of the chapel see E. Antoniades, 'E~cppautsTGS 'Ayiar Zopias, I1 (Athens, 1go8),pp. 169ff. For a different view see E. Mambury, "Topographie de Ste.-Sophie. Le sanctuaire et la solka, le mitatorion, le puits sacr6. Le passage de St. Nicolas, etc.," Atti del V congress0 intern. di studi bizantini, Studi bizantini e neoellenici, VI (IQ~o), pp. 197-209. Dr. Mango who has dealt with this problem in his forthcoming study on the Chalke believes that Antoniades' view is the correct one. 27 S. G. Mercati, "Santuari e reliquie Constantinopolitane second0 il codice Ottoboniano latino 169 prima della conquests latina (1204)," Rendiconti della Pontificia Accademia Romana di Archeologia, XI1 (1936),pp. 143ff.

G E O R G E P. G A L A V A R I S the solution of our problem must be sought Thingeyrensis, the Jew threw the icon into by direct consultation of the sources, and the "Samaritan" well, but his sacriligious act particularly of patriarchal documents, many was miraculously discovered and the image of which are signed by chartophylaces. Among was restored t o its former place. As for the Jew, he was converted to C h r i ~ t i a n i t y . ~ ~ the signatories of a synodic declaration of the patriarch Nicholas I11 (1084-IIII), issued on Thingeyrensis writes: Imago S.Mariae cum November 15, 1086 or 1101 (indiction 10) Jesu Christo, filio ejus; cuius iugzclzcm Judeus quidam cz.tltellovulneravit,et manavit s a n g u i ~ . ~ ~ T?S & Y I W T ~ T ]TOG ~ OEOGM~ydihqr' E K K ~ ~ C T ~ U S . ~ ~ None of these texts specifies how the Virgin This date agrees with that which I have held the Child, and it cannot be stated proposed on independent grounds for the first definitely that she was of the Blachernitissa three seals, so it is quite likely that their type; yet, if their accounts are taken literally, owner was the deacon and chartophylax of the statements that the stabbing was done this document. with a knife, rather than with a spear or an The same Peter is also known from arrow as in other instances, justify the epithet another document dated 1 0 9 2 , ~in~ which he paxaipw8~iua, and suggest that our seals solves certain difficulties of canon law. The copy the famous "stabbed" icon in St. Sophia, answers he provides betray some novel ideas to which Peter and Constantine, the deacons about salvation through and chartophylaces, may have been particular- Grace and an unusual view of confession, according to ly devoted. which the penitent should not enumerate his But who are Peter and Constantine? sins to God, especially if they are of a carnal Their titles show that they were chartophynature. For, Peter believes, if the sinner laces of the Oecumenical Patriarchate, memrecites his transgressions, his soul will be bers of the clergy of the church of St. Sophia. defiled; it is enough for him to invoke the This dignity, denoting the chief of the patriarof the Lord.34 name chall ibrary and archives, eventually became In a patriarchal document of Germanus I1 one of the most important in the patriarchal (1222-1240) entitled "Notice of the Most Holy hierarchy.30 Andronicus I1 (1282-1328) added the title pQya5in order further to dignify the Patriarch Germanus against John, the Metropolitan of Naupactus," dated July 1235, we rank of char to phyla^.^^ Thus, the absence of find the following signature : ' 0 xCfp~ocpfiha$ this epithet from our seals, apart from other factors, limits the search for Peter and Conmavrivos 6 A G A ~ v In ~ ~this . ~ ~instance, too, stantine to the period before Andronicus 11. Since no comprehensive work on Byzantine 32 PG, 119, 864. Cf. V. Grumel, Les regestes prosopography has as yet been published, des actes d u patriarcat de Constantinople, I , 3 The same story was told about an icon concerning Christ also located in the chapel of the Holy Well. See the Greek text in F. Combefis, Historia haeresis monothelitarum, sanctaeque in earn sextae synodi actorurn, vindiciae (Paris, 1648), pp. 648-657; E. von Dobschutz, Chvistztsbilder, Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, N.F., 111, (Leipzig, 1899), pp. 216**-ZI~**. Cf. Antoniades, op. cit., pp. 179ff. 29 Exuviae sacrae Constantinopolitanae, I 1 (Geneva, 1878), p. 215. 30 A discussion of the title and duties of the chartophylax may be found in Ch. Demetriou, M E A ~TTE~P ~TOG xap~ocpirhmo~ T ~ $v S Kwvmavr~vov-rr6hal M ~ y i x h qTOG ~ Xplmoir'E~~hquia5 (Athens, 1924). F. Dvornik, Les le'gendes de Constantin et de Me'thode (Prague, 1933), pp. 52 ff. Cf. L. Brkhier, Le monde byzantin, les institutions (Paris, 1949), pp. 501 ff. 31 Dvornik, ibid., p. 56. 28

(Bucharest, 1947), pp. 43, 44. E. Gedeon, T7cr-rpiapx1~ol-r r i v a u ~(Constantinople, 1890), p. 342. For the chronology of the patriarchs see V. Grumel, L a chronologie, Trait6 d16tudes byzantines, I (Paris, 1958)~pp. 436ff. 33 PG, 119, 1093ff. Gedeon, ibid., p. 347. Concerning the Peter of this document, V. Laurent ("Les bulles mktriques dans la sigillographie byzantine," 'EhAqvi~Ii, V [1932], p. 163) has raised the question whether Peter was the owner of the twelfth-century seal described by Schlumberger (Sigillografihie, p. 394) with the seated Hodegetria on one side, and the legend ( l l C ) ~ p opovaxbs ~ ~ a p8ya5 i o i ~ o v 6 p oon~ the other. To answer this question in the positive one ~vould have t o assume a promotion of Peter from chartophylax to great oeconomus. PG, lac. cit. Cf. G. Ralles, M. Potles, X6vraypa ~av6vwv,V (Athens, 1855)~pp. 369373. 35 PG, 119, 797, 801. Cf. F. Cabrol, H. Leclercq, Dictionnaire d'arche'ologie chre'tienne et de

MOTHER O F G O D , "STABBED WITH A K N I F E " the date of the document agrees with the one suggested for the fourth seal which bears the name of Constantine, and it is possible that he is in fact the Constantine Aulenos of the patriarchal document. Whether or not these suggested identifications prove to be correct, the fact remains that the owners of our seals were connected with the church of St. Sophia, where textual evidence indicates the existence of a "stabbed" icon. This connection leaves no doubt that the "stabbed" icon of that church was the prototype of our seals, though we do not know why Peter and Constantine, the deacons and chartophylaces of the patriarchal church, chose as their personal device an icon then existing in St. Sophia. The fact that, of the four seals pertaining to the same ecclesiastical office, the last is at least a hundred years later liturgie, 111, I, 1017; V. Laurent, "Les bulles mktriques dans la sigillographie byzantine," 'EAhqvt~&,I V (1931), P. 344.

233

than the first three suggests, however, that the choice of this iconographic type was not dictated by purely personal reasons. I t may be that, as the Virgin and Child on a Bij~o5 was the emblem of the patriarchs, and as the Virgin receiving the church of St. Sophia from Justinian was the device of the presbyters of that ~hurch,~B so the "stabbed" Virgin was the protectress of the chartophylaces of the Great Church. This remains, however, merely a suggestion, because another chartophylax of the Great Church, whose seal was published by Schlumberger, chose as his device the seated Virgin and Child flanked by an archangel and a 36 I hope to discuss on another occasion the seals of the Constantinopolitan patriarchs and those of the presbyters of the church of St. Sophia from the point of view of iconography. 37 Schlumberger, Sigillographie, p. 130. A more precise description in Konstantopoulos, op. cit., p. 184, no. 702.

Washington, D. C., Dumbarton Oaks Collection Lead Seals (enlarged twice)

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: The Mother of God, "Stabbed with a Knife" George P. Galavaris Dumbarton Oaks Papers, Vol. 13. (1959), pp. 229-233. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C229%3ATMOG%22W%3E2.0.CO%3B2-0

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 12

The Cult of Images in the Age before Iconoclasm Ernst Kitzinger Dumbarton Oaks Papers, Vol. 8. (1954), pp. 83-150. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281954%298%3C83%3ATCOIIT%3E2.0.CO%3B2-%23 16

The Cult of Images in the Age before Iconoclasm Ernst Kitzinger Dumbarton Oaks Papers, Vol. 8. (1954), pp. 83-150. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281954%298%3C83%3ATCOIIT%3E2.0.CO%3B2-%23 18

The Cult of Images in the Age before Iconoclasm Ernst Kitzinger Dumbarton Oaks Papers, Vol. 8. (1954), pp. 83-150. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281954%298%3C83%3ATCOIIT%3E2.0.CO%3B2-%23

NOTE: The reference numbering from the original has been maintained in this citation list.

The Evidence of Restorations in the Sanctuary Mosaics of the Church of the Dormition at Nicaea Paul A. Underwood Dumbarton Oaks Papers, Vol. 13. (1959), pp. 235-243. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C235%3ATEORIT%3E2.0.CO%3B2-J Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:43:14 2008

THE EVIDENCE OF RESTORATIONS

IN THE SANCTUARY RIOSAICS OF THE CHURCH OF THE DORMITION AT NIC-4EB

I

N a footnote to his paper prepared for the Eleventh International Congress for Byzantine Studies, held in Munich in 1958, Professor Kitzinger suggested the possibility that originally, under Hyakinthos the Founder, the conch of the apse of the Church of the Dormition at Nicaea was adorned with a standing figure of the Virgin, that this figure was later replaced by a cross which, in turn, was removed when Naukratios restored the status quo ante by reintroducing a figure of the Virgin to its former p0sition.l I n other words, it is suggested that in the history of the conch mosaics there were at least three stages-not two as has been universally assumed. With regard to the four angels in the arch of the bema, which some have reE. Kitzinger, "Byzantine Art in the Period between Justinian and Iconoclasm," Berichte zum X I . Internationa2en B yzantinisten-Kongress, TV, I (Munich, 1958) pp. 1-50. Cf. pp. 12-16 for a review of the question of the mosaics in the apse and bema, and especially note 59 for the proposed sequence of mosaics in the apse. For a more extensive bibliography and the positions taken by previous writers on the subject, the reader is referred also to the article of GCza de Francovich, "I mosaici del bema della chiesa della Dormizione di Nicea," in Scritti di storia dell'arte in onore di Lionello Venturi (Rome, 1956), pp. 3-27. Heretofore the mosaics of the apse (fig. 3) have been regarded (except by Kitzinger) as the work of two phases (one original, the other representing alterations). The dates ascribed to the two, and even the extent of mosaic surface representing each phase (cf. note 7 infra), have varied greatly. The first phase has been considered by some to have been pre-iconoclastic (VI-VII c.), by others iconoclastic; the second phase has been associated either with the Orthodox interlude (787-815), or with the period after 843. On the other hand, the mosaics in the arch of the bema (Etimasia and angels, but usually excluding the inscription between two of the angels, fig. 7) have been regarded as being of one epoch which has been ascribed t o one or the other of the two phases recognized in the apse.

garded as examples of pre-iconoclastic art," Professor Kitzinger was more cautious, but questioned whether the heads, at least, were not also r e ~ t o r e d . ~ The most reliable evidence for settling these questions consists of N. K. Kluge's original photographs, taken in 1912,~and their reproductions in the publications of Schmits and Lazarev.6 While the quality of this material leaves much to be desired, I believe that Schmit's plate X X and two of Kluge's photographs (our figs. 3 and 4) present clear evidence in support of Professor Kitzinger's suggestion regarding the mosaics of the conch. I n my opinion, however, he did not go far enough in questioning only parts of the angels, for here the evidence of Schmit's plates XI11 and XIV (our figs. 7 and g) is that the angels were restored from head to 2 This, despite the inscription of Naukratios between the angels in the south side of the arch (infra) which clearly refers to his restoration of the holy images. LOC. cit., pp. 14-16, especially note 55, where it is pointed out that the normal trimming of gold around the tops of some of the haloes is lacking, and other evidences of disturbance in the mosaics exist. I am greatly indebted to Dr. Alice Bank of the Byzantine Department of the Hermitage, Leningrad, for prints of Kluge's negatives used here in figures 3, 4, and 5. Theodor Schmit, Die Koimesis-Kirclze von Nikaia (Berlin and Leipzig, 1927), from which our figures 6-9 are taken. One could wish for more detailed comments by Schmit or Kluge on the condition of the mosaics. The total destruction of the church in the Greco-Turkish war in 1912 leaves us with the photographs and the published plates as the most useful record of the mosaics. Victor Lazarev, History of Byzantine Painting (in Russian), 2 vols. (Moscow, 1947). Lazarev used prints of Kluge's negatives in his publication. The pertinent illustrations appear on plates I, 11, and VI of vol. I, and figs. 6-8, and 38 of vol. 2.

236

P A U L A. U N D E R W O O D

foot. I t will be the purpose of these notes to determine as precisely as possible what can be learned of the history of these mosaics from a scrutiny of the photographic reproductions. For this purpose the plates were photographically enlarged, thus making it possible with the added magnification of binocular loupes to trace in many places the exact path of the seams, or lines of juncture, between areas of mosaic of different periods and to observe more clearly certain areas of disturbance in the mosaic surface where alterations occurred. Where junctures resulting from insertions into pre-existing areas were clumsily executed they are obvious, but in a number of places they appear to have been skilfully done and are either impossible to trace with exactness or remain entirely invisible in the reproductions. However, in the few instances where it is not possible to follow the complete contours of restored areas the distribution of clearly detectable seams informs us of their approximate course, or suggests definite conclusions as to the areas of replacement. The following demonstration will require constant comparison of figures 3, 7, and 9 with the drawings (figs, I and 2) and the outlining of areas of restoration in the arch of the bema (figs. 6 and 8). If one examines the photograph of the mosaics in the apse (fig. 3), it is apparent that, with the exceptions of the cross-shaped area containing the figure of the Virgin and the small area of the hand of God in the arc of heaven at the top, the remaining vast expanse of mosaic surface in the apse is original mosaic which we can call period I (fig. 2). With those exceptions, the mosaics of the conch are homogeneous and have all the features of having been done de novo in one continuous campaign. The rows of tesserae carry through in a consistently systematic way and the arc of heaven, the rays, footstool, zones of ground, and individual letters in the inscriptions are consistently trimmed by rows of gold tesserae to form a neat transition between the horizontally laid gold cubes of the background and the sharply defined edges of the objects made of different colors. Moreover, these mosaics are "signed" by Hyakinthos, the founder, whose monogram appears at each end of the principal inscription on the narrow

semi-circular facing between the apse and the arch of the bema. These characteristics of original work carry through the ornamental borders and the inscription into the mosaics in the upper part of the bema arch. The entire mosaic surface illustrated in Schmit's plate XII, which includes the great medallion of the Eti~nasiaat the top of the arch and the banners borne by the angels, is certainly a continuation of work done in the campaign that produced the original apse mosaics and presents the same consistency in its system of construction. A few cracks run through the Etimasin the full width of the arch, but these are not man-made. They result from structural breaks in the crown of the arch itself. In the arch it is only when we reach the areas between the inscriptions above the angels and the tops of their haloes that these conditions change and we detect irregularities that will be discussed below. In the apse it is clear, therefore, that the seams surrounding the figure of the Virgin indicate removal of mosaics from period I and the substitution of others which are necessarily later.' I t is perfectly obvious that at one time the central feature of the composition in the conch consisted of a monumental cross (figs. I, 2, and 3). The seams that surround the terminations of the lateral arms of the cross are very conspicuous and represent the edges of an area of mosaic that had been cut out. Indeed, within these areas, in their outer two thirds, one can still see the ghostly outlines of the terminations of the cross itself which precisely determine its contours, including the pointed serifs at the corners of the arms. The seams surrounding the tops, bottoms, and outer ends of these two areas (marked I1 in fig. 2 ) have comWulff's last opinion, summarized in his zu Altchristliche und byzantinische Kunst (Potsdam, 1g37), p. 72, that only the small fragments of the cross, the ornamental framework, the hand of God, and the monograms of Hyakinthos represent the remains of the original campaign is entirely unrealistic and on the face of it must be rejected. Such a view xvould mean that the post-Iconoclasts, intent upon replacing the cross by a figure of the Virgin, removed vast areas of mosaic, but spared extremely small areas of the object which they set out to remove.

Bibliographisch-Kritischer Nachtrag

S A N C T U A R Y M O S A I C S AT N I C A E A

monly been interpreted as cuttings made when the cross was removed in preparation for the insertion of the figure of the Virgin. This has led t o the conclusion that the original decoration of the conch consisted of a monumental cross. On the contrary, it can be shown that those particular seams represent the boundaries of areas that had been cut out of the gold background of the original mosaics (period I) in preparation for the insertion of the cross.s Those who have viewed these seams as the traces left after removal of the cross have overlooked, or have failed to account for, two things: first, the fact that traces of the cross still existed, and second, the very prominent seams, a t each side of the Virgin, that ran more or less vertically between the areas marked I1 and I11 in figure 2. If the mosaicists who executed area 111 (the Virgn) were the ones who cut out the areas marked I1 the mosaic surface would have been continuous and unbroken from one end of the cross arm to the other and there would have been no reason for seams across these areas, separating them from the figure of the Virgin. The seams between areas I1 and I11 can be accounted for only through their having been made as a result of the still later insertion of the mosaics of area 111 (the figure of the Virgin) within the area of the cross (period 11). The areas marked 11, therefore, are in large part the very tesserae that were put in place in period I1 and the seams above, below, and a t the outside of these areas represent the boundaries of areas removed from the original mosaic background in preparation for the insertion of the cross. These two areas, therefore, represent what was left after the central part of the extensive area of period 11, containing a cross (fig. I), had been destroyed for the purpose of inserting the figure of the Virgin (period 111). I n his description of the areas marked I1 in our drawing (fig. 2 ) Schmit touches on the

* In Hagia Sophia at Salonika, where the cross was original, but was replaced by the Madonna, its removal did not produce seams around the ends of the arms (see fig. 10). Had it not been original, seams would have defined the areas cut out for its insertion, just as they do at Xicaea. See infra, p. 238, for comments on the method used at Salonika, as well as at Kicaea, in obliterating the traces of the cross.

first of these two points, but merely states that the remnants of the cross itself were outlined with cubes of dark g0ld.O This explains the relative indistinctness of the outline of the cross in his photograph (our fig. 3), for the contrast is only slight between the darker gold of the outline and the lighter gold of the background surrounding it within the seams, and of the fill within the outline. I t is quite unlikely that the cross, when it was made, should have been outlined only in gold, even a gold of darker hue than its surrounding gold ground, for it would have remained very inconspicuous. Other comparable examples of monumental crosses set into gold grounds are outlined in tesserae of black (St. Irene, Constantinople, fig. II), or red (Hagia Sophia, Constantinople). Two alternatives present themselves: either the cross was indeed originally outlined in dark gold (in which case the entire surfaces of the areas marked 11, including the outlines of the cross, were left intact when the figure of the Virgin was inserted), or, as seems more likely, the mosaicists who substituted the Virgin for the cross merely picked out the individual tesserae that formed the outline of the cross and carefully replaced them with gold.1° I t is perhaps significant that the gold in the narrow strip of background that accompanies the figure of the Virgin seems also t o have been of a darker hue and very similar in color value t o Op. cit., p. 34 : "Innerhalb dieser horizontalen Arme des von der Risslinie umschriebenen Kreuzes sieht man Bruchstiicke einer Kreuzfigur, ~velchedurch Wiirfel dunklen Goldes gebildet wird." lo I t is the author's opinion, based on observation of the mosaics a t Hagia Sophia in Constantinople which had been removed and replaced, that the normal method of removing large areas of mosaic was to cut through the thickness of the three coats of plaster, a thickness of about 5 cm., to the masonry and then to build up with new plaster to the original thickness. However, a t Hagia Sophia an inscription was discovered in which the black glass tesserae had been individually picked out of their plaster setting, and tesserae of stone (matching the stones immediately surrdunding them in the background) had been substituted in an effort to remove evidence of the inscription. In small areas or lines of cubcs, such as the letters of an inscription or the outlining of the cross at Nicaea, this more me4iculous method could be used, but not in an area of any appreciable extent.

