$99 
 DBBS2 
 
 
 
 THE UNIVERSITY OF CHICAGO 
 
 ORIENTAL INSTITUTE PUBLICATIONS 
 VOLUME I 
 
 ORIENTAL FORERUNNERS 
 
 33] OF BYZANTINE PAINTING 
 
 JAMES HENRY BREASTED 
 
ett OF PRINCES. 
 ‘NOV 3 1994 
 
 
 
 
 fA 
 (75h \ 
 LOL agicn, sews 
 
 Fo lve 
 Division DS 99 
 
 Section a AB Ad 
 
 
 

 
 
 
 
 
 
 
 
 
 
 qe 
 fs it: 
 Ty ry , > r ‘ 
 ret ay ; nee i 
 7 a ae < ges F 7 ' , Te 
 a: yu ane 3 ie : i i d i AD TUN wih key am eb ' 
 J J i j ei hf; Mai i ry f 
 eras oan a" é rg oe a J t 1 ? P ‘ 
 : it et 
 - ; sa ath A " 4 ‘ j i 4 i e 
 vo" ay Oe ‘ An , heat Ye ‘ aI } 
 ‘ 4 : } Pus , ' ; 
 j FA ich vi bey i L y j ‘ i 
 eh) a fur dd : - . y 
 yom 7 a i . J | 
 a ’ i ¢ 
 i” Z ’ | 
 . i 5). ' i \ ri 
 ' - b 5 1 
 i : , . I 
 ar =f ‘| ' 1 
 ff Has I Ui ‘ } 
 grr h bly \ ) 
 A R's 
 p oy r \ 
 a HS h as , 
 | be { i 
 ead BS 7 A , . 
 nt i 3k : 
 H 1 t 1 a 
 ‘ a ‘ 
 i ae iz ; i 
 7 i 
 f % 44d 
 ‘ ' Y } 
 | ‘ 
 ee ! j 4 ‘ 7 f » 
 q al yn! ‘ 
 Ny 5 Hav ) ut ; ri 1 
 ; ; i " ‘ cy i 
 mb. ye 
 04 y Pes a i ; 
 1 , : ; " 
 LOhy v BA ‘ 
 m ot Ce ae \ , , ¢ A 
 : ae | ey 4 py . 
 7 * in i ‘ad ih f & ! | : f 
 oly ne se f ea ay he i 
 , ee ‘ oF b 
 al ya eS ee le } oe a | . ; 
 "i és as wtb nal oh J 7 i P é owe if ; i i 
 “4 f : , lhe ; : F rT i 
 ; Ma it: is Wands . 1 na eh hacks I , HN : tt vy 
 iv “ en at i ‘. At | ' ‘ é i ’ ol ; | ‘ ie * " 
 oe ig Were hand ee ey a | Wi ; . \ i ’ L i if tee | iy 5 
 ia z Me Wy ai Ne ee \ ea qv #3 i 1 ce ek) ey, mr i f Py at 
 4 Z P la KY Tor oe ty iba rs Au 
 iy . ae Ag 4 wh At 4 : \ ’ Ay i ; am) 
 od A : ¥ 
 t : 'G ek r ‘ Bet . * d on 
 ‘ ? . oy aa rif’ 
 Mes ney , thee ‘cir 
 ar ry: y nf , ' cr) ee y as 
 i¢ ‘ ? j ver 
 ' Y ( Li] rie ‘vi 
 r r ’ 4 a! 
 Nites (84) |) 5 (a 
 4 ‘ nes a4 yy ut ¢ 
 oy , 5 a } f ‘ m 
 ee Vt LA he % 7 ae, 
 ae) ' dn Menthe) ee , ‘ \ ae. 
 (ean Ade in uot 1) ee, t 
 - RAIN! ‘ “us fl at ‘ u , i 
 ‘ he t4 aie ' 
 ( ; i ms V 
 . ® Ag : e : 
 i ny oe on ( Y 
 f i i a nae | 
 Ab Biay Tias are 
 iy 4 i / vi Patt ) 5% nn ‘ 
 iY, ta ‘ oe j } 
 
 
 
 
 
 
 ve 
 tnt 
 
 » ; 
 i" ’ fia eu 8 A ‘_ we 7 } , ; i DR ih its i Ay 1 Aad F 
 Sie a Lg Bu 5) ts we eee ‘a iien , WG + ! 
 ey : i ot , 4 ; a 1) if 7 ye) eh — 2 a? ie a y 7 AY iw 4 ' Pas op yo ‘ ’ D cts : . bi ' ir 1 ‘2 
 fh ae inant } ee we 4 i ha 7 ASRS a | Po TAS , OW Pran, © \ is . we 
 re -~ : i » et ms! SiN, Rat ae il ie J et ; , 1 } ae. 
 i ¥ Apod dp vh a6 : 5 a : : a, 
 
“7 
 
 
 
 Rae: 
 
 ile ara y 
 1 } 
 
 
 

 
 THE UNIVERSITY OF CHICAGO 
 ORIENTAL INSTITUTE PUBLICATIONS 
 
 Edited by 
 JAMES HENRY BREASTED 
 
 
 
 
 

 

 
 
 
 
 
 ORIENTAL FORERUNNERS 
 OF BYZANTINE PAINTING 
 
THE UNIVERSITY OF CHICAGO PRESS 
 CHICAGO, ILLINOIS 
 
 ‘ THE BAKER & TAYLOR COMPANY 
 New York 
 
 THE CAMBRIDGE UNIVERSITY PRESS 
 London — 
 
 THE MARUZEN-KABUSHIKLKAISHA _ 
 .Tokyo, Osaka, Kyoto, Fukuoka, Sendai 
 
 THE MISSION BOOK COMPANY 
 5 Shanghai 
 
 
 
Digitized by the Internet Archive 
 in 2022 with funding from 
 Princeton Theological Seminary Library 
 
 
 
 https://archive.org/details/orientalforerunn0OObrea 
 
 
 

 
 Pl. X. — The Wall of BithnanaYa in Hall I] : Head of the Second figure 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 THE UNIVERSITY OF CHICAGO 
 ORIENTAL INSTITUTE PUBLICATIONS 
 VOLUME I 
 
 ORIENTAL FORERUNNERS 
 OF BYZANTINE PAINTING 
 
 FIRST-CENTURY WALL PAINTINGS FROM 
 THE FORTRESS OF DURA ON THE 
 MIDDLE EUPHRATES 
 
 By 
 
 JAMES HENRY BREASTED 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >) 
 
 Zif \2ERN 
 eo 
 
 
 
 THE UNIVERSITY OF CHICAGO PRESS 
 CHICAGO, ILLINOIS 
 
Coryricut 1924 By 
 Tue Universiry or Cuicaco 
 
 
 
 All Rights Reserved 
 
 
 
 Published March, 1924 
 
 Composed and Printed By 
 The University of Chicago Press 
 Chicago, Illinois, U.S.A. 
 

 
 
 ai 
 FRANZ CUMONT 
 ore le ek ae IN 
 
 ites ADMIRATION AND REGARD 
 
 4 
 ’ 
 i 
 R 
 j 
 { 
 = 
 iZ [ 
 ier 
 b 
 * 
 ' 
 oo 
 % 
 ~ : 
 oa 1 . 
 ‘ 
 s% ’ 
 4 y 
 
 
 
s 
 ' 
 = : 
 ’ 
 b 
 . 
 T, 
 ‘ } 
 ‘ 
 i 
 ye 
 ‘ 
 a4) 
 9 
 J 
 , * 
 cy 
 
 
 
 : é byw a f,' 
 , Oe al ; 
 
 i 
 We ADs: ¥ 
 
 eh) ' 
 
 
 

 
 PREFACE 
 
 Our only archaeological predecessors at the ancient frontier stronghold of the 
 Seleucids on the middle Euphrates, with which this volume deals, called it “the 
 nameless city.”! Three years ago, indeed, the name of Dura was known to us only 
 in a few unnoticed passages of Greek and Latin sources, and it is safe to affirm that 
 few if any students of the ancient world knew that such a place ever existed. Buried 
 in the heart of the Syrian Desert, the ruin to which this name once belonged had been 
 forgotten fifteen hundred years ago. Its situation, 140 miles beyond the desert 
 metropolis of Palmyra, saved it from the destruction which overtook the latter city 
 at the hands of Aurelian in 4.p. 273, and at the same time left it so far beyond the reach 
 of modern archaeological observation that it has remained a lost city for fifteen 
 centuries. 
 
 ‘Tn this lost city on the outer fringes of the Roman Empire in Asia were thus pre- 
 served the only surviving oriental forerunners of Byzantine painting, out of which 
 arose the pre-Renaissance painting of Europe. Moreover, the city of Dura itself is 
 a unique survival. <A city left like a wave-mark on the oriental desert by the receding 
 tide of Graeco-Macedonian invasion under Alexander and his successors, it became an 
 oriental home of Hellenistic culture, a center of Graeco-Syrian civilization, too inacces- 
 sible and too far from the Mediterranean to be built over and engulfed by later Roman 
 structures. Untouched, therefore, by the Roman demolition so common in cities 
 of its age, the Hellenistic strata lie almost on the surface, unaltered by Roman or 
 Byzantine occupation. Forsaken at last and buried under a protecting mantle of 
 desert sand, its monuments and works of art have suffered little from the relatively 
 scanty rainfall, which did not penetrate deep enough to do any damage below. 
 
 Dura lies in the debatable ground between Syria and the Mediterranean world 
 on the west and Mesopotamia on the east and north. It was the good fortune of the 
 University of Chicago expedition to make the first dash undertaken by white men 
 after the Great War across this desert region and the newly proclaimed Arab state, 
 from Baghdad to Aleppo and the Mediterranean. The story of this journey has been 
 briefly told in the first bulletin of the new Oriental Institute of the University of 
 Chicago. Creeping up the Eyphrates as quietly and as expeditiously as we could, 
 and making every effort to elude the treacherous and hostile Beduin, we reached 
 
 1“T)ie namenlose Stadt,” Sarre and Herzfeld, Archaeologische Reise im Euphrat- und Tigris- 
 Gebiet, II, 394f. Berlin, 1920. 
 
 2 See the present writer’s report, ‘The Oriental Institute of the University of Chicago: A Begin- 
 ning and a Program,” Oriental Institute Communications No. 1. University of Chicago Press, 1922. 
 
 1 
 
2 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Dura-Salihtyah just as the British were about to begin their retirement down river. 
 After a hasty preliminary inspection for which we ran up by automobile from the 
 British headquarters at Albu Kamal, we found ourselves at Dura with but a single 
 day which we could devote to making our records of the place. Without the protection 
 of the British Indian troops it was not safe to remain a moment longer at Dura, and 
 this first publication of the Oriental Institute therefore represents a single day’s field 
 work of the expedition. 
 
 Notwithstanding the unavoidably hurried character of our records at Dura, the 
 ancient wall paintings which we found there have aroused an interest which has ealled 
 for some report of them. The further fact that the most important of the wall paintings 
 recorded has since been so seriously damaged by the Arabs that our records are now 
 the capital source for knowledge of it, has emphasized the need of a publication of 
 these unique documents for the use of historians, archaeologists, and art students. 
 Their chief importance les in their evident character as cultural links between the 
 Orient and later Europe. ‘The reader who will examine the mosaics from the church 
 of San Vitale at Ravenna (Plate X XII) and compare them with the largest of the Dura 
 wall paintings (Plate VIII), will not long be in doubt that we have in these Dura 
 paintings a part of the heretofore lost oriental ancestry of Byzantine art. 
 
 At the invitation of the Académie des Inscriptions, therefore, the present writer 
 read a communication before its members at the session of July 7, 1922, briefly report- 
 ing on the paintings, and at the same time exhibiting enlarged photographs in colors. 
 The editors of Syria expressed such interest in the discovery that it seemed appropriate 
 to give to them the manuscript of this communication read before the Académie 
 for publication, together with the colored photographs. From the beginning the 
 eminent Belgian scholar Franz Cumont, a foreign member of the Académie, had 
 displayed the greatest interest in the whole question of the Dura documents, which 
 I had previously showed to him on his last visit to America. . With his customary 
 generosity he gave the enterprise unlimited time and placed his encyclopedic knowledge 
 unreservedly at my disposal. He translated into French the text of my communication 
 to the Académie, and added valuable suggestions and observations of hisown. He also 
 wrote a Note additionnelle further discussing the subject, and the whole was published 
 with the color plates in Syria (III, 177-218, and Plates XXXI-L).! For all his 
 kindness and invaluable aid it is a pleasant duty to express here my sincere thanks to 
 my friend, M. Cumont. 
 
 1The color plates were also published in Les Travaux Archéologiques en Syrie de 1920 a 1922 
 (Service des Antiquités et des Beaux-Arts) “Publication faite 41’ occasion de l’Exposition de Mar- 
 seille’”’ (1922), Paul Geuthner, Paris, 1923, accompanied by a discussion furnished by M. Cumont, 
 entitled: ‘‘Les Fresques d’Epoque Romaine Relevées par M. Breasted 4 Es-Salihtye sur l’Euphrate,” 
 pp. 48-54. An account of his first campaign of excavation at the place was also appended by M. 
 
 Cumont, ‘‘Les Fouilles de Salihtyeh sur lEuphrate,”’ ibid., pp. 55-75. These essays of M. Cumont 
 will be referred to herein as Les Travaux Archéologiques. 
 

 
 PREFACE 3 
 
 Interest in the matter did not stop here. I had asked the now lamented Clermont- 
 Ganneau whether he thought it would be possible for us to secure an air photograph 
 of the great Dura fortress by the aid of the French Air Force in Syria. He at once 
 wrote to General Gouraud at Beirfit and raised the question of the possibility 
 of further investigation of the whole ruin at Dura-Salihtyah. At the same time 
 M. Cumont was also very active in promoting interest in the matter, especially in the 
 possibility of excavating and recovering the other paintings of Dura, of which we had 
 seen traces but which we had not been able to excavate, as well as the wall of the tribune 
 which we had not been able to record satisfactorily because of our enforced departure. 
 General Gouraud responded with an offer to send troops to the fortress to aid in 
 excavation and also to protect the archaeologists in their study of the place. My 
 friend Cumont cordially urged me to go and gave me every opportunity to continue 
 the work which our hurried passage had only permitted us to begin. Obligations in 
 Egypt, however, did not permit me to accept the invitation, and M. Cumont, therefore, 
 undertook alone the commission of the Académie des Inscriptions to proceed to Dura 
 under the protection of French troops. With adequate time, and with the French 
 infantry available not only for doing the actual work of excavation, but also for the 
 military protection indispensable in this turbulent region, M. Cumont took up the 
 work at Dura with great success. He cleared a part of the Hellenistic city within the 
 fortifications; he found three more public buildings; he actually found a parchment 
 fragment containing a portion of the Hellenistic laws of the place; he found inscriptions 
 dating the temple paintings with certainty; and he excavated the other painted wall 
 which our hurried departure did not permit us to investigate. On these results of 
 his work at Dura he is to be heartily congratulated. He has published three reports 
 on this work preliminary to fuller publication.! I am indebted to his kindness for 
 sending me in manuscript his report in Syrza in advance of publication, from which I 
 have drawn valuable new facts resulting from his excavations. 
 
 While on his return journey from Dura-Salihiyah, M. Cumont sent me the following 
 letter which is of importance to readers of this book. 
 
 BETWEEN Hama AND Homs 
 23 November, 1922 
 
 My Drar FRIEND: 
 
 At the moment of passing Kadesh it occurs to me to give you a few details of my work at 
 Saélihtyah, whence I am returning. On my arrival there the 7th of November, I found 250 
 soldiers camped in the enclosure which you have described and before my arrival they had 
 already cleared the whole chapel. I found the great sacrificial scene (wall of Bithnanaia) 
 published by you, already terribly injured. The sand with which the Indian soldiers had 
 
 1 Comptes rendus Acad. des Inscr. (1923), pp. 12-41; Syria, IV (1923), 38-58; and Les Travaux 
 Archéologiques (see preceding note), pp. 55-75. 
 
4 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 covered it had not withstood the desert winds and the rains. The officers told me that on 
 their arrival they found all the faces mutilated by the Bedawin and the rest of the scene had 
 faded considerably in two years under the action of the sun and the rains. We found practi- 
 cally intact only the small figures in the lower part of the scene. 
 
 Your plates in color will remain the capital documents for the study of this painting and 
 it is fortunate that you were in a position to make such good reproductions of it before it was 
 destroyed. Of the picture of the Roman tribune, which you photographed in unfavorable 
 circumstances, we were able to obtain better reproductions. 
 
 In addition we were able to study, photograph and copy in color a large fresco, unfortu- 
 nately in rather poor condition, which occupies in the first hall the wall corresponding to 
 that of Bithnanaia (four personages sacrificing). 
 
 We also carefully recorded all the little pictures which you saw in the evening near the 
 tribune. 
 
 There are inscriptions which throw new light on the question of the date of these paintings. 
 One in Greek commemorates either the foundation of the sanctuary itself, or rather that of 
 one of its dependent structures. It is of the year 114 (a.p.) and states expressly that this is 
 a temple of Zeus(-Baal). The other dedication is in Latin and was placed in honor of 
 Alexander Severus by a Palmyrene cohort. 
 
 In the middle of the temple court rose an altar. This court was surrounded by the 
 habitations of the priests and hierodules and was contiguous to a large outer court or square 
 surrounded by a portico. 
 
 It is evident—and this is the chief result of our excavations—that the enceinte of Salihiyah 
 contained a Greek city which can only be Dura-Europos. The city is laid out with streets 
 at right angles (already the “block” system) and several of its buildings have been brought 
 to light: a sort of small theater or audience hall, a sanctuary furnished with steps each of 
 which bears the name ofshim who had the right to occupy it, etc. Near this point two statues 
 wefe brought to light of which one, of marble, is very fine. The earliest date which was 
 recorded is A.D. 31. 
 
 The most interesting of the graffiti contain lists of objects belonging to the temple. They 
 show that along with Zeus the Palmyrene gods Aglibol and Yarhibol were also worshipped. 
 
 Among the various objects found the most unexpected were some fragments of parchment 
 bearing the remains of writing which were exhumed in the course of a tour of the fortifications. 
 
 The necropolis lay outside the city beyond the west wall and contained a large number 
 of burial caves originally surmounted by circular or rectangular superstructures. 
 
 This in a few words constitutes what new facts we have learned about Dura. You can 
 see that our investigations have not rendered yours valueless but that on the contrary they 
 complete the results that you were able to obtain two years ago. I wanted to inform you 
 at once in these few lines of the result of my journey. As soon as I reach Paris I shall make 
 a report of my expedition to the Académie and I shall send it to you as soon as it is printed. 
 I hope to find on my return the number of Syria containing your article and I hope that 
 the color plates may be equal to our hopes. 
 
 Believe me always 
 
 Your cordially devoted friend, 
 
 Franz CuMoNT 
 
 P.S.: The frescos were carefully covered this time under a layer of sand retained by a 
 stone wall. 
 

 
 PREFACE 5 
 
 In a second Jetter M. Cumont adds several more important items: 
 
 Certainly your color plates [of Dura] ought to be republished with a commentary in 
 English. They are the only documents which we have that show the condition of the great 
 fresco at the time of its discovery. 
 
 If you wish to publish my letter I put it entirely at your disposition, but it does not 
 mention some of the most important things. I wrote it on the train before having studied 
 carefully the notes which I had made. I should add that one of the new frescos is signed and 
 that the name of the artist is Semitic and not Greek. His name is [lasamsos, ‘‘the Sun is 
 god.” There are also other points which you will perhaps wish to mention in the English 
 article. For example Clermont-Ganneau has expressed the opinion that in the frescos of the 
 tribune the three statues are not those of emperors but of the three Palmyrene gods, Baal, 
 Aglibol, and Yarhibol in war panoply, and I am inclined to think that he is right. 
 
 It will be seen, therefore, that the lamentable destruction of the great painting 
 (wall of Bithnanaia, Plate VIII) has made it impossible to secure any fuller or more 
 accurate records of it than we were able to make in the single day which the impending 
 British evacuation permitted us to make. Following M. Cumont’s recommendation, 
 therefore, the Oriental Institute is issuing as full a publication as possible of the work 
 it was able to do at Dura, and adding also the new data resulting from M. Cumont’s 
 excavations. Students of art and history will find in these documents a new vista 
 leading back from Byzantine art to an earlier oriental background. It reveals at the 
 same time with increased clearness the position of the Roman Empire on the Euphrates 
 both before and after the advance of Trajan, and especially the ceaseless interpenetra- 
 tion of Orient and Occident especially evident after the campaigns of Alexander the 
 Great, although evident to the orientalist as far back at least as the dawn of the Age 
 of Metal in Europe (about 3000 B.c.), and with hardly a doubt even much earlier. 
 These unique documents, therefore, illustrate very clearly the process which it is the 
 duty and function of the Institute to investigate and as far as possible to recover. 
 
 It would seem appropriate, therefore, that this first volume of the Oriental Institute 
 Publications should offer in this connection a brief statement concerning the origin 
 and purpose of the Institute itself. Founded by the generosity of Mr. John D. 
 Rockefeller, Jr., and the enlightened co-operation of the Trustees of the University 
 of Chicago, the Oriental Institute is a laboratory for historical research. It is intended 
 to furnish a means of organizing the members of the teaching staff of the Department 
 of Oriental Languages, together with members of the Institute from the outside, into 
 a body of investigators and at the same time to make accessible to them all the original 
 documents which can be made available, either by purchase, by copying, or by any 
 feasible method of collecting and gathering into the archives of the Institute the 
 surviving records of the ancient Orient. The region to which the Oriental Institute 
 purposes to devote its chief attention is commonly called the Near East, by which we 
 mean the eastern Mediterranean world and the adjacent regions eastward, at least 
 through Persia. It is now quite evident that civilization arose in this region and passed 
 
6 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 thence to Europe. In the broadest general terms, therefore, the task of the Oriental 
 Institute is the study of the origins of civilization, the history of the earliest civilized 
 societies, the transition of civilization to Europe, and the relations of the Orient to 
 the great civilizations of Europe after the cultural leadership of the world had passed 
 from the Orient to European peoples. 
 
 In illustration of the foregoing program, I may be permitted to mention some of 
 the enterprises which the Institute thus far has undertaken. In the winter of 1919-20 
 the Institute undertook a preliminary inspection of the accessible regions of the Near 
 Kast, for the collection of original monuments and documents and the study of the 
 monuments in situ. The journey included the voyage of the Nile as far as Thebes, 
 and in Western Asia a reconnaissance of the Babylonian Plain, the upper Tigris as 
 far as the mountains north of Mosul, where the party was stopped by the hostile 
 Kurds, the return to Baghdad, the ascent of the Euphrates and the return to the 
 Mediterranean via Aleppo, the coast of Phoenicia, and parts of Palestine. On this 
 journey a considerable body of valuable original sources, both written and unwritten, 
 was acquired by purchase. 
 
 The study of the documents of Western Asia is in pressing need of an exhaustive 
 Babylonian and Assyrian dictionary. With a resident staff of five people and with 
 the co-operation of a number of outside scholars the Institute has been engaged for 
 over two years in the compilation of this much needed dictionary, which is progressing 
 at the rate of nearly 100,000 cards a year. At present the files contain about 300,000 
 cards. It is hoped that the dictionary may be completed within eight years more. 
 