238

P A U L A. U N D E R W O O D

the gold in the outline of the cross. This could be regarded as an indication that the mosaicists who executed the figure of the Virgin (period 111) had no gold that matched the hue of the gold backgrounds of periods I and I1 and that in all probability they replaced the original outline of the cross (of black or red or some such color) with the darker gold cubes observed by Schmit. Good precedent for this method of removing the outline of a cross exists in the apse mosaic in the church of Hagia Sophia at Salonika where the cross, which seems to have been the original decoration of the conch (iconoclastic period ?), was replaced by a post-iconoclastic figure of an enthroned Madonna. Photographs made prior to restorations of more recent years (fig. lo)ll clearly show that only the borders and seriphs surrounding the arms of the cross were lifted out and replaced by gold tesserae, leaving intact most of the field within the borders and the gold background surrounding them. In all probability this is the method that was used also at Nicaea, but however that might be, the case for the cross as a second stage in the history of the conch mosaic does not depend on either of these alternatives and remains unaffected. Having concluded from an examination of the areas containing the terminations of the horizontal arms of the cross that there were at least three stages in this mosaic's history, and that the cross was an insertion into the original mosaics of period I, we find evidence leading to similar conclusions when we consider the seams that surround the nimbus of the Virgin. By comparing figures 2 , 3 and 4 (specifically,the area of gold ground immediately to the left of the halo), one can see that at each side of the halo there are two seams. The inner one closely surrounds the nimbus down to points close to the juncture of the halo with the shoulders. Its course is very ragged and extends upward into the central one of three rays. Near the shoulders this seam turns sharply to run outward in straight horizontal lines at each side that are only faintly distinguishable because they coincide with horizontal rows of the earlier gold background. At the outer ends of these lines they l1 Such as Boissonnas, negative no. 1783-B, a detailed enlargement of which is used in our figure 10.

form right angles again where they join the more or less vertical seams that carry downward the full length of the figure to the top of the front edge of the footstool. This continuous and rather symmetricalseam surrounding the figure and nimbus of the Virgin outlines area I11 (fig. z), and within it the entire surface of mosaic is of one piece and must represent the work of the last epoch, marked period 111. Returning to the area of the head, it is clear that there is a second pair of seams at each side of the halo, outside the first. These complete the boundaries that surround two small areas of gold background (represented by stippling in figure 2) which do not belong to periods I or 111. Further, it should be noted that the rows of tesserae in these two areas diverge in direction from the rows of period I at each side, just as the rows diverge in the two areas of the arms of the cross. The two small areas marked by stippling in figure 2 are the only parts that remained from the gold background inserted in period I1 around the upper part of the cross. The two outermost seams of these small areas mark the lateral extent of the cut that was made in this part of the conch for the purpose of inserting the cross. The width between themis far greater than was necessary to accommodate the upper arm of the cross (see fig. I), a fact which strongly suggests that a nimbus in the mosaics of period I pre-existed the cross in this area, and that its removal accounts for the great width of original mosaic that was destroyed there. On the basis of the preceding conclusions the form of the cross of period I1 can be reconstructed with a fair degree of certainty, and the extent of the mosaic surface of period I that was removed for the insertion of the cross can be determined in some places and conjectured in others. Figure I represents the cross surrounded by the seams made by the mosaicists who inserted the cross. However, only the seams drawn in solid lines around the ends of the horizontal arms and the two short lines at each side of the nimbus are fixed with absolute certainty. The broken lines indicate the conjectured extensions of the seams of period 11. Except for the line above the top of the cross, the conjectured outlines of the seams are drawn to coincide with the seams made by the mosaicists of

S A N C T U A R Y MOSAICS AT N I C A E A period I11 who would most likely have chosen to remove the mosaics up to the lines cut into the mosaics by their predecessors of period 11. In any case, if one allows for the figure of an earlier Virgin, the seams a t the sides cannot have been much closer together in period I1 than those represented in the drawing. In placing the cross upon the dais, or footstool, a conjectural three-stepped base was provided in the reconstruction patterned after that which supports the cross of the iconoclastic period in the church of Saint Irene in Constantinople (fig. 11). From this reconstruction three points become apparent which again indicate that the cross was not part of the original decoration of period I. The first is that the lateral arms of the cross were made to parallel the rows of tesserae in the background. These rows are actually horizontal, or very nearly so, but when seen from the usual point of view (from below) they appear to curve upward because of the curved surface of the conch. The arms of the cross, therefore, do not appear to be straight horizontal lines, but curve upward in a most unsatisfactory manner when seen from any point of view below their level. In the two instances of Hagia Sophia at Salonika (fig. 10) and Saint Irene in Constantinople (fig. IT), where the crosses were original to the mosaic decoration, the lateral arnis give, as nearly as is possible, the impression of straight lines when seen from below and consequently do not parallel the rows of cubes in the background, but actually curve downward.12 In other words, adjustments were made in the relation between the cross arms and the rows of tesserae of the background in order to provide for optical correction. This is the natural 12 The photograph of Figure 11 was taken from the floor. Although the vast scale of the mosaic does not permit one t o trace the individual rows of cubes in the background, a comparison of the upper edge of the green zone a t the bottom (which appears curved although it is actually horizontal) with the outlines of the horizontal arms of the cross (which appear straight although they are actually curved) illustrates the point. An elevation drawing such as the cross-section on pl. 5 of W. S. George, The Church of Saint Eirene at Constantinople (Oxford, n. d.), or his perspective view taken from a point well above the floor (ibid.,pl. 17), shows that in actuality the lateral arms are curved and the rows of tesserae of the background are horizontal.

239

thing for a mosaicist to do if he is executing a cross de novo in such a place. At Nicaea, where the cross was set into a pre-existing gold ground and was executed in a narrow and rather restricted area of insertion, such adjustments could not easily have been made, and in fact were not made, thus indicating again its make-shift character. Had the cross been made as the original element in the conch it would have been laid out so as to appear horizontal from below and executed before the background. The horizontal rows of the background would thus adjust themselves to the cross. The second point that becomes apparent in the reconstruction is the fact that the Nicaea cross appears to have been unique among examples of monumental crosses in that it is placed upon a footstool. Footstools of this sort appear in Byzantine art only when great personages (such as emperors) or the divine persons-Christ or the Mother of God -stand upon them or have their thrones set upon them. I t is a royal attribute which, to my knowledge, was never employed in conjunction with a representation of the cross. At Salonika, if the gap in the inscription at the base of the conch were filled in with the letters that were destroyed when the Virgin on her throne was substituted for the cross, there would have been no space available below the cross for a footstool comparable to the one at Nicaea. At Saint Irene (fig. 11)no footstool of any kind was used and the cross rests solely upon a small three-stepped base. This evidence alone strongly suggests that the footstool and the cross at Nicaea could not have been contemporary, as we have seen from other evidence. The original treatment of period I must have consisted of a figure of the Virgin standing upon the footstool. The third point is provided by the inscription surrounding the arc of heaven at the summit of the vault. This inscription is definitely original to the mosaics of period I and should, therefore, be appropriate to the iconography of the original decoration of the apse. Its source, though not its exact wording, is part of verse 3 of Psalm 109 (110) which reads, in literal translation: "I begot thee from the womb of the morning." The sense of the inscription cannot be regarded as a

P A U L A. U N D E R W O O D

reference to the Cross,ls but is perfectly compatible with a representation of the Virgin and with her role in the Incarnation. The evidence of the photographs of this mosaic thus firmly establishes the following conclusions. First, that the original mosaic decorations provided by the founder, Hyakinthos (period I), presented the standing figure of the Virgin upon a jeweled footstool. These mosaics were surely pre-iconoclastic both because of the subsequent history of the mosaic and because, as Gregoire has shown, Hyakinthos was known as founder of the church at the time of the VII Oecumenical Council, which met in Nicaea in 787, and must have flourished well before that date.14 The presence of the Virgin in the apse mosaics of period I prevents their being regarded as works of the periods of Iconoclasm. Secondly, we can conclude that this figure was removed by the Iconoclasts who replaced it with the image of the cross (period 11). Finally, the cross, in turn, was removed after Orthodoxy was re-established and an image of the Virgin was reinstated in the apse. I t was said above that the mosaics in the upper part of the arch of the bema (Etimasia and banners) were made during the same campaign of fabrication as the original parts of the apse mosaics, that is, period I which was pre-iconoclastic. In the two sides of this arch, to the south and north, there are two pairs of angels (figs. 7 and g). Each angel is inscribed and carries a labarum with the trisagion. The angels stand in a rather wide horizontal zone of green which extends to the lower ornamental border immediately above the cornice. At the bottoms of thesk zones, and extending the full width of the arch beneath the feet of the angels, - there are inscriptions. There are no evidences of seams of any kind immediately around or above the inscribed epithets above the heads of the angels, or around or above the inscribed banners which they carry. The two inscriptions beneath their feet, excepting the 13 In reviewing Schmit's publication, Wulff (in Repertorium fiir Kunstwissenschaft, 52 [1g31], p. 7) was troubled by the lack of relation between the inscription and the cross. l4 H. GrCgoire, "Encore le monastkre d'Hyacinthe a NicCe," Ryzantion V (1gzg-1930), pp. 287-293.

obviously later losses and patches of painted plaster at the left in the northern side (fig. g), show no signs of disturbance beneath them nor, for the most part, immediately above them. As Wulff and Schmit have remarked,16 all the inscriptions thus far named are epigraphically exactly like the inscription surrounding the arc of heaven at the top of the conch and like that on the narrow facing of the conch which, as we have seen, belong to period I. A careful comparison, especially of the alphas, epsilons, sigmas, and kappas, shows that they are all of the same campaign of fabrication, i.e., of period I. When we examine the inscription that runs vertically between the two angels in the southern side (fig. 7) we find, however, that the style of the letters is quite different and that they are made of blue glass as opposed to black in the other inscriptions.le The alphas no longer have the pendant cross bar, the epsilons and sigmas have become more rounded and less angular, and in the kappas the vertical stroke is detached from the rest of the letter, etc. As has been commonly acknowledged, this inscription records, in its implications, the restoration of the mosaics by one Naukratios, which makes the inscription later than those of Hyakinthos the founder, and later, therefore, than period I. Those who have attributed the angels to period I have assumed that the inscription was a later insertion, but this can be questioned, as Professor Kitzinger has done,l7 on the grounds that unless restorations had been carried out in the arch of the bema, and specifically in the areas oi the angels where his inscription was placed, Naukratios would hardlyhave recorded his work there. Moreover, since it is certain that the Iconoclasts did remove the forbidden image in the conch, it seems very unlikely that the angels would have been spared. On these grounds alone it appears quite clear that in part or in whole the angels are the Iconodules' replacements of earlier ones. This conclusion is confirmed, and the extent of the alterations is made evident, by a close examination of the areas immediately surrounding the angels (compare figs. 7 and g with figs. 6 and 8). With regard to the areas immediately Wulff, op. cit., p. 76; Schmit, 09. cit., p. 23.

Schmit, op. cit., pp. 23, 29, 33.

17 Op. cit., p. 14 and note 53.

l6

16

S A N C T U A R Y MOSAICS AT N I C A E A

surrounding Naukratios' inscription, Schmit, who recognized that it is later than the others and thought it to have been inserted, was unable, in substantiation of his opinion, to do more than to remarkthat the outline of the wing of Dynamis (fig. 7) follows a course which he thought was not original.ls An examination of the photograph, however, shows no traces of seams at either side of the inscription and the wings appear quite clearly to have been properly trimmed with single rows of gold and t o have remained undisturbed. Indeed, if one examines the area between the tops of the wings and the top of the green zone, and laterally through the bodies of the angels, no seams in the mosaics are evident, only certain areas of loss and later patches of crudely painted plaster. However, in each side of the arch a nearly continuous line of disturbance and seams runs almost horizontally at a level a little below the tops of the haloes and turns upward in curves at each halo to clear their borders (see figs. 6 and 8). This is most plainly evident in the northern side beginning to the right of Exousie (fig. 8) where an all-butstraight horizontal seam runs from the inner edge of the ornamental border to a point just short of the right side of the halo. Judging from Schmit's plate XIX (or Lazarev's fig. 7, vol. 11), once the repise reaches the border of the halo, just to the right of the top center, it appears to follow a very neatly executed joint between the outer edge of the halo's border and the original gold trim of period I for a very short distance to the left. At a point near the large patch between the two haloes, it departs from the contour of the halo, but owing to the patch we cannot pick up the seam again until we come to the top of the left wing of Kyriotites. For that reason this part of the conjectured path is indicated in broken lines in figure 8. A short distance to the left of the shaft of the labarum of this angel the seam again clearly reappears and runs horizontally toward the left to the right side of the nimbus. Here it breaks upward to follow a very ragged path around the top of the halo. A part of its course can be seen in figure 5 , where it can also be observed that around the top of the halo there is no gold trim; this begins only at the left below the seam where the restorer was able to provide Is

16

OF. cit., p. 23.

it. From the left side of the halo the seam runs in a straight line into the large patch of plaster at the far left. If one compares the gold background in the areas above and below this long line of seams and disturbances, one can see that, whereas the rows of tesserae above the line are relatively straight and continuous, their directions below the line diverge excessively from the horizontal-more than would be expected in a newly made mosaic in similarly constricted and irregularly shaped areas such as those between the haloes and the wings. This evidence indicates that in all probability a continuous seam, such as is indicated by the black lines in our figure 8, runs the entire width of the arch from border to border. The right-hand border was still preserved at the time of the restorations, and in examining its inner edge where it is joined to the gold of the background, slight irregularities in the juncture below the level of the horizontal seam become apparent. This can be seen by comparing the very neat line between border and background in the mosaics above the horizontal seam with the more irregular ones below that point. A straight and continuous line between such contrasting colors as those of the background and the border offers a logical place for the restorer to make a seam with the minimum of evident disturbance. Since the mosaics above the horizontal seam at the level of the haloes are work of period I, while those below the seam are of another period, and since the inscription in the lower part of the green zone is also of period I, it follows that a second horizontal seam must exist between these levels. In the northern side (fig. g), where it is more clearly evident than in the southern, this second seam follows a somewhat jagged line between, and at each side of, the angels along the upper edge of the green zone. (Contrast the rough lines of juncture between green ground and gold background in the arch-figs. 7 and gwith the neat and undisturbed juncture in the apse-fig. 3). The two angels, perhaps significantly, have lost their lower extremities and two rectangular patches exist beneath them which extend into the green zone. One can only conjecture that the general course of the seam in these areas corresponds to the broken line in figure 8. Were these areas lost

P A U L A. U N D E R W O O D

because the inserted mosaics did not adhere well around their edges where they lacked homogeneity with their surrounding mosaics ? In the southern side of the arch (fig. 7) seams can here and there be identified in positions that correspond closely to those we have traced in the northern side, although in general the restoration in the southern side was more expertly executed. These seams are indicated by the heavy black lines drawn into the photographs as illustrated in figure 6. By comparing their course in both figures 6 and 7, it can be seen that around the top of the halo of Arche there is no continuous gold trimming parallel to the rim of the halo and that the juncture here with the gold of the background is consequently rough. However, below the points at each side where the seams break away from the nimbus, the gold trimming, which is homogeneous with the nimbus and the figure of the angel, is carried down to the juncture of the halo and the shoulders. A similar situation exists above the halo of Dynamis where the seam is both jagged and clearly evident. Below the angels, again, a seam is evident along the top edge of the green zone between the two figures. I t is perhaps significant that the tri-lobed termination of the staff held by Dynamis, within the area of the green zone (period I), does not join onto the end of the staff in the area above (the restored area). The tri-lobed ornament is of a different period from the staff to which it should have been attached. The remainder of the seam, at the two extremities and below

the figures, is highly conjectural. But, as was said above, the clearly detectable portions of the seams inboth pieces of mosaic can and do indicate, by their direction and distribution, the approximate course of their complete circuit of the angels, and very definitely establish that the angels, in their entirety, had been removed and were later replaced. To a great extent, therefore, the angels shared in the history of the mosaics of the apse. Their predecessors were certainly the work of the founder Hyakinthos, and like the Virgin of period I were removed by the Iconoclasts. The angels, at least, were restored by Naukratios for whom no documentation appears to exist. In the light of the similar history of the mosaics in the apse and those in the bema, attempts to date the Virgin and the angels from two different periods should be reconsidered. Leaving aside questions of style,lg about which much has been written without serious regard to the evidence of restoration, and taking a realistic view of what would be possible in the environment of the iconoclastic controversies, the most logical conclusion would be that the figure of the Virgin in the apse and the four angels in the bema belong t o the same period; all of them restorations of Naukratios, and in all probability of a period soon after 843. l 9 In my opinion there are considerable variations in style among the four angels (which, however, must be regarded as contemporary works), some of which do not appear incom. patible with the style of the Virgin in the apse-

S A N C T U A R Y M O S A I C S AT N I C A E A

243

3. The Mosaic of the Apse

4. The Mosaic of the Apse, detail. Head of the Virgin

5. The Mosaics of the Bema, North Side, detail. Head of the Angel "Kyriotites"

. ,. -.

I

B. ---?"v-r*sugsl-,*

: 1 I

'\

&

\

\

\

& .'

..n"?gY*---*---

.'

.**-

7% "*

*$& " +.

.

\ " r'

>

a-

-. C>

6. The Mosaics of the Bema, South Side. Area of restoration

2,

<.

-

' ,

a r

-.,\

,.'I

:

,

\

I* \

\

,

4 -

7. The Mosaics of the Bema, South Side

i- 1

8. The Mosaics of the Bema, North Side. Area of restoration

9. The Mosaics of the Bema, North Side

The Date of the Narthex Mosaics of the Church of the Dormition at Nicaea Cyril Mango Dumbarton Oaks Papers, Vol. 13. (1959), pp. 245-252. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C245%3ATDOTNM%3E2.0.CO%3B2-T Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:43:34 2008

THE DATE OF THE NARTHEX MOSAICS OF THE CHURCH OF THE DORMITION AT NICAEA"

T

H E chronology of the Byzantine mosaic ensembles of the eleventh century, the "classicalage" of Byzantinemosaic decoration, is as yet somewhat imprecise. A fixed point is provided by Nea Moni, dated on the basis of reliable documentary evidence to the reign of Constantine IX (1042-56). The dating of the mosaics of St. Sophia at Kiev, after oscillating between 1017 and 1067, appears to have come to rest ca. I045 in the light of recent investigati0ns.l The decorations of Hosios Lukas and Daphni are undated, and although the consensus of opinion places the former in the early part of the century and the latter towards the end of it, voices are occasionally raised to contest this concl~sion.~ This dearth of exact dates confers particular importance on the narthex mosaics of the church of the Dormition (Koimesis) at Nicaea which are commonly assigned to 1025-28, and consequently regarded as the first link in the series of eleventh-century mosaic decorations. I t is the purpose of this paper to show that this date is unfounded, and that the Nicaea mosaics were probably made shortly after 1065. Since the church of the Dormition was completely destroyed in 1922,3 our discussion

* Figs. I, 4, 10, 11, and 14 are reproduced from photographs in the A. Kingsley Porter Collection, Fogg Art Museum, Harvard University, which derive from the Staatliche Bildstelle, Berlin. I should like to thank Miss K. B. Taylor, Assistant Librarian in the Fogg Museum, for supplying me these photographs. Figs. 11 and 14 have appeared in Th. Schmit's publication (see note 4). V. N. Lazarev, "Novye dannye o mozaikach i freskach Sofii Kievskoj," Vizant. Vremennik, x (19561, 164. Thus Lazarev, Istorija uizantijskoj iiuopisi, I (Moscow, 1947), 321, note 55, expresses the view that the mosaics of Daphni should be placed towards themiddle of the eleventh century. Cf. Comptes-rendus, Acad. des Inscr. et Belles-Lettves (1921)) 352-3 ; BZ, XXV (1925)~ 267-9.

must be based on the publications of Diehl, 0. Wulff, and Th. Schmit, and in particular on the photographs and architecturaldrawings accompanying Schmit's monograph which were executed in May- July 1912 by N. K. K l ~ g e I.t~is a matter of general agreement that the mosaics of the church fell into two groups: those of the conch and bema arch were manifestly of an early date, while those of the narthex, as well as two mosaic "icons" on the piers of the nave unquestionably belonged to the middle-Byzantine period. The date of the apse mosaics, especially of the four angels in the bema arch, has been the subject of heated debate among scholars, but may now be regarded as definitely settled, thanks to P. A. Underwood's paper in this volume. The great importance of these early mosaics, their high artistic quality, the profound theological conception of the apse decoration-all these factors have to some extent deflected attention from the rather more usual mosaics of the narthex, with the result that the date 1025-28, first proposed for them by Diehl in 1892, has never been subjected to a thorough examination, although Diehl himself, as we shall see, later changed his mind in this respect. The reader may be reminded that the mosaic decoration of the narthex was limited to the central bay, namely to a domical vault and to the lunette over the door leading into the nave (fig. I). In the lunette was a halflength figure of the Virgin orans (fig, 2); in the summit of the vault an eight-armed cross within a circle, surrounded by four medallions Ch. Diehl, "Mosaiques byzantines de NicBe,"

BZ, I (1892), 74-85, 525-6; reprinted with corrections in Etudes byzantines (Paris, I Q O ~ ) , 353-69; 0. Wulff, "Architektura i mozaiki chrama Uspenija Bogorodicy v Nikee," Vizant. Vvernennik, VII ( I ~ o o ) ,315-42.5 ; id., Die Koimesiskirche in Nicaa und ihre Mosaiken (Strasbourg, 1903); Th. Schmit, Die KoimesisKirche von Nikaia (Berlin-Leipzig, 1927).

246

CYRIL M A N G O

containing the busts of Christ, St. John the Baptist, Joachim, and Anna; and in the four pendentives of the vault the four Evangelists seated a t their desks. The date of this decoration depends on two inscriptions. The first, in mosaic (text figure), ran along the rim of

protovestiarius, the third, Symeon, became drungarius of the watch. All three received, furthermore, the important title of proedros. Another eunuch, Eustathius, was promoted to grand hetairiarch. Later on, Nicephorus retired to the Bucellarian theme in central

the lunette containing the figure of the orant Virgin, and was conceived as follows: K(iIp1)€ PoqBq (sic) T@uL;j80irhq N 1 ~ q 9 6 p q.rra-rp~~iq .rrpal. r r o o i ~ q P i m q ~ a ip ~ y a A q t ~ a ~ p t & p x qThis . inscription had been read almost correctly by Texier5 and reproduced in CIG, 8903. Diehl, however, in his 1Sg2 article, misread the words . r r a - r p ~ ~ripqa r - r r o o i ~ qPBmq as .rra-rpl~iq~ a i ~ r p w ~ o p B m qand , chided Texier for having given what was actually the true reading.6This mistake, which Diehl himself later corrected,' was to have unfortunate consequences. Under the influence of his earlier reading, Diehl tentatively identified the Nicephorus of the inscrip, eletion with a Nicephorus who, in I O Z ~was vated by Constantine VIII to the rank of proThis identification was accepted tovestia~ius.~ by Wulff9 because the second inscription, which we shall presently discuss, mentioned an emperor Constantine; Schmit and other scholars10 followed suit. Let us, however, look more closely at the evidence. Cedrenusl1 tells us that upon his accession Constantine VIII conferred high rank on the eunuchs of his entourage: his highest ranking chamberlain, Nicholas, was made domestic of the schools and parakoimomenos; the second, Nicephorus, was made

Asia Minor, and barely saved his life when that region was devastated by earthquake in 1035, whereupon he became a monk a t the monastery of Studius a t Constantinople.12 It is evident that this Nicephorus could not have been the dignitary who commissioned the Nicene inscription, since their respective titles were entirely different. The office of protovestiarius, the second most important in the hierarchy of eunuchs (after that of parakoimomenos) was quite distinct from that of vest&, which was created in the tenth century and was apparently a purely honorary one.13 Furthermore, had the Nicephorus of Nicaea held the exalted rank of proedros, it is inconceivable that he should not have recorded it in his dedicatory inscription." There is no reason to suppose, as Diehl has done, that the proedros Nicephorus succeeded Eustathius in the charge of grand hetairiarch; on the contrary, the appointment of Eustathius to that office makes it unlikely that any other person held it in the brief reign of Constantine VIII. I must admit that unfortunately I have not been able to identify the patrician, praepositus, vest&, and grand hetairiarch

Description de 1'Asie Mineure, I (Paris, 18391, 51. 6 Op. cit., 83. 7 Etudes byzantines, 366. Cedrenus, ed. Bonn, 11, 480. "Architektura i mozaiki," 421 ; Die Koirnesiskirche, 303. 10 S o G. Schlumberger, L'LpopLe byzantine, I11 (Paris, 1905)) 7 , note I ; 0. M . Dalton, Byzantine Art and Archaeology ( O x f o r d , I ~ I I ) 389; 0. W u l f f , Altchristliche und byzantinische Kunst, I1 (Berlin, 1g18), 558; E. Diez and 0. Demus, Byzantine Mosnics i n Greece (Cambridge, Mass., 1g31), 1 1 2 ; Lazarev, Istorija vizantijskoj divopisi, I , go. 11 Loc. cit.

l2 Cedrenus, 11, 514. C f . R. Guilland, " L e s eunuques dans ]'Empire byzantin," Etudes byzantines, I (1g43),204. l 3 O n these titles see J . Ebersolt, "Fonctions e t dignitCs d u Vestiarium byzantin," Me'langes Ch. Diehl, I (Paris, 1930), 84 sq. 14 T h e rank o f proedros was created i n 963 b y Nicephorus Phocas w h o made i t t h e highest i n t h e senatorial hierarchy. Originally i t was conferred o n o n l y one person. I n 1025, as w e h a v e seen, there were three proedroi. I n t h e second half o f t h e eleventh century t h i s rank became m u c h more common. See Ch. Diehl, " D e la signification d u titre d e 'prokdre' 8. Byzance," Me'langes G. Schlumberger, I (Paris, 1 9 2 4 ) ~105-17.

,

N A R T H E X M O S A I C S AT N I C A E A Nicephorus.15 Of his titles, the only one that could lead t o a n identification is that of grand hetairiarch, or captain of the emperor's personal guard, since the others were quite common, as was the name Nicephorus. Only a few hetairiarchs of the eleventh century are, however, known t o us. I n addition to Eustathius already mentioned, we find the following in the narrative sources: the eunuch Constantine, appointed in 1045 commanderin-chief of operations in Armenia,ls and later commander of the Byzantine army a t the disastrous battle of Adrianople in 1050;'~ David, a supporter of the infamous logothete Nikephoritzes, minister of Michael VII (1071-78) ;I8 under Nicephorus I11 (1078-81) we find Romanus S t r a b o r o m a n ~ s ;under ~~ Alexius I (1081-III~), Argyros Karatzas20 and Manuel S t r a b o r o m a n o ~The . ~ ~ numerous extant seals of grand hetairiarchs, among them one bearing the name of Nicephorus, which Schlumberger dates in the tenth or eleventh century,22 cannot unfortunately be used to determine the chronology of these officials. The second inscription, consisting of two iambic quatrains, accompanied a mosaic that was apparently over the door leading from the narthex into the south aisle. The mosaic represented the Virgin and Child flanked by an emperor and a Byzantine official. I t was seen in 1804by J. vonHammer, who copied the inscription in the following garbled manner :23 15 Professor R. Guilland has kindly informed me that he too had found no mention of this Nicephorus in Byzantine texts. On the office of the hetairiarch see his remarks in Byzantinoslavica, XIX (1958)~68 sq. Cedrenus, 11, 560-1. Cf. Ren6 Grousset, Histoire de 1'Armdnie (Paris, 1g47), 583. 17 Cedrenus, 11, 600 sq.; Attaliates, ed. Bonn, 33-4. Cf. Schlumberger, L'Lpope'e byzantine, 111, 583 sq. la Attaliates, 271 ; Cedrenus, 11, 734. David's seal has been published by Schlumberger, Sigillographie de Z'Empire byzantin (Paris, 1884), 34819 Attaliates, 286; Cedrenus, 11, 735-6; Anna Comnena, 11, 5. 5-7; Zonaras, ed. Bonn, 111, 726. 20 Anna Comnena, VIII, 7. 4. 21 His correspondence with Alexius I is preserved in Cod. Coislin 136. See R. Devreesse, L e fonds Coislin (Paris, 1g45), 128. 22 Sigillographie, 348, no. I. 23 Umblick a u f einer Reise von Constantinopel nach Brussa (Pesth, 1818), 112, 189-90.

I ANAIKPATAIOZ AEZllOTIZKONTTANTINOZ MONHNllPONOIAZAZH THNENOAAE A UPONAIAOZINQY KAEOlllATPI KIW BOYAITHIOKYPOtEN llEAU

El-WAEKAIAE t l l O I NANEIAIAKIONA TOYTON KAITONBUM U N A P X H l-ONENEQHNHMOINAYTOIN UZAEZllOTHAEKYPI ATHZ OIKIAt ETPAYATOAMM W N N I KHQOPOE llAPOENEIOY

This was reproduced in CIG,24 where a n attempt, largely unsuccessful, was made to emend it. The mosaic itself was seen again in 1813 by J. Nacdonald Kinneir,z5 in 182j by , ~ ~ in 1849 by A. Prokesch von O ~ t e n and , ~ ~ it disappeared some time A. M u r a v ' e ~ but before 1 8 g z . ~I~n 1834, however, while the church was undergoing restoration, a painted copy of this mosaic, which was then in a somewhat damaged condition, was made on the south wall of the narthex over the door leading into the adjoining square room (figs. 3 and 4). The original inscription was, fortunately, reproduced on the copy and 9 the emperor's recorded by W ~ l f f . ~Beside figure was written :

24

NOS.8782-3.

Journey through A s i a M i n o r , Armenia and Koordistan in the Years 1813 and 1814 (London, 25

1818)~26. 26 Denkwurdigkeiten und Erinnerungen aus dem Orient, I11 (Stuttgart, 1837)) 120. 27 P i s ' m a s Vostoka, I (St. Petersburg, 1851),

98-9. 28

See Colmar v. d. Goltz, Anatolische A u s -

Puge (Berlin, 1896), 440-1. 29 The photograph reproduced in figure 4 (taken in 1912 3 ) shows that in the early years of this century the entire narthex, including the composition that concerns us, was repainted and the inscription apparently obliterated.

CYRIL MANGO Next to the nobleman's figure it said: Eycj oa ~ a 6iorro1va(v) l ol6a ~ ~ ~ o p C r r w v ~ a 76, i ip6v &pxqybv BvBupqphwv &S 6ao-ir6~1v6i: wpiav TGS o i ~ i a s Eypaya T O E M UN N1~qq16pos rrapBiva.

As Wulff was quick to realize, this inscription had the same wording as that on the mosaic which Hammer recorded. What is more, the Greek painter of 1834 has given us an intelligible and, probably, a faithful copy of the text which lacks only the last verse of the first stanza. I t is likely that this line had been either destroyed or otherwise rendered illegible between 1804 and 1834. The double inscription may be rendered as follows: "The mighty lord, the Emperor Constantine, deeming this monastery worthy of support, gives it as a gift to the glorious patrician. . . .'j30 The second stanza, which in Byzantine terminology would have been called o-rixo~cbs BK rrpoocjrrou, was an address to the Virgin on the part of the patrician: "I acknowledge thee both as lady of my buildings and as leader of my thoughts; it is, however, as the lady and mistress of this houses1 that I, Nicephorus, have represented thee, 0 Virgin."3z These two stanzas provide the important information that the patrician Nicephorus had received the monastery as a donation from an emperor called Constantine, and that he had initiated some building activity in it. Attention should be drawn to the expression rrpovoias &c16v. There can be no doubt that the word pronoia is used here not just in its usual sense of "providence," "care," "solicitude," but as a technical term. In his detailed 30 T h e last verse, given as BO YAITHIOK YPO IENllEA U b y Hammer, is completed b y W u l f f (Koimesiskirche, 1 0 ) as pouhfj ~ ( b )K G ~ O S l v ~ i 6 4[, K ~ C ~ T OrEPOGhhl) ~ V ] T' ~ [ T E ] K ~Bv-ir664,. ~~oEv Neither restitution fits t h e metre, b u t t h e general sense, viz. t h a t t h e donation was confirmed or was to b e considered as binding, seems acceptable. 31 C f . t h e epithet ti o i ~ o ~ u papplied a to t h e Pharos church of t h e V i r g i n i n t h e Great Palace of Constantinople: Nikolaos Mesarites, Die Palastrevolution des Johannes Komnenos, ed. A. Heisenberg (Wiirzburg, 1 9 0 7 ) ~1 9 3337. ~ ~ ~ 3 2 I have l e f t i n t h e Greek t e x t t h e meaningless word 706' pwv as i t was transcribed b y W u l f f . Hammer read TOAMM UN. W u l f f himself prints t h e correction ~ o i a v suggested t o h i m b y J . Smirnov.

study of the pronoia-system, Ostrogorsky has established that this term occurs for the first time in the reign of Constantine IX (1042-55) concerning the donation of the Mangana (a region of Constantinople) to Constantine Lichoudess3In this case it is not clear whether the property consisted in the monastery of St. George of the Mangana, founded by Constantine IX who surrounded it with various dependent buildings, or in the imperial domain of the Mangana. In any case, it is generally admitted that the institution of the pronoia resembled in its initial stage, i.e. before the Comnenian dynasty, that of the charistikion (grants of ecclesiastical property to individual persons), and that no fast distinction was made at that time between these two terms. In the reign of Michael VII (1071-78) pronoia grants were sold in great numbers by the logothete Nikeph~ritzes,~~ who, it may be noted, had also held an influential place at court in the reign of Constantine X.35 Who then was the Emperor Constantine mentioned in the inscription ? In the eleventh century this name was borne by three emperors, Constantine VIII (1025-28), Constantine I X (1042-55) and Constantine X (1059-67). The first of these should most probably be eliminated since, as we have seen, the office of grand hetairiarch was held during his brief reign by Eustathius; furthermore, the use of the term pronoia suggests a somewhat later date. Thus we are left to choose between Constantine IX, in whose reign, however, the post of grand hetairiarch was occupied at least from 1045 till 1050 by the eunuch Constantine, and Constantine X. An examination of the structure of the church may help us to make this choice. Towards September 1065 Nicaea was shattered by a violent earthquake. This was 33 Pour l'histoire de la fe'odalite' byzantine, trans. H . Grkgoire and P. Lemerle (Brussels, 1954), 20 sq. T h e t e x t s are Cedrenus, 11, 645 and Zonaras, 111, 670. 3* Attaliates, 200-1. C f . Ostrogorsky, op. cit., 2 2 sq. ; SkabalanoviE, Vizantijskoe gosudarstvo i cerkov' v X I veke ( S t . Petersburg, 1884), 264; P. Charanis, " T h e Monastic Properties and t h e State i n t h e Byzantine Empire," Dumbarton Oaks Papers, 4 (1948), 69. O n charistikia and pronoiai, ibid., 72 sq. 35 O n t h e career o f Nikephoritzes see Guilland i n Etudes byzantines, I (1943),230-1.

N A R T H E X M O S A I C S AT N I C A E A actually the last of a series of shocks that occurred intermittently over a period of two years. The first quake occurred on September 23, 1063, and caused considerable damage in Constantinople and the coastal cities of the Propontis, especially a t Cyzicus, where Hadrian's famous temple ~ o l l a p s e d "After .~~ the two year period," writes Attaliates, "there having occurred an earthquake more violent than the repeated ones that had followed [sc. the first shock], but lesser than the first one, which was the greatest, Nicaea in Bithynia suffered a disastrous collapse and almost complete destruction. For its most famous and greatest churches, both the one dedicated to the Wisdom of God's Logos [i.e. St. Sophia] and serving as the metropolitan seat, and that of the Holy Fathers (where the Synod against Arius had been confirmed by the most-holy and orthodox fathers, and the true creed was proclaimed and shone forth brighter than the sun), were shaken and demolished; and the walls as well as private dwellings suffered the same fate. And from that day on, the quakes ceasedeH37 In his study of the city-walls of Nicaea, A. M. Schneider has established that about twenty towers as well as stretches of wall between towers were damaged by the earthquake of 106j and rebuilt in a very characteristic technique which he calls "verdeckte Schichttechnik "(fig. 5). Schneider has observed the same technique in the remains of the Dormition church, specifically in two massive piers which showed traces of the springing of arches. The lower part of these piers was built of normal brickwork (bricks 3.5 cm. thick; mortar joints j to 7 cm.), while the springing of the arches was in the "recessed brick technique," which led Schneider to conclude that the dome of the Dormition church and its supporting arches were rebuilt after the earthquake of 106j.38 Before discussing Schneider's statement, I should like to say a few words concerning the "verdeckte Schichttechnik." This system 36 Attaliates, 87-90; Cedrenus, 11, 657; Zonaras, 111, 679-80; Glycas, ed. Bonn, 605-6; Ephraem, ed. Bonn, 144. 37 Attaliates, 90-1. 38 A. M. Schneider and W. ICarnapp, Die Stadtmauer von Iznik (Nicaea), Istanbuler Forschungen g (Berlin, 1938), 40-1.

of construction is readily identifiable by what appear to be disproportionately thick mortar joints (ranging from about 8 to 15 cm., the bricks being about 3 to 4 cm. thick). Actually, the pointing of this joint, which may be flush or slanted ("weathered"), conceals an intermediate course of brick which comes just short of the vertical surface.39Whenever the pointing is intact, the recessed course is naturally invisible, but its presence may be inferred with certainty if the joints are sufficiently thick. The same technique occurs also in arches. The occurrence of the "recessed brick technique" is of great significance for purposes of dating, since its use appears to have been limited to the eleventh and twelfth centuries. The greatest number of examples is to be found at Constantinople. We may note the following: substructures and various dependencies of St. George, Mangana, probably dating from the reign of Constantine IX (1042-5 j);40 St. Mary Pammacaristos, central church and cistern under it (probably middle of the eleventh century);41ruins on top of the hill of the island of Antigoni (Burgaz), one of the Princes' Islands (probably middle of the eleventh century)42(fig. 6) ; monastery of the Chora (Kariye Camii), main church (ca. 1080); St. Saviour Pantepoptes (end of the eleventh century) ; the so-called "Prisons of Anemas" (ca. 1 1 0 0 ) ; ~St. ~ Saviour Panto39 See A. M. Schneider, Byzanz, Istanbuler Forschungen 8 (Berlin, 1936), 14. 40R. Demangel and E. Mamboury, L e quartier des Manganes et la firemiire rLgion de Constantinofile (Paris, 1939)~figs. 25, 26, 29, 30, 42, 47, 48. 41 The presence of this technique was revealed when the interior was stripped of plaster in the course of restoration work carried out by the Byzantine Institute. For the cistern, see Aziz Ogan, "Fethiye Camii," Belleten, XI11 (1g4g), pl. XLIII. The central church may perhaps be connected with the foundation of John Comnenus, curopalates and grand domestic (d. 1067). Cf. R. Janin, Les iglises et les nzonast2res ( L a ge'ografihie ecclisiastique de , 21, 1'Emfiire byzantin, 1,-3 l ~ a r i s 1g53]), 42 See J. B. Papadopoulos, "Les ruines de l'ile d'Antigoni," Byzantinisch-neugriechische Jahrbiicher, V (1gz7), 86; Montfaucon, Palaeographia graeca (Paris, 1708), 52. 43 B. Meyer-Plath and A. M. Schneider, Die Landnzauer aon Konstantinopel, I1 (Berlin, 1g43), 25

CYRIL MANGO half of the eleventh century),56as wellvas in crator (first half of the twelfth century) (fig.7) ; the Spaso-Preobraienskij cathedral of Cerniportions of the walls of Manuel Comnenus at gov (ca. 1036), the cathedral of St. Sophia the Blachernae (ca. 1150);~~ Odalar Camii at Polotsk (second half of the eleventh centu(twelfth century)45 (fig. 8); St. Theodosia ry), the church of SS. Boris and Gleb also at (Giil Camii), side apses;46 church at Aya Polotsk (early twelfth century),57etc. In fact, Mescidi, wrongly identiK a ~ i ; ~Sekbanba~i ' until about the middle of the twelfth century, fied with the monastery of Kyra Martha;4s a the recessed brick technique was the dominant ruined church near the village of Bostancik on the Asiatic shore of the Sea of M a r m ~ r a , ~ ~ type of construction in Kievan Russia. Outside of Russia, we find this technique in the etc. In Nicaea, in addition to the city walls restorations made by Constantine IX in the and the Dormition church, we find this technichurch of the Holy Sepulchre at Jerusalem que in the apse of St. Sophia, a repair that in the monastery of Nerezi (1164)~ (1048),~~ has also been connected with the earthquake and in the church of St. Nicholas at KurBumof 106j,50and in the ruins of another church.51 lija in Serbia (1168) (fig. g ) , which has been The recessed brick technique is also found recognized as a monument of almost pure Conoutside the immediate neighborhood of Constantinopolitan style.59 stantinople; in fact, the earliest dated exThus, the recessed brick technique may be amples come from Kiev, where this type of considered as a hallmark of Constantinopolitan construction must have been introduced by architecture, introduced early in the eleventh builders from the Byzantine capital. We find century and used until the end of the twelfth. it in the added ambulatory of the Tithe church To my knowledge, it is not found outside of (Desjatinnaja Cerkov') (ca. 1 0 3 g ) , ~in~ St. Sophia (1037-46), the Golden Gate (10371,~~ these chronological limits. The theory of an Anatolian origin, proposed by Brunov,BOmay the church of the Vydubickij monastery be safely discarded since, with the exception (1070-88),~~ the church Spas na Berestove of Nicaea and a ruined church, probably of (late eleventh or early twelfth century),55the the twelfth century, at Kur~unluin Bithynia ruins of a church on the property of the Kiev (on the southern coast of the Sea of MarInstitute of Art, excavated in 1947 (second m ~ r a ) , this ~ l technique does not appear to have been found in Anatolia. 44 Ibid. 4 j N. Brunov, " D i e Odalar-Djami v o n KonIf we now examine the extant photographs stantinopel," B Z , X X V I (1926),360; P. Schazof the Dormition church, we can easily rem a n n i n Atti del V Congresso Intern. di Studi cognize the two systems of construction noted Biz., I1 (1940), 376. by Schneider. The apse and bema, and 46 Brunov, "Die Giil-Djami v o n Konstantipossibly also the east walls of the projecting nopel," B Z , X X X (1929/30),560; id., " R a p p o r t sur u n voyage & Constantinople," Rev. des e'tudes rooms attached to the north and south ends grecques, X X X I X (1926),20-1. of the narthex, were built of normal brick4' Schneider, Byzanz, 53 and fig. 1 1 . work. According to Schmit's measurement^,^^ 48 Ibid., 63. C f . V . Laurent, " K y r a Martha," which agree with Schneider's, the bricks were Echos dlOrient, X X X V I I I (1939),296-320. 49 Brunov, " U n e Cglise monastique a u x environs de Chalckdoine,' Echos d'ovient, X X V I (1927),35. 50 Brunov, "L'Cglise de Sainte-Sophie de Xicke," Echos d'orient, X X I V (1925), 474; A. RI. Schneider, Die romischen und byzantinischen Denkmaler von Iznik-Nicaea, Istanbuler Forschungen 16 (Berlin, 1943), 1 2 . 51 Semavi Eyice, "Iznik'de bir Bizans kilisesi," Belleten, X I 1 1 (1949), 39 and pl. X V I I , fig. 6. 5 2 M . K . Karger, Archeologic'eskie issledovanija dvevnego Kieva ( K i e v , 195 I ) , 64-5,79 and fig. 5 I . 53 Akademija N a u k S S S R , Istorija russkogo iskusstva, I (Moscow, 1953), 127. 54 Karger, op. cit., 147 and fig. I I I . 5 5 Istorija russkogo iskusstva, I , 142-3.

Karger, op. cit., 223. Istovija russkogo iskusstva, I , 319. 58 H . V i n c e n t and F.-M. Abel, Je'rusalem, 11, 1-2 (Paris, 1914),fig. 9 7 ; W . Harvey, Church of the Holy Sepulchre Jerusalem, ( O x f o r d , 1 9 3 5 ) ~ fig. 68. 59 G. Millet, L'ancien art serbe. Les Lglises (Paris, 1919), 52-54; J . Ebersolt, Monuments d'architecture byzantine (Paris, 1934), pl. VIII, 2 . 60 Rev. des e'tudes grecques, X X X I X (1926), 2 1 ; B Z , X X V I (1926),361. 61 M. Ramazanoglu, " E i n e kleine Kirche i n Bithynien," l-le-rrpayykvaTOG O'A1~9vo0sBulavrlvohoymo0 Zuva6piou, I (Athens, 1955), 441 and pl. 109. 62 Die Koimesis-Kivche, 5. 56 57

N A R T H E X M O S A I C S AT N I C A E A 32 by 3.5-4 cm., while the mortar joints were 5 to 7 cm. thick. The recessed brick technique is, on the other hand, clearly indicated by the excessive thickness of the joints in the dome base (fig. 11) (the dome itself was rebuilt in 1807), and in both the north and south tympana. This may be observed in the brick mullions of the triple window and in the characteristically recessed arch of the tympanum of the south side (fig. IZ), and on the north side as well, though not as clearly.63 Thus the dome and its supporting arches must have been rebuilt following some serious damage to the church, and since the technique employed in the restoration is limited to the eleventh and twelfth centuries, it is reasonable to assume that this damage was caused by the disastrous earthquake of 1065. Now, the west faqade of the narthex was built in precisely the same technique as the dome base and tympana, and had recessed arches similar to those of the tympana. The very thick joints are clearly visible on Wulff's photograph of the entrance into the narthex (fig. 13) as well as on another photograph of the west faqade that does not appear to have been published heretofore (fig. 10). We may therefore conclude that the narthex was rebuilt at the same time as the dome base and tympana, hence after 1065. The plastic treatment of the west faqade with its wall arcades formed by several recessed archivolts (fig. 14) and its rounded niches within which the bricks are arranged in a fan-like pattern, is also indicative of the eleventh century, and may be compared to Kilise Camii and St. Saviour Pantepoptes at Constantinople, Kazancilar Camii at Salonica, St. Sophia at Kiev, etc. In fact, the plastic decoration of the faqade is enough to prove that this part of the building is much later than the original pre-iconoclastic church, notwithstanding Schmit's assertion of the c0ntrary.6~ If, Ibid., fig. 2 and pls. I and 11. Ibid., 18. Schmit states that the faqade wall of the narthex was contemporary with the katholikon because it was not parallel to the east wall of the narthex, but was exactly parallel to the east wall of the church. I am unable to see the cogency of this argument, since the faqade wall was probably rebuilt along its original line, and besides the discrepancy involved is quite negligible. Wulff also regarded the

however, it is true that the f a ~ a d ewall was rebuilt after 1065, then the narthex mosaics could not have been earlier. In this connection we may note the differences in the corbels on which the arches on either side of the domical vault of the narthex rested (fig. 4); the pair flanking the entrance into the naos were similar to one another, whereas those set into the west wall were quite different, the one to the south having a moulded profile with two fillets, while the one to the north had a plain bevelled face. The dissimilarity of these ~ ~ quite corbels was noted by W ~ l f f ,who correctly concluded that they were inserted in connection with a rebuilding (Wulff thought in terms merely of a remodelling of the central bay) in the eleventh century. In conclusion, I should like to suggest that the monastery of the Theotokos to which the Koimesis church belonged was given to the patrician Nicephorus as a pronoia-grant by the Emperor Constantine X (1059-67). In the earthquake of 1065 part of the church collapsed, and Nicophorus rebuilt the dome and its supporting arches as well as the west faqade (these are probably the ~ ~ i o p c r r a mentioned in the inscription). I t is possible that other parts of the church, e.g. the north and south walls, were restored at the same time, but in the absence of photographs or a detailed description of the masonry this cannot be determined. The narthex mosaics date from the same period, and since the "ktetoric" mosaic appears to have represented Constantine X as the reigning emperor, they must have been executed between 1065 and 1067. I must add that this conclusion is not entirely new. In fact, Diehl, whose incautious suggestion made in 1892 led to the adoption of the date 1025-28, later changed his mind concerning the date of the narthex mosaics. In a postscript to the revised version of his original article, published in 1905, he calls attention to the earthquake in the reign of Constantine X, and remarks: "On peut se

63

64

fayade wall of the narthex as part of the original structure. See the sketch-plan in his Bibliographisch-kritischer Nachtrag z u altchristliche und byaantinische Kunst (Potsdam, 1936)~5 7 , fig. 541. 6 5 "Architektura i mozaiki," 333 ; Die Koimesiskirche, 29. Wulff is mistaken in saying that three of the corbels were alike.