 “At the same time the administrative staff of the Institute has been at work endeav- 
 oring to organize the existent sources of knowledge regarding the ancient history of the 
 Near East. These materials take the form of a subject catalogue alphabetically 
 arranged on cards, and filed in drawers like a library catalogue. This project, there- 
 fore, aims at an encyclopedic organization of subjects which if completed would 
 enable the investigator to turn to any important subject in the range of Near Eastern 
 civilization and history, and to find collected under that subject all the material throw- 
 ing light upon it, whether in modern books and treatises or in ancient original docu- 
 ments and monuments. That such completeness never will be wholly attainable is 
 obvious, but even in an incomplete stage such an organization of materials is indis- 
 pensable to the purposes of the Institute. These archives now contain about 35,000 
 cards. In view of our lack of a sufficient Egyptian dictionary, one of the specialized 
 sections of these archives is devoted to a collection of all Egyptian words which have 
 anywhere been especially discussed by particular scholars in oriental journals and 
 treatises. 
 
 Among the great bodies of original documents surviving from the ancient Orient 
 none is more important or difficult than the religious compositions. The study of 
 such documents is still in its infancy. Our understanding of the Egyptian Book of 
 

 
 PREFACE if 
 
 the Dead is conditioned by our knowledge of the older Egyptian literature out of 
 which the Book of the Dead was built up. The current translations of the Book of 
 the Dead are quite worthless for this reason. It is necessary, therefore, to collect 
 and study the religious documents which were the predecessors of the Book of the 
 Dead. Of the Pyramid Texts, the archaic religious documents which are written in 
 the chambers and passages of the Fifth and Sixth Dynasty pyramids, and which contain 
 the oldest religious literature available in any language, we have had an adequate 
 edition since 1910. On the other hand the Coffin Texts, the extraordinary mortuary 
 literature inscribed in ink on the insides of the Middle Kingdom wooden coffins 
 dating from the half millennium of which the middle falls at about 2000 B.c., has 
 been but little collected and studied. The Institute has, therefore, undertaken the 
 collection, editing, and publication of all these materials. The work began in the 
 winter of 1922-23 in the great collection of the Cairo Museum. The task of making 
 a complete photographic record of the Cairo coffins is nearly finished and a considerable 
 beginning on the accompanying hand copies has been made. The European and 
 American museums will be included later, until all the known Coffin Texts have 
 been incorporated in the Institute records. Out of this complete body of copies 
 a standard edition of the Coffin Texts will be made, on the basis of which an 
 understanding of the Book of the Dead will be possible. This will, of course, be a 
 work of years. 
 
 Another important class of documents consists of those which embody proverbial 
 wisdom in animal stories. Probably few readers of the delectable Uncle Remus 
 tales have realized that these seemingly American stories found their way into the 
 cabins of our southern negroes from the slave markets of eastern Africa, whither 
 they had wandered from a remoter and an earlier Orient. Besides the East Indian 
 sources, the oriental originals are at present chiefly Arabic and Syriac manuscripts; but 
 such tales circulated as early as the Assyrian Empire in Western Asia, and fragments 
 of them are still found in cuneiform tablets, while in their oldest known form they have 
 survived in delightful sketches on papyrus by Egyptian artists who cleverly depict 
 human relations in the animal world. These amusing caricatures probably go back 
 to the fourteenth or fifteenth century B.c. Besides the line of descent to us through 
 the African slave markets, there is another through England, where the oriental animal 
 tales were translated by Sir Thomas North from Spanish and Italian versions in 1570. 
 The English translator is doubtless more familiar to most English or American readers 
 as the author of the version of Plutarch’s lives used by Shakespeare. This literature 
 of animal tales from the Orient has been translated into more languages than any other 
 book except the Bible and is a striking illustration of how culture influences have 
 passed from the Orient to the West. The Institute is now having photographed all 
 the known oriental manuscripts, chiefly Arabic, in which these Tales of Kalila and 
 Dimna, as they are called, have survived to us. 
 
8 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Parallel with this project in the field of literature the Institute is also at work in 
 the realm of the history of natural science. The extraordinary ancient Egyptian 
 surgical and medical treatise, now known as the Edwin Smith Papyrus, in the collec- 
 tions of the New York Historical Society, is really the earliest surviving scientific 
 document known to us. Its content has proved to be an epoch-making revelation, 
 and the Institute is preparing the translation and final scientific publication for the 
 New York Historical Society. A fuller account of the plans and activities of the 
 Institute and the scope of its work will be found in its first Communication’ issued 
 last year. 
 
 The researches and discoveries recorded in this volume were made possible solely 
 by the generous support of the Oriental Institute by Mr. John D. Rockefeller, Jr. 
 The second volume of this series, entitled The Annals of Sennacherib, by Professor 
 D. D. Luckenbill, is already on the press. It will contain all Sennacherib’s records, 
 including a full publication of the superb prism obtained by our expedition of 
 1919-20. It is an agreeable privilege to express here, on behalf of the members 
 of the Institute, our sincere appreciation of the liberal contributions which have 
 enabled us to begin this series of volumes as a tangible evidence of the debt of 
 science to the founder of the Oriental Institute. 
 
 It is a matter of gratification to its members that the first volume of a series to be 
 known as the Oriental Institute Publications? should contain these materials which so 
 unequivocally illustrate the process of culture transition from the Orient to Europe. 
 The recognition of this transition is daily revealing to modern men that there is no 
 sharp cleavage between the Near Orient and Europe. The successive rise of both 
 from prehistoric savagery has been one evolutionary process, the recovery of which will 
 enable us to write the coherent and unified story of mankind. Writing these words, 
 as I do, overlooking the hills of Tuscany but a few hours away from Ravenna, and 
 contemplating the roofs and towers of Florence spread out below this historic villa 
 where it is believed by many that Boccaccio found the scene of his immortal 
 Decamerone tales, it seems peculiarly fitting that these prefatory words should be 
 penned in the midst of surroundings which reveal at every turn how great a part 
 in the revival of. European culture was played by the men of Tuscany. For the art 
 which developed so richly here in Florence, and especially the art of painting, was 
 based to no small extent upon that of Byzantium, of which we must evidently recognize 
 
 the ancestry in the wall paintings of Dura. 
 JAMES HENRY BREASTED 
 VILLA PALMIERI 
 FLORENCE, ITALY 
 June 1, 1923 
 
 1 Oriental Institute Communications No. 1 (to be abbreviated as O.I.C. 1). University of Chicago 
 Press, Chicago, 1922. 
 
 2 This series may be referred to as O.J,P. I, ete. 
 

 
 
 
 TABLE OF CONTENTS 
 
 ‘ PAGE 
 INTRODUCTION BY FRANZ CUMONT. ; : ‘ 5 , f i : ‘ 15 
 I. Tae History or Dura-SAuiniyag . : é ; tpl 
 II. Toe Discovery or THE DurA-SALIniyAH PAINTINGS é : : Ay 
 Ill. Tae Crry anp Fortress or Dura . ; : ? ; : : ' 62 
 IV. Tue Tempe or Zeus-Baat 8N THE Fortress oF Dura . : : : 2 68 
 V. THe WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE . j f ; : 15 
 Hau II, Sout Watt, C-D (Plates VIIJ-XIX) 4 ‘ : LO 
 
 Hatt II, West Wat1, D-E (Plate XX, 1) : ‘ ; : . 88 
 
 Hatt II, East Watt, B-C (Plate VIII) . ; " 2 ORs ev: : 24-89 
 
 Hauu I, NortH Watt (Plate X XT) : : P . . 94 
 
le 5r 
 
 * aie ie ° eos 
 
 ; ae hice ia ese + a ar) 
 
 he 
 
 / ry 2 rs A 
 “) Va en As | 
 eh aa Psi 
 
 DA 
 = 
 ’ 
 
 . * eae histeah 
 ad ; -_ pede ith ae hee 
 
 cz " 7 Ay 4 
 To me a sf! ie) 
 tenes Hane Nott eh 
 
 ne 
 
 hi bs 
 
 ¥ 
 Mu a 
 “ay 
 
 Batis Ae mat eat oh 
 
 : | ae 
 G f Sr wy BAe ne ker an a msi ‘a i ee. ty, 
 a 
 
 iH me a it a ‘ee ‘ | 
 ving " se 
 
 : Ae 
 
 aaa e ee * ies 
 oie mints ca ht ant i 
 Pa Khel a <—o 
 a) Cae ae” ~ A, 
 - 3 ne y 
 ee 
 Noe ' ond ‘ Ki 
 ale ¥, bai, TET hs ie av 
 7 ie) cine, ', ioe 
 ip mie 
 
 7! oo y 
 — ah’ 
 
 pr igrs Hale. js 
 Py pir, 
 Pay OMS j ie) 
 
 ‘Th: = * ara 
 
 Fm! 
 ; : c ob iis - A " a? 
 tae ae Mies 
 | Bae 
 NY Tete 
 
 rt) : 7 sa 
 
 Wy Mey 
 ina ictal oe v4 
 oon) t 
 
 (yu é ae 
 
 f #) a ¥ Meat 
 
 sa if i ly ir 
 ‘tein gee Ae 
 . a is eee 
 
 
 

 
 
 
 LIST OF ILLUSTRATIONS 
 PLATES 
 
 PLATE 
 I. Fortress oF SALInfyaH 
 
 1. Looking Southwest in Northwest Wadi 
 2. Gate near Middle of Southwest Wall 
 II. Fortress or SAuIniyAn 
 1. Looking Northwest along Southwest Wall 
 2. View along Inside of Same Wall 
 III. Fortress or SALmiyan 
 1. Inside View Looking East along Massive Castellum and North Wall 
 2. Northwest Angle of Castellum 
 IV. Fortress oF SAuIniyAn 
 1. Outside View of North End of Castellum, Looking East 
 2. Northwest Bastion of the Castellum, Looking Northeast to the Euphrates 
 V. Fortress or SALIHTYAH 
 1. Cross-Section of Wall, Showing Evidences of Two Structural Stages 
 2. Outside View of Extreme West Angle Containing Chapel, Looking Southeast 
 VI. Fortress or SALIBTYAH 
 
 1. Outside View of Extreme West Angle Containing Chapel, Looking West 
 2. Shrine in Hall II, Looking Southwest 
 
 VII. Caape. or SAuIniyAH 
 1. The Two Columns at West Side of Court 
 2. Steps Leading into North End of Hall II 
 VIII. Tue WatL or Bironanata IN Hatt II 
 GENERAL VIEW (Color Plate) 
 IX. Tae WALL or BiTHnaAnatrA IN HAtt II 
 Tue THREE MInistrants (Color Plate) 
 
 X. THE WAtL or BITHNANAIA IN Hatt II 
 HEAD OF THE SECOND Figure (Color Plate, frontispiece) 
 
 XI. Tar WALL or BITHNANAIA IN Haut II 
 BITHNANAIA AND Her Group (Color Plate) 
 
 XII. Tae Watt or Birananata In Hatt II 
 HAD oF THE SEcoND Figure, SHowine INsJury To Lerr Eye (RESTORED IN PLATE X) 
 
 XIII. Tae Wat or Birxananata In Hatt II 
 Tur THREE MINISTRANTS 
 11 
 
XVII. 
 XVIII. 
 XIX. 
 
 XX, 
 
 XXI. 
 Xoo 
 
 XXIII. 
 
 MAPS 
 
 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 . THe WALL oF BITHNANAIA IN Hatt IT 
 
 BITHNANAIA AND HER Group 
 
 . Tue WALL oF BITHNANAIA IN Haut II 
 
 Upper Part or FIGURE OF BITHNANAIA 
 
 . Tae WALL or BITHNANAIA IN Hatt II 
 
 Rieut Enp: THE Four MEN witH GREEN BRANCHES 
 
 Tue WALL or BITHNANAIA IN Hatt II 
 Nints FiGure, THE First Boy In THE FOREGROUND 
 
 THe WALL or Birananala IN Hatt IT 
 Trento FiGurE, THE GIRL IN THE FOREGROUND 
 
 Tae WALL or BiTHnanalA IN Hatt II 
 ELEVENTH FIGURE, THE SECOND Boy IN THE FOREGROUND 
 
 CHAPEL OF SALIH{YAH 
 
 1. West Wall of Hall II with Desert West of Fortress on Right 
 2. Partly Excavated North Wall of Court, Looking Northeast to the Plain of Khana- 
 Mari 
 
 Tuer WALL OF THE TRIBUNE 
 
 Srxtu-Century Mosaics 1N THE BasILica oF 8. VITALE AT RAVENNA 
 
 1. Emperor Justinian with Bishop Maximian and Suite 
 2. Empress Theodora and Suite 
 
 SYRIAN RELIEFS 
 
 1. Priest Offering Sacrifice to the God Bel 
 2. Mortuary Relief from Palmyra 
 
 MAPS 
 
 PAGE 
 
 1. WesteRN, AsIA, SHOWING THE ROUTE OF THE ORIENTAL INSTITUTE EXPEDITION OF 
 1919-20 AND THE SITUATION OF DURA AND THE PLAIN oF KHana-Mari following 20 
 
 2. THe Puain oF KHAna-Mari . : 40 2 eid aa, ae following 24 
 
 oo :SSr 
 
LIST OF ILLUSTRATIONS 13 
 
 TEXT FIGURES 
 
 PAGE 
 
 1. TyprcaL View or EuPHRATES VALLEY IMMEDIATELY ABOVE ‘SANAH ’ 4 hienoe 
 2. TypicAL WALL OF THE DESERT PLATEAU ALONG THE EUPHRATES . : “ieee 
 3. TypicAL Desert Vauipy ‘Want’ on Ricut BANK oF EUPHRATES BELOW DEIR EZ- 
 DOR, (2 ; : : y : : : : ; ; ; : ; ; ; . 24 
 4. Mopern Hit, tHe Cuassicau Is, tHe ANCIENT SOURCE OF BiruMEN . ....._—_ 25 
 5. THE BriruMEN Basin aT HIT. ; 3 , : : ‘ ; : ; : ; A p45 
 6. A Brrumen WELL aT Hit. : ; ; ' AOE 
 7. Lookine up THE EuPpHRATES AT HApDiTHAH ‘ é ees 
 8. Scanty FIELDS ALONG THE EUPHRATES Two MILes aBove Hapiruan. : . ae 
 9. IRRIGATION WATER WHEEL IN THE HUPHRATES AT HApDiTHAH : , . - 80 
 10. THE VALLEY OF THE EUPHRATES AT ‘ANAH tly 2h ; d j [ JMS 
 11. ISLAND CULTIVATION IN THE EUPHRATES AT SANAH . ; ; a Oe, 
 12. TAMARISK THICKETS ALONG THE HUPHRATES AT EL-KAIm ; , Wh eda 
 13. THE EUPHRATES AT TIBNI . : ; ‘ ; , ‘ : 7 $00 
 14. Toe EvpHRATES VALLEY AT HALABIYAH  . : . ; y : Ae) ook 
 15. UNCULTIVABLE TAMARISK THICKETS ALONG THE EUPHRATES AT SABKHAH . ee SG 
 16. Tor EvpHRatTEs AT MESKENAH . ; : : : : ; : ; : od 
 17. Tot EUPHRATES AND THE PLAIN OF DEIR Bz-ZOR_ ; we oS 
 18. THe PLAIN AND City oF DEIR 52-Z6rR. i ; ! : ; ( 39 
 19. SHEEP FEEDING ON THE UNCULTIVATED TRACTS OF THE DEIR EZ-ZOR PLAIN . 40 
 20. A Woot CARAVAN IN THE STREETS OF DEIR Ez-ZOR. : ; ; } : . Al 
 21. Across THE Roors oF DEIR Ez-ZOR : : ; , : ’ : ye eZ 
 22. THe BrinGce at Derr £z-Zor. ; : ; ; ; , eas 
 23. A CoRNER OF DEIR EZ-ZOR AND THE GARDENS ON THE ISLAND. ; . 44 
 24. AraB Boys or DEIR 5z-ZOR . ; é ? ; : : : . : 2° +45 
 25. Mpyapin AND THE PLAIN oF KHANA-MaARI ! : : 4 ; ’ . 46 
 26. THE ARID PLAIN or KHANA-MARI AND THE PLATEAU Horizon : ; . AT 
 
 27. Tar EUPHRATES AND THE PLAIN oF KHANA-MARI FROM THE PLATEAU AT SALINiYAH 48 
 
 28. GLIMPSE OF THE BAZAARS IN BAGHDAD : ; ; { ; : ; 5 : 49 
 29. Tur Brings or Boats OVER THE TIGRIS AND PALM GROVES AT BAGHDAD . ; 50 
 30. British GUNBOAT ON THE LOWER EUPHRATES AT FALLUJAH . p ’ : : 54 
 
 31. Tur SEvEN Cars oF OuR CARAVAN BETWEEN FALLOJAH AND RAMADI ON THE LOWER 
 EUPHRATES : : : s : : : : 3 : ; ¥ 5 ‘ . ; 55 
 
 32. BEGINNING OF THE PLATEAU ABOVE RAMADI AND BELOW Hirt. . : : ; 56 
 
14 
 
 53. 
 54. 
 
 55. 
 56. 
 57. 
 58. 
 
 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Our Cars STALLED NEAR HapiTHAH IN A DESOLATE STRETCH BETWEEN HIT AND 
 CANAH. 
 
 . Our British Convoy Hattine at NAHTYAH JUST ABOVE ‘ANAH 
 . CAMELS OF THE BritisH Miuirary TRANSPORT BELOW ALBU KAMAL 
 
 . British HEADQUARTERS AT ALBU KAMAL 
 
 STARTING FOR AN AIRPLANE HXAMINATION OF THE KHANA-MARI PLAIN 
 
 Our Wacons DRAWN UP FOR DEPARTURE FROM THE FORTRESS OF DURA 
 
 . OuR Wacons LoApING FoR DEPARTURE FROM Mzyapin, AFTER THE First Day’s 
 
 Marcu 
 
 . SHEIKH RAMADAN-Bue [pn SHALLASH AND A Group oF His TRIBESMEN OF THE ALBU 
 
 SARAI ABOVE DEIR Ez-ZOR 
 
 . A Group or ARAB RIFLEMEN OF SHEIKH RamMADAN-Bgc IBN SHALLASH 
 
 Tue Heap or OuR CARAVAN DESCENDING FROM THE PLATEAU TO THE HUPHRATES 
 NEAR TIBNI 
 
 . OuR WAGONS IN THE KHAN AT SABKHAH 
 
 . SHEIKH SuwAN OF THE SABKHAH ARABS ABOVE DEIR §Ez-ZOR ON THE UPPER 
 
 EUPHRATES 
 
 . Our Horses BEING WATERED IN THE EUPHRATES FOR THE Last TIME, ABOVE M&s- 
 
 KENAH, Two Days’ JoURNEY FROM ALEPPO 
 
 . MopERN ALEPPO AND THE MounpD oF THE ANCIENT CITY 
 
 . THE Court oF A Mosque IN ALEPPO . 
 
 Tur VALLEY OF THE ORONTES IN NorTH SyRIA. 
 
 Tur WELCOME MEDITERRANEAN SHORES Nort oF BEIROT . 
 
 . Recorps oF MopERN CONQUEST AMONG THE MEMORIALS OF ANCIENT CONQUERORS 
 
 ON THE MEDITERRANEAN CLIFFS AT THE DoG River Norts or Berrtr 
 
 GENERAL CUNNINGHAM WITH OFFICERS AND Part oF OuR EXPEDITION AT THE SOUTH- 
 wEst GATE oF THE DurRA FORTRESS 
 
 Our CAMP IN THE ANCIENT FortTRESS OF DuRA 
 Our First GLIMPSE OF THE PAINTINGS 
 
 British Hast Inp1an TRoops at Our DISPOSITION FOR HWXCAVATING THE GROUND 
 PLAN OF THE TEMPLE IN THE ANCIENT FortTRESS OF DuRA 
 
 AIRPLANE VIEW OF THE ANCIENT Fortress or DuRA-SALIHfYAH . 
 PLAN OF THE ZEUS-BAAL TEMPLE IN THE FortTRESS OF DURA 
 CoLuarR, NECKLACE, AND PECTORAL ORNAMENTS OF THE Lapy BITHNANAIA 
 
 INSCRIPTIONS ON THE WALL oF BITHNANAIA 
 
 PAGE 
 
 101 
 

 
 INTRODUCTION 
 By FRANZ CUMONT 
 
 The discovery of which Mr. Breasted gives an account in the following pages is 
 one whose importance can hardly be exaggerated. It throws a vivid light upon 
 numerous questions of which science still awaits the solution. It raises also new and 
 perplexing questions, and it may be predicted that it will provoke abundant com- 
 mentary. In these preliminary remarks, I would like simply to indicate briefly in 
 what particulars it especially deserves our attention. 
 
 In the first place, this discovery is of great interest for our knowledge of the 
 political history of ancient Syria. It furnishes us with authentic proof that the 
 Romans established a military stronghold in a region which we have not heretofore 
 known to have been permanently occupied by the legions. A massive fortress com- 
 manded the passage of the Euphrates at the point where the route from Palmyra 
 crossed the river (see p. 26). This post was the most advanced of all those which 
 are known along its banks. To what epoch must we carry back the establishment of 
 a Roman garrison at Sdlihiyah? It is certainly not to be dated before the conquests 
 of Trajan. Before the reign of this emperor neither Palmyra nor Damascus nor 
 Petra were subject to the imperial legates, and the desert of Syria was left outside the 
 limits of their jurisdiction. But after the reduction of Mesopotamia to the condition 
 of a province (114-16), Rome necessarily experienced the need of holding a strategic 
 position which up to that time had been without value to her. The troops stationed 
 there were able to assure the communications across the desert with the newly annexed 
 territory, and to guard the transports descending the Euphrates against the enterprises 
 of pillaging Arabs. Nevertheless, the occupation of Mesopotamia by Trajan was 
 only ephemeral. Hadrian abandoned it immediately after the death of his predecessor 
 and we do not know whether he constructed a fortress here, as he did elsewhere, for 
 guarding the frontier and observing the hostile country on the other shore. It is 
 therefore possible that the fortress of Sdlihtyah is subsequent to the expedition of 
 Lucius Verus against the Parthians (162-65), or even to that of Septimius Severus 
 (197-99).! It was only then that a large portion of Mesopotamia became definitely 
 Roman. 
 
 There is the same uncertainty as to the duration of the occupation. While 
 awaiting new excavations we can do no more than suggest hypotheses as to the circum- 
 stances by which it was terminated. But we know that Dura was desert in the time 
 
 1 This was written before the date of the paintings was established. Their date has important 
 bearing on the date of the fortress. See pp. 88 and 64-65.—J. H. B. 
 
 15 
 
16 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 of Constantine (p. 48). It is probable that it had been abandoned when Diocletian 
 fixed the frontier of the Empire at the course of the Khabtr and fortified Circesium 
 at its mouth. It was at this time, one might suppose, that the legionaries evacuated 
 their camp in the forsaken town.! None of the places mentioned in the Notitia 
 Dignitatum (Or. xxxii) seems to be identifiable with Salihtyah. 
 
 The paintings of Salihtyah have even greater value for the history of religion 
 than has the castle for political history. The scene (Plate VIII), in which one sees 
 the family of Konon assisting at a double sacrifice offered by two Syrian priests, is 
 unique in its kind and all its details will deserve to be studied. I confine myself 
 here to a few rapid remarks. 
 
 The two celebrants are clad in white robes and wear conical caps of the same color. 
 This headdress, which seems to be of Hittite origin, and the long robe gathered at the 
 waist, appear like those of the king, Abd-Hadad, pictured sacrificing as far back as 
 the fourth century before our era on the coins of Hierapolis, and Lucian informs us 
 that in his time the priests of that city? were clad in a garment all of white, and that 
 they wore on the head the z7?dos,’ that is to say, a felt cap of conical shape. In a 
 bas-relief from Ciliza on the north of Aleppo, the priest Gaios appears after the same 
 fashion making an offering to the god Bel for himself and his family, exactly as at 
 Sdlihtyah. The tall white cap has, furthermore, remained to the present day as 
 part of the costume of the dervishes. 
 