CYRIL MANGO basilica of St. Demetrius at S a l ~ n i c aThe .~~ demander si, aprks un tel dhsastre, l'hglise original Antiphonetes icon was kept in the de la Koipqois aussi n'eut point besoin church of St. Mary Chalkoprateia at Cond'Ctre restauree, et si le grand hCtCriarque stantinople, and was associated with a famous NicCphore, qui vivait, nous le savons par legend, found in numerous manuscripts of l'inscription signalee plus haut, sous un the eleventh and later centuries, concerning empereur Constantin, ne doit pas en cons4 the Jewish moneylender Abraham and the quence Ctre reportk, lui et son oeuvre, & la . ~ is ~ difficult to tell in merchant T h e ~ d o r e It seconde moitih du XIe sikcle."66 Later on, what period the Chalkoprateia icon acquired Diehl seems to have adopted this latter a certain measure of celebrity, since the alternative, since in his Manuel d'art byzantin67 Abraham legend (which, in part, may go back he says without any further explanation: to the seventh century) was not originally "Les mosaiques qui dkcorent le narthex concerned with it; in fact, the association appartiennent . .. certainement au milieu du with the Chalkoprateia is made only in the XIe si&cle,et le grand hCtCriarque NicCphore, title of the legend as found in our manuqui les fit exkcuter, vivait sans doute sous s c r i p t ~However .~~ that may be, the cult of Constantin X (1059-1067)." Christos Antiphonetes was popularized by In the above demonstration I have rethe empress Zoe (d. 1050). Psellus, in his frained from mentioning any consideration account of Constantine IX, relates under of style, since our knowledge of eleventhG Zoe the heading Frr~pl TOG ' A v ~ ~ g w v q ~ othat century mural decoration is not sufficiently had a precious icon of Christ which by slight detailed to provide exact chronological changes in the color of its complexion forecriteria; for example, the fact that the warned her of the future: when the Lord's narthex mosaics of Nicaea are less linear than face appeared pale, that portended disaster, those of St. Sophia at Kiev does not neceswhen it looked rosy, that was a good omen.73 sarily imply an earlier date. I should like, Zoe also built a church of Christ Antiphonetes however, to mention one further considerin which she was b ~ r i e d . 7One may be ~ ation which may have some significance. On tempted to suppose, therefore, that the apthe eastern piers of the Dormition church pearance of this unusual epithet at Nicaea had were two mosaic "icons" that have been some connection with the patronage extended published by Schmit. They represented Christ with the unusual epithet 6 ' A V T I ~ ~ V ~ T to ~ SChristos Antiphonetes by the Empress Zoe, and to consider this as a further confirmation and the Virgin Eleousa. These mosaics of the date suggested here for the mosaics. have been dated in the tenth or eleventh century and, although direct proof is lacking, 70 G. A. and h f . G. Soteriou, 'H paolhl~fi TOG it is quite likely that they were contemporary &yiou Aqpq~piou@ ~ o o a h o v k(qA~t h e n s ,1952),209 with the narthex mosaics.68 The epithet ( n o t illustrated). 7 1 O n t h i s legend see esp. Nelson and Starr, Antiphonetes is not mentioned in the " T h e Legend o f t h e Divine Surety and t h e Painter's Guide,es and was exceedingly rare Jewish Moneylender," A n n . de l'lnst. de Phil. in Byzantine iconography; in fact, the only et d'Hist. Orient. et Slaves, V I I (1944),289-338. other instance of which I am aware is a fresco 72 I h a v e discussed t h e history o f t h e Antiof the fourteenth or fifteenth century in the phonetes icon i n a s t u d y entitled The Brazen Etudes byzantines, 369, note I . G. Rfillet i n A. Rfichel's Histoire de l'art, I , I (Paris, r g o j ) , 196 also states t h a t t h e church o f t h e Dormition was restored after 1065, b u t h e incorrectly applies t h i s date t o t h e mosaics o f t h e presbytery. 6 7 2nd ed. (Paris, 1926), 521. 68 SO Lazarev, Istorija vizantijskoj iivopisi, I , 307, n o t e 47. 1 3 ~D ionysius o f Fourna, Manuel d'iconographic chre'tie~zne, ed. Papadopoulo-KCrameus ( S t . Petersburg, 19og), 227, 281.

House, t o b e published b y t h e Royal Danish Academy i n Arkaeologisk-kunsthistoriske Meddelelser, I V , 4 (1959). 73 Chronogr., ed. Renauld, I (Paris, 1926), 149-50. 74 Sathas, Bibl. graeca medii aevi, V I I , 163. C f . Janin, Eglises et monastkres, 520. There is also a monastery tou Antiphonitou, probably o f t h e t w e l f t h century, near Kyrenia, Cyprus. See G. J e f f e r y , A Description of the Historic Monuments of Cyprus (Nicosia, 1918), 336-8 ; G. A. Soteriou, T a p u l a v ~ ~ vpvqpeia a Kimpov, I ( A t h e n s , 1 9 3 j ) , fig. 14, pls. 24 and 59b.

1. Nicaea, Church of the Dormition. Narthex, Vault Mosaic

2. Nicaea, Church of the Dormition. Narthex, Lunette over Central Door

5. Nicaea, City Walls. Detail of Tower

6. Island of Antigoni (Burgaz), Byzantine Ruin, detail

7. Constantinople, St. Saviour Pantocrator, West Fa~ade,detail

8. Constantinople, Odelar Camii. Detail of Brickwork

9. KurHumlija, Church of St. Nicholas 10. Nicaea, Church of the Dormition. West Facade

11. Nicaea, Church of the Dormition. Dome

12. Nicaea, Church of the Dormition. View from Southeast

13. Nicaea, Church of the Dormition. West Faqade, Censtral Door

14. Nicaea, Church of the Dormition, West Faqade

http://www.jstor.org

LINKED CITATIONS - Page 1 of 1 -

You have printed the following article: The Date of the Narthex Mosaics of the Church of the Dormition at Nicaea Cyril Mango Dumbarton Oaks Papers, Vol. 13. (1959), pp. 245-252. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C245%3ATDOTNM%3E2.0.CO%3B2-T

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 34

The Monastic Properties and the State in the Byzantine Empire Peter Charanis Dumbarton Oaks Papers, Vol. 4. (1948), pp. 51+53-118. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281948%294%3C51%3ATMPATS%3E2.0.CO%3B2-X 34

The Monastic Properties and the State in the Byzantine Empire Peter Charanis Dumbarton Oaks Papers, Vol. 4. (1948), pp. 51+53-118. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281948%294%3C51%3ATMPATS%3E2.0.CO%3B2-X

NOTE: The reference numbering from the original has been maintained in this citation list.

Greek Terms for "Flax," "Linen," and Their Derivatives; And the Problem of Native Egyptian Phonological Influence on the Greek of Egypt Demetrius J. Georgacas Dumbarton Oaks Papers, Vol. 13. (1959), pp. 253-269. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C253%3AGTF%22%22A%3E2.0.CO%3B2-4 Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:44:02 2008

GREEK TERMS FOR "FLAX," <'LINEN," AND THEIR DERIVATIVES;

and the Problem of Native Egyptian Phonological Influence

on the Greek of Egypt*

I. ANCIENT, KOINE, MEDIEVAL, AND MODERN GREEK TERMS AND NAMES

F

LAX and LINEN. The Indo-European peoples knew flax and linen, as is shown by the term *lino- "flax, linen" : Greek Aivov, Slavic *16na, Baltic lina; (all with short i ) , Latin linum, Celtic (Irish lin), and Germanic (Gothic lein, Old English lin, etc.) (all with long i).l The peoples living around the

* I t is a pleasure t o acknowledge m y indebtedness t o Professors J o h n L. Heller and Herbert C. Y o u t i e for their helpful criticism o f t h i s paper, and t o Dr. Donald C. Swanson and Dr. E d m u n d Berry for stylistic improvements. I should also like t o t h a n k Dr. Abdullatif A. A l y (Cairo), Mr. Alexander P. Clark, Curator o f Manuscripts o f t h e Library o f Princeton University, and Dr. J . R. Ashton, Librarian o f t h e University o f N o r t h Dakota, for help i n securing papyrological material, and Professor K u r t L a t t e (Gottingen) w h o m I consulted o n a reading i n Hesychius. S o m e modern Greek material derives f r o m t h e Archives o f t h e 'lrropi~bvA E ~ I K-rijs ~ VNBas 'EAAqvl~iis ( A c a d e m y o f Athens) and o f t h e M ~ u a l u v t ~ b'Apx~iov v (ibidem). I a m t h a n k f u l t o Dr. J o h n Kalleris, Director o f t h e former institute, for supplying information f r o m t h e Historical Lexicon, as well as t o Mr. Nicholas Kontosopoulos (Historical Lexicon) and t o Dr. Phaedon Bouboulidis (Medieval Greek Archives). Further modern Greek material was received f r o m friends i n various parts o f Greece i n answer t o t w o questionnaires sent b y m e i n December 1956 and October 1958, w i t h kind support f r o m t h e office o f Dean R. B . W i t m e r o f t h e University o f N o r t h Dakota. T h e names o f these informants appear i n another s t u d y (Memorial Volume for Manolis Triandaphyllidis, Athens, 1960), b u t I wish t o mention here m y obligation t o Dr. Demetrios Loucatos (Folklore Institute, A t h e n s ) , Mr. Christodoulos Papachristodoulou (principal, Venetokleion G y m nasium for Boys, Rhodes, Greece),Mr. Panayotis Bournelos (Carystos), Mr. Panayotis Stamos (Polygyros), and Mr. Constantine Psychogios (Lechaina). J . Pokorny, Indogermanisches etymologisches Worterbuch (Bern, 1948-58), p. 691 ;

Mediterranean also cultivated flax2 and its processing and products were known as early as the Mycenaean age, as the Mycenaean ~ r e k kterms of the fourteenth century B.C. that occur on tablets of Pylos and Knossos adequately attest : Zinon "flax," "linen," "linen thread," etc. (KN 222, PY 184)~ lineiai "female flax workers" ( P Y ) and lineiaon (gen. plur., PY 8), a derivative of hlv~rjs (as Baoiheia fern. "queen" from BaalIn Homer hivov is frequently used and 0. Schrader-A. Nehring, Reallexikon der indogerman. Altertumswissenschaft2, 1.323 ff. ; A. Ernout-A. Pl'leillet, Dictionnaire Ltymologique de la langue latine3 (Paris, 1 9 5 1 ) p. ~ 6 4 3 f . ; A. W a l d e J . B. H o f m a n n , Lateinisches etymologisches Worterbuch3 (Heidelberg, 1938-56), 1.810 f.; c f . C. D. B u c k , A Dictionary of Selected Synonyms i n the Principal Indo-European Languages (Chicago, 1g4g), p. 4001. Victor Hehn, Kulturpflanzen und Haustiere i n ihrem Ubergang aus Asien nach Griechenland und Italien, 8. Aufl., neu herausg. v o n 0.Schrader (Berlin, I ~ I I ) ,p. 192. - For more complete information o n flax and linen see Olck, i n Pauly-Wissowa, Real-encyclopadie der class. Altertumswissenschaft, s.v. Flachs, 6 ( ~ g o g ) , cols. 2435-2484; Hugo Bliimner, Technologie und Terminologie der Gewerbe und Kiinste bei Griechen und Romern, 1 2 (Berlin, I ~ I Z ) pp. , 191199; 0. Schrader-A. Nehring, Reallexikon der indogermanischen Altertumskundee, I (1g17-23), 323b-326b, s. v. Flachs; H . Thkdenat, i n Ch. Daremberg e t E. Saglio, Dictionnaire des antiquitis grecques et romaines, (Paris), 3.2 goo), 1260b-1263b, s.v. linum. C f . H . Michell, The Economics of Ancient Greece, ( N e w Y o r k , 1g40), pp. 5 8 f . ( f l a x ) , 1 8 4 f . (linen and silk), z g o f . (linen); J o h n Kalleris, Al -rrpGral SAal - r i j ~ ir~avrovpyias €15 T ~ V ~ T O ~ E H ~ A~ l pK m~ o~v V, ( A t h e n s , I952 = ' E ~ a - r q p lAaoypapl~oG ~ 'Apx~iov o f t h e Academy o f Athens, 6 [1g50-511). See Michael Ventris and J o h n Chadwick, Documents in Mycenaean Greek, (Cambridge, 19-56),pp. 131,15g,zg6f., 320, and (Vocabulary) 408 a.

254

DEMETRIUS J. GEORGACAS

it occurs as "linen garment" through the Hellenic period to the New Testament and through the middle ages. Egypt as well as the Byzantine Empire4 produced linen. .The cultivation of flax and the linen industry were wide spread in Greece itself, as is shown by terms designating production of flax and manufacture of flaxen articles and linen clothing in all periods of the Greek language and by place-names derived from the terms for "flax" etc. This is made sufficiently clear by the following table of terms. I t is worth noting that Egyptian and Greek flax is that known as linum usitatissimum L. I.

Aivocp96po~

h~voxirwv(Hesych.)

&uaihlvo~

tKKal6~~ahlvo~

Ivvedihlvo~

etc.

2.

Koine Greek

hivov (and Aivov, name of a place on the Propontis with ethnicon A~voiroio~) hlvaio~ A~vdrpiovand Aivdrptv A~vdrpios(Lat. linarius) Aivdrpuevov

Ancient Greek

linon (Mycenaean) and hivov (Iliad) lineia "woman flax worker" (Mycenaean) hive05 (Attic hlvoijs) "of flax, linen" and Alvlq f. "tape measureJ'; also hivaios piros (Souda) hivepyfis, h i v ~ p y i j(-la) A~voyavfi~ A1v66eapos ~ I V ~ ~ E T O S

A1v6lworo~ hivo9fipa~m. h1vo9Gpa$ (-pqt) A I V O K ~ ~ ~ U KOI E S T& ' hlva

TTO~O~~VTES

(Hesych.) ~IV~KPOKOS

h1~6.rre.rrhos hivo.rr6po~ Hesych., h~vomdropai) h l v 6 m q ~(h~vomdrle~ hlvoppaafis hlvoolvfi~ hiv6moho~ hivoupy6~

h~vllyiov hively6~ h l ~ l ~f.f i Aiv~vo~ h1v66pus hlvoepyfis A~V~~EUKTOS hlv0Kddryq hlvo~ahapis hlvo~&lapov A~vo~p~Sfi hlvo56~ ~IV~T~~EKTOS

hlv6Tihq~~o~

j21vo.rrhfig

hlvo.rrh6~0~ "linen-weaver" (cf. 69ovlo.rrh6~os)

h1vo.rrhirvas [-.rrAirvras ?] ' rpii3eir~(Hesych.)

hivo.rrhirq~(not hivo.rrhvri~with LSJ)

h1vo.rroi6~(cf. 69ovio.rro16s)

hiv6mspo~

h~vomlpu$

4 For the Roman Empire see E d i c t ~ w hiv6mrpo~

Diocletiani, ed. Th. Mommsen (Berlin, 1893 = K~S,

6Soviohlvo.rrbhqs (cf. ~ S O V I ~ 690vto.rrpdTTq~, CIL,3, Supplement), $$ 26-29, pp. 39-48 on .rrbhqs "linen merchant") hivov. For Byzantine times cf. Leo Philos., Tb h r a p x ~ ~ bPl~hiov, v ed. J. Nicole, 2.1; 9.1,6 hives m. (= hivov) and 7: 'Aooilai K ~ T ~ OC. U ,Sathas, M ~ o a ~ o v ~ ~ f hlvbS i "made of linenH (p.Mas*. 67006, BlphioSfi~q (Venice), 6.489 (linum brought to [6th cent.] -rrpoo~~qahea hiva rpia is misacByzantine markets from Egypt); Adam Mez, cented as hiva by Preisigke, W&terbuch, The Renaissance of Islam (London, 1937); 2.24 instead of A~vdr); h1v6v Etym. m. pp. 458-462; cf. G. Zoras, Le corfiorazionz bizantine ; studio sull h r a p x l ~ b vPiPhiov dell' im566.36-38. This adj. is from anc. hlvoijs, as peratore Leone VI (Rome, 1g31), pp. 179-182 txTih6s from txrrhoij~,x p u o 6 ~from xpuooijs, etc. (on 69ov1o.rrphal [not 6~avio.rrp6ral]); Ph. Koukoules, Bulav-r~vijv Pi05 ~ a lrroh1~iap6~,2.2 h1v60.rra~T0v

(Athens, 1948), 22f. See also the following note. hiv6omppa (cf. hivou o.rrlppa)

G R E E K TERMS F O R "F'LAX" AND " L I N E N "

A1v6u.rrsppov Zha~ov h~vom~pplvov hivomaoia Aivomcc-rG (-&a) hiv6mqpa (Lat. linostema) hlvomohia h1v6mpogo~ hlvo-rslxqs hiv0~6poi h1voir8lov (see below) hlvohm (see below) h~voupy~iov hlvoupyia h1voupy6~ h1voupyG (-iw) hlvoGs h~voijq[acrvrsia] hlvoijcpfi~ hlvoijcp1~6~ h1v6ijcpos AivoGxo~ hlvocpa~6s h~vocpttvnqs(from h~vovcp&vrqs) hlvocp6po~yij h1v6~ha1vo~ ~ I V ~ X O ~-ov TOS, hivucpav-r&p~o~(and hsvuqavr&p~os under the influence perhaps of hivriov) hivucpavreiov hlvucpavrqs hivucpavri~bs h~vucpap~os hlvucpeiov hl~cp1~6~ hivucpos hivwvia aIv6h1vo~ &~p6hivo~ 3hivos Sqpa &.rrohivG(-6w) &.rrohivwoi~ Gtdlvilw, -opal (not G~ah~vdrw, -vGpai) i~hlvilw(not M ~ v a w ) E~rlhivsirw EITI~IVEUT~~S I.rr~hivEv'P.rrl.rropsirsoSai T& 6rnB.r~Aiva ~ a ErrtPMl TEIV

(Hesych.)

hew6h1vov ~ov6h1vov, TETP&AIVOV, ~pihivov 6.rr6hlvov. ~b dpplvov (Hesych.) bp6hlvaS~h Ciypia b96vla (Hesych.) bp6Alvos (synon. ~T~.TTIVOS)

etc.

3. Medieval Greek hivov in medieval Greek is a learned (Koine) term;s as is also the plur. ~h hiva meaning "white linen" and "hunting net" (Eustathius, k p ~ ~ p o h a1452.60; i, 574.30). hlvap~ov Const. Porphyrog., De cer. 658.13, 673.5 and g ;Leo Philos., Tb hrapxi~bvPlPhiov 13.1 (ed. J. Nicole, Geneva, 1893, p. 47); AchmetisOneirocr. 219 (ed. F. Drexl, p. 172, lines 12 and 15) h~vapiou219 (p. 172, line 14), 210 (p. 165, line 2) ;in the latter passage xdpav Pmapp6vqv hivapiou should be emended t o read hlvaplv, which mss. rBS actually give; Michael Choniates, 2.365.13 (ed. S. Lambros) & hivapiou aipeolv 6rro~pivs~al; -rb 61~aiwpavTOG hlvapiou Assises B, o q l (ed. C. Sathas, MEUU~COVIK~~ BlPhioSfi~q,6 118771, P. 489, line 9 ) ; the plur. -rh AivCtpla means also "hunting nets" = anc. Greek -ra hiva (see the preceding). The spoken medieval form was no doubt hiv&plv, which is recorded in this form by Achmet (see above) ; Prodr. 1.96 ~ a p v w~b hlvaplv, 2.35 hivap~v ~ a PapPCXKiv; l also Steph. Sachlikis in his 'Acpqyilos~s461, 463 (ed. W. Wagner, p. 95). hlvap&s m. Du Cange, Glossarium Graecit., s.v. ; see the following. h~vapo.rrouhq~fis m. Du Cange, ibid., S.V. = Koine hiv&y.rropo~, h~vorrdhq~. i hlvih~v:EIS ~ q p o r n o i r ~ ~adi vGaGiv, EhaGlv ~ ahivih~v Prodr. 2.53 (ed. Hesseling-Pernot). Perhaps for hiv6ha1ov. hiv6s "made of linen": &p&Aiahlvh [not hivB] ylha 81m6om Theophan. Contin. (ed. Bonn), 318.16; E ~ T Ehlva [not hlvE] E ~ U E~ ~VT E Epiov fi hivou G~aqavoGs 6cpaopcc-r~Gregorii Nyss. Oratio V I I (A.D. 368/9), Migne, PG, 3 5.7 73 D ; TO~SEK hivou ~ a oqpijv 1 &piois ircpaopaol, idem, Orat. XI V (A.D. 3731, ibid., 877 A ; ~ G W K E Vb O s b ~TOG hivou TT)V xpijotv E ~ .Sr r h m l p a v &.rr6hccvo1vAsterii Amaseae episc., Homiliae (ca. A.D. 400), Migne, F'G, 40.16 j A ; s i ~ ~a~ ~ a ~ a i v o u o l v~&o. rhr bw hv i -rh ~ I K $K T ~hivou K C $ ~ pias Exov~aT?V K U T Q U K E U ~ VS. Nili, Liber de monastics exercitatione (A.D. 450), Migne, PG, 79.804B; -rh ZK hivwv 6cpCtopma Leo Philos., Tb E ~ a p x t ~ bPlPhiov, v o . ~ ( e d J. . Nicole, p. 22) ; iTGVhsmGv hivwv ~ aGiacpavGv l KUT' bocpirv pkv ovvaywyi ~ a ovvilqo~g l Michael Psellos i n C. Sathas, Msoalwvl~fi B I P A I o S ~ ~ 5 K ~(18 , j6), p. 73; 06 Gioohs EvG~birmouoaxhaiva~E$ Epiou ~ a hivou l E{ucpaopiva~ Nikolaos Mesarites, Die Palastrevolution des Johannes Konznenos, ed. A. Heisenberg, (Wiirzburg, ~ g o j )p. , 41, line 36; ~ S DemeGlacpavij 6cpaopa-r~jlivou ~ a Pl E T & ~ bpoG trios Chrysoloras in S. Lambros, 17aAa1ojl6ys1a ad T T ~ h o ~ ~ o v v q u3~ (Athens, a~&, 1926), p. 230, 1.5.

D E M E T R I U S J. G E O R G A C A S

i ~ i p ~a~ ~ A TLeo ) s Philos., Tactica, inMigne, PG, 107.724 C; 15x6 TQV~ioccy6vrwvEX~WSEV T&TOI&a Aivdc; idem, Tb h a p x l ~ b vptphiov, 9.1 (ed. J. Nicole, p. 39) ; 'Ehv . . .~ioi?&J~ p a y y c n a i a AtvGv il p6Ai~osKTA. ibid., 9.6 (p. 40) ; Oi Ev TS n6A~tIpyaZ6yevot A t v a KTA. 9.7 (p. 40) ; fi Z06xa drxb AtvoG ~ apappmos i AchmetisOneirocr. 225 (ed. Drexl, p. 177, line 9); &-rrb Epiou fi &-rrb htvo6 ij pappcntos ibid. 229 (p. ISO, line 17) ; E\ 68 pappdclctov V ~ S Efi IAtv6v ibid. 262 (p. 215, line 15) ; ravviv Atv6v F. Trinchera, Syllabus

graecarum me.lrtbranarum (Naples, 1865), p. 324, text no. 240; 01 6i Aivbv X I T ~ V ofU ,68 C I ~ A W T ~Glossae V Graecobarb. (Du Cange, Glossar. Graecit., s.v. yaAAo-r6v, 858).

Atvopappmo5 "made of linen and cotton": Atvop&~pmov IpCrrtv Prodr. 1.93 (ed. HesselingPernot). Cf. dt-rrb AtvoO ~ a (4) i p&ppm05 in Achmet (see preceding). ~ t v o ~ i v a ~ neut. ov "blue linen cloth" : Z-repa ~ithw-rh A I V O ~ ~ V E T UE"xov-ra pahhbv ~ ~ a v t m 6 v

Const. Prophyrog., De cer. 465.17;

~XEI

TTIAWT& ~av~ouxhii'va EvSaSupiva ~ I V O P ~487.6. VET~ AtvoatSi5 "like flax" : ZouS&pta t a i oqputivSta, dtpcp6~apa vopilo AtvoatSfi aivat Ammon.

Alexandr., Migne, PG, 85.1576 A. Atv6pahAov: Ixt~paxkltov,Atv6paAov mantile CGL 3.269.50 ; lilomallon [for htv6paAAov] mantile ibid. 3.193.13; cf CGL 7.572 S.V.*Atv6paAov et -AAov mantela (mantile). h t v o p a ~ ~ o ~ & p ineut. ov "linen paAAw~&piov": AtvopaAhw~&ptaConst. Porphyrog., De cer., 469. I ; Atvo~aAAw~CrptaS K ~ T ~ VT heophan.

Contin. (ed. Bonn), 318.15. The simplex paAAo~aptovis Koine and medieval. Atvo-rrMto5. linarius, CGL, 2.361.19 ; see Koine htvoTih6~05 A t v 6 x h o ~ o ~(learned) : t a i OT&AIKES &yp~6ouoi hl~6Tih0~01 T&S & P K T O U ~ Const. Manasses (ed.

Bonn), p. 275, line 6477. ~ t v o " ~ a v r oE 5 pyaoia Leo Philos., Tb t-rrapxutbv PtPAiov, 3.1 (ed. J. Nicole, p. 39), translated

res lintearia. htv6~puoos "linen interwoven with gold": Atv6~puoa I p h i a Malalas, Chron., 457.16, and Atv6~puoovcpmi6Atv 18; Atv6~puoaIpdcrta

thius, n a p ~ ~ p o h a i574.30. , 1 suggest the

spelling Gtdivioat, &tahiviomo, (with -vi-)

in place of transmitted 6iaAivfloa1, Gtahivq-

o m o since the derivative verb is possible

in -ilw, not in -am. The word is a Koine

verb (Pbrynichus, Praeparatio Sophistica,

ed. I . de Borries [Leipzig, 19111, p. 64B =

Anecdota Bekker, 36.23, GtaAtvioat, 61acpu-

y ~ i v Bv ~ o i s ~uvqysoiots y~vbp~vov,which

is also traditionally written 61aAtvijoai).

The entry for Koine and med. G i d i v a o as

well as the medieval E ~ A t v d r omust be struck

out of our lexica.

EKAIVIZW "escape from a hunting net": B
lexicographers have unnecessarily perpetuated. In the thirteenth century Georgius Cyprius, nupotpiat, 3.3 (LeutschSchneidewin, Paroenziografihi Graeci, 2. I 10.15) EKTEp61Kio~t'[Kcti] $~hlvioat' TGV TOGS Sqpau~&5.It is modern

StaStSpau~6v~wv Greek that helps us to understand in this

and numerous other cases which form

actually existed and which was a ghost

word-form. Cf. mod. Gr. EaAtvilo (not


ExtAtv~Oo: gloss Atvo-rrrCrlai. Exthtvair~~, ~iaptpAkxat Hesych. Many of the terms listed above in 2 (Koine) continued to be used in medieval Greek, even if they do not appear in the texts so far published, and were also, therefore, medieval Greek words. Likewise many of the terms listed in 4 (modern Greek below) were already current in medieval Greek. I n general, most modern Greek lexical elements have been inherited from medieval Greek, whether or not they are recorded in texts so far published and in our dictionaries. In many cases it is impossible to stamp a word either Koine or medieval, for it may be both.

Theophan., Chronogr. (ed. de Boor), 244.22, and Atv6~puoovcpa~16hiv25. GtdtviZo : glosses ~ICXTE~SIK~OU~' Stahtvioat. t a i Stacpuyaiv Hesych. (ed. K. Latte, 1.438) Ph. I
~~

G R E E K TERMS F O R " F L A X " A N D " L I N E N "

4. Modern Greek &-rroAtvapi6~~f.

plur. Cythera "the pulpy remnants of flax" Iu-ropt~bv AE~IK~V~ i j NBas 'EAAqvt~fi~,2.5122~).Synon. Aiv6{vAa, (I

5

htvo~vhi6es. dt.rr6htvos m. Crete (ibid). dt-rrohivhvw Pontic (ibid.); see Koine &.rrohtvG. &.rroAivwpav neut. Pontic (ibid). Atvaia f. "thin linen rope," "cord" Epirus

(Chimara), Pontos (Chaldia, Kerasounda, Liveri), ine'a Epirus (Drymades, etc.), AtvtCr Epirus (Droviani, Schoriades, Tsamandas, etc.), Corcyra (Argirades), Paxi, Peloponnesus (Achaia, Arcadia, Elis, Olympia, Corinthia), A d a " t h r e a d Euboea (Avlonari, Kourouni, Kymi), AeviCt "piece of rope" Euboea (Carystos, Ochtonia, Vryses), etc. Some of the more usual meanings are: a. thin rope, string (synon. m C r y y o ~ ) ,b. strip of cloth, c. thread. Another meaning occurs in modern Cretan: Alvaia or htvai f. "odor of burning linen cloth" (synon. A~vovGi&, 6 Aiv65, ~b Atv6v). Cf. Koine Atvaios. Atvapaptdc f. "field of flax" Epirus (Kourentochoria, etc.). The suffix -apt&, earlier -ayaia, was detached from nouns such as KaAayaia, & r r o ~ d a p a i a , pptlo~ahapaia, etc. Cf. also the derivatives ppilapidc (ppila), ppo~apiCr (ppbprl), ~p19aotb( ~ p t s i ) , cpma~ia ( ? m i ) , etc. The 'Iu-ropi~bv AE~IK~V ~ f Nda5 i ~ 'Ehhqvt~fis, 1.512a, gives -age& as if the latter suffix were -6a; see, however, D. J. Geor-

gacas in Glotta, 31 (1951)~p. 211ff. m. "flax grower or dealer" common (also in the lexica of da Somavera, Weigel, Brighenti, etc.) from med. Aivap65, q.v. I n Euboea AlvapSi5 is the designation of a bird (=cpAGpos ?). hivap6vios "of flax," "flaxen" Elis, Naxos, Thrace, (inare'rius) etc. AivapjSpa f., name of a weed that grows in flax fields, perhaps delfihinium junceum, in Messenia (Maniaki, Papoulia, etc.), Zacynthus, Constantinople. hivapjotos "of flax" in various places: .rrpZya Atvapj o i o "Linen cloth" Corcyra ;~b Atvapj o i o as a noun "place in which flax is cultivated" Thrace (Saranda Ekklisies). Aivdrpi neut. "flax": (1722) K. Amantos, Atvap&s

" ' H aixpahooia TOG Ni~ohaov AIKIV~OV,'' 'EAAqvt~h, 11 (1939), p. 153; 'lcrropia TGV~a-rh ~ j vOiryypopAaxiav ~ ~ A e o S 6 v ~ o nap& v TOG ~ ~ T ~ o T ~ o ~Mvpdwv ~ T o L I~vpoG Ma-rSaiov, in E.

Legrand, Bibliothdquegrecquevulgaire, 2.256, line 706; K. Dapontes, A6yot ~ a v q y w p t ~ o i , (Venice, 1778), in E. Legrand, Bibliographic helle'nique, 2.274. The plural htvtipta : Stabv this r, 1.137 (K. N. Sathas, K p q ~ t ~ Ola-rpov, [Venice, 18791, p. 156) ; (1713) F. Miklosich and J. Miiller, Acta et difilomata, 2 (1887)~ p. 199.-Today Atvtipt is common and dialectal (Bova, Otranto, Kerasounda, Trapezounda),htvaptv (Cyprus, Rhodes, etc.), inar in the northern Greek dialect group, Atvlp and vkp in Cappadocia (Gurtonos, Aravan), h t v ~ r hin Tsakonia. From Koine and medieval htvaptv (see above med. Atv&ptov). AivaptCc f. "field

of flax" from Atvapaia. "water in which flax is retted" Crete. There is also the noun htvapi~q5m. as the name of a mineral (Eleftheroudakis, ' E y ~ v ~ h o . r r a t S ~AE~IK~V, ~bv 8.719a). htvapo~orravilw "beat flax" in Mani, Maniaki in Triphylia and in lexica. Synon. o-rovp-rrilw ~b htvdtpt (Cynouria). hlvapo~o.rraviu-rpa f. "woman worker who pounds flax" in Mani and Maniaki. htvapo~6.rravo~ m . "wooden beater of flax" in Naxos, Mani, Maniaki, Adrianople, and in lexica. Synon. K~ITUVO~. htvapo~ovrdiu-rpa f. "woman worker who removes the stalk of the flax as it is being pounded" (synon. ~ o v r d i u - r p a ) . Cf. the phrase ~ o v r a h i l w~b ~ivdtpi. Aivap6AaSo neut. "linseed oil" (Eleftheroudakis 'Ey~wAo.rraiSt~bv/\ E ~ I K ~S.V. V ; etc.). Synon. m. Alvapi~T)~

Atv6haSo. htvapoydryyavo

neut. "machine [pqydtvt] by which the stalk of the flax is crushed" in many regions. Synon. pCryyavos (Epirus, Ionian Islands, Peloponnesus, Pontos, etc.), ~6cp~pa Cf. . yayyavilw ~b htvdrpt . hivapovlpi neut. "water in which flax is washed" Maniaki in Triphylia. Aivap6~ravo neut. "linen cloth" in Messenia (eparchia of Olympia). Synon. htv6.rrav0, Aivb -rravi. Aivap6mo1vo

neut. "rope made of flax" Peloponnesus, Crete, Cos, etc. Synon. htvttx,

Aiv6moiv0, .rrdap&pi. Atvap6movro neut. "coarse

rug of linen" in Messenia (eparchia of Olympia). htvap6mopo~m. "flaxseed, linseed" common (inardspurus in the northern Greek dialect territory). Synon. Atv6orropo~.

DEMETRIUS J. G E O R G A C A S

h ~ v i h a ~ oneut.

"linseed oil" is from the katharevousa term hivihaiov (synon. Alvap6Aa60), which occurs in a codex of the sixteenth century (K. Amantos, 'ASqvE, 43.151). But modern hlviha~ov "linoleum" was adapted from Neo-Latin linoleum ( = Greek ~ q p w ~ 6 v ) . ~ I V ~ V I O S "flaxen; linen" in South Italian Greek. hlvop&ppmos "~nadeof linen and cotton" (a IT~OS katharevousa term), ~ I V O ~ I T & ~ common T ~ O ~from rned. mod. Greek ( ~ I V O I T & ~ Idial.), Aivopappmo~ (see above). In the dialectal speech of Cyprus htvo-nap~mot m.pl. is a nickname for crypto-Christians. h~vopp&~v neut. "men's linen drawers" Pontic, hlvopp& Pontos (Oenoe, Ophis, Trapezounda) and Thrace; plur. A~vopp&~a in Prokonnesos. hivoppox~io neut. in Cephallenia, Laconia 6 Cephallenia, (Mani), h i v o P p o ~ ~ iAmorgos, Cythera, Peloponnesus (Chatzi in Pylia), etc., hlppoux~i6 Aetolia, h1voppax~16Crete (Viannos) "pit filled with water in which flax is retted" (cf. h l v a p i ~ q ~~, i v a ~ o v i p i ) . Cf. ppoxialw TO hlvapi, Aivoppox~&lo,N O V ~ E C I W TO Aiv&p~,etc. hlvopp6~1 neut. Aetolia, Athens, Ionian Islands, Peloponnesus (Messenia), fnovrd~ Leucas = (I) Alvoppox~16 (see above) and (2) "a pit formed by the sinking of the ground" (Cephallenia). h~voppox~dtlw "ret flax" in many places. Synon. ppoxldtlw TO h ~ v & p poum~Qw ~, TO AIV&~I, Aivo!3poxilw. hlvoppoxilw "ret flax": ~novrogizoin Leucas. hlvoypaqia f. "printing on cloth" (a learned

creation). "bound with linen": piphiov -ov (a katharevousa term). h i v o ~ d a p i neut. "flax straw" (from med.

~ I V ~ ~ E T O ~

*Alvo~crh&plv). h ~ v o ~ d a p l af. "flax straw" in Zacynthus (from hivov and ~crhapaia). AIVOK~KKI neut. "linseed" in Epirus (Chimara,

Kourentochoria), Ionian Islands (Zacyn-

thus), Peloponnesus (Laconia, Cynouria,

Messenia, Pontos [Sinopi]), jnokdk Leucas,

inukdk Jannina, iinokdh Artaki, etc. From

~ ~ V O K ~ K K I dimin. V, of ~ ~ V ~ K O K K O S . ~ I V ~ K O K K O Sm. "linseed" in Naxos (Komiaki), etc. h1v6ha8ov neut. "linseed oil" in the early

seventeenth century (Agapios Landos, Biphiov ~ d o O p ~ v orew-rrov1~6v, v [Venice, 17961 p. 186, hlv6Aa60 common (Ionian Islands, Adrianople, etc.). Synon. A1vap6haS0, Iha~ov. Atviha~o,Koine A~vomipp~vov A~vop&ih~vo~ "made of linen and wool" in Triphylia (Siderokastro) hiv6paAhos "made of linen and wool" (cf. med. EK hivow ~ a$pias i or $c $piow ~ ahivow i ; see note

5). A~vopi-rago~ "made of linen and silk" common q ~ note ; 5). (cf. med. hivou ~ api ~ ~ a t see A1v6cwha neut. plur. "the pulpy remnants of

flax after it has been beaten" in Erotokritos A 282 (ed. S. Xanthoudidis, p. 11); Erofihile A I34 (ed. S. Xanthoudidis, p. 94); so in Crete, Cythera, Peloponnesus (Dimitsana, Mani), Thrace (lndksla), etc. Synon. &TCO~Ivapi6as, hlvocuhiS~~. h1vocuhi6asf. plur. Cythera, Mani =hlv6~uhaq.v. hlvtrrravo neut. "linen cloth" in Macedonia, Rhodes. Synon. hlvb ~ a v i , Alvap61~av0, A1vap6mowro. h i v 6 ~ 1 r r af. "cake made of linseed after the

oil has been squeezed out" (used as feed for cattle) in some places. hlv6s "made of linen," "linen" common (h~v6s in Voi'on, Aiyvbs Vithynia, Aixv6s Carystos) : hlvb ~ a v i"linen cloth," (synon. ~~Ctvlvos) h ~ v b I T O I J K & ~ I"linen ~O shirt," hivh poijxa "linen clothes" all common; as nouns 6 h1v6s "the smell of burning cloth" in Rhodes and TO A1v6v "idem" in Pontos (Oenoe) (synon. Aivai and A~vouG~a;see above) ; T& h1v6 "linen clothing, linen suit" common, and "cloth for bed sheets, tablecloth, cloth for towels made of cotton or even silk" in Crete (Prinias of Monofatsi) ; hlv& ~ a tihlva l Gyparis A 21 (ed. E. Kriaras [Athens, 19401, p. 215). A1v6om~af. "sack made of coarse linen" and hlvoo&~a "sack for flour" Megara. Synon. ~avaf3drro0,A~vdrroa,-rooup&i~. Aiv6moivo neut. "rope made of coarse linen" Cos. Synon. Aivap6mo1vo. h1v6muyyos m. in some places and A l v b m q o s

in Amorgos "string of flax." m. "flaxseed," "linseed" in the early seventeenth century (Agapios Landos [see above, S.V. h1v6ha6ov], p. 8) and in Crete, Cythera, etc., lilzdsporo in Otranto, and hiv6mopov neut. in Pontos (Trapezounda). Synon. A1vap6o.rropo~.

Aiv6mopos

GREEK TERMS FOR "FLAX" AND "LINEN"

~ l v o m o h f i f. "linen suit"

(a katharevousa term) ; in military language it means "linen underwear" (for both m ~ h i aand OrroG6-q~). Cf. synon. &mp6pouxa, iocjpouxa. Cf, also Koine h~vomohiaand anc. Greek h1v6o-rohos. AlvobGia neut. plur. "underwear" and especially "petticoats made of linen cloth woven by women" in Crete (see below). AivouGia f. and Aivowidr (see below). hlvo~opCj"wear linen clothing" : the participle h~vocpop~pivo~ in Thrace (Adrianople) and in lexica (e.g., da Somavera: "vestito di lino"). Cf. Koine Alvocp6pos. A I V ~ X O ~ Tn O eut., name of the plant erythraea maritima, in Corcyra ; from K o i n e h 1 v 6 ~ o p ~ o v . payyavilw TO AlvCrp~ "work out the flax by means of a machine called pdryyavos" (6ouhabo TO

Alvapi m b p&yyavo): payyaviopivo AlvCrpi

(synon. mouhhiv in Cyprus, mouMi in Cos, o~ovhhoi~6i in Crete and Cythera, ~ o p ~ b G a in Crete); opposite &payy&v~moAivCrp~. Cf. hivapopdryyavo. ~ i h i v a ;see s. v. A1v6s. E ~ h i v i Z"beat ~ the dry flax so that its seeds

fall away" in Mani (Kambos of Avia); figuratively "rejoice greatly" in Crete (Imbros of Sfakia), probably with the earlier meaning of "escape from a hunting net" (cf. <~61x-niZw:ksidixtizu "escape from a net" in the speech of the island of Samos). This word derives from med. it~hivloaaor. of IKhlvlZw "to escape from a hunting net [= Aivov]" (see med. i ~ h ~ v i labove). w yihohiv~ neut. "thin flax" in Triphylia (Gargaliani) from y ~ h b vhivov. I t should be noted that the sixty-five modern Greek terms given above do not constitute a definitive list since it is impossible to provide an exhaustive list of terms from writings that have been neither excerpted nor completely published, or from unrecorded oral Greek speech which constitutes such a rich source for medieval and modern Greek. 5. Names

Family Names : The family name AivapBs m. occurring in many sections of modern Greece must also have been medieval and post-medieval, and derives from the occupational term AivapBs m. "flax grower" and "flax dealer." Hence the place-name A i v a p h l ~ a neut. plur. on the island of Paxi.