 In the same way the nakedness of the feet, which distinguishes the two officiants 
 from the mere assistants, reminds us of the Old Testament and of Islam. ‘‘Take the 
 shoes from off thy feet, for the place where thou standest is holy ground,” commands 
 the voice which speaks to Moses from the burning bush,® and the interpreters of this 
 verse see in it the origin of the obligation imposed on the priests, of fulfilling the cere- 
 monies of the cult and of offering sacrifice with bare feet.6 In reality the custom of 
 removing the shoes before entering the temple is derived from the practice still common 
 in the Orient, of removing them at the door of the house in order to avoid defiling the 
 interior with mud and filth, and the Jewish proscription of shoes is likewise found in 
 many cults of antiquity.’ It is common knowledge that even today the Moslems 
 still demand this mark of respect of all who cross the threshold of a mosque. 
 
 1 This conjecture has been confirmed by Cumont’s further excavations.—J. H. B. 
 
 2 Babelon, Monnaies de la Bibliotheque nationale. Perses Achéménides, 1893, Plate LIII; Dussaud, 
 Notes de mythol. syrienne, p. 97. Cf. my Htudes syriennes, 1917, p. 261. 
 
 § Lucian, De Dea Syriae, ¢. 42: ’EoOis 5¢ abréowst raca \evk} Kal wtdov emi rH Kepadf Exovow. 
 
 4 tudes syriennes, p. 257. 
 
 5 Exod. 3:5; ef. Vigouroux, Dict. de la Bible, s.v. ‘“Chaussures,”’ p. 634. 
 
 6 Theodoret, Ad Ex. quaest., 7 (Patrologia graeca, LX XX, p. 231): Tupvots tepets root ras Necroupylas 
 éreréedouv Kal Tas Bvaias. 
 
 7 Notably in that of the Great Mother (Prudentius, Peristeph., X, 154 ff.). Cf. Graillot, Culte de 
 Cybele, 1912, p. 139, n. 3. Some other examples are cited by Gruppe, Griechische Mythologie, p. 912. 
 
INTRODUCTION 17 
 
 We are reminded of still another ceremony of Judaism by the branch or palm 
 carried in the left hand by the four men, while they raise the right hand as a sign of 
 adoration.! It naturally suggests comparison with the lulab,? a bouquet composed 
 of a palm and twigs of myrtle and willow carried by the Jews at the Feast of 
 Tabernacles, which was an ancient agricultural feast of the vintage. At the present 
 day the branches of myrtle and willow are still distributed to the assistants at the 
 feasts of the Nosairis, and an analogous custom exists among the ?Ahle Haqq of 
 Kurdistan.’ 
 
 The ceremony depicted on the wall of the chapel (Plate VIII) is exceedingly 
 curious. At one side a priest who holds a ewer plunges a plant into a vase filled with 
 water. This water is probably that of the Euphrates,‘ a sacred river already invoked 
 in the ancient litanies at Babylon, whose waters were prescribed in preference to all 
 others and were as holy as those of the Nile in Egypt. Beside this priest, a second 
 makes an offering on an altar, whereon a fire is burning. We find here united the 
 cults of the two opposed elements; for the adoration of the elements, and in particular 
 that of water and fire, is characteristic of the religion of Mesopotamia in the Roman 
 period as it was practiced by the ‘‘Chaldeans” or the Magi, and it is this worship 
 which the polemic of the Christian apologists especially attacks.® 
 
 It is surprising to find this ceremony, wholly oriental—as is also the plan of the 
 temple according to the remark of Mr. Breasted—close by the other, wholly Roman, 
 of the tribune offering sacrifice to the emperors® before his legionaries and in the 
 presence of the flag—a picture which calls up before our eyes with a vivacity singularly 
 striking a scene from the life of the Roman military camps such as learned research 
 enabled us to reconstruct in thought. How could these two cults fundamentally 
 so different coexist in the same chapel?’ The explanation which first suggests itself 
 to one’s mind is that the garrison of SAlihtyah was made up of legionaries and a nwmerus 
 
 1 This gesture, of which Mr. Breasted has given the significance (p. 85), is found also in Syria, 
 for example on one of the stelae of Nerab (at the Museum of the Louvre). 
 
 2 Cf. Jewish Encyclopedia, s.v. ‘‘Lulab.” 
 
 3 Dussaud, Les Nosaires, p. 89 ff.; Minorsky, Notes sur la secte des >Ahlé-Haqq, p.93. Paris, 1922. 
 
 4 On the cult of the Euphrates in the Roman epoch and the use of its waters, cf. Etudes syriennes, 
 pp. 251 ff. 
 
 5 Cf. my Textes et monuments figurés relatifs aux mystéres de Mithra, I, 105 (n. 1), 108. The 
 mysteries of Mithra drew this cult of the elements from the religion of Babylonia. 
 
 6 As M. Cumont in agreement with Clermont-Ganneau now holds, the statues depicted in the 
 paintings are militarized Syrian gods and not emperors, a conclusion with which I agree.—J. H. B. 
 
 7 This is true of an epoch well anterior to that of Odenathus and Zenobia. M. Réné Dussaud 
 calls my attention to an inscription of Palmyra (Répertoire d’épigr. sémit., I, 285) which is a dedication 
 made in 132 by a Nabatean cavalryman in the camp of <Anah. This camp is situated on the Euphrates, 
 down-river from Dura, and was undoubtedly abandoned by the Romans along with the rest of Mesopo- 
 tamia in 117. The Nabatean cavalryman must have been in the service of Palmyra, which was 
 holding <Anah on the caravan route. 
 
18 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Palmyrenorum, and that the former worshiped the deified Caesars according to Roman 
 rite, and the latter according to Syrian rite. But the ceremony portrayed on the 
 wall of Bithnanaia contains nothing of a military nature. The head of the family 
 causes it to be celebrated in the presence of his descendants. His daughter and 
 granddaughter assist in it, as do his sons and grandsons. It is a family sacrifice on 
 the part of some powerful house, not an official sacrifice on the part of migratory 
 soldiers in the service of the Empire. We are also led to suppose that the Romans 
 took possession of a native sanctuary in existence prior to their arrival, and that they 
 set up therein the cult of the Emperor by building there the isolated shrine which must 
 have contained the statue of the prince and the standards of the garrison (pp. 70-71). 
 But then we are not obliged to assign the same date to the two compositions, one of 
 which adorns the first and the other the second room, and the Roman decoration of 
 the latter would be later than that at the back of the chapel. If one might be allowed 
 to hazard a guess in a question still buried in such obscurity, one would be tempted 
 to assign the first painting to the epoch of the autonomy of Palmyra when her authority 
 extended as far as the Euphrates. Konon must have been some chief of a tribe in 
 subjection to the great city. The second painting would, on the contrary, date from 
 the time when the Romans were occupying Saélihtyah and commanded the respect of 
 the desert Arabs. We are thus again confronted by the question of chronology upon 
 which we touched at the outset.! 
 
 In spite of all thejr historical and religious interest, the surprisimg works which 
 Mr. Breasted has analyzed with such minute precision would seem above all of especial 
 value for the history of art. Though we know Alexandrine painting relatively well, by 
 means of the Pompeian frescoes, which sprang directly out of them, and though Egypt 
 has furnished us by means of her mummies a considerable number of portraits, never- 
 theless up to the present time we have had only a very vague idea of what Graeco- 
 Syrian painting might have been like. We are hardly able to study it in more than 
 the decoration of a few tombs, of which the most remarkable is the one which 
 Dr. Sobernheim has photographed at Palmyra.? The paintings recovered by M. Alois 
 Musil in the castle of ‘Amra, in Arabia Petraea,’ belong to the epoch of the Omayyad 
 Caliphs, and they give us but a poor idea of what must have been those of antiquity. 
 For the first time we find here at Dura large compositions enabling us to grasp the 
 process employed by Graeco-Syrian artists in drawing and grouping their figures. 
 
 1 The inscriptions since found by Cumont place the two paintings here discussed at two widely 
 separated dates: the ceremony of Plate VIII in the first century, and that of the tribune in the early 
 third century.—J. H. B. 
 
 2Last published in Djemal Pasha’s collection, Alte Denkmdler aus Syrien, Paldstina und 
 Westarabien, Plate 68. Berlin, 1918. Cf. Etudes syriennes. 
 
 3 Alois Musil, Kusejr <Amra (published by the Academy of Vienna), 1907. The Dominicans of 
 Jerusalem will give us shortly some new reproductions of these curious paintings. 
 
 
 

 
 INTRODUCTION 19 
 
 In the detail of their clothing and their ornaments, these figures display close 
 relationships with those of the Palmyrene sculpture. Such a woman as the one whose 
 image has been preserved on a sepulchral stone seems to be a twin sister of Bithnanaia, 
 her head dressed with a similar embroidered toque, her person enveloped by the same 
 veil and laden with the same jewelry (Pl. XXIII, 2).!_ But the paintings of Sdlihtyah 
 in the matter of technique are quite superior to the ordinary products of Palmyrene 
 funereal sculpture. Once having cast one’s eyes upon him, one will not soon forget the 
 serious expression, the look of ecstasy of the sacrificing priest, and his bronzed face in 
 which the ethnical characteristics are drawn with a touch so sure that one might be 
 looking upon the portrait of some Beduin chieftain encountered in the bazaar at 
 Damascus. 
 
 Mr. Breasted has already noted the resemblance of these pictorial works to the 
 mosaics at Ravenna. Figures seen from the front, ranged in one line, their feet 
 resting so insecurely on the ground as to give the impression of floating in the air, 
 all these special points are common to certain Byzantine compositions and to our 
 paintings. We may be certain that these documents of capital importance which 
 have just been furnished to us will henceforth be constantly called upon whenever 
 the complex question of the oriental origins of the art of the Middle Ages is under 
 consideration. 
 
 One last question presents itself. Why was this sumptuous decoration executed 
 in the modest chapel of a fort? 
 
 Should one believe that the custom of covering walls with frescos was, in Syria, 
 so widespread that even in a house hidden away in the desert there was the desire to 
 embellish its walls with such ornamentation? Or did some special reason here really 
 justify an unwonted luxury? M. Clermont-Ganneau, who has offered us in the study 
 of this enigmatical problem the benefit of his ever ingenious learning, has at this 
 point advanced a suggestion which deserves being taken into serious consideration. 
 Near Dura, which, as has been noted, is situated on the Euphrates opposite Saélihtyah, 
 there was enacted a great historical drama. In 244 at this place the young emperor 
 Gordian III was treacherously assassinated by an officer of his own army, a man 
 whom the biographers of the Caesars call Philip the Arab. This artful Beduin caused 
 divine honors to be rendered to his victim and his remains to be taken to Rome; but 
 the soldiers, in memory of the leader whom they had just lost, set up a monumental 
 cenotaph on the spot where he had perished. ‘This tomb, ‘‘visible from afar,’’? which 
 must therefore have been placed upon an eminence, was situated between Zaitha and 
 Dura, and was still in existence in the time of Julian, who, as he descended the left 
 
 i Unpublished bas-relief of the Louvre (A.O. 2198) of which M. Réné Dussaud has kindly had a 
 photograph made for us. Compare Clermont-Ganneau, Htudes d’archéologie orientale, I, 112 ff. 
 M. Gabriel Millet has pointed out to me another statue whose ornaments resemble those of our 
 paintings, Marshall, Catalogue of Jewellery, p. 320. 
 
20 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 bank of the Euphrates, presented funeral offerings there.t May we concede that this 
 sepulcher, where rites were practiced in behalf of the deceased emperor, was in reality 
 situated on the other side of the river at Salihtyah? Or at least, must one believe 
 that a memorial of the tragic death which had occurred in the vicinity, in the territory 
 which had once again become Persian, was celebrated in a sort of expiatory chapel in 
 the Roman camp nearest at hand? Does the name Kan-kaleh, ‘“‘House of Blood,” 
 which the fortress bears in Turkish, preserve an echo of the murder perpetrated in 
 its neighborhood and recalled by the annual ceremonies within its walls? These 
 are hypotheses which only further excavations can verify or disprove. 
 
 One needs must hope that the pacification of the country will permit of undertaking 
 excavations which will bring to light all the paintings of this ‘‘House of Blood.” 
 Under the most unfavorable of circumstances Mr. Breasted, with his collaborators, 
 has secured such an account as one would hardly have expected could be obtained 
 in a single day, and we owe him warm appreciation for the revelations which he brings 
 us. But the task now presents itself of completing his investigation and making 
 known in its entirety the extraordinary monument which will henceforth make famous 
 the name of Saélihtyah.? 
 
 1The position of this tomb cannot be established with absolute certainty. Zosimus (iii. 14) 
 states that it was at Dura; Ammianus (xxiii. 5. 7), that upon arrival at Zaitha one saw ‘‘Gordiani 
 imperatoris longe conspicuum tumulum’’; Eutropius (ix. 2. 3), on the contrary, the place to be 
 ‘‘vigesimo milliario a Circesso”’; but it is probable that he is giving the distance from the station 
 at Zaitha, indicated in the itineraries, instead of that of the neighboring tomb. For other, still less 
 precise references, cf. Pauly-Wissowa, Realencycl., s.v. “Antonius,” No. 60, col. 2627. [The inscriptions 
 afterward excavated by Cumont have shown that Clermont-Ganneau’s suggestion is not tenable.— 
 ds 1GL Jey} 
 
 2 As above noted, M. Cumont has begun with the greatest success the continuation of the excava- 
 tions which circumstances have not permitted the Oriental Institute to carry further.—J. H. B. 
 

 
OU he 
 
 34 
 
 
 
 
 CARCHEM one 
 RoMEY a oe 
 & Kh» 
 
 
 
 
 
 
 
 
 
 
 he NICEPHORIUM 
 SR Sa Ba f 
 
 Aden 
 el “ 
 wy - 4 
 
 
 
 
 
 
 
 
 
 34 tb 
 
 
 
 
 
 
 
 
 
 30 + 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 v 
 PDES " an 
 = Nebifiishaah Pe wueiecdl 
 ee BYBL OS wo Albu Aamal EEO 
 Oe, Beirdts¢ ¢ Laalbele !) 
 | 
 aa a 4 Damascus 
 LJ \ 
 \ 
 ’, 
 1) Z 
 7 1) 
 KP x A 
 te 
 He/s8 
 Jerugalems 
 
 lehem 
 
 G. of t 
 
 Akaba rt 
 
 36 38 
 
 
 
 Map 1.—Western Asia, showing the route of the O1 
 of Dura and the 
 
Mhogsabad 
 N INEVEH 
 
 © Balawat 
 Nimrdd 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 46 48 
 
 WESTERN ASIA 
 
 from 
 
 THE PERSIAN GULF 
 
 to the 
 MEDITERRANEAN 
 
 showing the route of the 
 
 UNIVERSITY OF CHICAGO EXPEDITION. 
 and dhe position of 
 
 DURA AND KHANA-MARI 
 
 VIGHFOBOUS tee 
 
 
 
 36 
 
 Froads and Caravan Foutes 
 
 
 
 
 
 froute of the University of Chicago Expedition —=—_=-=—== 
 Scale 1: 4.000000 134 
 Ow 220 #4071 260% 60 = 106 miles 200 
 fo! “, 
 
 "fe ,07 
 ite 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 44 
 
 One 
 is « 
 hte, Us fr 
 MV WN Ny, 
 Sti 
 vie, 
 as 
 M ial ies - 
 NUCL AT a, 
 aig “Bascag= re G 
 Abu Shahrein : 
 R i | 
 si Gs = ae 
 Son 
 i 48 
 
 Htal Institute Expedition of 1919-20 and the Situation 
 
 fin of Khana-Mari. 
 
“ae, 
 9 , 4 
 a} 
 4 
 e 
 ae | (<= 
 ' 
 i : 
 \ 
 Fa 
 + 
 ‘ 4 
 aT 
 a 
 4 
 i] 
 j 
 ‘ 
 * 
 ° 
 » 
 i | 
 ‘ 
 fj 
 . 
 #¢ 
 ° 
 “f 
 . 
 <— 
 e 
 
 ve 
 
 
 
 bina 
 ‘ 
 \ 
 ‘ 
 ‘ 
 ‘ 
 a 
 j i { 
 j ‘ Les 1° 
 foe “p 
 4 ‘ 
 id aed Any Pi 
 f oP 
 w 
 ’ bo 
 ‘ 7 
 A 
 AL, 
 ~ ar 
 4 
 a’ 
 ‘ 
 ‘ ’ 
 i "t 
 ' 
 ! ' 
 ' 
 rs 
 i 
 1 
 ‘ : 
 * 
 34 
 F 
 * 
 ' 
 ' 
 ‘ 
 qt yy 
 a 
 i 
 ‘ 
 . 
 ’ 
 1 a ‘ 
 i 
 ; 
 be 
 h 
 . 
 Ve 
 ' 
 a 
 - 
 Py Ria i 
 Ve 
 s 7 
 i 4 
 1 
 ' i 
 i] 
 * 4s 
 ‘ \ _ 
 sf 
 : thi) 
 ’ Ay 
 ’ ‘ 
 a | 
 * 
 j i ~~ 
 
 
 
 
 
 
 ° xy . a =i ' j y 
 - AIRS Os 
 6 ml id if 
 
 
 
 7] ( i +d Te - “Ye 
 \ ee t- " i io Nae Ue foh) vd ys 
 | ea A oe, SPOT ME salen (Ss 
 ok rks ae eae tae 
 ‘oe se : 1 De a ae NS x ay * i, : Le ; Pn 
 ij | Di al = : ry *, yi. PP ics a . i, < 
 A a) ' ’ 1)” 
 { },) ‘ . 7 oe 
 4) ‘ ? } ee bs y a" " ry . 
 j he eo ue w ma 
 « 4 ea «* f- ‘ vi a e 7 a 
 ; - ; * “7 Ps nS | ‘Pls ‘ 
 ; 7 a x 
 ; Py ~ PL ay >, i ¥ ert 
 ’ ' r %, . a ’ j* 
 e = 7 BE 8.) wy nae 
 4. Men, ed?) ee 
 ‘ WAS ' ap si Jt Ts I “hn 
 7 oat pe : en 7 : 
 . he ling o< ( P 
 ie ree i - Po, af ' aD 
 Bayes cs oe ee 
 AES, Dy a? + oe wer eth a 
 7 ' 7 a a AA us , ar i. a, » he 
 i | ‘hae, 
 ; } i “ , : 
 ; » 4 | I ’ a 
 j ee Vi 4 oF / ¥ 
 4 { as ras oi 
 f L’s c ; 
 i ae i ' { ‘ ys ‘ty , "7 
 ‘- TONE Me he ics are 
 : ey ite J iy 4 yee ‘ee ef ie 
 2 rhe 3 7 To) . ae ns 
 -, f v4 4 
 ; A >). +1, _ Oe 
 =| 
 a 
 ' 
 Ava 
 \ 
 ! 
 ' 
 J ‘ 
 ire 1 
 “ 
 re, 
 
I 
 
 THE HISTORY OF DURA-SALIHIYAH AND THE MID- 
 EUPHRATES KINGDOM OF KHANA-MARI 
 
 The northern extension of the Arabian Desert, which we commonly call the 
 Syrian or the Syro-Mesopotamian Desert, reaches a latitude about that of the northern 
 shores of the Mediterranean. ‘This desert shoulder thus thrust northward separates 
 the eastern shores and adjacent territory of the eastern Mediterranean (Syria-Palestine) 
 on the one hand very completely from the world of the two rivers, the Tigris and 
 Euphrates, on the other. This geographical barrier of the Syro-Mesopotamian Desert 
 has always caused a political and cultural cleavage, though it has never been an imper- 
 vious barrier, because cultural influences have been able to cross or go around its north- 
 ern end, like the armies which were so often the expression of political power. In this 
 desert region the Euphrates in its westernmost course approaches to a point about 100 
 miles from the northeast corner of the Mediterranean. Thence it turns southeastward 
 toward the Persian Gulf. Its southward and southeastward course through the 
 Syro-Mesopotamian and then through the Arabian Desert has been the path of many 
 a western army leaving the Mediterranean world and endeavoring to reach that of 
 Babylon and Nineveh, the region now popularly called Mesopotamia, which did not 
 originally extend to the Persian Gulf or include the Babylonian plain. But such a 
 march down the Euphrates across the desert was a difficult and dangerous enterprise. 
 It is no accident that the successors of Alexander in Syria and Asia Minor early lost 
 the Tigro-Euphrates world beyond the Syrian Desert; nor that the Roman Empire 
 definitely relinquished this region and placed the eastern frontier at the Euphrates. 
 
 As a result of this strategic situation a stretch of some 250 to 300 miles of the course 
 of the Euphrates, from a point below Carchemish-Jerablis (near where the river is 
 closest to the Mediterranean) down to the vicinity of ‘Anah, formed a kind of inter- 
 mediate fringe lying between the Mediterranean world on the west and that of the 
 two rivers, or Mesopotamia and Babylonia, on the east. This stretch of Euphrates, 
 therefore, was sometimes held by Babylonia, Assyria, Persia, or Parthia as a western 
 frontier, and again by a western power like that of the Macedonians or Romans as 
 its eastern frontier. Its alternating political connection is suggestive also of the 
 commingling of cultural elements in this region, which in its civilization clearly 
 betrayed influences received from both East and West. The composite civilization 
 which thus grew up along this stretch of Euphrates, like the political history of the 
 region, is little known; and the discoveries resulting from the work of the University 
 
 21 
 
22 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 of Chicago Expedition and the mission of M. Franz Cumont have demonstrated the 
 fundamental importance of exploration and excavation among the ruins which still 
 survive there. 
 
 The plateau of the Syro-Mesopotamian Desert has many points of resemblance 
 to the Sahara Desert plateau of Northern and Northeastern Africa, of which indeed 
 the great deserts of Western Asia are the geological continuation on the Asiatic side 
 of the Red and Mediterranean seas. Entering the Syro-Mesopotamian Desert from 
 the north, as we have noted, at about the latitude of Aleppo, the Euphrates has never 
 succeeded in eroding a valley of continuous breadth. It is in some stretches a narrow 
 defile or canyon; again the valley expands to a width of some miles. Throughout 
 
 
 
 Fig. 1.—Typicat View or EUPHRATES VALLEY IMMEDIATELY ABOVE ‘ANAH 
 
 Plateau cliffs dropping to the water on opposite (Mesopotamian) side; narrow cultivated strip 
 between river and cliffs on near side. 
 
 all the desert course of the Euphrates, however, its valley never widens so as to create 
 an alluvial’ floor of cultivable soil sufficiently extensive to furnish the agricultural 
 basis for sustaining a large population, as in Egypt above the Nile delta; or for a 
 corresponding development of culture and of stable political power. For stretches of 
 many miles the cliffs drop almost to the water’s edge, and in such places there is 
 absolutely no population at all. Of the region between ‘Anah (Fig. 10) and El-Kaim 
 (Fig. 12) the engineer Cernik, who passed through it in 1872, says: “Hs ist soviel als 
 gar nicht bewohnt.”! We made the same observation on our journey through this 
 district in 1920. The series of Euphrates views distributed through this discussion 
 will serve to demonstrate this character of the Euphrates Valley better than an 
 elaborate geographical description. After three typical glimpses of the Euphrates 
 
 i Cernik, Technische Studien-Expedition durch die Gebiete des Euphrat und Tigris. Erginzungs- 
 heft No. 44 zu Petermann’s Mittheilungen, 1875. 
 
 a 
 
 ee 
 
 _— 
 
THE HISTORY OF DURA-SALIHIYAH 23 
 
 (Figs. 1-3), the views begin at Hit (Fig. 4), where the alluvial plain of the lower 
 Euphrates practically ends as you ascend the river, and moving upstream they show 
 a series of characteristic points as far upriver as Meskenah (Fig. 16), where the road 
 from Baghdad to Aleppo leaves the Euphrates. 
 
 Omitting smaller and less important areas of cultivation, where there is a very 
 limited expansion of the valley floor, there are two chief regions where the alluvial 
 plain is of such proportions as to permit a cultural development of some importance. 
 These are the district of ‘Anah about 180 miles above Baghdad, and the plain below 
 Deir ez-Zér, some 320 miles from Baghdad, that is about midway between this city 
 and the Mediterranean. We are dealing in this book with the latter of these two 
 
 
 
 Fic. 2.—TypicaL WALL OF THE DrserT PLATEAU ALONG THE HUPHRATES 
 
 Arid floor of the Euphrates Valley and a part of our caravan in foreground 
 
 districts, the plain below Deir ez-Zér, the modern metropolis of the middle Euphrates 
 region. We shall hereafter refer to this town as Deir. With some interruptions from 
 intrusive promontories of the plateau, this plain extends for about 75 to 80 miles 
 southeastward of Deir. In width it varies from nothing at all to a maximum of some 
 35 kilometers, about 22 miles; but by no means all of this maximum width is cultivable 
 alluvium. Around Meyadin the grassy plain extends at one place some 25 miles 
 westward and furnishes grazing in spring for the flocks, herds, and mares of the 
 Anazah Arabs; but the cultivation never extended so far. Somewhat over 10 miles 
 below Deir the plateau thrusts forward to the river for several miles on the west side 
 
24 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 opposite the mouth of the Khabir, while on the east the Mesopotamian Desert sends 
 out the tip of a tongue of mountains in a long promontory which completely fills the 
 interval between the parallel channels of the Euphrates and its tributary the Khabtr 
 coming in from the north and east. The heights on each side of the Euphrates here 
 form a narrow defile, and it was this strategically strong gateway which the Romans 
 selected as the most easily defended frontier barrier of the Empire on the east. Here, 
 therefore, on the east side of the Euphrates, at the tip of the above-mentioned narrow 
 promontory, between the Euphrates and the Khabfir, Diocletian built the frontier 
 
 
 
 Fre. 3.—Tyricatn Desert VAuury ‘Wapbt” on RicgHT BANK OF EUPHRATES BELOW Derr Ez-Z6r 
 
 Part of our caravan climbing to the top of the plateau because the desert cliffs left no room for 
 passing along the river bank below. 
 
 fortress of Circesium toward the end of the third century of our era. This narrow 
 defile at Circesium cuts off the northern district of the plain for some 10 miles below 
 Deir, and the intrusive plateau cliffs occupy several miles more. The remainder of 
 the plain stretches southward from the defile at Circesium on the north in a coherent 
 whole along the river for something like 55 to 60 miles to modern Albu Kamal on the 
 south.? 
 
 This plain from below the Circesium defile, that is, from the mouth of the Khabar, 
 to Albu Kamal, never in Graeco-Roman days received any name which has descended 
 
 1 Exact measurements along the windings of the river are not available. The measurements 
 given follow the general trend of the plain. 
 

 
 
 
 
 
 
 
 IR 6ZZ6r 
 
 \\ = “ Gs 
 B \Abu Hassan 
 I aN 
 So) arr al 
 Ws 4 "S 
 
 NG 
 x 
 
 
 
 Crs Ne 
 con 
 AbuNuhud, 
 
 in iy 
 2 Ah 8 Snelich Abd eb-L4 i 
 
 \. EAS oe /-Aweljnah 
 eS oes R favs WEEE . Ap 
 
 AE iUlage \ 
 on Ad, 
 
 J 
 Tell Zubagr ACIRGESIUM Busairah-Kerkes pen 
 Kubbet el-lmam Aji 
 
 \ 
 D y 
 \Z 
 Bogrus® oy /-K, ‘Oo 
 > gene RC Au pA se ‘a 
 F ” Meyadi at 
 
 
 
 MA 6 
 
 
 
 AV spear 
 aw fs 
 
 “Tell Jama 
 
 
 
 \O 
 WS 
 
 
 
 [p\Telleateiyah 
 
 OU 7) = 
 K aud 
 y anarisk 
 Anan — Lhickets 
 x Ros 
 
 Tell es) a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 PLAIN OF KHANA- MARI 
 
 
 
 
 
 
 
 
 
 AND | os : : ; Irzi, TW”? tHE ‘a 
 DURA-SALIHIYAH | ROR eT 
 Scale of Miles mam Shave obo S 
 i0 15 20 Aasr e/- Kain » ah Hy 
 30° Al 
 
 Map 2.—The Plain of Khana-Mari 
 
 
 

 

 
 
 
 THE HISTORY OF DURA-SALIHIYAH 25 
 
 to us; -nor have the Moslems ever given it a designation, political or geographical, 
 which has become current. It is difficult, therefore, to refer to it in terms intelligible 
 to the reader, for modern historians have likewise failed to give it a name. Indeed, 
 enveloped as this region is in the heart of the Syro-Mesopotamian Desert, few historians 
 are acquainted with it, and fewer still have ever seen it. For reasons which will be 
 evident as we proceed, we shall call this plain “‘Khana-Mari,” a term compounded of 
 two names which were evidently the earliest to be applied to the region or to parts 
 of it, in so far as we know. 
 
 The river wanders for the most part along the western edge of the river “‘ bottoms,” 
 so that the plain of Khana-Mari lies chiefly on the east side of the Euphrates (Fig. 17). 
 This leaves its eastern portion far from the river, and at an early date advantage was 
 taken of the fact that the northern boundary of the plain is the Khabar. From this 
 
 
 
 Fic. 4.—Mopern Hirt, tap Cuassican Is, tae ANCIENT SOURCE OF BITUMEN 
 
 tributary an irrigation canal was dug along the eastern boundary of the plain to the 
 Euphrates at El-Werdi (see Map 2), where the Mesopotamian plateau advances 
 to the river in a lofty promontory to form the southern boundary of the plain.1 The 
 plain falls into two districts, the result of a natural division caused by the advance of 
 the western plateau to the river, not far south of the middle of the plain. This salience 
 of the western desert pushes forward directly to the margin of the river for nearly 14 
 miles, and for some 9 miles the rock walls of the plateau escarpment fall sheer almost 
 to the water’s edge. On the west side of the Euphrates, therefore, the plain is com- 
 pletely cut in two; on the east side there is no such division. This intrusion of the 
 
 1 See Gertrude L. Bell, From Amurath to Amurath, pp. 78 and 82; Sarre-Herzfeld, Archdologische 
 Reise im Euphrat- und Tigris-Gebiet, II, 387. While we are not able to date the origin of this canal 
 with precision, the early history of the plain would indicate that it was in existence before the Moslem 
 Age and probably in old Assyrian and Babylonian days. 
 
26 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 plateau on the west side reduces essentially the cultivable area of the plain, but in 
 its maximum length of 60 miles it may have included at times as much as 400 square 
 miles of cultivated fields—an area equal to a good-sized American county. 
 
 With relation to the larger world of the Near East as a whole, Khana-Mari occupied 
 a favorable strategic position, in spite of its situation in the heart of a desert, separated 
 on all sides by hundreds of miles of waterless waste from the nearest communities 
 of men (see Map 1). It was a veritable island in the sandy ocean. One line of 
 connection, however, was the river itself, which furnished a highway to Babylonia 
 on the southeast and to Asia Minor and Armenia on the northwest and north—not a 
 navigable highway to be sure, but at least a desert road amply supplied with water. 
 Westward stretched over 200 miles of desert to the valley of the Orontes and the 
 
 
 
 Fig. 5.—Tur Bitumen Basin at Hit 
 
 Mediterranean kingdoms. On this western desert road nine or ten days’ journey 
 distant (140 miles) lay the oasis of. Tadmur, more familiar to us as Palmyra, which 
 supported a caravan road between the Mediterranean kingdoms and the middle 
 Euphrates region. On the east was the Mesopotamian Desert with a caravan route 
 leading to Nineveh and the upper Tigris region. Khana-Mari lay thus like an island 
 haven on the desert line of communication between Persia, Babylonia, and Assyria 
 on the east, and Asia Minor, Egypt, and the Mediterranean world on the west. It is 
 evident that in such a situation a sufficient population, maintaining themselves on an 
 agricultural basis, would find sustenance here to develop a kingdom powerful enough 
 to seek more than merely agricultural prosperity and to control the traffic up and down 
 the river as well as the caravans of merchandise passing east and west through the 
 deserts on each side. 
 
 Today little of this ancient prosperity survives, as we shall see. The irrigation 
 canal is clogged and has fallen into disuse. It contains water for a time in some places 
 

 
 THE HISTORY OF DURA-SALIHIYAH 27 
 
 after the season of flood. In the north, around Deir, skin buckets operated by animals 
 raise the water to irrigate the fields. In the south, beyond the limits of Khana-Mari, 
 and especially above ‘Anah,! huge water wheels driven by the current of the river serve 
 the same purpose much more efficiently. In Khana-Mari itself, however, cultivation 
 is at present limited to the region immediately around the towns, and of these 
 there are very few. There are some mounds marking the ruins of ancient cities, 
 though they are far from being as numerous as in the region around Aleppo or on the 
 Babylonian plain. Even our hurried journey through the country made it evident that 
 no great and powerful nation native to this region and rooted in its soil ever gave 
 
 
 
 Fig. 6.—A Brirumen WELL at Hit 
 
 Hot water bubbles up through the liquid bitumen 
 
 impressive expression to its power in imposing buildings and massive strongholds. 
 Such structures now surviving in Khana-Mari were the work of great empires intruding 
 from without. Such is the stronghold of Circesium which the Roman Empire erected 
 at the northern end of the plain, fixing the imperial frontier securely in a region of 
 copious water, plentiful provender for a garrison, and strong natural defenses, although 
 surrounded by vast reaches of desert fatal to an army once involved in them. It was 
 an outpost like an island or a bridge pier which the Romans held secure but failed to 
 advance permanently to the other shore. In the same way Alexander’s successors 
 made use of the place and continued to hold it. Just as under Greeks and Romans 
 
 1The water wheels begin a few miles below El-Kaim and extend to Hit. 
 
28 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 it was a safe frontier outpost of the West against the East, so it had earlier been a 
 similar stronghold of the East against the West, and a number of important battles 
 in the age-long struggle of Kast and West were fought on the plain of Khana-Mari. 
 Until recently, however, we have not been able to trace the history of the region 
 farther back than the Seleucid occupation, nor even as far as this without serious gaps. 
 
 Concerning the period preceding Alexander’s successors little was known of 
 Khana-Mari until the last years of the past century, when cuneiform documents 
 containing references to the lands of Hana (Khana) and Mari began to appear. 
 Some references to Mari had been observed in documents previously known,! but its 
 situation had at first been wholly problematical. 
 
 At length indications were noted in the cuneiform documents that Mari was 
 situated along the middle Euphrates near the steppe called Suhz (Sukhi).2 The first 
 
 
 
 Fig. 7.—LooxktneG up THE EUPHRATES AT HApiTHaH 
 
 The narrow strip of vegetation stops in the immediate background where the cliffs of the plateau 
 may be seen descending to the river on both shores, 
 
 document mentioning the land of Khana was published in 1897.3 It was a cuneiform 
 tablet containing a deed of gift of a house situated “‘near the palace” in the city of 
 Tirka, and it bore the seal of ISar-lim, king of Khana. For ten years this remained — 
 the sole document from Khana known tous. Then in 1907 Johns published a marriage 
 contract! from Khana, which showed that this country was situated in Mesopotamia 
 in the region of the river Khabtr. At about the same time there appeared a tablet 
 which came from a temple erected in Tirka by Shamshi-Adad, an Assyrian patesi 
 
 1 Thureau-Dangin, Die sumerischen and akkadischen Kénigsinschriften, 1907, pp. 22-23, 170-71. 
 
 Compare also King, History of Sumer and Akkad, plate opp. p. 102, and p. 98; Ed. Meyer, Geschichte 
 des Altertums, I, 2% (1913), §386. 
 
 2 Weissbach, Babylonische Miszellen, pp. 9 ff.; and Herzfeld, Revue d’Assyriologie, XI (1914), 137, 
 
 3 Thureau-Dangin, Revue d’ Assyriologie, IV, 3. livr. (1897), pl. XXXII, no. 85; and Orientalistische 
 Litteratur-Zeitung, XI (1908), col. 93 (note 1). 
 
 4C. H. W. Johns, Proceedings of the Society of Biblical Archaeology, X XIX (May, 1907), 180; 
 cf. Ungnad, Zeitschrift fiir Assyriologie, XXII (1909), 9. 
 
 
 
 
 

 
 
 
 THE HISTORY OF DURA-SALIHIYAH 29 
 
 who ruled about 1900 8.c.!. The evidence for its provenience indicated that the tablet 
 was found at Tell <Ashara in the plain below the defile of Circesium. Shortly after 
 this a tablet purchased near Deir el-Far in the same plain near the great ruin of RAahaba 
 proved to be a gift by “King Ammi-Bail” of land situated partially in the city of 
 Tirka.2 In 1909 another of these documents from Khana was published by Thureau- 
 Dangin, who had first observed and identified the references to it. The new tablet® 
 recorded a contract concerning a sale of land in the territory of Tirka under a king 
 called Kashtiliashu. Summarizing the evidence of content, handwriting, ete., 
 Thureau-Dangin concludes that all these documents from Khana came from the same 
 body of archives.* 
 
 Finally this rapidly appearing series of documents regarding the kingdoms of 
 Khana and Mari culminated just before the Great War in a report by Herzfeld of his 
 
 ee 
 
 
 
 Fig. 8.—Scanty FIeLtps ALONG THE EuPHRATES Two MILES ABOVE HapiTHaH 
 
 The water wheel of Fig. 9 in the background 
 
 discovery that the tiny Arab village called <Ashara in the plain of Khana-Mari, 
 15 kilometers south of Meyadin, was built on a mound of ancient ruins dating far 
 back of Graeco-Roman antiquity. He acquired here from the natives ancient Baby- 
 lonian sculptures of very early date reaching far back in the third millennium B.c. 
 The side of the mound out of which these pieces undoubtedly came has been cut away 
 by the action of the Euphrates current, and in this vertical section exposed by the 
 river Herzfeld himself dug out of the ancient rubbish a tablet dating from the First 
 Dynasty of Babylon and recording the fact that a king of Mari named Zim{ri] had 
 
 1P, Condamin, Zeitschrift fiir Assyriologie, XXI (1908), 247; and important discussion by 
 Thureau-Dangin, Orientalistische Litteratur-Zeitung, XI (1908), col. 193. 
 
 2 Ungnad, ‘‘Urkunden aus Dilbat,” Beitrdge zur Assyriologie, VI (1909), No. 5, 26 ff.; Vorder- 
 asiatische Schriftdenkmdler ....zu Berlin, VII, no. 82; and Herzfeld, op. cit., p. 138. 
 
 8 Thureau-Dangin, Journal Asiatique, 10° Série, XIV (1909), 149-55. 
 4 Thureau-Dangin, zbid., p. 155. 
 
30 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 erected a building in Tirka.! It was then evident that the ruins of Tirka were. buried 
 under the mound of Tell ‘Ash4ra, and that the little Arab village on the mound was 
 built on the ruins of the capital of Khana, for such Tirka must have been. That the 
 king of Mari should have been erecting a building in the capital of Khana would 
 suggest that he had conquered Khana and was ruling it. The position of Tirka in 
 the midst of the plain around modern Meyadin shows clearly that the kingdom of 
 Khana included at least that plain stretching toward Deir ez-Zér, however much more 
 it may have comprised. 
 
 Shortly after this Clay? discussed the occurrence of the name of Mari in various 
 forms and suggested that the middle Euphrates town of Merra or Merrha, included 
 
 
 
 Fig. 9.—IRRIGATION WATER WHEEL IN THE HUPHRATES AT HapiTHAH 
 
 Compare Fig. 8. The current turns the wheel and raises the submerged and filled buckets or 
 jars to be emptied into the aqueduct leading to the field on the left. These wheels begin a few miles 
 below El-Kaim (Fig. 12) and extend to Hit. 
 
 by Isidor of Charax in his lists, should be identified with Mari of the cuneiform docu- 
 ments. This equation of Mari with Merrha is not at all unlikely; though Merrha 
 has commonly been placed at Irzi, on the left bank of the Euphrates surmounting 
 the salience of the plateau which terminates the plain of Khana-Mari on the south. 
 But it is probable that the town of Mari will be found nearer to Dura, for Isidor 
 makes Dura his nineteenth station and Merrha his twentieth. The interval which 
 he gives between Dura and Merrha is 5 schoinoi. I have never marched it myself, 
 but, measured with no allowance for tortuous terrain, the interval on the British 
 
 1 Herzfeld, op. cit., pp. 131-39. | 
 
 2 Clay, Yale Oriental Series. Babylonian Texts, Vol. I, p. 4. .New Haven, 1915. 
 
 
 

 
 
 
 
 
 THE HISTORY OF DURA-SALIHIYAH 31 
 
 War Office map from Salihtyah to Irzi is over 35 miles, a distance considerably ex- 
 ceeding 5 schoinoi as used by Isidor. Merrha is therefore much more likely to have 
 lain somewhere within the Khana-Mari plain—a situation which would make its 
 identification with Mari much more probable. Thus our identification of Sélihtyah 
 as ancient Dura enhances the value of the identification of Isidor’s Merrha with Mari, 
 for it places Mari in accordance with Isidor’s arrangement of his stations 5 schoinoi 
 from a known point, and connects it with the plain where the discovery of Tirka has 
 shown it must belong. 
 
 
 
 Fig. 10.—THEe VALLEY OF THE EUPHRATES AT CANAH 
 
 We look downstream from the north end of the <Anah district. Although several miles long, the 
 cultivated strip at <Anah is so narrow that the whole of it would not exceed the area of a good-sized 
 farm in the Mississippi Valley. 
 
 We are now in a position to discern something of the course of history on the 
 Middle Euphrates from a very early date. The little kingdom of Mari was already 
 subject to the Babylonians as far back as the supremacy of Lagash around 3000 B.c., 
 when Eannatum, the king of Lagash, was overlord of Mari and there was intimate 
 intercourse between the kings of Mari and those of Babylonia. A king of ‘‘Ma’er 
 or Mari” in this same early epoch seems to have offered to Shamash of Babylonia an 
 archaic seated statue now in the British Museum.! In these early days when ephemeral 
 political leadership was shifting frequently from one Babylonian city to another, 
 even the little kingdom of Mari might aspire to such supremacy. ‘The early dynastic 
 
 1 Herzfeld, op. cit., p. 137; King, op. cit., pp. 97-98, and plate opp. p. 102, right-hand figure; 
 Thureau-Dangin, Die suwmerischen und akkadischen Kénigsinschriften, 1907, 170-71. 
 
32 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 lists preserved a tradition from an age so early as to be half mythic,! that Mari gained 
 the leadership of Babylonia and maintained it for at least a generation through the 
 reigns of several kings.2, We can only conclude that the presumably Semitic chieftain 
 of Mari was a military leader of ability who was able to march his desert troops down 
 the Euphrates and conquer the alluvial plain below, just as the ancestors of Ham- 
 murabi, many centuries later, were able to do. The overthrow of Mari after its 
 leadership had weakened seems to have made an impression, at the time of its 
 occurrence, as an event of sufficient importance to furnish the current and official form 
 of the year name used in dating: ‘year when Ma’er (Mari) was destroyed.’ 
 
 
 
 Fig. 11.—Istanp CULTIVATION IN THE HUPHRATES AT CANAH 
 
 The walls of the desert plateau are seen descending to the river on the other (left) bank, as in 
 Fig. 10. 
 
 With the rise of the powerful Sargon I, his dominion extended up the Euphrates 
 for a time into Syria, even to the silver-bearing mountains of Cilicia, and his records 
 affirm that he held ‘‘the upperland, Mari, Yarmuti and Ibla.’* Mari seems to be 
 
 1 Delaporte, La Mesopotamie, p. 65, places the dynasty of Mari in “‘Chronologie mythique”’ (in 
 L’ Evolution de V Humanité, Vol. VIII, Paris, 1923). 
 
 2L. Legrain, Historical Fragments, pp. 18-24 (in ‘University of Pennsylvania, The University 
 Museum, Publications of the Babylonian Section,” Vol. XIII); again, Museum Journal (December, 
 1920). See also Gadd, The Early Dynasties of Sumer and Akkad, pp. 8-12. 
 
 3 Legrain, ibid., pp. 62-63. 
 
 4Poebel, Historical Texts, pp. 177-78 (in “University of Pennsylvania, The University 
 Museum, Publications of the Babylonian Section,” Vol. IV, No. 1). My colleague Professor Lucken- 
 bill calls my attention to the fact that Sargon’s possession of Mari is confirmed by an Assyrian copy 
 of Sargon I’s records made under Sargon II of Assyria. See Schroeder, Keilschrifttexte aus Assur 
 verschiedenen Inhalts, Text No. 92. It places Mari on the Euphrates. See also Poebel, op. cit., 
 pp. 222 ff. 
 
 
 

 
 THE HISTORY OF DURA-SALIHIYAH 33 
 
 used here vaguely both for the middle Euphrates region and for territory which it 
 controlled for an uncertain distance up the Euphrates. It is evident that Mari has 
 been absorbed in the northwestern expansion of Babylonia as it pushed up the 
 Euphrates into Syria. 
 
 Very much later, under the Ur Dynasty, there was a governor of Mari named 
 Izi-Dagan, as we learn from his seal. Not long afterward, about the middle of the 
 twenty-fourth century B.c., we find a man of Mari, probably a hardy military 
 adventurer, who went down the river to the Babylonian plain and made a success- 
 ful career at Isin, where he was strong enough to establish a dynasty. His name 
 was Ishbi-Irra and he was the first king of the well-known Dynasty of Isin.! 
 