The family name A~vCrpqs m. occurs in fewer places and derives from the term Aiv6pt "flax," given to persons as a nickname. Aivapiqs m. as a nickname occurs in Rethymno and as a surname in some villages of the eparchia of Mylopotamos, both in Crete.6a

Place Names: Place names derived from the terms for "flax" and from their derivatives are found in great abundance; for example the following :' A~vaptnaneut. in Pylia (Kondogoni) and as

the name of a village in Evrytania; plur. A~vaptcrcia in Naupaktia, Trichonia, and Pylia. AivapBs m. name of a locality in Chios (Pyrgi). ~ O A~vCrpq, G name of a locality in the Peloponnesus (Dirachi). A ~ v a pneut. ~ in Aegina, Calymnos, Euboea (Carystos, Platanistos), etc.; the plur. A~vdrpia in Aetolia (pronounced Znlirja), Epirus (Konitsa), Calymnos, Symi, and of a village in Argolidocorinthia. A ~ v a p ~f.a name of localities in Elis, Aetolia, Rhodes (infive villages),Thessaly (Pertouli), Scyros (harbor and village) and as the name of a place previously called r ~ o p - r t ain Elis ; plur. Alvap~is in Thessaly (Pertouli), Triphylia, Rhodes, Arvapi~sin Euboea (Episkopi), etc. Also Alvapf f. in Icaria, plur. Alvapis in Crete (Siteia), AlvapCr in Carpathos and Crete (Viannos). All from the noun AlvaplCr, hivapaia (see above). A~vapiGianeut. plur. in Rhodes (Vati). Atvapimpa f , in many places; from the noun *hivapimpa (cf. ppllimpa, ppopimpa etc.). Alvapi~oaf. in Epirus and Evrytania. Alvap6~aprro~ m. in Naupaktia. Cf. /\lv6~apno$. Alvapoxcjpacpo neut. in Aetolia. A a 6 f. : o n j Aiv6 an area in Scyros where flax was cultivated. Alvo661a neut. plur. name of localities in Rhodes (near Lardos). Aivobpa f , in Nisyros. Aivoppox~iov neut., a medieval place-name (M. Goudas, "Bvlavr~valyypacpa .rfjs Iv ' A 9 9 f ~ p 6 p~ovijs TOG Ba-rosr~Gtov," 'Em-rrlpis 'E-ra~paias BvlavrlvGv Zrrou6Gv, 4 [1927], p. 213, line 21 r o c . . . ~ b c x o sTOG i-rr~h~yopivou 6" This I learned from Mr. Andreas Stavroulakis, in a letter dated zo December 1958. For some of these names cf. Kalleris, op. cit. (note 2), p. 188, note 5.

260

D E M E T R I U S J. G E O R G A C A S

A ~ v o p p o ~ ~ I TOG o u MEAIOOIVOG, and p. 214, line 34 T ~ V61dqcp9iv~aPiraxa, sbv Evovopaand 16pavov A l v o ~ p o ~ ~ i TOG o v MEAIOOIVOG) modern in Volissos of Chios; A~voppox~io in Cythera and Mani; / \ I V O P ~ O XinE IPylia ~

(Mitioti) and in Triphylia (three times), Cephallenia, Amorgos, Crete, Nisyros, Rhodes (Arnitha: linovro;d) ; plur. AlvoppoXEI& in Zacynthus, and as the name of a winter stream in Seriphos. A1vopp6~1n eut. in Elis, Messenia, Carpathos, Acarnania (pronounced ~ ~ u v r d FLeucas ), ; plur. A1vopp6~1a in Cephallenia, Ithaca, Nisyros, Acarnania, Aetolia. J \ I V ~ K ~ ~m. T I in O $Zacynthus. Cf. Alvap6~ayrros.

11.

A1vorr6sqs m. name of a pond in Cos.

A1v6po~af . in Cephallenia.

AIVOT~TII ( T ~ SKamopia~,year 1622) and T ~ S

K G ~ J\IVOTOTT~OV ~ s (T?$ K a o ~ o p i a ~ year , 1617)

(E. Poulitsas, "'Errlypacpai ~ a $vSupjo~l$ i EK ~ f j $Bopaiou ' Hrr~ipou," ' E r r ~ ~ q p i'sE ~ a ~ p ~ i a s BuLavr1vi3v Zvou6Gv, 5 [1928], pp. 66 and 61) ; Atvo~6~t.1 is also the name of a locality in

Elis, of a place in Macedonia through which

the river Aliacmon flows, and of a village

in the Koritsa area in North Epirus (South

Albania).

Alv6xwpa neut. in Lesbos (pronounced indxuma). etc.

AINOYAION, AINOYTIN, (GHOST-WORD)AINOYTION, ETC.;

AlNOYAlA (FEM.), AINOYTIA; KYAIAION; ENWAION-ENOIZI, ETC. I.

ilivoZ;diov

The noun Alvoir61ov is only a late Koine word, being found in papyri of the second, third, fourth, fifth, and sixth centuries A.D. So Alvoir61ov va161~6v p . 0 ~ ~ 1066.10, . slpfi i tv Alvou6iwv P. Stud. XX 106.1, ~ a oapavov ~ a Alvoir61a i E T T I X G 660 ~ I ~ P . Masp. 67006.66 (vol. 11, 1913), etc.8 Its meaning must have been "an item of linen clothing" or "linen garment" ( P . Oxy. I 114.8 A1vo661ovip1~6p~upov [second-third century]) and subsequently "linen shirt.jJ9 The same meaning "linen garment," "linen shirt" is carried by the ancient Greek XITGV"linen shirt for women," Koine term AlvoGs (or Alvbs) XITGV and X I T W V ~ U K O ~ . camisia (CGL, 3.272.61). The linen shirt became a customary garment; a parallel is Lat. Zinea (actually from vestis Zinea), which was "a woman's shirt" while camisia was " a man's shirt."lO The noun is a F. Preisigke, Wdrterbuch der griechischen Papyrusurkunden, 2 (Berlin, 1927), p. 24. Preisigke, ibid., gives "linnenes Hemd" as t h e only meaning. 10 O n t h e linen garment worn after baptism (vestis linea in Jerome, Epistula L X I V , 3 19 tunc induemuv veste linea nihil i n se mortis habente, sed tota candida, ut de baptism0 consurgentes cingamus lumbos i n veritate et tota pristinorum peccatorum turpitude celetur [Corpus Script. Eccles. Latin., ed. I . Hilberg, 64 ( I ~ I O ) , p. ~ I O ] ) see , J . Quasten, " A Pythagorean Idea i n Jerome," American Journal of Philology, 63

derivative of the noun A1v6v neut. "anything made of flax," "linen cloth," "linen garment" with the diminutive suffix -06610~. This Koine noun has survived in modern Greek : A~voirS~aneut. plur. "linen underwear" is in use in Mesara and Chania on the island of Crete and A1vo661 "linen petticoat" on Crete.ll This has also yielded the placename Alvoir61a neut. plur. in Rhodes and specifically in the area of Lardos.lZ This placename probably derived from the flax that was cultivated on the island in past times, (1942), pp. 207-215. C f . Leo Spitzer, " A d ditional Note o n ' W o o l and Linen' i n Jerome," ibid., 64 (1g43),p. 9 8 f . Also tpdrr~ov~ I T T I ~ ~ I V O V Levit. 13.47 (fv ipcrriq ips@ iEv Ipcrriq cmrrrrrvIvq), vestis lintea i n t h e Vulgate. Communicated b y Mrs. Evangelia Frangaki (Filothei', A t h e n s ) ; c f . Ms. 663, p. 10, o f t h e ' l a r o p ~ ~ bAv E ~ I K ~ V'EAA~vIK~~s rjlwooqs, Acade m y o f Athens. Also i n a Cretan folksong: "Av h a $EGTEPI&, E I V P~ yI ~ ~ i ?ha pi T& h~vciQOU,I EAa pi Th h1vo661aUoU, pi ~h ~ETC((WT&(TOW (pylci=~ir6ia 'fine weather'), i n which s h Alvoir61a has perhaps t h e meaning "linen clothes," "linen suit" ( t h i s distich was communicated t o m e b y Mr. N . Kontosopoulos). l2 Ch. I . Papachri~todoulou,T o ~ o v u p ~ T ~~b S P66ou, (Rhodes, 1g51),p. 107. T h e author spells this n a m e Aqvoir61a as i f i t were derived f r o m Aqvoir61v ( f r o mAqvb~m. " v i n e press"). However, i n response t o m y inquiry he has informed m e ( b y letter o f z February 1958) t h a t he, too, had dimin. thought o f connecting i t w i t h h~voirS~a, o f hivov "flax."

G R E E K T E R M S F O R "FLAX" AND " L I N E N "

and is not Aqvoir61a from AqvoirGlv "small vine press" (:Aqv6s m. "vine press"). For, whereas vineyards do not exist in that area, it is likely that flax was cultivated along the banks of a winter stream called Oovt&s which flows near Lardos. The Rhodian place-names A~vapiaf., AlvapiG~aneut. plur. and A~voppo~ € 1 6neut. corroborate this thesis. From similar terms we may infer production of flax also for the islands of Calymnos (AIV&~I~),Cos (AIVOIT~T~S),Nisyros (A~voirpaf., A1vopp6xla neut. plur.), and Carpathos. I n any case, though the noun A1voir61v or (without 6) A~voirlv (both suffix variants -zi8in and -din are the rule in the modern Rhodian dialect) is not in use in the speech of Rhodes but hlv&plv neut. (with dimin. suff. -ap~v)is used in its place, yet AlvoirGlv must have been Rhodian, as the Rhodian place-name indicates. 2. Ilvov'zlv

I should like to show now that in Greek Koine there existed a word A~voirG~ov(and AIVO~~GIV) and in Egypt a local Coptic form Alvoin~ov, and spoken A~voirnov,but that the word h~voiry~ov,which figures in Greek dictionaries and has been used as an example of an alleged consonant change of 61 to y ~ ,is in fact a ghost-word form. The following papyrus texts come into discussion: P. Masp. 67006, P. Princ. AM 8961, and P.Got. 14.

3. Ilvo?jyla?

I n a papyrus document of about A.D. 522, from Theodosiopolis, located near Minieh in Upper Egypt, the word AlvoirG~a is clearly read: P. Mnsp. 67006.66 ~ a o&pavov l ?v ~ a l AlvoirG~a CITI / ~ G p l aGiro, which Maspero explains in a note "A1voir6la: Ctoffes de lin (?)."I3 In the same papyrus, line 85, Maspero read ~ a 1&Aha Alvoiry~a 6ir0 t m x G p ~ aand in the apparatus he tentatively restores the word Alvoiry~a to A1voirpy~a.l4 Reil gives Alvoby~a with a question mark.15 13 Jean Maspero, P a p y r u s grecs d'ipoque byzantine. Trouvailles de K8m-Ich-qaou, in Catalogue giniral des antiquitis igyptienxes d u M u s i e d u Caire, I ( I ~ I O ) , p. 28. l4 Maspero, ibid., p. 30. 15 Theodor Reil, Beitrage zur Kenntnis des Gewerbes i m hellenistischen Agypten, Dissert. Leipzig (Borna-Leipzig, 1913)~p. 118.

The first objection to Alvoirpyla would be The its difference in meaning from A~voiry~a. second is that a noun A~voirpylacould be a derivative of a substantivized adjective ~lvoupyfis - i s (cf. ~alvoupyfis and therefrom ua~voirpy~os on papyri) and, in fact, there are adjectives A~vspyfis and Alvospyfi~ "flaxen," of linen,"16 but there is really no basis for a word Alvoirpylov in Koine or in medieval or modern Greek as there is no basis for Alvoliy~ov. I t is, moreover, difficult to restore Alvoirpyla from hyoiryla, in which four out of eight letters cannot, according t o Maspero, be safely read. But the word is not AlvoirG~a,as Preisigke17implies by listing Maspero's AlvoOy l a under the entry A1voir61ov without further comment. Through the mediation of Professor H. C. Youtie, Dr. Abdullatif A. Aly (Cairo) has kindly read for me the papyrus P. Masp. 67006 and reports that the reading in line 85 is Alvoiryla (XINO~I-IX).~~ This is in fact the only instance of the word being written Alvoiry l a instead of AlvoirGla or Alvohla. In line 66 of the same papyrus the word is AlvoirG~a,so it would be reasonable for us t o expect the same form A~voirGlain line 85. Yet, because the spoken form was perhaps Alvoirr~awith -T-, the scribe may have intended t o write h ~ v o i r n a in line 85. In either case, Alvoiry~ais a scribal error. This judgment, with which Mr. Youtie concurs, is based on the writing and spelling errors of the text, of which Maspero himself sayslS that the long fragment of a marriage contract that appears on the verso of papyrus 67006 (pp. 23-33) is written "en un grec barbare dCfigurC par une abondance insolite des fautes d'orthographe. Ecriture assez soignke, mais maladroite (tout B. fait nkgligCe B. partir de la ligne IOI), sorte d'onciale grossikre m&lCede quelques ligatures cursives. " Though I was myself unable, without the 16 On Alvospyfi~ (Oppian) and A l v ~ p y q ~ (Lycophron, Dionysius Periegeta) without the stem vowel -0- cf. nouns in -0epy6~ and -spy65 and adjectives in -oepyfis and -spyis such as K ~ ~ T O ~ P YX~PSV!O E ~ Y ~ Sx, ~ u ~ ~ ~ etc. yfi5, 17 Preisigke, op. cit., 2, p. 24. L. R. Palmer, A Grammar of the Post-Ptolemaic P a p y r i , I , I (London, 1946)~pp. 14, 82, 89, lists only AIVO~~IOV,obviously depending for his material on Preisigke. 18 By letter from Mr. Youtie (May, 1958). l9 Op. cit., p. 22.

262

D E M E T R I U S J. G E O R G A C A S

reproduction of the original papyrus, to check the eighty-nine examples of the ligature vr and the ten of the ligature vy in order to ascertain whether the one is confused with the other, which I hope to do in the future, I am for the present reasonably certain that A~vorjy~ain this papyrus represents AlvoirS~a or Alvoir~~a. Professor Youtie reminds me of p a ~ p d o u p a and errors such as ~ ~ o ~ ? o u for ~ppu[A]ouyov for ~pTihoupov which corroborate the view that A~vovy~ais a miswriting of Alvovr~a.~~

4. I~voin/~cc (non-existent) I n another papyrus from Lycon Polis (or Lycopolis) in Upper Egypt of A.D. 481, the text of a dialysis between Cyrus, Bishop of Lycopolis, and two brothers, first published in 1922 by the late Henry B. Dewing, the editor reads A~voiry~adtvGp1~2rGiro and translates it as "two men's linens,"21 whereas it actually means "two men's linen shirts," and further Alvoby~a 1 ~ a p a j ~ c n j 6 w ~[accent a -T&] ~ a ~ v o i r p ~y ~p ai apaAAw~6v , .rrapmdGw~ov[accent - T ~ vZV, ] Alvoiry~a&AAa p a h h w ~ hGiro translatedby the same editor as "three new. ..linens,one woolen, two other linens mixed (?)with but actually meaning "new linen shirts trimmed on the edges, one woolen towel trimmed on the edges, and two other coarse linen towels." Dewing reproduces in his article a photograph of the part of the papyrus that includes lines 34-52; in his C ~ m m e n t a r yhe~ ~ makes no remark whatsoever on the reading hlvoiry~a,as if it raised no difficulty.

...

In his letter cited in note 18. Henry B. Dewing, "A Dialysis of the Fifth Century A.D. in the Princeton Collection of Papyri," Transactions of the American Philological Association, 53 (1922)~p. 117, line 40. This papyrus was not reedited in Papyri in the Princeton University Collections, I, ed. A. C. Johnson and H. B. van Hoesen (The Johns Hopkins University Studies in Archaeology, No. 10, ed. David M. Robinson [Baltimore, 19311); 11, ed. Edmund Harris Kase, Jr. (Princeton University Studies in Papyrology, No. I, ed. Allan Chester Johnson [Princeton, 19361); 111, ed. A. C. Johnson and S. Pullman Goodrich (ibid., No. 4 [1g42]). A revision of lines 1-18 of the text and several new readings of other lines was given by E. H. Kase, Jr., in No. 2 (1936), p. 78f. 2 2 Dewing, op. cit., 122, lines 43-45. 23 Ibid., 124, line 7. 20

21

This has been the source for all subsequent treatment of the term Alvoiry~a. ~ i l a b ac~l cepted Dewing's text without raising any .~~ has also question about A ~ v o i r y ~ aEnsslin reproduced the text and praised Dewing's exemplary t r a n ~ c r i p t i o n He . ~ ~remarks26 that ~ I ~ Masp. 67006.85 the reading A I V O ~in~ P.Cair. isnow substantiated by the Princeton papyrus, and mentions the entry AlvoOG~ov "linnenes Hemd" from Preisigke's lexicon. He then explains AlvoOy~adtvGp~~ci as "Mannerhemden" ~ a l v o i r p y ~ aas and Alvoiry~a 1~apmcn3-rwGa~~ "neue bortenverzierte Leinenhemden," adds that "die hlvorjy~a&Aha pahhw-ra (2. 44) fugen sich nicht ohne weiteres ein," and suggests that A~voiry~ap ~hhwT&would be "als lintea villosa, 'rauhe linnene Tucher' als Tisch- oder Handtiicher erklart." Paula Wahrmann, in her bibliographical survey of works on the Greek language published in 1926, likewise states that the word hlvoiryla had been previously known only from P.Cair.Masp. 67006.85, and reminds her readers of A~voirG~ov "linen shirt. "28 Finally, Liddell-Scott- Jones have taken over this word from Sammelbuch (7033.40),and explain Alvoiry~ovas equivalent to A1voO61ov with the meaning "linen shirt." This word is recorded in the reverse indices.29 24 Friedrich Bilabel, Sammelbuch griechischer Urkunden aus Aegypten, 3 (Berlin and Leipzig, rgz6), no. 7033, pp. 190-192. 25 Wilhelm Ensslin, "Ein Prozessvergleich unter Klerikern vom Jahre 481 (Zu Papyrus Princeton 55)," Rheinisches Museum, N.F., 75 (1926)~ pp. 422-446. The text is on pp. 422-426. 26 Ibid., p. 442, note I. 27 Read: rrapwaw€io-r& (as it is in the text). This word occurs also in P.Masp. 67006.80 (Maspero, op. cit., p. 29) pacpopla yuvalK1a ~ a (Maspero, ibid., un1 ~ a p w c r v 6 o ~1 a6 ~ Tpla necessarily comments "pour .rrapcryccv6o6q ( ?) ") and 83 K a l aAAa o-rlxapla -rrapmawGoSa K u S ~ ~ E plva I ITEVTE. The normal spelling of the word was ITapayaW6wTa (pron. parayav6ota'). On the word 1~apaycnj6q~ (Lat. paragaudis, paragauda) l cf. Ph. Koukoules, Bulavr~vGv Pi05 ~ a -rroA~TIU~~S2 , .2.49 (with bibliography); on an Iranian etymology, L. Th. Lefort, "Le copte source auxiliaire du grec," MLlanges Bidez (Brussels, 1934)~p. 574, note 2. Medieval .rrapaycnj61a neut. plur.: Const. Porphyrog., ed. A. Vogt, I. R. 142 (p. 132, lines 25-26). 28 P(aula) W(ahrmann), "Literaturbericht fiir das Jahr 1926," Glotta, 17 ( ~ g z g )p. , 21gf. 29 Otto Gradenwitz and F. Bilabel, E. Pfeiffer, A. Lauer, Heidelberger Kontrarindex der griechischen Papyrusurkunden (Berlin, 1g31), pp. 26d and 12oc; E. Locker, Rucklaufiges Worterbuch

263

G R E E K TERMS F O R " F L A X J JA N D " L I N E N " Thus the word-form htvo\iytov, read by Maspero in one papyrus in 1910 and misread in another by Dewing in 1922, has been perpetuated up to the present and now stands unquestioned. Ensslin and Bilabel were content to rely on Dewing's reading of this same papyrus, so that no verification of the word hivoiry~a was made. I t was H. Idris Bell who, in a report on papyri published in 1922-23, noted that Dewing's "transcription contains some obvious errors of reading or restoration, and he [Dewing] has been kind enough to give me his opinion of various corrections I proposed, several of which he accepts; but as he is at present unable to refer to the original I prefer not to note them here."30 I t is regrettable that we do not have these corrections.

once in the text and has the shape w (in Avyoumou, line I). In this script there is no confusion between the letters y and T because no gamma has the shape of tau, although a few taus resemble a lower-case gamma. Thus, my own reading hivohia is certain, and the inevitable conclusion is that the word Alvorjyia which has been perpetuated for the last fifty years is a ghost-word caused mainly by Dewing's misreading, and that it should now be permanently eliminated?2 I t should be added that confusion between the letters T and y also occurs in manuscripts other than papyri; thus CGL 3.430.26 &pTiJpCa7a (which had been read &pyjpa-ra), 3.431.24 urjaypos (previously read oh-rpos), 2.31.4 h t 9 a u ~ v r o v (previously read &yt-rdro~vrov),e t ~ . ~ ~

6. Aivo6zia (seventh century)

5. Aivo2jtiu (fifth century) Having had some doubt myself about the matter (see above section 2), I checked the reading on the photograph given by Dewing, and became convinced that the reading Alvoiryla was not correct. The ligature M is exactly the same in the word that Dewing reads h i v o ~ i a(lines 40, 43, 44) as in the words ~ o h o l s(line 45) and ~ o h w v(line 47). Having subsequently received from the Princeton University Library a photostat of the entire papyrus,31 I was able to study the handwriting of the main text (lines 1-75) in which the word occurs. My findings are as follows: in thirty-six instances ( m - , om-, TOW-, EW-, TW-, o u 6 a ~ o v r m - ,povovrou, audtxrrov) the ligature for vr is written in the shape ~r (that is VI with a crossbar -at the upper right), whereas in two cases it is written VT (in ~ a p w a v r w 6 o v , lines 38 and 44). On the other hand, the letter y (gamma) is written thus: r r ~ yor, very nearly so. The ligature for v y occurs but der griechischen Sprache (Gijttingen, 1944), p. 147; C. D. Buck and W . Petersen, A Reverse Index of Greek Nouns and Adjectives (Chicago, 19441, P. 64, col. bH . Idris Bell, "Bibliography: GraecoR o m a n E g y p t . A. Papyri (1922--23), " The Journal of Egyptian Archaeology, 10 (London, 1924), p. 160. 31 Princeton Collection AM 8961 (so given b y E. H . Kase, J r . [see above, n o t e 211 and n o longer under no. 55 as given b y Dewing, Bilabel, Ensslin, and others).