 
 
 Fig. 12.—TaMARISK THICKETS ALONG THE EUPHRATES AT EL-Kamm 
 
 Such tracts furnish only food for camels, seen feeding in this view 
 
 During the career of Isin arose the First Dynasty of Babylon, the famous line of 
 Hammurabi, a family which also pushed down the Euphrates from the west and 
 established its fortunes in the Babylonian plain. It is evident that these upriver 
 Semites like Ishbi-Irra and the ancestors of Hammurabi were aggressive and 
 energetic men of action. Their drift down the Euphrates and their easily won 
 supremacy in the alluvial lowland were not unlike the barbarian intrusions into the 
 later Roman Empire, whose advent brought in new blood and efficient leadership. 
 There are indications that the same type of Semitic chieftains also pushed into Egypt 
 more than once. What they brought with them was chiefly aggressive leadership, 
 for there is no evidence that the regions from which they came possessed anything 
 but the crudest stages of incipient civilization in the period preceding 2000 B.c., the 
 period with which we are dealing. Like the barbarous Persian shepherds of Cyrus’ 
 
 1 Poebel, op. cit., p. 223, 
 
34 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 time, they had to learn civilization in the more advanced communities whose territories 
 they invaded. But the unlettered Persians, who could not write and did not even 
 possess a system of writing when they first advanced against the civilized world of the 
 Ancient Orient, produced Cyrus and Darius, two of the greatest organizers that ever 
 lived. In the same way these upriver Semites of Mari and the region above it brought 
 forth the founders of dynasties and such able rulers as Hammurabi.! 5 
 Poebel has suggested that Mari was the political center of the middle Euphrates 
 region in the days when Hammurabi himself attacked it? and destroyed its wall. 
 We should expect a sagacious ruler like Hammurabi to push up the Euphrates and 
 prevent further invasions of Babylonia from this quarter, such as his fathers had long 
 been carrying on. When he edited his code he seems still to have held the Khana-Mari 
 region. After Hammurabi’s successors failed to hold his dominions together and lower 
 Babylonia fell away, it is probable that Mari likewise escaped from the control of the 
 Babylonian dynasty for a time, and thereupon we lose sight of it for long centuries. 
 In the foregoing rapid survey we see that the ancient documents afford us glimpses 
 of the situation on the middle Euphrates for some nine centuries after 3000 B.c. After 
 the beginning of the twenty-first century B.c., it is nearly a thousand years before 
 we gain another glimpse of the plain of Khana-Mari. In the long interval the civiliza- 
 tion of Babylonia had declined as a result of foreign invasion, and her political fortunes 
 had likewise suffered eclipse. ‘The expansion of Assyria had meantime carried her 
 armies westward to the Euphrates, where Khana-Mari did not escape the growing 
 power of the mighty kingdom on the Tigris. From the new documents discovered 
 in the excavation of the old Assyrian capital at Assur, whence the kingdom took its 
 name, we learn that in the thirteenth century B.c. King Tukulti-Inurta I traversed 
 and captured ‘‘Mari, Khana, Rapiku, and the mountains of Akhlami. ... . "8 The 
 1Jt is a curious perversion and misunderstanding of existent evidence which would find in the 
 region whence these able leaders came, the original source of civilization—a region which excavation 
 has shown was far inferior in civilization to both Babylonia and Egypt in the period preceding 2000 
 B.c., and indeed for long after this date. Moreover, alarge part of the region is arid desert, furnishing 
 no economic basis for the development of civilization. The theories of an “‘Amorite Empire” have 
 been formulated in complete misunderstanding of the elementary facts in the economic geography 
 of the country. For example, we are told that “‘the northern part of this territory, known as Syria and 
 Mesopotamia, is fertile” (A. T. Clay, The Empire of the Amorites, p.50. Yale University Press, 1919). 
 The actual facts are that, with the exception of the lower part of the valleys of the Baltkh and the 
 KhAébir, Mesopotamia is a desert, or waterless steppe, and much of Syria likewise; for the desert 
 extends from Mesopotamia westward beyond the Euphrates into Syria including the hills overlooking 
 
 Homs in the Orontes Valley; while eastward it even invades the east side of the Tigris. Needless to 
 say that civilizations do not arise nor empires develop in deserts. 
 
 2 Poebel, op. cit., p. 223, year name of Hammurabi’s fourth year. 
 
 3’ Schroeder, Keilschrifiterte aus Assur historischen Inhalts, 2tes Heft (‘‘ Veroffentlichungen der 
 Deutschen Orient-Gesellschaft,” 37, Leipzig, 1922), Text No. 60, ll. 69 ff. I am indebted to my 
 colleague Professor Luckenbill for this reference. The list of the king’s foreign conquests following 
 the foregoing regions includes thirty-four geographical names and among them appears the ‘land 
 Duri,”’ which can hardly be our Dura. 
 
 
 
THE HISTORY OF DURA-SALIHIYAH 35 
 
 
 
 Fig. 13.—Tue Evpurates At TIBNI 
 
 The view is eastward, showing the scanty vegetation of the arid steppe 
 
 
 
 
 
 Fic. 14.—Tue EvpHrates VALLEY AT HALABLYAH 
 
 For over 5 miles above this point there is no cultivation along the river. The succession of 
 views here omits the Deir ez-Zér plain and its region, which will be found in Figs. 17-24. Photograph 
 from Sarre-Herzfeld, Pl. LX XII. 
 
 
 
 
 
36 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 list seerms to be proceeding from east to west, that is, up the Euphrates, and thus places 
 Mari below Khana, in accordance with the position of Isidor’s Merrha below Dura 
 which we have discussed above. In so far as we know, this was the first appearance 
 of the Assyrians in Khana-Mari, and Assyrian power was therefore first felt here in 
 the thirteenth century B.c. 
 
 Almost four centuries later we find the Assyrians again on the middle Euphrates 
 led by another Tukulti-Inurta, the second of the name.! On his sixth campaign 
 Tukulti-Inurta II? (890-885 B.c.) marched his army in a great circle southward down 
 the river Tartar (parallel with the Tigris on the west) and then up the Euphrates and 
 
 
 
 Fig. 15.—UNcuLTIvVABLE TAMARISK THICKETS ALONG THE HUPHRATES AT SABKHAH 
 Part of the khan of Sabkhah in the foreground 
 
 the Khabar. His annals call the region “Sirqu”’ and ‘ Rummunidi,” and it may be, 
 as Scheil has suggested, that we are to recognize the oriental ancestor of the Roman 
 Circesium in Sirqu. 
 
 It is likely to have been during this Assyrian domination that the rulers of the 
 Khana-Man plain noticed the natural strength of the bold promontory opposite the 
 middle of the plain, where the western plateau thrusts forward to the river as we have 
 
 1 This name was long read “ Tukulti-Ninib.”” The proper reading has been much discussed, for 
 example by Clay (Yale Oriental Series. Babylonian Texts, Vol. I, pp. 97-99), but it is not yet 
 entirely certain. 
 
 2 The annals of this king turned up in the hands of antiquity dealers in Mosul at the very time 
 when the foregoing documents concerning Khana-Mari were being recovered. This great tablet, 
 93> 104 inches in size and bearing 147 lines of text, has been rendered available by the excellent 
 publication of V. Scheil, Annales de Tukulti-Ninip IT roid’ Assyrie 889-884, Paris, 1909 (“‘ Bibliothéque 
 de l’Ecole des Hautes Etudes,” 178™° fasc.). 
 
 
 
THE HISTORY OF DURA-SALIHIYAH 37 
 
 already stated (p. 25). At any rate the Semitic rulers of the region placed a stronghold 
 on the crest of this promontory, and called it simply the “burg” or “castle,”’ which is 
 in Assyrian daru.!. We know nothing of the history of the place which thus gained 
 the name of Dir or later Dura. Complete excavation over the whole top of the 
 promontory might possibly reveal remains of the Assyrian stronghold and records 
 of its history, but they are not yet available. Hence we cannot trace the fortunes 
 of this remote little kingdom on the middle Euphrates during the Assyrian domination; 
 and although it was traversed again by the campaigns of Assurnacirpal IIT (885-860 
 B.c.), we learn next to nothing of Khana-Mari throughout the course of the Assyrian 
 
 
 
 Fig. 16.—Tue Evpurates at Meskenau 
 
 At this point the Baghdad-Aleppo road, branching off toward Aleppo, 54 miles distant, forsakes 
 the river. 
 
 Empire or of the Chaldean and Persian empires which succeeded in the control of 
 the Euphrates world. 
 
 With the fall of the Persian Empire we again catch a glimpse of Khana-Mari. 
 The strategic advantages of making this plain and its river buttresses the frontier 
 gateway of the Syrian desert, protecting their Syrian dominions on the east, was at 
 once observed, as we might have expected, by Alexander’s successors. Fortunately 
 for our investigations, the able Macedonian commander Nikanor was favorably 
 impressed with the natural strength of Dura itself. Isidor of Charax calls Dura 
 “a city of Nikanor, a foundation of the Macedonians.’? This colonization must 
 have occurred while Nikanor was governor of Mesopotamia under Seleucus I, and 
 
 1 There can be no doubt that the Greek name of the place, Aotpa, has descended from the Assyrian 
 daru, “‘stronghold;” so also Cumont (Syria, IV [1923], 53). 
 
 2 See the following footnote. 
 
38 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 previous to 312 B.c., when he revolted against his sovereign. Apparently it proved 
 impossible to give the place an exclusively new name, for the old Assyrian designation, 
 Dura, continued to be used and persisted to the end. Perhaps as the result of an 
 unsuccessful attempt on Nikanor’s part to give his new foundation a name which would 
 mark it as a Macedonian settlement in the Orient, we find that the place was “called 
 Europos by the Greeks.”! The recovery of the records of this Macedonian colony 
 on the Euphrates has very strikingly revealed a new example of the penetration of the 
 Near Orient by Greek influences after the campaigns of Alexander the Great. For 
 these colonists from the West maintained their essentially Greek culture for genera- 
 tions, and the earliest monuments which Sarre and Cumont found in the place are 
 
 
 
 Fic. 17.—Tsue EuPHRATES AND THE PLAIN OF DEIR £z-ZOR 
 
 The cliffs on the horizon far back from the river are those on the left, or Mesopotamian, side of 
 the Euphrates. The canal (p. 25) skirts these cliffs. 
 
 purely Greek in character, including even the code of laws which the colonists used, 
 and of which a fragment written on parchment was recovered by Cumont. 
 
 The earliest houses in the town, which go back to these Macedonian colonists, 
 and one of which had been uncovered by a British officer who showed it to us, were also 
 partially excavated by M. Cumont.? They are laid out with streets at right angles, 
 forming a city with square-block arrangement such as the architects of the Hellenistic 
 
 1Tsid. Char., Mans. Parth., 4: Aotpa, Nuxdvopos dds, xricua Maxeddver, bard 5& “EA\QVOY Epwros 
 xadetrat (C. Miller, Geographi Graeci Minores, I, 248). 
 
 2 They were first noted by Sarre and Herzfeld in their Archdologische Reise im Euphrat- und 
 Tigris-Gebiet, II, 391. 
 
 
 

 
 
 
 THE HISTORY OF DURA-SALIHIYAH 39 
 
 period introduced in towns like Priene—a system commonly attributed to one of the 
 architects of Alexander the Great, the Macedonian Deinocrates, who employed it in 
 his plan of Alexandria. 
 
 Dura long continued to be a frontier stronghold of the Seleucid Empire and 
 Polybius' tells how, almost a century after its foundation, our Europos was taken by 
 Molon in his insurrection against Antiochus (221 B.c.), as the limit of the rebellious 
 commander’s advance down the Euphrates. 
 
 The situation of Dura-Europos in the heart of the northern desert, however, was 
 too remote from the Mediterranean for it to remain permanently attached to the 
 decaying Seleucid Empire, and after the rise of the Parthians (third century on) it 
 was held by them. Cumont notes evidence that it was then culturally cut off from 
 the west. While Asia Minor and Syria adopted the Julian calendar early in the 
 
 
 
 Fig. 18.—TuHe PLAIN AND City or Drtr £z-Z6R 
 
 The cultivation at present extends but a short distance from the town, whose minarets are 
 visible on the horizon. 
 
 Christian Era, Dura continued to employ the Graeco-Macedonian calendar with its 
 inconvenient intercalary months, one of which is mentioned in the temple graffiti 
 of Dura. For generations the place must have been completely under the domination 
 of oriental influences, which would be felt the more in view of the fact that Arab 
 and Aramaean settlers had long shared the town with the Macedonian colonists and 
 mingled their life with that of the Western immigrants, as the Semitic names among 
 the inscriptions clearly demonstrate. 
 
 It was probably at this period of the Parthian occupation that the city received 
 at least some of its massive fortifications. At any rate it is quite evident that the 
 extreme northwest angles of the city wall, where the townsmen later built their temple 
 of Zeus-Baal (see pp. 68-74), were erected before the latter part of the first century of 
 
 1Polybius v. 48. 16, especially as explained by Eduard Meyer in Pauly-Wissowa, Realencycl., 
 
 s. v. “Europos,” No. 5, against the correction of Droysen and of Benzinger. See Cumont, Les 
 Travaux Archéologiques, p. 71. 
 
40 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 the Christian Era, that is, before this region had ever seen a Roman soldier. Sarre 
 and Herzfeld, who wrote under the impression that the city was never occupied by 
 the Romans for any length of time, evidently regarded the fortifications as pre-Roman. 
 If so, it would seem more probable that the Parthians erected them than that they 
 should have been the work of the Seleucids. Stretching at its greatest length some 
 1,400 meters (over four-fifths of a mile) along the heights overlooking the Euphrates, 
 and expanding over 1,000 meters (three-fifths of a mile) at its greatest breadth, the 
 immense inclosure of the city walls is still a very impressive sight. Commanding the 
 whole, a massive castle 50 by 390 meters (about 165 by 1,270 feet) was thrown along 
 the brow of the cliff above the Euphrates. If done by Parthian builders, this building 
 is a great achievement of Parthian architecture. If indeed we should attribute the 
 fortress to them, it should be noted that we have in the vast stronghold of Dura- 
 Salihtyah the most imposing, if not the greatest, fortress left us by the Parthians. 
 
 
 
 Fic. 19.— SHEEP FEEDING ON THE UNCULTIVATED TRACTS OF THE DEIR £Z-ZOR PLAIN 
 
 In such impressive buildings they were following traditions inherited from their 
 predecessors, the Assyrians and Babylonians, but especially the former. The impres- 
 sions made upon the Romans when they first beheld these gigantic buildings of Western 
 Asia are evident in such Roman monuments as the triumphal arch, an architectural 
 form unquestionably derived from the Parthian palace front, which was itself in turn 
 the offspring of the Assyrian palace fagade.? 
 
 But the strong city of Dura, guarding the western approaches of the Parthian 
 Empire, was long immune from western aggression. The eastward advance of Rome 
 after 200 B.c. fell far short of the plain of Khana-Mari, and the Romans had no dealings 
 with the Parthians until 92 B.c., when the latter made overtures to Sulla desiring the 
 friendship of Rome. The conquest of Syria by Pompey in 64 B.c. did not even 
 include Palmyra, which is 140 miles nearer the Mediterranean than Dura. Neverthe- 
 less the absorption of the lands around the eastern end of the Mediterranean by Rome 
 re-established the ancient commercial connection between the Mediterranean lands 
 
 1 See the author’s Ancient Times, p. 611, Fig. 248. 
 
 
 
THE HISTORY OF DURA-SALIHTYAH 41 
 
 on the one hand and the Tigro-Euphrates world on the other. On the desert roads 
 already discussed above, the two most important stations were Palmyra and Dura. 
 The blood of the ancient Aramaean and Phoenician traders who had carried the 
 Aramaean tongue throughout the vast territories of the Persian Empire from the 
 frontiers of India to the Aegean shores of Asia Minor and had distributed throughout 
 the ancient world the first alphabet—the one from which our own is descended—was 
 the dominant strain in these two remote desert towns. They could not fail to take 
 advantage of the great market which the Roman conquest had opened on the west. 
 While Palmyra is separated from the Euphrates by over 100 miles of difficult and 
 
 
 
 Fic. 20.—A Woon CaRAVAN IN THE STREETS OF DEIR £z-ZOR 
 
 Deir is the modern metropolis of the middle Euphrates, and the road from Aleppo to Mosul 
 crosses the Euphrates here. 
 
 dangerous desert, this situation proved her salvation from the Parthians on the east. 
 At the same time, when approached across the desert from the west, the oasis of 
 Palmyra is almost equally difficult of access and hence the clever Palmyrenians long 
 succeeded in escaping the clutches of the Roman Colossus. Situated thus in a position 
 of enormous commercial importance on the eastern fringes of the Roman Empire, 
 Palmyra flourished prodigiously. Her position in the east was analogous to that which 
 had been occupied in the western Mediterranean by Carthage. Both cities were of 
 Semitic origin. In both the traditions and racial instincts growing out of ages of 
 commercial domination and reaching back to the dawn of history impelled them to 
 economic leadership. As the Phoenician merchant princes of Carthage long threatened 
 
42 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 the leadership of Rome in the western Mediterranean, so the commercial power of 
 Palmyra was a serious obstacle to Roman expansion in the Orient. Antony failed 
 in an attempt to raid the rich markets of Palmyra with his cavalry in 41 B.c.; and, 
 long surviving its western prototype, this eastern Carthage of the desert pushed 
 eastward to the Euphrates, absorbed Dura, and by the end of the first generation 
 of the second century of our era was holding the Euphrates Valley far down toward 
 the Babylonian plain on the route to the Persian Gulf at SAnah.! 
 
 In this prosperous commercial kingdom of Palmyra, Dura occupied an important 
 strategic point commanding the caravan route of the Euphrates crossing, which united 
 
 
 
 Fic. 21.—Across THE Roors oF DEIR EZ-ZOR 
 
 Palmyra with the East. As I have already stated, Dura is about 140 miles almost 
 due east of Palmyra, and in a brief airplane flight which I made along the Euphrates 
 above and westward from Albu Kamal, I could see that the country west of the 
 Euphrates toward Palmyra was quite easily traversable. The engineer Gernik also 
 reports the practicability of the route across the desert between Palmyra and Dura- 
 Sdlihtyah.2 Dura thus enjoyed strategic advantages in traffic which gave it an impor- 
 tant share in the commercial prosperity of the Palmyrene kingdom to which it was 
 subject. Public buildings, the customary expression of such prosperity in the ancient 
 world, were not wanting. The earliest important building of this kind at Dura was 
 
 1Cumont, Syria, III (1922), 210 (note 1). 
 
 2 Cernik, op. cit., p. 18. The later British military surveys also show a road from SAlihtyah to 
 Palmyra. 
 
 
 

 
 
 
 
 
 THE HISTORY OF DURA-SALIHIYAH 43 
 
 a sanctuary or place of assembly, presumably of stone masonry. Its stone seats for 
 the aristocracy were found, when Cumont cleared away the rubbish, still bearing 
 carved upon each seat the name of its regular occupant, like the seats in the theater 
 of Dionysus at Athens. Each name was preceded by the date 373 of the Seleucid 
 Era, which corresponds to A.D. 61. The names display a curious mixture of Greek 
 and Semitic, naturally resulting from the commingling of races which had been going 
 on at the place for nearly four centuries. The building itself had been put up thirty 
 years before the introduction of the seats, that is, in A.p. 31, by one Ammonios, son of 
 Apollophanes, son of Seleucus, as a pious contribution to the life of the town.! Two 
 
 
 
 Fic. 22.—Tue Brings at Derr £z-ZéRr 
 
 This bridge connects the town with an island in the Euphrates. A bridge connecting the island 
 with the other (Mesopotamian) shore has long been planned but never built. 
 
 other buildings, close by, erected for purposes of public assembly, also testify to the 
 prosperity of the town at this time.” 
 
 Not more than two generations after the erection of the first building just men- 
 tioned, that is to say, not long before a.p. 100 and perhaps earlier, the flourishing 
 town of Dura erected an oriental temple to the Palmyrenian gods, the great triad of 
 Palmyra, Baalsamin (Zeus-Baal, or Zeus Megistos), Aglibol, and Yarhibol. In doing 
 so they left the group of three buildings enumerated above and withdrew to an angle 
 of the city wall at the extreme western salience of the fortifications. In a 
 
 1See Cumont, Comptes rendus (January 12, 1923), and Les Travaux Archéologiques, pp. 59-60. 
 2 See Cumont, ibid. 
 
44 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 quadrangular area where the city wall, returning upon itself, surrounded the proposed 
 building spot on three sides (see Fig. 55, p. 93), they erected the new sanctuary on a 
 ground plan which tradition had transmitted in Syria from old Babylonian temple 
 buildings. Prominent in the pious enterprise was the family of one Konon, who may 
 have been the most powerful citizen of the place. The family of Konon employed 
 Ilasamsos, an oriental painter doubtless living at Dura, to adorn the walls with 
 surprisingly ambitious paintings. That they should have had at their disposal an 
 oriental painter of such ability as Ilasamsos is striking evidence of the high level of 
 Palmyrenian civilization in the first century of our era. In the sanctuary and looking 
 down upon the temple shrine containing the sacred images of the gods, this painter 
 draughted upon the wall in splendid colors a group of eleven life-size figures, including 
 
 
 
 Fig. 23.—A Corner or DEIR £2-Z6R AND THE GARDENS ON THE ISLAND 
 
 at least three and perhaps four generations of the family of Konon engaged in worship 
 (Plate VIII, and pp. 76-88). To most of the leading figures in the scene the artist has 
 appended their names, from which we learn the identity of these forgotten worthies 
 of the Euphrates world who died eighteen centuries ago. It is a curiously unex- 
 pected fact that, in the first century of our era, a family of renown in such a 
 remote provincial city should have had themselves thus depicted on the walls of the 
 sanctuary, precisely as noble families fifteen centuries later were wont to do in the 
 Christian churches of Europe, as so commonly at Florence. It is also a very suggestive 
 fact that we should find among them, occupying a position of central importance in 
 the scene, a princely lady gorgeously appareled and splendid in adornments of gold 
 and precious stones (Plate XI). In her we must without doubt recognize just 
 such a noble Syrian lady as arose a century later in neighboring Palmyra in 
 the person of the familiar Zenobia; and it is of interest to note that in the name 
 
THE HISTORY OF DURA-SALIHTYAH 45 
 
 of Bithnanaia borne by our noble lady of Dura, we have a name of the same 
 formation as that of Zenobia (see pp. 86-87). In costume and appearance on state 
 occasions the famous Zenobia of Palmyra will have been a counterpart of our 
 Bithnanaia of Dura. 
 
 Perhaps about a quarter of a century later, that is, in A.D. 114, Lysias, a man of 
 the third generation of this noble family, erected along the north side of the temple 
 court, and abutting on the fortress wall immediately behind, a dwelling presumably 
 for the priests of the temple, as an act of piety toward the gods of the place. He 
 recorded his pious deed in an inscription in the court which was recovered in the course 
 of Cumont’s clearance. 
 
 
 
 Fic. 24.—Aras Boys or DretIr £z-Z6R 
 
 It is a question of whether such a development of the life and civilization of Dura 
 could have taken place under Parthian domination. It isnot unlikely that the growing 
 power of the flourishing town had already thrown off the Parthian overlordship before 
 the first arrival of the Roman legions. When the people of Dura looked down upon 
 the Roman eagles for the first time, as Trajan passed along the opposite (left) bank 
 of the Euphrates in a.p. 115, it is therefore probable that the Parthian garrison had 
 already long evacuated the place. But the fine strategic wisdom of Hadrian would not 
 permit him to hold for Rome a region so dangerously exposed as were the new oriental 
 conquests of Trajan. 
 
 1Cumont, Comptes rendus (January 12, 1923). 
 
46 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Later on, however, when Lucius Verus won his victory over the Parthians in A.D. 
 162, Dura! must have fallen again into Roman hands; but if a Roman garrison 
 occupied the place at this time, there is now no surviving trace of their presence. 
 Two generations later, however, in A.D. 229, a cohort of Palmyrenians numbered the 
 XXth left a dedication in the temple'to Alexander Severus, at a time when Rome was 
 endeavoring to check the rising power of New Persia under Ardashir. Of this probably 
 brief Roman occupation but a single Latin inscription survives in the place, in so 
 far as excavation has disclosed its records. It is probable, as Cumont has already 
 noted, that these Roman forces were thrown into the fortress of Dura in the time of 
 Septimius Severus to protect his advance eastward into Mesopotamia. We may 
 
 
 
 Fic. 25.—Mryabtin AND THE PLAIN oF KHAana-MarI 
 
 perhaps attribute to the Roman commanders of this period the thickening and strength- 
 ening of the city walls still observable at some points (Plate V, 1). Quite evidently 
 belonging to this period is a memorial of their sojourn left by these troops at Dura 
 on the wall of Hall I in the Zeus temple (Plate XXI)—a wall-painting in which we 
 have the easternmost representation of Roman troops now surviving. Their com- 
 mander, the tribune Julius Terentius, has had himself represented sacrificing at the 
 head of his followers to the divinities of Dura and Palmyra. The legionaries shown 
 in this painting worshiping the Tyche of Palmyra will without doubt have been 
 levied at the latter city, and are thus evincing their desire to enjoy the favor 
 of the Fortuna of their own home city, together with whom they also reverence the 
 Fortuna of Dura. 
 
 1 The evidence that this battle took place near Dura-Europos rather than at Jerablis-Europos, 
 much farther up the Euphrates, has been set forth by Cumont, Les Travaux Archéologiques, pp. 73 f. 
 
 
 

 
 THE HISTORY OF DURA-SALIHIYAH 47 
 
 We cannot discern the fortunes of the frontier stronghold of Dura during the 
 destructive wars which made the third century of our era a counterpart of the terrible 
 third century B.c. Nevertheless, some of the catastrophes of that disastrous century 
 took place not far from Dura. The young Emperor Gordian III was assassinated 
 hardly more than a day’s march from the city in a.p. 244, and the massive monument 
 which his troops raised to his memory was long a feature of the landscape in the Dura 
 region on the left bank of the Euphrates, although it has not been identified with 
 certainty in modern times. The Arab assassin Philip, who thus thought to usurp the 
 throne of the Caesars by the murder of Gordian III, was a predecessor of other and 
 similarly ambitious orientals in the Near East. At Palmyra the able prince Odenathus, 
 taking advantage of the preoccupation of the Roman emperors with wars which hurried 
 
 
 
 Fic. 26.—Tue Arip Puan or Kuana-MArRI AND THE PLATEAU HorRIZOoN 
 
 The plain is cultivated at present only on the fringes of the towns and hamlets, leaving most of 
 it an unproductive steppe. 
 
 the legions from Syria to Gaul, succeeded with the help of his gifted queen Zenobia 
 in making his desert city a formidable power. Dura must have shared in the prosperity 
 of Palmyra and felt the growing power by which Odenathus was striving to make his 
 kingdom an oriental empire like that of the Seleucids or the kings of Parthia. When 
 Aurelian’s triumph in distracted Gaul enabled him to turn his attention to the East, 
 the defeat which he inflicted on Zenobia (a.p. 273) was transformed into complete 
 destruction by the emperor’s orders to massacre the entire population of Palmyra. 
 The famous kingdom of Zenobia disappeared, and the destruction which overtook 
 Palmyra has left surviving but very scanty wreckage of that important Syrian civiliza- 
 tion of the Aramaeans of which it had been the expression. 
 
 Half a generation after the fall of Zenobia and the destruction of Palmyra, Diocle- 
 tian determined to place the frontier stronghold of the Roman Empire on the Euphrates 
 
48 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 at the north end of the plain of Khana-Mari, at the narrow defile above the mouth 
 of the Khabar (p. 24). The fortress of Circesium which he erected here is some 
 40-odd miles above Dura. Left thus over two days’ march outside the boundaries 
 of the Empire, Dura was thrown back upon itself and its desert surroundings. 
 Stripped of all protection from the desert barbarians, the people of Dura seem slowly 
 to have forsaken the place. In striking contrast with Palmyra, therefore, Dura 
 suffered no violent destruction. It lay much too far out in the desert for the legions 
 of Aurelian to carry their devastations thither. Nor did the legionaries ever occupy 
 it again. As Cumont’s excavations have disclosed, its buildings and monuments 
 fell gradually into decay and betray no sign of any violent overthrow or destructive 
 
 
 
 Fic. 27.—Tur EvpHRaATES AND THE PLAIN oF Kuana-MaArRI FROM THE PLATEAU AT SALIBtYAH 
 
 conflagration. A Christian ascetic was dwelling as a hermit in the ruins of the city 
 as early as the reign of Constantine (A.D. 324-337).1 A generation or more later, 
 when Emperor Julian passed the place on his ill-fated expedition of a.p. 363, the place 
 was already a desert. Although like other Roman commanders he marched down 
 the other (left) bank of the river, he was able to observe the complete desolation of 
 the place, where lions and gazelles had already taken possession of the dwellings of 
 men.? 
 
 In course of time the roof of the forsaken temple of Zeus-Baal fell in, and the 
 rubbish from the fall rose sufficiently high to cover and to preserve the lower zone of 
 
 1G. Hoffmann, “Ausziige aus syrischen Akten persischer Miartyrer,’” Abhandl. der deutschen 
 Morgenl. Ges., VII, No. 3 (1880), p. 28. 
 
 2 Ammianus Marc. xxiii. 5. 8: ‘‘Duram desertum oppidum’’; xxiv. 1. 5: ‘Prope civitatem 
 venimus Duram desertam.’”’ Similarly Zosimus iii. 14: "H)ée .... évredOev els Aodpa, tyvos wey ds apa 
 more WONLs HY Hepovoay, TOTE 5é Epnuov. 
 
 
 
THE HISTORY OF DURA-SALIHIYAH 49 
 
 paintings from destruction at the hands of fanatical Arabs. Over the whole place 
 a mantle of sand carried in by the desert winds slowly collected until it lay over the 
 town—in some places as deep as 2 meters. Along the southwest side facing the 
 
 
 
 Fiq. 28.—GuIMpPseE OF THE BAZAARS IN BAGHDAD 
 
 desert the great drifts of sand rose almost to the parapet of the fortress walls and 
 today mask the height of the masonry and reduce it to ignoble proportions (Plate I, 2). 
 
 With the decay of all centralized political power, the plain of Khana-Mari, lying 
 as it then did outside the Roman Empire, rapidly declined. Its absorption by New 
 Persia and its conquest by the Moslems in the seventh and eighth centuries of the 
 
50 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Christian Era left it without leadership, and its resources were not worth the attention 
 of the Moslem conquerors. The desert encroached upon its fields, the irrigation canal 
 from the Khabir was choked up, there was insufficient grain and other food, and the 
 population steadily decreased. Thus the towns were forsaken as the fields shrank 
 to a scanty fringe around each village, and the solitude of the prehistoric wilderness 
 from which it had originally come settled upon the plain of Khana-Mari. Upon this 
 desolate wilderness, once green with far-stretching fields, the forsaken walls of the great 
 fortress of Dura looked mutely down for centuries. As the Arabs wandered in from the 
 desert and strolled over it, they naturally associated a fortress so imposing with the 
 heroic memories of Saladin (Sal4h ed-Din), as they commonly do, and like a number 
 of other such places they called it Salihtyah, ‘‘place of Salah.”’ Thus its ancient 
 
 
 
 Fig. 29.—Tue Brince or Boats OVER THE TIGRIS AND PALM GROVES AT BAGHDAD 
 
 Assyrian name of Dura and its later Macedonian name of Europos were finally forgot- 
 ten, and survived in modern times only in the few scanty historical references in ancient 
 Greek and Roman writers which have been cited above. Nor, as we have seen, had 
 modern historians any idea of the identity of the ancient stronghold. For many 
 centuries since its first establishment by the Assyrians or possibly the Babylonians 
 and for some six hundred years more after its foundation by the great Macedonian 
 general, it had guarded the crossing of the Euphrates where the traffic of East and 
 West passed over the historic river, and thereafter it was a lost city for sixteen centuries. 
 
 Into its silent defenses General Cunningham, the commander of the British Army 
 on the Euphrates in the Great War, after a sharp engagement with the Arabs, threw 
 his farthest outpost early in 1920.1 These British representatives of the greatest of 
 
 1 There was a small scouting detachment which was stationed a few miles farther up the river 
 
 and for a time there was British occupation in Deir ez-Zér, but Sélihtyah marks the uppermost 
 position of the British Army on the Euphrates in 1920. 
 
 
 
THE HISTORY OF DURA-SALIHIYAH 51 
 
 modern empires probably little dreamed of the significance of this place in the history 
 of the greatest of ancient empires. Nor could they or anyone else, when they entered 
 the fortress of Sdalihtyah, have had any premonition of the unfolding vista of ancient 
 human life and development which was to result from the British occupation of the 
 venerable stronghold. All about them lay a rich treasury of human documents from 
 remote ages. In the relatively dry climate which this region enjoys, the winter rains 
 had not penetrated far into the mantle of protecting sands. Beneath this covering 
 lay almost perfectly preserved the monuments of this ancient border land between 
 East and West. Their recovery has revealed to us something of the history of Khana- 
 Mari as we have sketched it above, and disclosed to us monuments of art which 
 anticipated and which contributed essentially to the development of art in Europe. 
 
 The story of the recovery of these monuments furnishes the final episodes in the 
 history of Dura, accompanied unhappily by the melancholy fact of the destruction or 
 serious damage of the great painting which forms the chief subject of this volume. 
 The vandalism of Arab ignorance and fanaticism, to which this destruction is due, is 
 now only held in check by temporary French occupation, while Cumont continues for 
 a time the effort to save what may yet be rescued from a place so remote and difficult 
 of access. In the long and often repeated alternation of eastern and western power 
 in the plain of Khana-Mari, which has gone on now for some five thousand years, 
 the occupation of Dura by the British and its present permanent tenure by France 
 under her Syrian mandate are the renewed extension of western supremacy in the 
 ancient Near East, so often asserted and so often lost. 
 
II 
 
 THE DISCOVERY OF THE DURA-SALIHIYAH 
 PAINTINGS 
 
 The route down the Euphrates, as we have observed (p. 48), followed the east 
 or left bank of the river throughout ancient history until the Arab occupation. 
 Modern explorers desirous of following the ancient road have commonly passed 
 along the east side of the river, and Dura-Salihtyah, lying on the west (right) bank 
 has usually found but casual mention, as various travelers have recorded their impres- 
 sions in viewing it across the stream. Thus Sachau went up the east bank and Miss 
 Gertrude Bell went down the same bank in 1909. In 1872, however, the engineer 
 Cernik visited the site and published a brief summary of his observations.! It was 
 not until 1898 when F. Sarre visited Salihtyah that archaeological examination of the 
 place was begun. He inspected it several times subsequently—for the last time in 
 1911-12—but did not publish his results until 1920, in combination with E. Herzfeld 
 in their monumental work on the ancient remains in the Tigro-Euphrates world.? 
 
 On the return of the Mesopotamian expedition of the Oriental Institute of the 
 University of Chicago from the upper Tigris and its arrival in Baghdad on April 23, 
 1920, the Civil Commissioner, Colonel A. T. (now Sir Arnold) Wilson, and General 
 (now Sir) Perey Hambro, the Quartermaster General, sent me information to the effect 
 that they would be glad if we could undertake to ascend the Euphrates on an archaeo- 
 logical mission on their behalf. Colonel Wilson explained to me the circumstances 
 which prompted his request and handed me a small file of papers which he said would 
 further inform me of the reasons for the desirability of our journey up the Euphrates. 
 
 On opening the papers I found an unfinished drawing in colored crayon of four 
 human figures standing in a row and beside them a series of letters indicating the places 
 of four additional figures in the same row. Under the sketch was a written note 
 reading: ‘‘Rough sketch of a wall decoration excavated at Salihiyeh” [Salihtyah]. 
 The drawing and the note were made by General Cunningham, who was in command 
 on the upper Euphrates. My curiosity was naturally at once aroused and, looking 
 farther through the papers, I discovered a dispatch addressed to Colonel Leachman 
 
 1 Cernik, op. cit., p. 17. 
 
 2 Friedrich Sarre and Ernst Herzfeld, Archdologische Reise im Euphrat- und Tigris-Gebiet (4 vol.), 
 Vol. II, pp. 386-95, Abb. 371-77, and Vol. III, Plates LXXXI-LXXXIII, together with a plan 
 of the fortress by Schulz. I regret that when my article in Syria, first publishing the Salihtyah 
 paintings, was written I was not yet acquainted with this valuable work, as it had not reached our 
 library. 
 
 52 
 

 
 DISCOVERY OF THE DURA-SALIHIYAH PAINTINGS 53 
 
 and signed by Captain M. C. Murphy. It was dated March 31, 1920, at ‘ Albukemal”’ 
 and read as follows: 
 
 While at Salihiyah I discovered on 30th inst. some ancient wall paintings in a wonderful 
 state of preservation. The paintings are in the west corner of the fort and consist of life-size 
 figures of three men, one woman, and three other figures partly obliterated. The colours 
 are mainly reds, yellows and black. There is also some writing which I have tried to reproduce 
 below. 
 
 I should be glad if you would forward this to the proper quarter. 
 
 (signed) Murpny, Carr. 
 Re ASV. 
 
 This dispatch was covered by a note from Lieutenant Colonel Leachman to the 
 Civil Commissioner which I read with unusual interest: 
 
 As a result of our occupation of the old fort at Salahiyeh and the digging of trenches, a 
 certain amount of finds have been made. The paintings to which the attached refers are 
 most interesting & should, I think, be seen by an expert. If your American archaeologist 
 is still about, it would well repay him to come & see this. The films enclosed are of the 
 pictures. Could you please have them developed. If anyone comes up, it should be soon 
 
 for obvious reasons. . 
 (signed) G. LEACHMAN 
 
 The papers had evidently been sent to Miss Gertrude Bell, the intrepid explorer 
 and well-known archaeologist, now holding an important post in the British administra- 
 tion in Mesopotamia. Miss Bell had promptly responded as follows: 
 
 Look at the enclosed—it seems to me to be most curious & interesting. A. T. [Wilson] 
 
 says if Prof. Breasted will go & look at it he’ll provide him with transport & would be very 
 grateful for his advice. We both beg him to fit it into his programme if he possibly can. 
 
 The whole correspondence was covered by a letter from the Civil Commissioner 
 to General Sir Perey Hambro requesting me to call and stating that the Civil Commis- 
 sioner would ‘‘be very glad if he [Breasted] could be sent, presuming that G. H. Q. 
 agree to his entering what is very much of a war area.” 
 
 As the British authorities had thus far thought it unsafe to allow our expedition 
 to go up the Euphrates more than 100 miles at most, for reasons disclosed in the 
 foregoing correspondence, I seized the opportunity with the greatest pleasure; but 
 asked for a fortnight to be spent among the monuments on the Persian border first. 
 I had noticed the statement in Colonel Leachman’s letter that “If anyone comes up 
 it should be soon for obvious reasons.”’ In strict confidence the Civil Commissioner 
 then informed me that there would be no time for us to make the trip to Persia because 
 a withdrawal known only to the High Command was about to be undertaken—an 
 operation which would shift the British frontier on the Euphrates almost 100 
 miles farther down the river. This withdrawal was occasioned by difficulties in a 
 line of transport communications excessively long. If we went to Persia first, therefore, 
 
54 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 the paintings would by that time le far out in Arab territory beyond the British 
 lines, where they would be absolutely inaccessible without the protection of the 
 troops, who could not be dispatched thither again. Furthermore the paintings 
 would be exposed to the fanaticism of the bigoted Arabs, who would undoubtedly 
 destroy them, as indeed in spite of precautions they have done since our departure 
 from the place. It was evident that we should leave for the upper Euphrates at 
 once. 
 
 On April 28 in seven automobiles kindly furnished us by the British Army and 
 Civil Government we left Baghdad for the upper Euphrates.! The accidents and 
 
 
 
 Fic. 30.—BritisH GUNBOAT ON THE LOWER EUPHRATES AT FALLUJAH 
 
 This is the usual point for crossing to the south (or west) side of the Euphrates on the Baghdad 
 Aleppo route (see Map 1). When we reached it, the bridge of boats (like that at Baghdad) had 
 been carried away; we were obliged to ascend to the bridge at Ramdédi, where we were overtaken by 
 darkness among the hostile tribes who shortly afterward murdered the intrepid Colonel Leachman. 
 
 delays of desert travel by automobile were such that the nearly 300-mile trip from 
 Baghdad to the British frontier on the middle Euphrates occupied an entire week. 
 This was particularly unfortunate, for it deprived us of the time which we had hoped 
 to devote to the study of Salihtyah and its paintings. Twenty-seven miles from 
 Salihtyah, that is, below it, we found the British headquarters at Albu Kam4l, where 
 we were very kindly received by General Cunningham and his officers. Lieutenant 
 
 1 This journey has been briefly described in Oriental Institute Communications No. 1, pp. 25-32. 
 
 
 
 
 
DISCOVERY OF THE DURA-SALIHTYAH PAINTINGS 55 
 
 Colonel Leachman, of whose tragic death at the hands of the Arabs on his way down 
 the Euphrates a short time after our arrival we had no premonition, even cleared 
 out his office to make room for our five field beds. To all these gentlemen at Albu 
 Kamal, whose many kindnesses we remember with gratitude, it is a pleasure to 
 express our appreciation here. To Lieutenant Colonel Leachman it was chiefly due 
 that we were able to continue our journey beyond Salihtyah and up the Euphrates 
 to Aleppo and the Mediterranean. We were thus, I believe, the first western expedi- 
 tion to cross the Arab state from Baghdad to the Mediterranean after it was proclaimed. 
 It is not necessary to repeat the journal of this, for us, memorable journey, but it 
 will contribute to a correct understanding of the physical background of the civilization 
 
 
 
 Fig. 31.—Tue SEVEN Cars oF OuR CARAVAN BETWEEN FALLtJAH AND RAMADI ON THE LOWER 
 EUPHRATES 
 
 with which we are dealing to include a rapid panorama of the route from Baghdad 
 to the Mediterranean, which the reader will find in the accompanying photographs 
 (Figs. 28-50). 
 
 As soon as possible after we reached Albu Kamal General Cunningham kindly 
 drove with me the 27 miles from the Albu Kamal headquarters to Salihtyah for a 
 preliminary examination of the paintings on the afternoon of May 3. On the way 
 he showed me the battlefield on which the British had but a short time before inflicted 
 a severe defeat on the Arabs, who were nevertheless still maintaining a hostile attitude. 
 Sniping and desultory fighting, such as had preceded the battle, were still going on. 
 Indeed after we left ‘Anah on our way up, the hostile tribes had been firing across the 
 river at various points, and we realized that Sir Perey Hambro was quite right in 
 
56 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 calling the region ‘very much of a war area.’’ Driving in at the eastern or south- 
 eastern angle of the great fortress of Salihtyah, we found British East Indian troops 
 camping within, alongside the vast northern castle, under the command of Major 
 Wright-Warren. I was surprised at the extent and massiveness of the enormous 
 stronghold as we drove along its entire length parallel with the Euphrates and drew 
 up near the western angles of the wall. Descending from the car, General Cunningham 
 led the way over the rubbish piles commonly found in such ruins and around a jutting 
 corner of massive masonry. Suddenly there rose before us (Fig. 53, p. 89) a high wall 
 covered with an imposing painting in many colors depicting a life-size group of eleven 
 
 
 
 Fig. 32.— BEGINNING OF THE PLATEAU ABOVE RAMADI AND BELOW HIT 
 
 In this region the Baghdad-Aleppo road passes out of the alluvial plain of Babylonia and ap- 
 proaches the narrow valley of the Euphrates eroded in the desert plateau. The cliffs of the plateau 
 are seen in the background; in the foreground, trenches from the Great War. 
 
 persons engaged in worship. My surprise at the extent of the vast fortress now gave 
 way to amazement, as I gazed at these wonderful figures looking gravely down upon 
 us, and as suddenly disclosed as if they had been conjured up by magic from the silent 
 wastes of the desert which stretched out far below us. It was a startling revelation 
 of the fact that in this deserted stronghold we were standing in a home of ancient 
 Syrian civilization completely lost to the western world for sixteen centuries. A 
 hundred and forty miles nearer the Mediterranean lay the desolate ruins of Palmyra, 
 which Zenobia and her husband had made the beautiful capital of a powerful com- 
 mercial empire. Here on the painted wall (Plate VIII) I saw, emerging from what 
 
 
 

 
 DISCOVERY OF THE DURA-SALIHIYAH PAINTINGS 57 
 
 might have been a palace doorway, a queenly lady arrayed in royal splendor, with 
 gentlemen richly clad ranged on her either hand. Could this be Zenobia herself 
 looking out upon us from the sanctuary wall in one of her dependent cities of sixteen 
 centuries ago? Then my eye fell upon written words traced across the front of her 
 skirt, and I read in Greek: “Bithnanaia, daughter of Konon.” It was not Zenobia, 
 but she and her family were obviously heirs of Graeco-Syrian culture represented 
 here by precious memorials, for recording and preserving which I realized we had 
 but a few brief hours before the British evacuation would leave us without protection 
 and force us to depart. 
 
 We stood in a roofless structure of massive masonry, and it was evident that the 
 wall paintings were in a transverse hall, which must have been the holy of holies of a 
 
 
 
 Fig. 33.—Our Cars STALLED NEAR HADiTHAH IN A DESOLATE STRETCH BETWEEN HiT AND ‘ANAH 
 
 sanctuary of some kind. The roof, undoubtedly an arch or vault, had been relatively 
 high and above the great painting there were still clearly visible the lower portions of 
 a second upper range or zone of paintings, of which almost all had been carried away 
 by the falling of the plaster (see Plate VIII and p. 76).! It was evident that, besides 
 studying the painting, we must also investigate the building. At General Cun- 
 ningham’s request Major Wright-Warren agreed to furnish several squads of East 
 Indian troops for the further excavation of the building, in the endeavor to clear 
 the ground plan and also to lay bare other painted walls, which we could see in the 
 
 1J7¢t must have been these traces of the upper row which were seen by Sarre. He remarks that 
 he saw ‘“‘Reste eines Fresko-Gemildes in einem Gebaude nahe der W.—Ecke der Stadt, aus rustica- 
 aihnlichen Gipsquadern errichtet’’ (Sarre-Herzfeld, op. cit., p. 392). 
 
58 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 hall preceding that of Bithnanaia. Having made these arrangements, we returned 
 to spend the night at headquarters in Albu Kamal. 
 
 Karly on the morning of May 4 we bade our kind hosts at headquarters farewell. 
 We were not to see them again. Major Wright-Warren’s detachment was about 
 to abandon its outpost at Salihtyah and return to headquarters at Albu Kamal, 
 whence all were to withdraw down the river. They wished us success and good luck 
 on our coming dash through the Arabs and up the Euphrates to the Mediterranean. 
 It was still early when we drew up alongside the sanctuary of the paintings. The 
 East Indians were already at work clearing the walls, and, in the first chamber (Hall I), 
 we saw to our surprise a small scene in which a Roman tribune was depicted at the 
 
 
 
 Fig. 34.—Our Britiso Convoy Harting at Nantyau JUST ABOVE CANAH 
 
 In this region between <Anah and Salihtyah the hostile Arabs on the other (left) shore of the 
 Euphrates were shooting across the river, and the protection of a military convoy was necessary. 
 
 head of his troops, engaged in the worship of what looked like three statues of 
 Roman emperors painted on the wall. There was the tribune’s name written beside 
 him in Roman letters: “Julius Terentius, tribune,” and before him was the red battle 
 flag of Roman troops, the only one of which the color is preserved, reminding us of 
 Plutarch’s tale of the red flag hoisted over the tent of Varro before the battle of Cannae. 
 Here were these tokens of Roman occupation full 35 miles outside of the well-known 
 Roman frontier at Circesium by the mouth of the Khabir, which we were not expecting 
 to see until we had put a two days’ march behind us. We had before us the eastern- 
 most Romans ever found on the Euphrates, or anywhere else for that matter. Mean- 
 time the ground plan, as the Indian troops continued to work, was beginning to emerge. 
 