The word h i v o h ~ ais also found in a later papyrus and listed in the Heidelberger Kontrarindex and the Reverse Index of Buck and P e t e r ~ e n , while ~~ Liddell-Scott-Jones have ignored it. The papyrus in question (the end of a private letter), of the seventh century A.D. and of unknown provenience, now located in the collection of the Municipal Library of Goteborg (Sweden), reads as follows: (P. Got. 14.4) ~ a ~b i tvoili uai 6rjo ah& h t v o h ~ a~ a ~uAi-r![a] l ~ a 1 .. .35 "and the 3 2 On ghost-words (French mots-fant6mes) see W . W . S k e a t , " R e p o r t u p o n 'Ghost-Words', or words which h a v e n o real existence," in Transactions of the Philological Society (London, 1885-87), pp. 350-374. T h i s s t u d y deals w i t h ghost-words and ghost-phrases in English. Examples f r o m L a t i n and other languages are given b y Max Niedermann, " G h o s t Words," M u seum Helveticum, 2 (1g45), pp. 123-136. Greek ghost-words f r o m papyri are given b y Ph. Kout s TOGS 'Ehkoules, llapa-rqpfio~t$~ a6 i1 o p ~ h u ~EIS ~ ~ V I K O-rradpou~ ~ S ( A t h e n s , I ~ I I ) , 14,and ~ . "'Eu

TQV 'EAhqvi~Gv -rr&pov,"

Bulanis,

2

( I ~ I Z ) ,

pp. 485, 487, 495, 502. 33 See Koukoules, Byzantinische Zeitschrift, 20.417 and n o t e 5. 34 0. Gradenwitz et alii, Heidelberger Kontririndex der griechischen Papyrusurkunden, pp. 34b and 1 2 0 ; Buck-Petersen, A Reverse Index of Greek Nouns and Adjectives, 116a and (s.v. AivoirG~ov)6 4b. 35Hjalmar Frisk, Papyrus grecs de la BibliothBque Municipale de Gothembourg* (Goteborg, 1929), p. 29 (Goteborgs Hogskolas Arsskrift 3 5 : I ) . T h e same word v e r y probably occurs also i n line I o f t h e same papyrus: A[I]~&TI~.

D E M E T R I U S J. G E O R G A C A S

ear-ring and two fine linen shirts and cups [?I and . . . ." Professor Frisk, the editor of the papyrus, is, however, dubious concerning the word Alvohla, and only confuses the issue by the following suppositions : A . He considers the spelling ~ I V O ~ T I as U a lapsus for h1voir6la, also as ~uhi-rla for whi61a, E i p a i ~ a for EipaiGa, and Evoil~ for tvh161ov; consequently in the apparatus: Frisk says " 1. [= lire 'read'] Alvoir61a," "1. CvhtG~ov," "1. wAi61a." I t is probably for this reason that Liddell-Scott- Jones have ignored the word-form Alvoir~la. Such corrections are, of course, admissible only when corroborated by, not when contradicted by, additional evidence. Indeed, the incorrect spelling of words in many cases reveals t o us their actual pronunciation a t the time when the document containing them was written. So the spelling A ~ v o h l a does reveal its actual pronunciation in the spoken language of the seventh century A.D. in the region (probably Egypt) where the papyrus was written. B. Frisk further says in his C ~ m m e n t a r y ~ ~ that Bvoil~has l1 for 610 and that its final v has dropped out; in other words, he postulates here not a misspelling due t o carelessness, but a word-form reflecting contemporary pronunciation. Contrarv t o Frisk's statement, A ~ v o h ~ a cannot be a casual misspelling because it occurs three times in P.Princ. AM 8961 and also in P.Got. 14.4. I n f a d , the occurrence of this word in a papyrus of the seventh century actually corroborates my reading of the Princeton papyrus. I n this connection it should be stressed that in P.Got. 14 not every 6 is changed into T, so that the change is not really phonetic. E.g., in the same papyrus the following words with 6 are found: 1.4 Giro ; 1.6 &6~Aqbv; 1.8 &SEA~ j v ;1.9 llaAAa6ia; 1.10 -rra16ia and dr6~Apbs. Only specificterms, therefore, show T for 6, and these must have undergone this change under the influence of the native non-Greek speech. a. KUA~TIU: I t must be observed that ~ u h i ~ l a is a doubtful reading (yuhi~![a]) and that, if it stands for K W ~ ~ a~ Iterm U , found in two papyri of the fourth century,37 it does not

mean " c ~ p , " ~but a is equivalent t o ~ o t A i 6 1 o v ~ ~ (before A.D. 1000 both 01 and w represented the phonetic value [ii]). Frisk's explanation of ~uAi61aand wAi~1a as diminutives of ~ i r h ~ t (which occurs in P. Strassb. 99.11, second century B . c . ) ~is~ not likely, for the normal diminutive of ~ i r h ~f. t"cupJ' would have been ~ w h i ~ l"small a cups" and a further diminutive would have been *KUhlKi61~. b. BvoiI~: Finally, BvoiL was developed from BvoiSlv [pronounced endidin], and in fact Cvoi61a with 01 is attested in an inscription of H a l i c a r n a s ~ u s ,in ~ ~which oi is t o be taken as a real diphthong [oil; the latter is from Bv+61ov or Bvh161ov~~ (cf. pb161 from fioi61ov). Attic Iv+61ov in turn is not from Ionic Evh-rlov but from *Evh161ov.~~ Parallels of 01 from 01 are ~oi61ov P.Cair.Zen. 20 (third century B.c.) for ~+61ov "sheepskin, fleece" and K U A O ~ ~ I OP. V Zen. Col. 43 (third century B.c.) and P.Columb.Inv. 209 for ~aA+Gtov "cord." I n papyri there also occur such 1, dt~o61aCcrw1, spellings as E'crro~,6 ~ ~ 6 61~.ohoy61, ywvfil, etc. However, the spelling lv661ov, Bvo6iwv is attested in papyri from Tebtynis, for which the editor lists this word in the index as l v h 6 1 o v . For ~ ~ I from 6 [a] cf. ovv~op16jv P. Hamb. 62.9 (Fayum, A.D. 123), but 38 Preisigke, ofi. cit., I , p. 847, 'Becherchen' with a question mark. The same author translates K O I ~ ~ ~ IasO V"Bauchfleisch (Speise)." 39 SOLiddell-Scott- Jones, 1.967 s.v. 4O Frisk, op. cit., 30. 4 1 CIG 2.2663.6-8 ~-rrsv6wo-rraAh1wv I lauyo~ K a I EVOI I61a K a l xo~povKT)\. The editor, August Boeckh, reads dv[+]61a. The word ~vot61acontains not 01 [ = u ] but a real diphthong [=&I. 42 I G 22.1377. I 6 (Traditio Cimeliorum Parthenonis) EvwtGio 61aAi00 xp[uuh]; z2.1388 A 17

K~]q~hfi, ~ E Q & V BvoI~[~]w, TI, 8pp05, h o 6 ~ p i 5 ,ijhw 6iro/18 xp[uuw (39918 B.c.); z2.1j44.20 (Tabula

curatorum templi Eleusinii, ~ZZ/I-32918)BTEPU

8 m p a xpuoia Giro, Ivh161a 600 xpI2l wo6 ~ V T E ~ I Aaopiva, xpwoia hem&; I G 11.2.199 B 46 (Delos, 274 B.c.) ivh161a xpwati O & a ~ and a B 66 (Delos, 274 B.c.) ~ a 6m~iih101 i Giro, 6 1€ 5 EXOV hi0ov ~ a l Bvh161ov ~ a 6iAAa xpuvia -rrav~o6a~r& ~ a 6AAov i B ~ T ~ ~ A IKTA o v P. Petr. 3, p. 37 (third century B.c.);P. Ryl. 124.30 (first century A.D.). 43 On epenthesis of I (lv6161ov from Ivh61ov) see E. Mayser, Grammatik der griechischen Papyri, I (1go6), p. 131, also 73 and 106. For 6 from T see J. SVackernagel, Philolog. Anzeiger, I 5 (18851, P. 199. 36 Ibid., p. 30. 44 A. E. R. Boak, Papyri from Tebtynis, I, Michigan Papyri, I1 (Ann Arbor, Mich., 37 Bilabel, Sammelbuch, no. 1941, K O I ~ ~ ~ I O VPart ; P. Lond. 3.1259.38 ~whi61ov. 19331, PP. 59 and 246a.

G R E E K TERMS FOR "FLAXJJA N D " L I N E N " a u v ~ o p ~ lP. q ~Hamb. 68.27 (Aphrodito, after

550 A . D . ) . ~ ~

7. Explanation of the form A L V O ~ T L V What precisely is AIVOWIOV or hlvowlv, whose plural hlvorj-r~awe have encountered ? Since the meaning of hlvofr-r~ov in its context is obviously "an item of linen clothing," this word must ultimately be a derivative of hivov n. "anything made of flax," "linen cloth," "linen garment," and related to the adjective AlvoGs or the late Koine adjective A1v65 "flaxen," "linen." One might be inclined to understand A~voir-rlovas a diminutive of a substantivized h1vow6v neut. of the adjective Alvovr6s for Atvo~6s-fi-6v. There is a parallel adjective ~ ~ A A w Tas,~ e.g. s ~ 696v1ov ~ paAhov~6vin P. M a s p 67006.65 (sixth century), where there is no reason for adopting Maspero's suggestion to read pahhw~6v ("il faut peut-&tre lire p d w T ~ v " ) ~; ~ a s~ ~ e i o tentitive 's correction has been accepted as a certainty by Reil,4sand the reading 686v1ov paMw~6vhas been reproduced

265

by E n s ~ l i n Preisigke, .~~ on the other hand, correctly adopts paAAow-r6~"mit Wollbesatz versehen."5O Though the adjectival suffix -w-r65 is the normal one in Koine as well as in medieval and modern Greek, the form -ow65 is possible, the -24- representing the ancient -w- as in ~bppowvovfor ~itppovov,oiqouva for oiqwva, etc. ; cf. ancient iapow~65(IG, 9.2.461. 38). Consequently the papyrus reading pahhowbv needs no correction. We find, however, no such form as pahhohlv or the like, so that ~ I V O ~ I T I Vcan hardly be explained directly from Alvowbv. The only satisfactory explanation seems to be that hlvofrrta represents the Egyptian pronunciation of h1voir61a. In fact, in Ptolemaic papyri the letter T sporadically replaces 6, and the syllable TI is found instead of 61, as in the case of Atvoir-r~a.~~ This is so in the ~ ~ in many other papyrus P. M a s p . 6 7 0 0 6 and papyri.53 T is also found instead of 6 in

Ensslin op. cit. (above, note 25), p. 443. Preisigke, op. cit., 2, p. 49. 51 SO 6-r' for 66' ( = 8 8 ~ ) {TE , for @E,T O X ~for d5 Griechische P a p yrusurkunden derHamburger Goxi, 6 6 ~ w K for ~ s 868w~as,T W ~ E K ~ O(for W 8o6-), Staats- u n d Univevsit&tsbibliothek, ed. Paul M . etc. ; aiyi-r~for aiyi61, pa-rile~v(for -rw6~~apfivow, Meyer, I, Heft 3, Urkunden 57-117 und Indices paGilelv), oepi-r~ov(oepi61ov=.rrw~i61ov), -r16irpwv (Leipzig and Berlin, 1924). (6t6irpwv), etc. (Edwin Mayser, Grarnmatik dev griechischen P a p y r i a u s der Ptolewaerzeit, I 4% On the adj. pahhw-r65 and the noun derived [Leipzig, 19061, p. 176. Cf. Karl Dieterich, Unterfrom it see Liddell-Scott- Jones, s.v; Preisigke, 09.cit., 2, p. 49; Sophocles, Lexicon of the R o m a n suchungen zur Geschichte dev griechischen Sprache and Byzantine Periods, s.v.; Du Cange, Glossawon der hellenistischen Zeit bis z u m 10. Jahrh. n. Chr. [Leipzig, 18981, p. 84); EqJo-riow for EqoGiow r i u m ad scrifitores mediae et infimae Graecitatis, 858 ; ! . ~ a h h o ~villosus: 6~ Corpus glossariorurn (Ostvaca Mich., 102.8 and 17; from Karanis, latinovum ( = C G L ) 3.269.40 ; pahho-rfi abolla, early fourth cent. A.D). ibid., 3.272.59; malloti bellata 3.197.9; Buck52 I noted the following: line I I E-rva, 33 ETVOV Petersen, A Reverse I n d e x of Greek N o u n s and (but cf. 128 E ~ V U 1, 12 E ~ V O V I 2) ,E T O ~ E(but 71, Adjectives, 526b; K.Amantos. "Mahhw~6v-pahho117, I25 E ~ O ~ E39 V )T,O ~ E V T U ( = ~ o ~ ~ v T77 u), 'EITET~PIS'E-ra~p~ias Bwlav-r~vijvZITOWT&PIOV," T O ~ E V T 41 W Vand , 54 T I ~ O I P O V (but 43 G~po~pov), 6812, 2 (1g25), p. 278f. I n medieval Greek 45 -rlaqopqS (=61aq6po1~),47 ol~powv(=u16qpahhw-r6v (Prodromos 1.95 ~ a ~&pvw l ~ a -ra i (=ir1~01~661ov), poGv), 47 w-rro~~o-r~ov 67 OKTOOV pahhw-ra) and pdhw-r&p~v(Theophan. Contin., (=dy6oov), 69 I T E ~ I Uand -rre-rIa (inadvertently 617.4 mpayhopaAhw-rapla) "woolen blanket" repeated), 76/77 a-rro-roolv (= &-rr66oolv),86 E ~ T O ~u, p-rrav(synon. mod. Greek P E ~ ~ v To~owhhhq PqKOVTa, I07 and I I 0 - ~ ~ E V T E K ~ ~ T( = E K-rrEV-rEKal~ ~ (see above, p. 256,a). -ravia); cf. A~vopahho~apla B E K ~ ~etc. s), I n modern Greek !.~ahhw-r6~ (Zacynthus, etc.) 5Y E.g: drrro-rc.ho~v P . Princ. 111 142.9 (A.D. 23) and pahAovr6s (Thrace; with the vowel u in the for & I T O ~ C ~ V E I V northern dialect group for unaccented o [here w]), G~a-roxfiv P. Vindob. 1.16 (Fayum, after A.D. 87) pahho-rfi f. "the amount of wool from a shorn for 61a6o~fiv sheep," "one shearing" (Euboea), pahhw-r6 neut. 2 ) Ec660u E
50

DEMETRIUS J. GEORGACAS

borrowed w0rds.~4Likewise in Greek loanwords in C o p t i c S Stenuis ~ [t] stands for Greek 6 The reverse phenomenon, namely 6 Jernstedt (Tiflis, 1927)~p. I O O ~ . for 6C16iow, drqSlZ6pqv TET~UTT~T Les ~ V papyrus The'odore Reinach, ed. P. Collart, I1 (Cairo, 1940), no. 84 for TETPUrr68av E~T~TOP S . Hamb. 68.47 (Aphrodito, after A.D. 550) for € 1 8 6 ~ 0 ~ i~66osP S I 7.736.13 (A.D. 208) for ~ 1 6 6 ~ 0 s &po~~Ijopa~ lustro C G L 3.485.23 for &?06~\5o(Ph. Koukoules, B Z , 20 uai t &pqoS~ljo~a~

[1911], P. 390). P. Princ. AM 8961.38 hw~i~lov, P. Cair. this is from M a s p . 2.6713g.Va 23 ho-rl~=hw~i~la; hw6i~lovwhich is found in papyri of the second, third, and sixth centuries A.D. (Preisigke, op. cit., 2, p. 44 S. VV. hw6iKlv, hw6i~tov;cf. also C. Wessely, Wiener Studien, 24 [ I ~ o z ]p. , 137); the form hw8iKlv (second-third centuries A.D.) with -iv is the genuine spoken form (see below, note 65), a derivative of A061c (found in papyri of the first and second centuries A.D.. ; cf. hCj81~~~06 srohhai, &rrhoi [probably &rrha7] TE ~ a IVT~ITIOI, i K-rh. Perip.$s mar. Erythraei, chap. 24, ed. H. Frisk [ G H A 33 (1927), I , p. a]), borrowed from Lat. lodix "coverlet," "blanket." 55 Coptic or New Egyptian comprises two dialects: ( I ) the Sahidic dialect, i.e. the dialect of Misr in Upper Egypt, or more specifically the one spoken in the North Nile Valley (from Old Cairo to Asyut with Hermoupolis [Antinoe] as its center; Arabic es-salad "the upper land," i.e. Upper Egypt) recorded from the third century on; and (2) Bohairic or Lower Egyptian spoken in the province of Bahirah, i.e. the dialect of Alexandria and its surroundings (Arabic buhaira "Western Lower Egypt"), recorded from the ninth century on. The oldest document in Old Coptic is of the second century A.D. See Georg Steindorff, "Bemerkungen iiber die Anfange der koptischen Sprache und Literatur," Coptic Studies in Honor of Walter Ewing C r u m (Boston, 1950), pp. 189-213. Loan-words, both of Christian religious content and of everyday use constitute about one eighth of the Coptic vocabulary. Sahidic has more Greek words than Bohairic, over goo in all (L. Th. Lefort, "Greco-Copte," Coptic Studies in Honor of W . E . C r u m , 69). Coptic was influenced by Greek pronunciation at various stages of the evolution of the Greek language (W. F. Albright, Language, 10 [1934], p. 220). What is more, Egyptian scribes who were ignorant of Greek often wrote Greek words phonetically as they heard them, a procedure that may account for native traits in recorded Greek texts. On the relation of Coptic to Greek cf. E. Schwyzer, Griechische Grammatik (Munich,1938-53), 1.16of. 56 SO X O ~ O T I O C (A. Van Lantshoot, "A propos du Physiologus," Coptic Studies in Honor of W . E. Crum, 348, no. 8) for xapaGpi6~;

in the place of T , may be observed in the Egyptian Greek of the second century B.C.~' and in later papyri,S8 including P. Masp. 6 7 0 0 6 , ~as ~ well as in Christian inscriptions@ and in Coptic.61The same phenomenon applies ~ohocBohairic (K. Wessely, Die griech. Lehnworter der sahidischen und boheirischen Psalmenversion Vienna, 1910 = Denkschriften d. A k a d . d. W i s s . in W i e n , phi1os.-hist. Kl., 541, p. 246) for 66hos; a19apo~oc( E . W . Budge 18.22) beside o19apo80~ (ibid., 14.2) for ~ 1 9 a p q 6 6and ~ Bohairic AIATOXOC for 61dr6oxo~; A. Rahlfs, "Griechische Worter im Koptischen," Sitzungsber. d. Preuss. Akad. d. Wiss., 2. Halbband (Berlin, 1912), 1038; TCCnOTA for 8iorro~a (H. P. Blok, "Die griechischen Lehnworter im Koptischen," Zeitschr. f . agyptische Sprache und Altertumskunde, 62 [1927], 54); AYeATHC Bohairic for air9drSq~, Bohairic TAnANH and A- for Gmdrvq, Bohairic KATATIKC1 for ~a-raGi~q, and Bohairic

nOTHPION for sro6fipqs (A. Bohlig, Die

griechischen Lehnworter i m sahidischen und

bohairischen Neuen Testament jMunich, 19541,

1.1o6f.).

5 7 SO & T T O ~ I U ~ W(&ITOTIU&TW), I ~ ~ K T W (TBK-

V ~ o v ) Sihos , (~ihos), sri6aaos (rrhaaos),~ 6 ( T8~ T~E;) see E. Mayser, Grammatik der griechischen P a p y r i aus der Ptolemaerzeit, I. I 75. 58 6 6 ~ 0 ~P. s Princ. 111.141.3 (from OxyISKIWS~V S; P . Princ. rhynchus, A.D. 23) for T ~ K O W AM 8961, line 38 ( P a p y r i in the Princeton University Collections, 11, 82.38) for U K I U T ~ V; 66pow P . Grenf. 2.38.2 for ~6pow;yi8avos for ysi-roves P. Mich.Tebt. I1 282.4 (C. B. Welles, American Journal of Philology, 67 [1946], p. 86) ; etc. 59 So 25 and 38 E-rEpa as well as 27 ETEPU, 49 ~~Sap-rov (~i-rap-rov),60 and 87 Gwhap~ov (~haplov), 94 -row6 EUTIV (TOW&JTIV), 94 E ~ O I ~ W S EXEIV (hoipas ZXEIV), 100 T ~ ~S E ~ E (TT ~~ T SS E ~ E T ~ S ) , 83 ~apmcrv6aSaas well as ~ r a p a ~ a v 6 w(for ~a rapayav6w~a). 130 SO I T O ~ I U O V (for IT~TIUOV) and QSaSos (for 6 6 ~ ~ 0 s G. ) ; Lefebvre, Recueil des inscriptions grecques-chre'tiennes d'dgypte (Cairo, 1907), p. 128f., No. 663. 131 AAZlC (for T~CIS), AlMWPlA (for papia), T~OSEKTOP G~~paywvov, as well as T ~ O T E K T~~, ZqGqc~c, (once) as 9~a6pov,U ~ X I ~ E K - ~ ~ VcaSavac , well as cmavac, 6owp.rravov (~psravov), xlpo6ov1a (XEIPOTOV etc.; ~ ~ ) ,see Th. Hopfner, "uber Form und Gebrauch der griechischen Lehnworter in den koptisch-sa'idischen Apophthegmenversion," Akad. d. W i s s . in W i e n , $hilos.hist. Kl., Denkschriften, 62, Abh. 2 (Vienna, 1918), p. 7; Blok, op. cit. (above, note 56) 53; Bohlig, loc. cit. (above, note 56); Wessely, 09. cit. (above, note 56) gb.