 
 

 
 DISCOVERY OF THE DURA-SALIHIYAH PAINTINGS 59 
 
 In spite of two columns discernible in front of the two painted halls, it was now evident 
 that in plan the building was clearly identical with that of the old Babylonian temples 
 of much earlier age. 
 
 With this archaic architecture following traditions which reached back to remote 
 ages of Babylonian culture, and its rich mural decorations revealing both Graeco-Syrian 
 and Roman civilization, the building was a historical document of the first rank; 
 while the enormous stronghold in which it was imbedded equally required to be 
 recorded and made available. Here was a task which might occupy a large expedition 
 for months or years; and we had, before the British evacuation, but a single day in which 
 to record and rescue this great body of documents. A moment’s reflection made it 
 evident that we must devote ourselves to the great painting on the wall of Bithnanaia 
 
 
 
 Fig. 35.—CaMELS OF THE British MILITARY TRANSPORT BELOW ALBU KAMAL 
 
 in preference to everything else. One of the faces in this remarkable painting (Plate 
 XIX)—and General Cunningham had informed me it was the finest face in the entire 
 group—had already been barbarously scratched out, presumably by some prowling 
 Arab who had succeeded in gaining entrance to the place. This disfigured face 
 which had survived for eighteen centuries, to suffer destruction after it seemed to be 
 safely in our hands, was a distressing demonstration of what would happen after we 
 left, and admonished us to save the rest if possible. I therefore asked Dr. Luckenbill 
 to take our largest camera, which was only 5X7 inches, and figure by figure to 
 make photographs of the great painting, on as large a scale as possible, while I made 
 detailed archaeological notes on the entire composition, including full data as to the 
 colors and their distribution. Although I had never used the method before, it 
 seemed probable that my field notes would make it possible, after our return to civiliza- 
 tion, to insert the approximate colors of the original in enlarged prints of the field 
 negatives. This at least was the only method available in the lamentably brief 
 
60 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 time at our disposition. The other gentlemen were asked to make a sketch plan and 
 to insert in it all the dimensions which they found it possible to measure, as fast as the 
 Indian troops laid bare the ground plan. 
 
 The sun was low in the west when Dr. Luckenbill and I had finished our work on 
 the wall of Bithnanaia and the other fragmentary paintings in Hall Il. We then 
 turned our attention to Hall I, and especially to the painting of the Roman tribune. 
 The excavation of the hall was confined to a trench skirting the painted walls, and the 
 deep rubbish which filled the room prevented placing the camera properly. Further- 
 more, the great painting had exhausted all our 5X7 inch plates but one. We dug 
 a recess for the camera as best we could in order to get it low enough and far 
 enough from the wall, but were unable to secure a remove from the wall sufficiently 
 
 
 
 Fig. 36.—BritisH HEADQUARTERS AT ALBU KAMAL 
 
 far to insure the inclusion of the whole painting: one of the three emperors (as we 
 thought them) and one of the two goddesses fell outside the range of the instrument. 
 We had been racing with the declining day and the fading light. It was not possible 
 to dig out more of the rubbish and secure a better position. The exposure must be 
 made at once or there would be no light with which to make it. With the sun on the 
 western rim of the desert the slide was drawn and the last plate we had took a faint 
 twilight impression of the easternmost Romans ever found depicted on a wall. 
 
 There was nothing more to be done in securing a fuller record of the paintings, 
 except to edit one’s notes while one’s memory was fresh and vivid. The younger 
 members of the party made a tour of the city walls, doing as much pacing as possible 
 in order to gain some impressions of the size and extent of the stronghold.! 
 
 1J7t was impossible in this hasty inspection to secure data for an accurate plan of the fortress, 
 and the plan which I first published in Syria needs serious correction. See p. 63. 
 
 
 
DISCOVERY OF THE DURA-SALIHIYAH PAINTINGS 61 
 
 Realizing fully the hasty character of our observations and the unparalleled 
 opportunity presented by the place, we were a weary and rather discouraged party 
 as we crept into our field beds that night. Nevertheless we stowed away the plates 
 and the notebooks with the consciousness that more could hardly be expected from 
 a single day’s work. In the gloom preceding the dawn we devoured a hurried and 
 chilly breakfast, and, bidding Major Wright-Warren farewell, we climbed into our 
 Turkish arabanahs, joined presently by several mounted Arab rifles, an escort of 
 friendly Arabs which we owed to the thoughtfulness of Lieutenant Colonel Leachman. 
 The arabanah, drawn by horses, is not a vehicle suited for speed, and it seemed tanta- 
 lizingly slow after the automobiles, which we had of course surrendered to our British 
 friends on leaving their lines. Having left the protection of the British troops, 
 therefore, we did not find a dash across no man’s land in an arabanah a wholly reassur- 
 ing experience, and, for the sake of our Salihtyah records at least, we were not a little 
 relieved when we were met by an escort sent out by the pasha from Deir ez-Zér to 
 receive us. We did not then know what we afterward learned, that a Jewish merchant 
 approaching Deir ez-Zér had just been relieved by his escort of some thousands of 
 pounds in gold which he had in his bags. It was, nevertheless, evident that our 
 situation was too unsafe to permit us to visit the little Arab village of Tell ‘Ashara, 
 perched on the mound covering Tirka, the ancient capital of Khana. Leaving it at 
 some distance on our right, therefore, we traversed the plain of Khana and spent 
 our first night among the Arabs in Meyadin. It does not fall within the purpose of 
 this volume to recount our further adventures on this memorable journey through 
 the tribes of the upper middle Euphrates, through whom we carried our records in 
 safety. In another fortnight we were safe within the zone of civilization at Beirit, 
 where Dr. Luckenbill developed his excellent negatives and we contemplated with 
 great satisfaction the figures of our ancient Syrian friends from the walls of Salihiyah, 
 figures which my field notes would enable us to endow with the hues and colors of life. 
 
CUT 
 THE CITY AND FORTRESS OF DURA 
 
 The plain of modern Deir ez-Zér, which furnished the agricultural basis for the 
 support of a tiny kingdom, has already been discussed in the historical sketch of the 
 city (Chapter I). A salience of the desert plateau from the west and southwest, 
 thrusting forward across the plain clear to the river at Dura-Salihtyah, forms a 
 buttress, quite isolated and protected on three sides by its natural situation. This 
 buttress is about 80 meters in height above the level of the river.! As may be seen 
 very clearly in the airplane view (Fig. 55, p. 93), the promontory drops on the north 
 sheer to the channel of the Euphrates, which here flows east by south, curving at the 
 same time directly to the east. The actual bed of the river has in modern times 
 shifted away from the city, under the walls of which it must once have flowed. The 
 two flanks of the natural buttress on northwest and southeast are sharply cut off from 
 the rest of the plateau by two precipitous wadzs (AB and CD in Fig. 55), leaving only 
 one side of the inclosed promontory, that is, the southwest (BC), accessible to attack. 
 We have already seen that the strength of the position was not overlooked by the 
 Assyrians, whose occupation of it gave it its name of “Stronghold” (dar). It is 
 perhaps not an accident that the only wall of the fortress whose direction is not dictated 
 by the local topography, that is, the wall on the weak southwest side toward the plateau 
 (BC), extends in a line exactly northwest to southeast, which is the customary direc- 
 tion for the sides of buildings and cities in ancient Babylonia and Assyria. 
 
 It was on this buttress of the plateau that the Macedonian Nikanor laid out his 
 city, on the customary rectangular, four-square plan. Our hurried stay afforded us 
 no opportunity to examine the area within the fortress. The credit for discerning 
 the date of this settlement, on chiefly archaeological grounds, must be given to Sarre 
 and Herzfeld.2 The preliminary study of the city by Cumont, who was able to employ 
 the spade there, amply confirmed this date and, although it has not yet proceeded far 
 enough to reveal the arrangement of the city plan, it has furnished an account of three 
 
 1 This figure is taken from Cernik (op. cit., p. 17). My own impression would not have made 
 
 the place as high as Cernik reports; but he made careful profiles across the valley in this region and 
 as an engineer is without doubt to be relied on. 
 
 2 They describe it as “eine Ruinenstadt, die nur kurze Zeit bestand; die hellenistische Schicht 
 ist die einzige und liegt unmittelbar unter der Oberfliche und auf dem gewachsenen Felsen. Die 
 Zeit dieser Besiedelung ist . . . . die Wende unserer Aera, das erste christliche Jahrhundert. Das ist 
 aber in dieser Landschaft die Zeit Palmyras”’ (op. cit., p. 394). Without the evidence of the paintings 
 and additional inscriptions, not yet available at the time of Sarre and Herzfeld’s studies, it was 
 impossible to discern that the city had endured in reality some six hundred years; but otherwise 
 
 62 
 
 
 

 
 THE CITY AND FORTRESS OF DURA 63 
 
 public buildings and several houses (pp. 42-438 and 38). Neither are we as yet in a 
 situation to offer any final discussion of the great walls which surround the city. 
 Nevertheless, in view of the fact that it is likely to be long before any final investi- 
 gation of the entire fortress can be made, a summary of the available information 
 may be of some service. 
 
 The great stronghold stretches across the entire promontory from wadi to wadi 
 (AB to CD). The walls therefore follow the margin of the desert buttress, skirting 
 the Euphrates side and the two wadzs; but on the southwest side crossing the promon- 
 tory from the upper end of the northwest wad (B) to a point near the upper end of 
 the other or southeast wadz (C). This was the only side on which the fortress was open 
 to attack from the level face of the desert plateau on the southwest (see Fig. 
 55 and Pl. IL). It is some four-fifths of a mile from wadz to wadi and the longest 
 dimension of the great inclosure in a general northwest-southeast line is about 1,400 
 meters as measured by Sarre and Herzfeld.! Its greatest width is about 1,000 meters. 
 Along the river front (AD) the northern line of the fortress is about 850 meters in 
 length, while the opposite wall, facing the plateau on the southwest side (BC), is about 
 1,175 meters long. A large part of the inclosed area along the southwest wall and 
 inward toward the center of the city is much higher than along the river wall, so that, 
 as the ground falls away, the north or northeast wall along the Euphrates (AD) is 
 much higher than elsewhere. On this somewhat lower ground along the height 
 overlooking the river was erected an enormously massive castellum of vast size. It is 
 50 meters wide and 390 meters, that is, almost a quarter of a mile, in length. Its outer 
 corners are strengthened by salient towers. The only well-preserved gate (H) at 
 present observable is in the southwest wall (BC). It is square in ground plan like 
 the towers distributed at intervals along this wall and distinguishable in the airplane 
 view. The masonry is composed of white, calcareous stone, called by Sarre and 
 Herzfeld ‘‘Gipsstein.”’ The blocks are cut with rectangular faces and are of the same 
 height throughout the same course. In size they vary greatly but are most frequently 
 about one fourth to half a meter in height. 
 
 The date of this important stronghold is still not entirely settled. Regarding the 
 gate (£) in the southwest wall, Sarre and Herzfeld state that in its arrangement it is 
 
 their determination of the date of the place is entirely correct. Similarly their conclusion that it 
 was never included in the Roman or Byzantine Empire, but that it had flourished as an outpost of 
 Palmyra, like the town Zenobia on the northeast of Palmyra, is only slightly to be modified by the 
 new data now available. : 
 
 1Qp. cit., pp. 388-90. All the other dimensions of the fortress employed in the foregoing discus- 
 sion are likewise from this useful survey, which includes the first plan of the walls ever issued. The 
 hasty sketch-plan which I included in my article in Syria is very incorrect and should now be super- 
 seded by the plan of Sarre and Herzfeld (surveyed by Schulz) and the airplane view above published 
 for the first time. At this writing Cumont’s new plan, which will displace all but the airplane view, 
 has not yet appeared. 
 
64 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Byzantine in type. It is possible that the city walls were not the work of any one 
 generation, and that this gate is a later construction. Two different constructive 
 stages are quite certain in the cross-section of wall shown in Plate V. For the portion 
 of the city wall overlooking the northwest wadi (AB) and containing the Zeus-Baal 
 temple, however, a date not later than the latter half of the first century of our era is 
 certain. 
 
 The structural chronology is clear at this point (B). An examination of Plate 
 VI, 1, discloses that the temple of Zeus-Baal at the extreme western corner of the fortress 
 (see B, Fig. 55, p. 93) was erected in a salience of the fort which is protected by walls 
 of great height and massiveness. The builders of the temple made as much use of the 
 
 
 
 Fig. 37.—STARTING FOR AN AIRPLANE EXAMINATION OF THE KHANA-MARI PLAIN 
 
 Through the kindness of General Cunningham, Dr. Luckenbill and the author were given an 
 opportunity to study the region above Albu Kamal from the air. 
 
 fortress masonry as they could. As we shall see in examining the temple (chapter iv), 
 the south wall of Hall IT was already in position when the builders erected the temple. 
 They utilized it as the south wall of the temple at this point. They plastered this sec- 
 tion of the city wall and Ilasamsos covered it with his great painting of the family 
 of Konon and Bithnanaia (p. 76). The city wall at this point, therefore, was already 
 in existence before the execution of this painting of Ilasamsos, dated to the latter part 
 of the first century of our era. As we have already indicated (p. 45), no Roman had 
 ever seen this place before the invasion of Trajan in a.p. 115. It is quite evident, 
 therefore, that this part of the city wall is pre-Roman in date, and we must therefore 
 conclude that the massive defenses along the northwest wadi were already in existence 
 
 
 

 
 
 
 
 
 THE CITY AND FORTRESS OF DURA 65 
 
 before the Roman invasion of this region. Furthermore, these northwest walls 
 seem to me in character precisely like the other ramparts of the city. Dating the 
 fortress as pre-Roman, then, we must attribute it either to the Parthians or to the 
 Macedonians, and I am inclined to attribute the great stronghold of Sdlihtyah to the 
 Parthians. 
 
 As far as the history of the frontier of the Roman Empire on the Euphrates is 
 concerned, acceptance of the Parthian origin of the fortress of Dura would relieve 
 us of the troublesome problem of accounting’ for the existence of such an important 
 Roman stronghold in this region a generation before it was conquered by Roman 
 
 
 
 Fig. 38.—Our WaGcons DRAWN UP FOR DEPARTURE FROM THE FortRESS OF DuRA 
 
 It was too early in the morning for successful photography. The five Syrian wagons (arabanahs) 
 replaced our seven automobiles which were the property of the British and were returned to them at 
 Albu Kamal. In these wagons our expedition journeyed from Dura-Salihtyah to Aleppo. Apology 
 is due Major Wright-Warren for the unfair flank movement executed by the camera squad before he 
 was able to don his uniform. 
 
 troops. It is quite clear that the place was occupied by a Roman garrison in the third 
 century, and at that time the walls may have been strengthened or their thickness 
 increased as shown in Plate V, 1. But there is nothing in the discoveries at Dura to 
 indicate that a stable and lasting Roman frontier was established here two days’ 
 march outside the permanent boundary established by Diocletian 35 miles farther 
 up the Euphrates at the mouth of the Khabir. 
 
 Regarding the identity of our fortified town with the Dura of the ancient sources, 
 let it be noted that, according to Isidor of Charax, Dura was situated at a distance 
 
66 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 of 10 schoinot below the mouth of the Khabir,! that is, a matter of some 35 miles, 
 just the distance between Salihtyah (Dura) and the mouth of the Khabir (see above, 
 p. 24). In a.p. 363 when the Emperor Julian followed the left bank of the Euphrates, 
 after passing the Khabfr he arrived in one day’s march at Zaitha or Zautha, after 
 which the second day’s march brought him to Dura.2 While the ancient sources 
 thus speak of the town as if it were on the left bank of the river, it is quite evident 
 that there could not have been more than a settlement guarding the Mesopotamian 
 shore and making the bridgehead secure. Such references to the arrival of a traveler 
 
 
 
 Fic. 39.—Our Wacons LoapDING FoR DEPARTURE FROM Mbryap{N, AFTER THE First Day’s Marco 
 
 at a city on one shore of a river, while the traveler himself may really have been on the 
 other shore, are quite comprehensible and occur elsewhere. In view of the reference 
 to Dura which we found in the painting of the tribune, and the evidence found by 
 Cumont disclosing also Europos as the other name of the town, there can be no doubt 
 regarding the identity of the place now called Salihtyah. 
 
 1Jsidor of Charax, Mans. Parth., 4 (ed. C. Mueller, Geogr. graeci minores, p. 248): From the 
 walled town of Nabagath, past which the Khabar flows, 4 schoinoi to Asicha; thence to Dura, 
 6 schoinot; total, 10. 
 
 2See Ammianus Marc. xxiii. 5. 8 and xxiv. 1.5; also Zosimus iii. 14, and compare Isidor of 
 Charax, loc. cit. G. Hoffmann (op. cit., p. 166) discusses the position of Dura and arrives at the 
 conclusion that it must be found nearly opposite the ruins of Kan-kaleh by Salihtyah. For the 
 verification of the foregoing reference I am indebted to Cumont’s kindness. 
 
 
 
THE CITY AND FORTRESS OF DURA 67 
 
 Later.—Since the above was written, and just as this book goes to the press, I 
 have received the following important communication from M. Cumont: 
 
 Since your departure I have received the report of Commandant Renard on our excavations. 
 It contains especially a study of the fortifications of Dura, accompanied by plans. In com- 
 paring it with the treatise of Philo of Byzantium (toward B.c. 200) I have gained the con- 
 viction that the actual walls, except certain partial rebuildings, are those which Nicanor 
 raised when he founded the Greek colony. I made a communication concerning it to the 
 Académie last month. It will ultimately appear in Syria. I did not fail to call attention 
 therein to your observation on the date of the tower where the great fresco is painted. It 
 had, furthermore, also been made by Commandant Renard. 
 
 There can be no doubt, therefore, regarding the pre-Roman date of the fortress 
 of Dura. The doubt concerns only the choice between the Macedonians and the 
 Parthians. M. Cumont believes the ‘‘walls’ were the work of Nikanor, intending 
 probably to leave the date of the vast castellwm undetermined as yet. In any case 
 it is obvious that the stronghold of Dura was built on oriental plans. It is oriental 
 in architecture and displays just the characteristics which are found in Parthian 
 buildings. If the fortress was built by Nikanor, he was following oriental models 
 and we have in it, for example, one of the few Greek buildings which contains the arch. 
 
IV 
 
 THE TEMPLE OF ZEUS-BAAL IN THE FORTRESS 
 OF DURA 
 
 The builders of the Zeus-Baal temple selected a site quite removed from the 
 center of the city, whose rectangular block plan is clearly discernible in the airplane 
 photograph if not in the halftone print employed in this volume. Almost at the apex 
 of the westernmost angle of the fortress is a rectangular bastion which was chosen 
 to contain the temple of the gods of the city. The walls of this bastion at this point 
 inclose a rectangular area oriented almost exactly to the four cardinal points.) In 
 order to utilize the fortress walls as those of the temple, the latter was also oriented 
 in the same way. ‘The main axis of the temple therefore lies in an east and west line, 
 with the entrance at the east and the shrine at the west. In this particular it is unlike 
 the old Babylonian and Assyrian temples, which lie with their corners to the cardinal 
 points. In ground plan, however, this temple of Zeus-Baal is identical with that of 
 the old Babylonian temples. For it will be seen that a temple having an entrance 
 from a court to transverse halls like this is clearly a sanctuary of the old Babylonian 
 type, like the temple of E-mach at Babylon, for example.? Our hasty excavations with 
 the East Indian troops (Fig. 54, p. 91) revealed enough of the structure to make these 
 main arrangements of ground plan fairly certain. Since then these results have been 
 confirmed and extended by the excavations of Cumont, which have disclosed a series 
 of rooms on the right (north) of the court, as well as also on its front (east) side. In 
 the center of the court, as we would have expected, Cumont’s excavations also revealed 
 an altar, while a dittle naos was found by him at the northwest corner of the court. 
 We owe to M. Cumont’s kindness the inclusion of the new data completing the ground 
 plan (Fig. 56, p. 95). 
 
 On the west side of the court we had found two columns in front of the inner halls 
 at the left (south) of their entrance, and I concluded there must have been two more 
 on the right (north) of this entrance. These two columns have since been found by 
 Cumont. The four columns form a narthex in front of the temple halls. Within 
 the narthex four steps in the middle lead up to the front door of Hall I. This door 
 is not in the axis of the court. It will have been crowned by an arch, without doubt, 
 like the doors of the Babylonian temples. Hall I to which it gives access averages 
 
 1 Tt was this fact which led us to suppose that the walls of the city as a whole were also oriented 
 to the cardinal points, as in the plan published in Syria, Vol. III. The airplane view shows the 
 incorrectness of this conclusion. 
 
 2 See Koldewey, Das wiedererstehende Babylon, Abb. 38, p. 56. 
 68 
 
 
 
 
 

 
 THE TEMPLE OF ZEUS-BAAL 69 
 
 4.04 meters by 7.38 meters in size! We were not able to clear this hall, which lay 
 deeply incumbered with rubbish (see Fig. 54). We cleared the north wall and found 
 there the painting of the Roman tribune (Plate XXI) and the small paintings sur- 
 rounding it (pp. 101-2). On the front wall of Hall I, north of the entrance door, we 
 found also small paintings since reported fully by Cumont. We made no clearance 
 at all of the south wall but could discern at the top, above the rubbish, traces of a 
 larger painting later uncovered by Cumont and discussed herein (p. 90). 
 
 
 
 Fie. 40.—Suetkn RamapAn-Bra Ipn SHALLAsSH AND A Group or His TRIBESMEN OF THE ALBU 
 SARAI ABOVE DEIR EZ-ZOR 
 
 The sheikh is the tall white figure at the left. He has turned away from the camera because of a 
 disfigurement received in the Great War, part of his nose having been carried away by shrapnel. 
 He is a shady character who at this time was receiving a subsidy from King Faysal and spending it 
 on behalf of the Turkish Nationalists. Although he practically forced the writer to carry a con- 
 fidential letter for him to Aleppo, he was most kind and hospitable and showed every kindness to the 
 expedition in his great black madif (guest tent), visible here behind the people. Note the Euphrates 
 cliffs at left. 
 
 Between Hall I and Hall II there is a large arched opening with a width of nearly 
 6 meters, that is, much too wide to be called a doorway. The spring of the arch 
 which crowned it is preserved in the masonry near the top of the south abutment 
 (see Plate VIIT). It is almost certain that the roof of Hall I was likewise a vault. 
 
 1 Our hurried measurements varied somewhat from these figures, which are those of Cumont’s 
 survey. His dimensions are used throughout this discussion of the temple. 
 
70 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Hall II, a transverse hall like the one before it, was the sanctuary proper. It is 
 4.30 meters wide and in length about 8.60 meters. It is therefore longer than Hall I, 
 with which it almost merges owing to the great width of the arched connection between 
 the two halls. The walls of the entire northern half of this Hall II have been quarried 
 out or have fallen down into the wadi immediately below-(Plate VI, 1). The south 
 wall, which is preserved to a height of 7 meters (Plate VIII), is not a wall constructed 
 in the course of erecting the temple. It is on the contrary a part of the original wall 
 of the fortress in this west salience. It is clear at this point, therefore, that the fortress, 
 
 
 
 Fie. 41.—A Group or ARAB RIFLEMEN OF SHEIKH RAMADAN-Bga IBN SHALLASH 
 
 or at least this part of it, is older than the temple. This is especially evident in examin- 
 ing the wall between Hall I and Hall II, as seen at the left in Plate VIII. For the 
 wall between the two halls mentioned is built of smaller blocks than the masonry of 
 the fortress at this point, and, moreover, it does not engage with the fortress wall, 
 i.e., the south wall of Hall II! Traces at the top of this south wall would indicate 
 that this hall was roofed by a lofty vault. 
 