G R E E K T E R M S F O R "FLAX" A N D " L I N E N "

to the other stop-consonants K and y, .rr and p, e.g. & ~ d p (for a &yahpa),&p~irp~ov (for &pyirplov), s r a o t h t ~ f j(for ~ ~ u ~ I A I K etc. ~~s), Considering that Greeks and Egyptians lived together for five hundred years, it is natural that numerous Greek words should have been introduced into the native Egyptian tongue. The Greek elements adopted by Coptic constitute a significant source for the Greek Koine of the papyri,sa so that the treatment of sounds in Coptic (e.g., the transformation of y into K,and 6 into T) should be considered in conjunction with the same phenomenon in Greek papyrus texts, as P. KretschmerB3has observed. Such linguistic phenomena evidenced by carefully written papyri which display no obvious barbarisms deserve to be recorded and explained. The form Alvohla is a case in point. The noun Alvoir6lov neut. "linen shirt" (P.Oxy. 114.8 [second century A.D.]; 1066.10 [third century]; etc.)a became A ~ v o h v [ l i n ~ i t i n ] Professor .~~ Hjalrnar Frisk did, to Cf. Lefort, op. cit., 70 note I. Glotta, 26 (1938), 42f. H. P. Blok (loc. cit. [above, note 561, p. 60) had also stressed that "die griechische Dialekt- und Papyrusforschung wird auf die Dauer des reichen koptischen Materials nicht entbehren konnen, wenn man einmal zu einem einheitlichen Uberblick iiber ihr ganzes Gebiet gelangen will." Preisigke, op. cit., 2, p. 24, s.v. A1voi~61ov; cf. Liddell-Scott- Jones, s.v. A1voir61ov. 86 On the termination -1v see D. J. Georgacas, "On the Nominal Endings -1s -IV in Later Greek," Classical Philology, 43 (1948), p. 243 ff. It is certainly true that the written language does not always keep pace with the contemporary form of speech. When, however, colloquial elements appear in papyri, they should not be "corrected" by the editors. Thus, in late Koine texts nouns terminating in -1v (neut.) should not be corrected into -IOV. E.g. T E I T ~ V I V in an account of unknown provenience of the third century A.D. should not be read T ~ E I T ~ Vas IOV, Naphtali Lewis ("Ostraca grecs du Musee du Caire," Etudes de papyrologie, 3 [1936], p. 106) suggests; K I ~ ~ W T I Vand &A1oi6lv in a papyrus of unknown provenience, perhaps of the fourth century, should not be read KIPWTIOVand &Awoi61ov,as Edmund Harris Kase, Jr. (Papyri in the Princeton University Collections, I1 [1936], nos. 95.15 and 22, p. 87f.) suggests: &Awoi61v and KIPWTIV are well-attested forms like K ~ E poirplv in the same papyrus (95.17). Thus also ~b Avrr&61v (Ostr. 5.718; fourth or fifth century A.D.) which Herbert C. Youtie reads as Ao1srCr61ov (Transactions of the American Philological Association, 81 [1g50], p. 106), should probably 62

63

be sure, observe in the papyrus he editede6 the occurrence of T for 6 in the words hlvohla, w h i ~ t a ,and E i p a l ~ a ,but he seems to have been unaware of the phonological importance of this change. The transformation of ancient Greek 6 [dl or Koine 6 [dl into [q], that is, in the first instance, the devoicing of d, resulting in t, and, in the second, the substitution of t for d, appears to have been a local phenomenon in the Greek of Egypt and never affected the language used in Greece. But even in Egypt this phenomenon was not at all general (note the presence of both 6 and7 in the same words: ~ ~ T U K Ufor S ~ ~ S U K U~t6irpov S, for 616irpwv, 61Crroxos for 61&60xos, e t ~ . ) , and ~ ' must have been due to the influence of native Egyptian on the Greek language of Egypt,@ both in the case of the syllable 61 and in other combinations. The Koine term Amoir61ov and kvo66lv used by the Greeks of Egypt probably became Coptic lindtin; then the Greeks, hearing the Copts use it, took it back in its new form Alvoirrlv, a Riickwanderer into Greek speech.

8. There is no Change 6 1 4 7 ~ 1 We need not consider the possibility of an early forerunner of the diminutive suffix -0riy1, found in the modern Greek dialectal speech of Kymi (in Euboea),s9 or of the transformation of d j into j before a vowe1,"J be sb A O I T ~ & ~ I V(pronounced liipdain). P. Oxy. 1858.5 (sixth-seventh century A.D.) yoyaplv should be given under lemma yop&plov, yoptrplv and not with Prof. Kiessling (Worterbuch der griech. Pa+yrusurkunden, 4.437) under yoyaplov. The noun 696vqv ' I V ~ I K ~ VPeripl. mar. Erythr., 31, should not be corrected to 696v1ov ' I V ~ I K ~ V (so H. Frisk, Le Pe'riple de la Mer Erythre'e [Goteborg, 19271, p. 10, line 23) but to 696v1v. See above, p. 263, bf. O7 H. P. Blok (above, note 56) considers such examples to be the result of dissimilation, and others such as Swva66~(for Gwvcrrb~),G ~ o ~ r 6 6 q ~ (for ~ E U T ~ T ~ etc., S ) ,the result of assimilation. 68 Cf. Eduard Schwyzer, Griechische Grammatik (Munich, 193g-53), 1.160f. OD S. Karatzas, ' Y T T O K O ~ I ~ TOG; IK& i61Gymo~ Kirpq~ ~ a lsr~plxbpov (Diss. Athens, 1944, published in 1954: Collection de l'lnstitut Frangais d'Athknes, 84), p. 21. Cf. my review in BZ,50 (1957),P. 153. 70 SO D. Vayacacos, 'A9qv6, 60 (1956), 335, following Karatzas.

DEMETRIUS J. GEORGACAS

not because of the gap between Koine Greek in Egypt and Greek speech in Greece, but simply because there is no Koine Greek word Aivoi-yiov, but only a ghost-word form by which my distinguished colleagues S. Karatzas7l and A. Tsopanakis7=were led t o believe that the change of [ d j ] into [ j ] was an early phenomenon. As far as I know, the substitution of [ j ] for [ d j ] is merely a medieval dialectal phenomenon which came into being after the shrinking of two syllables into one (synizesis or yodization) entered the structure of the language. Indeed, in some dialects d j appears as [ j j ] ,i.e. as a double consonant that resulted from the assimilation of d to j. On the other hand, the use of j instead of d j is familiar to speakers of modern Greek who may observe children pronounce djo for ddjo (written &S~io),vdja for vddja (written P661a), pejd for pedjljri (written .rraiS~a),etc. Karatzas' statement that the synizesis 6jd( in the preposition Sia is found in ancient Greek73cannot be considered correct, except, of course, that it occurs for metrical reasons.

I t is not clear whether Hesychius' gloss (sic), given in this meaningless form in the Diogenianian part of He~ychius7~ can possibly be h l v o h ~ aneut, plur. or AIVOWTI& fem. sing. In any case, the gloss cannot be

A1vam.a

h~vcnrr~&.~~

I take the noun to be A~vovridcf. because the gloss is so accented. Such a noun is not attested as a Greek nominal derivative, yet it is possible, as we shall see below. The word hivovridc or h1vow61dc would, to be sure, be out of its alphabetical order in Hesychius' lexicon and this, according to Professor whom I K. Latte, the editor of Hesy~hius,7~ consulted about this item, is improbable. Karatzas, op. cit., pp. 2 2 and 64. A. Tsopanakis, BZ, 48 (1955),p. 70. 73 Karatzas, op. cit., p. 2 2 . 74 Hesychii Alexandrini lexicon, editio (altera) minor, ed. Mauricius Schmidt, ( J e n a , 1867), col. 686. In a footnote t h e gloss A1yvirs.~mv6~ is separated f r o m th1vaw7.a. 75 T h e gloss is listed i n Liddell-Scott- Jones, s.v., 1051a, and is also included i n Buck-Petersen, A Revevse Index of Gveek Nouns and Adjectives, p. 165a. 7s B y letter o f 6 November 1956. 71

72

Nevertheless it seems to me that the gloss h~vovridcis possible.

The lexicon of Hesychius, preserved in a single manuscript of the fifteenth century ( M a r c i a n u s Gr. 622), is not now in the form in which its author wrote it. Whereas the Marcianus is completely alphabetized with some displacements and stray entries, the earlier arrangement of Hesychius and Diogenianus applied alphabetical order to only the first three or four letters of each entry. Successive recopyings of Hesychius' lexicon brought about two errors, (I) frequent separation and rearrangement of glosses, and ( 2 ) systematic interpolation of the work, especially with glosses drawn from the collection of Cyrillus (an Egyptian Christian of the fifth century).77 The order in columns 686-687 is as follows: hiv hiva th~vay~p~owpivq thivapa~ thivm.dc h1vSia9a1 Aiv61oi Aivaov hlvairs ~ ~ V O K ~ ~ V K E S

Aivov hivq, o v v d r m ~ ~ v h l v o ~ h (6V ) T ~ S h~vom&lei ~IV~TT(T)TJS hivo-rrhi-jyes ~ I V O T ~ ~ O I

hivo~i-rwv hit

etc. Thus, if the original gloss was ~ivovrldc or hivow61a it may originally have been placed between the glosses hlv0~6~101 and h~voxi-rwv but, after it had been miscopied as hivamidr, it could easily have been transferred in a later manuscript to its present place. 7 7 See t h e preface o f K u r t Latte, Hesychii Alexandrini lexicon, vol. I , A-A, (Hauniae, 1953); and J o h n L . Heller's review i n American Journal of Philology, 76 (1955), pp. 216-218. C f . also J . L. Heller's review o f The Herbal of Rufinus b y L . Thorndike and F. S . Benjamin, Jr. (Chicago, 1946), i n Classical Journal, 43 (19481, P. 445-

G R E E K T E R M S F O R "FLAX" AND " L I N E N " I t may be noted in passing that the gloss A I ~ v ~ I S ' K Cwhich ~ ~ ~ V is ~ Salso , out of its alpha-

betical order and was separated by Musurus, is a repetition of the Cyrillian gloss 976. One may cite hundreds of such glosses transferred from their original place to another place in the lexicon as a result of having been first miscopied and then alphabetized anew by subsequent users and/or scribes. I cite one example: The gloss *yop6s'wp~6s, found in Hesychius (as well as in Cyril and Zonaras) is alphabetized between ~ O T ~and V [y6po~os and] *yoCv; *yop6s, however, seems to be a scribal error for y v p 6 ~ 7which ~ was realphabetized by a user and copied in this fashion in a subsequent manuscript. In the modern dialect of Crete and specifically of Rethymno and its surroundings there exists the noun AtvouGttr f. [linu8jd] "mass of flax processed in a machine called manganos [see Atvapoptxyyavo above]."7BAdmittedly, this word does not explain the form Atvowrta which I am suggesting in place of t h t v m . t r , but an informant from Arta (Epirus) told Mr. D. Loucatos that she knew the word A~vowr~a,"flax processed in a manganos," as a synonym of the expression 78 Cf. Liddell-Scott- Jones, s.v.; Hesychius, ed. Latte, 1.388, gloss 863. '9 There is also AtvouGtCc [linuBjL] f. "odor of burning linen cloth" (synon. Atvaia or Atvai; see above, p. 257, a) in Asi-Gonia near Rethymno, Crete, as reported by Mr. N. Kontosopoulos (Athens), which is a derivative of Atvo661 with the suffix -Qa(originally -aia).

269

payyavlopivo htvapt also used at ArtasO (see above p. 259, a). The form Alvov~lais, in any case, unrelated to Coptic A~voirrlvand may have arisen under the influence of an adjective *Atvw~6s "of linen" ;so paAAw~6~ "woolen" (synon. ptdhlvos ; med. ~b ~ ~ A w T ~0v1 )~ ,~ 6 5XITWV . h b ipiwv (Hesychius), I T T I A ~ T ~ S "winged," "stuffed with feathers," 8 v o a v w ~ 6"tasseled," ~ "fringed", ~ E T ~ ~ W"ofT silk" ~ S (Herodian, Byzantine and mod. Greek ; synon. mod. Greek ~ET&
The results obtained in this study, namely the elimination of the ghost-word AmoiIytov and the explanation of the form Atvoirrlv, may appear to be rather minute. I t is, however, by means of such detailed investigations that the study of language is gradually built up. I n particular, I have attempted to illustrate the importance of medieval and modern Greek for the correct understanding of papyri. By neglecting the later evolution of the Greek language, papyrologists have been induced t o commit numerous errors both in the editions of texts and in dictionaries. This deficiency can be remedied only by a thorough acquaintance with Byzantine and modern Greek, which form the direct continuation of the Koine. However, Mr. Stephanos Pappas, principal of the Gymnasium for Boys No. I at Arta, has kindly made inquiries about the terms AtvoiI61, etc., but without result (as he AtvouGt&, A~vom~tr, reported to me in his letter of November 1958).

http://www.jstor.org

LINKED CITATIONS - Page 1 of 2 -

You have printed the following article: Greek Terms for "Flax," "Linen," and Their Derivatives; And the Problem of Native Egyptian Phonological Influence on the Greek of Egypt Demetrius J. Georgacas Dumbarton Oaks Papers, Vol. 13. (1959), pp. 253-269. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C253%3AGTF%22%22A%3E2.0.CO%3B2-4

This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes] 10

A Pythagorean Idea in Jerome Johannes Quasten The American Journal of Philology, Vol. 63, No. 2. (1942), pp. 207-215. Stable URL: http://links.jstor.org/sici?sici=0002-9475%281942%2963%3A2%3C207%3AAPIIJ%3E2.0.CO%3B2-I 10

Additional Note on "Wool and Linen" in Jerome Leo Spitzer The American Journal of Philology, Vol. 64, No. 1. (1943), pp. 98-99. Stable URL: http://links.jstor.org/sici?sici=0002-9475%281943%2964%3A1%3C98%3AANO%22AL%3E2.0.CO%3B2-M 55

Review: [Untitled] Reviewed Work(s): Coptic Sounds. Part I: The Main Currents of Their History by William H. Worrell W. F. Albright Language, Vol. 10, No. 2. (Jun., 1934), pp. 220-224. Stable URL: http://links.jstor.org/sici?sici=0097-8507%28193406%2910%3A2%3C220%3ACSPITM%3E2.0.CO%3B2-Q

NOTE: The reference numbering from the original has been maintained in this citation list.

http://www.jstor.org

LINKED CITATIONS - Page 2 of 2 -

65

On the Nominal Endings -##, -##, in Later Greek Demetrius John Georgacas Classical Philology, Vol. 43, No. 4. (Oct., 1948), pp. 243-260. Stable URL: http://links.jstor.org/sici?sici=0009-837X%28194810%2943%3A4%3C243%3AOTNE--%3E2.0.CO%3B2-Q 77

Review: [Untitled] Reviewed Work(s): Hesychii Alexandrini Lexicon by Kurt Latte J. L. Heller The American Journal of Philology, Vol. 76, No. 2. (1955), pp. 216-218. Stable URL: http://links.jstor.org/sici?sici=0002-9475%281955%2976%3A2%3C216%3AHAL%3E2.0.CO%3B2-R

NOTE: The reference numbering from the original has been maintained in this citation list.

Byzantium in the Seventh Century: Report on a Dumbarton Oaks Symposium Ernst Kitzinger Dumbarton Oaks Papers, Vol. 13. (1959), pp. 271-273. Stable URL: http://links.jstor.org/sici?sici=0070-7546%281959%2913%3C271%3ABITSCR%3E2.0.CO%3B2-L Dumbarton Oaks Papers is currently published by Dumbarton Oaks, Trustees for Harvard University.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. 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/journals/doaks.html. 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.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 07:46:25 2008

BYZANTIUM IN THE SEVENTH CENTURY Report on a Dumbarton Oaks Symposium

T has become a tradition for Dumbarton Oaks to hold in the spring of each year a symposium on a theme within the general field of studies to which the institution is dedicated. The present volume includes several of the papers which were read at the symposium of 1957. In order to enable the reader to form an idea of the context in which these papers were presented a brief report is appended here, outlining the program of the discussions and summarizing in general terms the problems with which they dealt. The symposium, which was directed by the present writer, bore the title "Byzantium in the Seventh Century." Its scope was somewhat larger than the title suggests, for in fact the program encompassed the entire period between the death of Justinian I and the outbreak of Iconoclasm, a period which is coming to be recognized more and more as one of the most critical in the history of Byzantium. Scholars working on different aspects of Byzantine civilization have concluded independently that is was during the late sixth and seventh centuries that the foundations were laid for the Byzantine Middle Ages. Accordingly, the program was designed to range over a number of different fields; to define the significance of the period in each of them; and thus to throw light on the relative importance and the interaction of different factors in bringing about so critical a development. Of a total of eight papers the first four were concerned essentially with political, social, and economic history, the remaining four with intellectual and cultural history. The program opened with a lecture by George Ostrogorsky dealing with the drastic territorial upheavals of the period and their effect on Byzantium's position as a world power. This was followed by a paper by Peter Charanis illustrating concomitant changes

I

in the ethnic composition of the peoples dwelling within the Empire and shifts in the relative importance of different ethnic groups. While both these papers were concerned with elements of change, the third lecture, again by Ostrogorsky, gave more emphasis to factors which persisted amidst catastrophes and transformation. Dealing with the internal organization of the Empire, and particularly with the survival of urban life, the paper led to the conclusion that, contrary to the views expressed by some scholars, cities and commerce continued to play an important role during this period. Robert Lopez' lecture, which followed, dealt more specifically with commerce and trade and showed that, while in regard to the pattern of international trade the period must indeed be presumed to have brought about far-reaching developments, Byzantium's role was essentially a conservative one. All four of these papers are published in the present volume in the same order and substantially the same form in which they were de1ivered.l Next came a paper by Anatole Frolow which was designed to form a link between general history on the one hand and intellectual history on the other. Based on the author's extensive researches in the history of the cult of the relics of the True Cross, the paper (since published in St. VZadimir's Seminary Quarterly, N.S. 2 [1g58], pp. 13-30) showed to what extent during the latter half John L. Teall's study of the Byzantine wheat economy, while not presented at the Symposium, supplements the paper by Lopez which it follows in the present volume. Dealing with a sector of the Byzantine economy in which the catastrophes of the seventh century are often assumed to have had particularly grave consequences, Teall's study shows that the extent of the adjustments called forth by the political and military events of the period should not be overrated.

ERNST KITZINGER

of the sixth century this cult began to enjoy official and governmental sponsorship, particularly in military contexts. Pursuing a related line of investigation the present writer then attempted to define a similar and concurrent change of attitude on the part of the secular government in regard to religious imagery, and to attribute these changes to a subtle shift in the imperial government's concept of its own power and authority. Entitled "Emperors and Images," this lecture was an expansion of a thesis put forward in a study which appeared in Dumbarton Oaks Paflers, Number 8 (see especially pp. 121-128). AS a supplement to this paper James D. Breckenridge offered a short report dealing specifically with the coinage of Justinian I1 and summarizing the conclusions of a forthcoming monograph (Numismatic Iconografihy of Justinian 11 [N.Y., 19591). Further light was shed on the new spiritual and intellectual climate of the era in a paper by Paul J. Alexander entitled "Byzantine Literature and the Spirit of the Seventh Century," which the author hasnot yet been able to prepare for publication. The last lecture, again by the present writer, dealt with the major stylistic trends in Byzantine art of the period. An expanded version of this paper was subsequently submitted as a report to the Eleventh International Congress of Byzantine Studies held in Munich in September, 1958, and has been published in the Acts of that Congress. In general, the papers confirmed for many aspects of Byzantine life the critical character of the period and its important role in laying the foundations of the Byzantine Middle Ages. On the other hand, they also served to underline the essential conservatism of Byzantine civilization, the staying power and regenerative strength of long-established traditions-Graeco-Roman traditions in particular-even amidst all the changes. This is true, for instance, in the political sphere where the concept of a universal empire survived the tremendous challenges of the age; it is true in the social and economic sphere where the adjustments called forth by political and military events were relatively far less drastic than in the West; and it is evident also in the continued strength of the classical element in literature and art.

The most obvious and tangible changes are those that took place on the political map, the catastrophic losses of territory in Italy, in the Balkans, in Asia, and in Africa, and the inroads made by barbarian invaders even in what remained of the imperial domain. One point which the discussions at the Symposium particularly served to emphasize was the great importance of the invasions of the Slavs. Depriving Byzantium of a firm hold on the Balkans, these invasions were like a deep wedge that split asunder the ancient world. Thus their effect was graver perhaps than that of any of the other territorial changes in East or West. Some of the other developments and innovations that were discussed stand in a direct causal relationship with the physical transformation of the Empire. This is true, for instance, of the intensified contacts with the Eastern barbarians and their increased influence on Byzantine manners and customs; or of the reorganization of rural life on a semi-military basis, a development which in turn involved far-reaching adjustments in the ownership of land and consequently brought about major changes in the class structure of Byzantine society. The repercussions of all this in the cultural and intellectual sphere are bound to have been very great, although due caution must be exercised in any attempt to establish direct causal connections. Thus the new pervasiveness of the religious element, which, in the light of several of the papers read at the Symposium, stands out as one of the most characteristic traits of the age, may be assumed to be due, in part at least, to the manifold stresses and strains to which all strata of Byzantine society were subjected. The fact that the Empire was forcibly brought into intimate contact with barbarians and unbelievers, and was on the defensive as it had never been before, may likewise have helped to produce an unprecedented emphasis on Byzantium's Christian identity. Moreover, the events seem to have induced the rulers themselves to promote and emphasize religious factors in their own sphere. I t was a means of expressing their own subordination tobut also their role as representatives of-the only Power that amidst all the catastrophes remained truly universal.

BYZANTIUM I N T H E SEVENTH CENTURY

The symposium, however, also made one aware of the limitations of a method which seeks to interpret cultural and ideological developments in terms of material ones. A synoptic view, such as was attempted here, of Byzantine history and civilization in the late sixth and seventh centuries suggests, in fact, that an inner crisis, a kind of spiritual malaise, was already in the making when such a state of mind was as yet hardly justified by political and military events. In the visual arts, for instance, it seems that symptoms of change and disintegration can already be observed in the time of Justin 11, if not in the closing years of the reign of Justinian himself. There seems to be a kind of slump, a relaxation of effort after the great aesthetic achievement of the first half of the sixth century. I t may well be that in other respects too the spiritual crisis was a crisis before the event, an anticipation of stresses to come rather than their product. The general picture, conveyed by the

papers, then, was one of an age of insecurity, of suffering and mortal peril, but also of inner strength and creative opportunity. In three spheres the Empire was shown to have been particularly resilient and resourceful: the economic, the military, and the spiritual. In this last respect, and particularly in regard to religious practices, there clearly were excesses. Hence, a search for the causes of eighth-century Iconoclasm does not seem to neccessitate going as far afield as some scholars have done. The overt cause-excessive addiction by the people to the cult of holy images-was a very real thing in seventhcentury Byzantium. The imperial court's eventual reaction must be seen in the light of the fact that the emperors themselves, in the stress of the times, had helped to bring into vogue. On the other hand, one must not lose sight of the positive forces set free in the crisis of the seventh century, for it is they that were to give mediaeval Byzantium its strength and its character.