 In the axis of the building, in the middle of the sanctuary (Hall II), is a shrine 
 of semicircular ground plan, with the open side of the semicircle facing toward the 
 
 1 This last observation was made from the photographs after my return from Mesopotamia, 
 and does not, I regret to say, appear in my notes made on the spot. 
 
 
 
 
 
THE TEMPLE OF ZEUS-BAAL 71 
 
 front of the building, that is, eastward (see plan, Fig. 56, and Plate VI, 2). The 
 niche in the center of the shrine was found empty as shown in the photograph. In 
 the same shrine are two small niches, one on each side of the central niche. Of these 
 two smaller niches the one on the south side is still preserved and may be seen in the 
 photograph (Plate VI, 2). Both were likewise found empty. No door closing the 
 front of the shrine, if it ever had one, is discernible in any of the traces we could find. 
 This shrine closely resembles the niche containing the emperor’s statue as it is found 
 in the surviving legionary chapels.! Our shrine, preserved for only about one-third 
 of its presumable original height, should doubtless be restored with a half-dome at 
 the top, as in the examples shown by Domaszewski (op. cit.). It is remarkable to 
 
 
 
 Fig. 42—Tue Heap or Our Caravan DESCENDING FROM THE PLATEAU TO THE HUPHRATES NEAR 
 TIBNI 
 
 find here a shrine so strikingly like those of the Roman legions of the west in a temple 
 of unmistakably oriental ground plan. 
 
 The presence of the legionaries in the wall paintings of the temple, engaged in 
 worshiping a group of divinities including the two local Fortunas (p. 97) so com- 
 monly worshiped by the Roman troops, shows that this temple served during the 
 Roman occupation in the third century as a garrison chapel. It should be noted, 
 however, that Cumont did not find under the shrine any traces of a subterranean 
 treasury, so frequently placed there by the legions of the west. At the same time the 
 wall graffiti found by Cumont show clearly that the temple was devoted to the worship 
 of Zeus Megistos, or Baalsamin. Associated with this Zeus-Baal in the graffiti were 
 also Yarhibol and Aglibol, the three forming the well-known triad of gods of Palmyra. 
 The people of Dura, like the legionaries later forming its garrison, were no doubt 
 largely of Palmyrenian origin and carried to Dura the gods of Palmyra. 
 
 1See Domaszewski, Die Religion des roemischen Heeres, Plate II, Fig. 4; also Cagnat, in Saglio- 
 Pottier, Dict. des Antiquités, s.v. ‘‘Praetorium.” 
 
72 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 We must conclude, therefore, that this temple was originally erected by the people 
 of Dura as a sanctuary sacred to the Palmyrenian triad: Zeus-Baal, Yarhibol, and 
 Aglibol, and that the three niches in the shrine in Hall II were occupied by statues of 
 these three divinities. The paintings which we found in this hall likewise show no 
 connection with Rome or with legionary customs or ritual. They reveal furthermore 
 the prominent part played by Konon, a noble civilian of the town, and his family in 
 the original construction of the temple. These paintings, dated to the latter half of 
 the first century of our era, and the inscription of Lysias (pp. 87-88), a member of 
 Konon’s family, recording his erection of a dwelling for the priests connected with the 
 
 
 
 Fig. 43.—Our WAGONS IN THE KHAN AT SABKHAH 
 
 temple in A.D. 114, disclose to us quite unmistakably the fact that in this sanctuary 
 we have an oriental temple which originally had nothing whatever to do with Rome 
 or the worship of the emperors—a temple which had long existed and carried on its 
 oriental ritual before the Roman occupation of the third century. 
 
 In finishing the interior walls of the structure the builders employed a lime plaster 
 spread directly on the surface of the masonry, in preparation for the mural paintings 
 with which the building was to be decorated. These painted wall decorations extended 
 almost from the floor to the ceiling. At least the lower portion of an upper range of 
 paintings was observable above the lower painting on the south wall of Hall II and 
 may be seen in Plate VIII. That such mural paintings in the so-called pagan temples 
 of Syria later furnished the models for the wall paintings of the Christian churches of 
 
 
 
THE TEMPLE OF ZEUS-BAAL 73 
 
 the same region cannot be doubted. When the overthrow of Zenobia and the destrue- 
 tion of Palmyra left Christianity a freer hand for the rapid conquest of Syria by the 
 church, it is obvious that the temples which the Christians appropriated were decorated 
 with just such paintings as these at ancient Dura. Indeed it is not difficult to imagine 
 that such painters as Hlasamsos of Dura, after the disappearance of Palmyra, would 
 easily accept Christianity. Under the control of the new faith these artists would 
 no longer paint Zeus-Baal, Yarhibol, and Aglibol; but, following the same traditions 
 in style, technique, and composition, the successors of Ilasamsos must have painted 
 the virgin, the Savior, the apostles, the great Church Fathers, and the newly converted 
 
 
 
 Fic. 44.—SHEIKH SUWAN OF THE SABKHAH ARABS ABOVE DetrR Ez-ZOR ON THE UPPER EUPHRATES 
 
 The sheikh is the second figure to the left. The head of a powerful group of Arabs, he made a 
 very straightforward impression. He based great hopes on President Wilson and the fourteen points, 
 knowledge of which had reached him even in this far-away Arab wilderness. His sturdy son, very 
 proud of a new Mauser rifle, is the fourth figure. 
 
 sovereigns, precisely as we have them in the mosaics of Ravenna, which are the lineal 
 descendants of these mural paintings of ancient Dura. 
 
 Architecturally we could gain but fragmentary impressions of the building. It 
 evidently, like the architecture of Babylonia, displayed only structural forms. Its 
 apparent unity of plan is deceptive. The intrusion of so un-Babylonian an element 
 as a frontal colonnade at the rear of the court in front of the inner halls suggests 
 more of a facade than such buildings of Western Asia originally possessed; but this 
 colonnade was not really along the front of the building, which should properly be 
 
74 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 found in front of the court. The building was not considered enough of a unity to be 
 exempt from extension and enlargement. The rooms immediately on the north of 
 Halls I and II were not sufficient for the practical needs of the temple ritual, and a 
 series of rooms along the north side of the court were afterward added. Outside the 
 court on the east also a portico was erected, which turned eastward away from the 
 temple and followed the city wall. One can easily imagine this outer porch as a place 
 where much of the daily idling and gossiping of customary town life in the Orient 
 commonly took place. Like the mosque, which is the descendant of the ancient 
 oriental house and temple, the building really possessed no fagade, properly speaking, 
 and these extensions which it suffered did not violate any real unity of design. 
 
 
 

 
 Vv 
 
 THE WALL PAINTINGS OF THE ZEUS-BAAL 
 TEMPLE 
 
 The interior of Halls I and II was elaborately decorated with wall paintings. 
 Before proceeding to a detailed description of these paintings it may be well to dispose 
 of the technical questions involved, in so far as the notes from my brief examination 
 will permit. The walls are covered with a white lime plaster, rather smooth, but with 
 considerable deviation from plane, and on the average about 2 to 24 centimeters 
 thick. The pigments are all water colors which have been mixed with some adhesive 
 gum of such strength that the paint is still firm and hard. I noted the following 
 colors and shades: black, white, maroon, red, brown, orange, purple, magenta, 
 several shades of green, yellow, pale blue, and gray. Two distinct methods are notice- 
 able. The contours of the figures are sometimes run in with black lines, like the early 
 Italian primitives; other figures, however, are entirely without such contour lines. 
 It is also noticeable that the chzaroscuro is not consistent in the same picture, but the 
 light may be received from different directions in the same scene. Here and there, 
 however, there is a total lack of any attention to the light and the tones are laid on 
 flat. This seems to be the case especially in the figures in which the contours are 
 plainly marked with black lines. 
 
 The collapse of the roofs has exposed the paintings to the rains—not directly, for 
 the halls were filled with rubbish, largely made of dried mud. This rubbish, resting 
 against the walls, was doubtless converted into soft mud by the rains, with the result 
 that the surface of the paintings was covered by a grayish black film of dried mud, 
 which it was impossible for us to remove. This film of mud may also have been 
 deposited when the roof leaked before its collapse, as the mud of the sun-dried brick 
 vaulting was washed down. The mud film may be seen especially on the white 
 garments in Plates IX and X. Under this obscuring surface the colors were well 
 preserved and bright, except as dimmed by the film of mud. Without being cleaned 
 the paintings were exceedingly difficult to photograph. Professor Luckenbill, to 
 whom I owe the photographs, struggled manfully with this difficulty and his results 
 are most creditable. J made very full notes on the colors, and the color plates herein 
 are reproduced from enlarged photographs, hand-colored in accordance with my notes. 
 In the main the tones are probably approximately correct, although the modern color 
 printer has in places put too much brightness in the tone. Our colored photographs 
 are the only surviving records of this painted wall, now much mutilated by the 
 Arabs, especially the row of heads at the top. 
 
 75 
 
76 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 THE EARLIER GROUP OF PAINTINGS 
 HALL II, SOUTH WALL, C-D (PLATE VIII) 
 The Wall of Bithnanaia 
 
 It was the wall C-D in Hall II which was first noticed by Captain Murphy as 
 bearing paintings and it was this wall which we found cleared when we first began work. 
 The paintings which it bears are in two registers, beginning almost at the floor. They 
 probably extended entirely to the ceiling, or at least to the spring of the vaulting. The 
 upper register, however, has entirely disappeared, with the exception of the feet of 
 one figure at the lower left-hand corner (Plate VIII). The plaster bearing the upper 
 painting has fallen off but the wall survives to a height of nearly 7 meters above the 
 chapel floor. The lower register bears the largest and most important painting in 
 the chapel (Plate VIII). Indeed it is the most important monument thus far found 
 
 
 
 Fia. 45.—Our Horses BEING WATERED IN THE EUPHRATES FOR THE Last TIME, ABOVE MESKENAH, 
 Two Days’ JOURNEY FROM ALEPPO 
 
 at Dura. It is an impressive composition 4.30 meters long and nearly 4 meters high. 
 The top of the cornice over the whole scene is almost 4 meters from the floor. 
 
 The fallen plaster has carried away the upper right-hand portion of the lower 
 painting, depriving us of one head, the major part of two more, and the top of a fourth. 
 Otherwise the lower painting is almost intact. We shall call it the wall of Bithnanaia 
 after the gorgeous lady who is so prominent in the painting. This ambitious composi- 
 tion consists of a row of eleven figures, eight filling the entire breadth of the available 
 wall (4.30 meters) with three more in front of the eight mentioned, the whole group 
 of eleven figures being flanked by an architectural background. A gorgeously 
 appareled lady stands a little at the left of the middle, while on the right are ranged 
 four men bearing green branches and on the left stand three ministrants who are 
 carrying on religious ceremonies. Numbering these people from left to right, the 
 figures one to eight, who form the main group, are curiously disconnected from 
 
 
 

 
 THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE <7. 
 
 the architectural background against which they are placed. Let us consider first 
 this architectural background, which is in several planes. In the main it is 
 made up of three doorways. Out of the central doorway the lady (the fourth figure) 
 seems to step forth. The two door leaves in this doorway are yellow and paneled and 
 the one on the right is open. On the right of this door the yellow paneling continues 
 to a double door behind the seventh and eighth figures. The right leaf of this last 
 door is also open. On the left of the double door behind the lady, at the left end of 
 the painting, is a porch with two piers, behind which is an open door. It also is a 
 door of two leaves, one leaf being closed. While it would be difficult to determine 
 what kind of a building the painter has here endeavored to represent, it is likely to have 
 been some part of this chapel itself, in view of the ceremonies depicted. 
 
 
 
 Fic. 46.—Moprern ALEPPO AND THE Mounp oF THE ANCIENT CITY 
 
 Aleppo was a small city kingdom as far back as the fifteenth century B.c. The mound covering 
 the ancient walled town now rises in the midst of the modern city. It is here seen from the east, as 
 we approached it fourteen days out from Baghdad. 
 
 With regard to the relation of the figures to the architecture, the first and second 
 figures are depicted standing with feet nearly if not accurately on the floor, but the 
 third figure places one foot on the base of a tall vase which belongs to the second figure. 
 As this position of the foot is inconceivable, it raises the question whether the feet of 
 the first and second figures are actually on a pink floor. This rectangle of pink may 
 conceivably be intended to represent a vertical surface, a doorstep under the threshold 
 (see Plate IX), a doorstep in a plane in advance of the black and white floor which 
 begins on the right of the blue vase. The fourth figure (the lady) stands partially 
 inside the middle doorway and evidently leans forward, bringing her head outside 
 the door lintel (see Plate XI). The floor on which both she and the third figure stand 
 is black, or black and white. The first to the fourth figures inclusive show some rela- 
 tion to the architecture against or in which they are placed; but the four figures at 
 
78 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 the right, that is, the fifth to the eighth, stand, all four, with their feet far above their 
 black and white floor, as if floating ike phantoms (see Plate VIII). This is especially 
 noticeable with the fifth figure, whose right foot projects over and far above the 
 
 
 
 Fic. 47.—Tue Court or A Mosque IN ALEPPO 
 
 black floor behind the central doorway (see Plate XI). In front of the central doorway 
 there seems to have been at least one step, which probably appears again also below 
 the seventh figure and just behind the young girl (tenth figure, Plate X VIII) and may be 
 in the same plane with the possible pink doorstep at the left (Plate IX). At the right 
 
 a! 
 

 
 
 
 THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 79 
 
 of this black step and below the fifth and sixth figures (Plate XI) one can discern the 
 base of a square pier or a pilaster, over which the left foot of the fifth figure and the 
 right of the sixth have been painted. Behind the seventh figure (Plate XVI) the 
 lower edge of the panel of the (closed) door leaf cuts through the instep of his right 
 foot. It is evident that these four figures (fifth to eighth) have been painted in either 
 very unskilfully or as an unforeseen after-thought. 
 
 The three young people, the ninth, tenth, and eleventh figures (Plate VIII), who 
 seem to stand below the row of eight adults, are in a very advanced plane and no traces 
 of anything below their feet, nor even of their feet themselves, have survived. 
 
 If there is no homogeneity as between the architectural background and the figures 
 placed against it, there is also a similar lack of proper correlation as between the figures 
 themselves. This latter lack of correlation is so noticeable as to indicate that the 
 painting is the work of two different men, working presumably at two different times. 
 The first and the second figures, with the exception of the faces, disclose almost no 
 shadows on the flesh forms, so that the legs are finished in flat tones, but the light in 
 the faces is well treated, especially in the second figure (see Plates IX and X). Now 
 in the architecture the light comes from the right and in the first and second figures it 
 comes from the left. Turning to the others beyond the second (that is, to the third 
 to eleventh figures), it is noticeable that the flesh forms do not display the black 
 contour lines of the first and second figures. They are well treated as to light. The 
 legs look round, with shadows along the edges and light in the middle. This is so 
 different from the first and second figures that we must conclude that these two were 
 done by a different hand. This conclusion is perhaps confirmed by the architecture, 
 if we are correct in concluding that the floor under the first and second figures is pink 
 while under the rest it is black or black and white. With regard to the composition as 
 a whole, the grouping of the tall figures in a long line across the observer’s line of vision 
 strikingly suggests the similar grouping in the oldest Byzantine compositions at 
 Ravenna (see Plate XXII). Furthermore, just as in our Dura painting, in the Ravenna 
 mosaics all the figures face the observer. This question can be better understood, how- 
 ever, when we have examined the wall of the tribune (Plate X_ XI, and see pp. 94-101). 
 
 Turning to the details of the figures, the following matters are of consequence: 
 
 FIRST FIGURE (PLATES VIII, IX, AND XIII) 
 
 Figure of a man standing with weight on left foot, and right foot forward. He 
 wears a sleeved garment hanging from the neck and shoulders almost to the ankles. 
 It falls from the left shoulder down to the right leg and returns upward in front to 
 fall over the extended left forearm. It is white with pink borders down each side of 
 the middle of the front and is gathered in at the waist by a girdle with heavy folds. 
 There is some effort to paint the ight on the drapery. The head and the upper part 
 of the ears are entirely covered by a close-fitting pink cap with a loose, full peaked 
 
80 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 tip which falls over to the left. The man has a slight beard along the jaw and chin 
 and a scanty moustache. His feet are shod with white shoes gathered in at the 
 instep by a black cord over a tongue which rises in the middle in front. The flesh 
 hue is a deep brown. A necklace, which he holds in his right hand, hangs down in a 
 straight line; it is pink with a pale blue lotus flower forming each end. The cords for 
 fastening the necklace issue from the lotus flowers like stems. A dumb-bell-like object 
 (thunderbolt ?) in his left hand has pink balls and pale blue stem. 
 
 SECOND FIGURE (PLATES VIII, IX, X, AND XII) 
 Figure of a tall man standing with weight on left foot, and right foot forward. 
 He wears a sleeved garment which hangs from the neck to the lower calves; it is white 
 
 
 
 Fic. 48.—Tue VALLEY OF THE ORONTES IN NoRTH SYRIA 
 
 Our party in returning followed this valley southward to the sources of the river 
 
 with a narrow girdle knotted in front. There is an effort to paint the hight on the 
 folds of the garment. On his head he wears a tall conical pointed white cap painted 
 in one flat tone and covering the tops of the ears. His feet are bare. The flesh hue 
 is a deep brown. A vase beside him has a tall fluted base mounted on three legs, 
 the whole being colored pale blue with black contour lines. The vase is filled with a 
 transparent fluid, presumably water, into which the man inserts a branch or plant 
 held in the right hand. The plant, which is rather faint, is done in black and, ascending 
 with several drooping branches, crosses a pitcher held in the man’s left hand. In 
 this left hand, besides the pitcher already mentioned, the man holds a plate bearing 
 two knives. The pitcher, the plate, and the knife blades are all pale blue, while the 
 knife handles are pink. 
 
 
 

 
 THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 81 
 
 THIRD FIGURE (PLATES VIII, IX, XI, XIII, AND XIV) 
 
 A tall man standing with weight on left foot, and right foot forward. He wears a 
 sleeved garment hanging smooth without any of the folds of drapery observable in 
 the second figure. On his head is a conical cap painted with noticeable observance 
 of chiaroscuro and in this respect better than the similar cap worn by the second 
 figure. The ears are not covered at the top by the cap as in the case of the second 
 figure. The feet, which are bare, are without the black contours defining the feet of 
 the second figure and, as in the case of the cap, there is an evident effort to paint the 
 light and give the flesh forms roundness. The face exhibits soft contrasts without 
 the hard lines observable in the face of the second figure. The eyes are fine; there is 
 no sharp line marking the lower lid, as in the face of the second figure, only a soft 
 
 
 
 Fig. 49.—Tur WeLcomMEp MEDITERRANEAN SHORES Norra or BreirttT 
 
 dark shade. Like the first two men, he wears a scanty beard. The flesh hue is a deep 
 brown. Beside him is a burning censer on a tall slender standard. The man extends 
 his right hand above the censer so that it is enveloped in the rising flames, a gesture 
 probably indicating that he is dropping incense into the fire. The flames are, strangely 
 enough, indicated by black lines; the column is white with black contours. There is 
 a vertical scratch which mars the left edge of this hand just where it is marked off 
 from the yellow of the door in the background. In his left hand, just as in the case 
 of the second figure, this man carries two knives, supported, in this case, however, 
 upon a gray bowl. This vessel has a red interior, which may have been meant for 
 either blood or wine. The knife blades, which are marred by the chipping of the 
 plaster, are a graduated gray and totally different from those held by the second 
 figure; the handles are red. Both the blades and the handles are skilfully done with 
 treatment of the light and without the black contours shown in the second figure. 
 
82 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 The points of the knife blades and the right-hand portion of the bowl are covered by 
 the full sleeve of the fourth figure (the lady)—a curious error in composition, because 
 the fourth figure clearly stands in a plane farther back than that of the third. In 
 this connection it may be noted that the sleeve covering the man’s upper arm just at 
 the level of the knife handles has not been properly correlated with the yellow of the 
 background and draws in too suddenly toward the man’s left side. There is some 
 indication that this bad drawing has been hastily corrected by inserting a thin tri- 
 angular area of white just touching the left angle of the knife handles. 
 
 t at % Pe 
 eS. * . . os : 
 
 oie 
 
 
 
 Fic. 50.—Recorps or MopERN ConQquEST AMONG THE MEMORIALS OF ANCIENT CONQUERORS ON 
 THE MEDITERRANEAN CLIFFS AT THE Dog River NortuH or BEtrttT 
 
 FOURTH FIGURE (PLATES VIII, XI, AND Xv) 
 
 A gorgeously dressed woman standing with weight on left foot (barely visible) and 
 right foot forward. Her right hand is raised to the level of her neck and held palm 
 outward. The left arm apparently hangs down and is entirely concealed by the full 
 folds of her mantle. She wears a long white overmantle with full sleeves hanging 
 over a bodice of magenta-colored stuff. On her head is an elaborate cap with a 
 rectangular front covered by a magenta-colored headcloth draped and hanging to the 
 shoulders. Only the front of the cap shows under the draped headcloth. Her ears 
 are adorned with earrings. About her neck she wears a collar having the appearance 
 of four rings of metal. Hanging down on her breast is a necklace with pendants 
 along the lower edge. Four of these pendants are visible on the right of the upraised 
 
 
 

 
 THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 83 
 
 hand and two on the left of the thumb of the same hand. Below this necklace and 
 perhaps suspended from it is an oval medallion in elaborately wrought metal work. 
 The reticulate border is black, dotted with white spots, the whole border being against 
 a gray background. The oval interior is purple, as if made up of some large precious 
 stone like amethyst. Below the oval medallion is a round one with interior design 
 made up of mixed colors in red, green, white, and purple, so mingled that the design 
 cannot be discerned. It is possibly some precious stone of highly varied coloring. 
 To this round medallion is appended an elaborate pendant in the form of a rec- 
 tangular frame made up of three vertical rows of seven spherical beads alternately 
 white, red, and green. In these details the color plate (Plate XI) is not strictly 
 accurate. They may be discerned very faintly in Plate XV. Two pendants hang 
 down from this rectangular bead design. Each of these two is made up of two eylindri- 
 cal beads, a red one at the top and one of pearl gray just below it, while each of the 
 two terminates in a globular pearl-gray bead. An effort has been made in Figure 57 
 (p. 99) to make the details of this elaborate Palmyrene jewelry more clear; but my field 
 sketches were too hasty to make details always certain. As we have already noted, 
 the sleeve of the lady’s right arm curiously interferes with the bowl and the knives 
 held by the third figure, almost as if the knives were thrust into her side. She wears 
 low shoes painted white, which show partially below the white overgarment. The 
 flesh hue of the lady is somewhat lighter in tone than that of the three ministrants 
 (first, second, and third figures). The relation of the lady’s figure to the architecture 
 has already been discussed above, but it may be noted that she has not stepped forth 
 so as to be entirely outside the open door. Her left foot, on which her weight seems 
 to rest chiefly (her left shoulder is lower than her right), is within the doorway although 
 her head is entirely outside of the lintel and is even projected against the ceiling of the 
 portico. 
 FIFTH TO EIGHTH FIGURES (PLATES VIII AND XVI) 
 
 Four standing men, all with weight on the left foot, with right foot forward. 
 Each has the right hand raised, palm outward, in a posture like that of the lady 
 (fourth figure). The left hand, projecting from the garment, holds a green branch 
 conventionally painted, which might be either the tip of a palm or an olive. The 
 head of the eighth figure is entirely gone, those of the sixth and seventh nearly so, 
 and also the upper part of the head of the fifth. No traces of a cap or headdress are 
 visible but two masses of heavy black hair hang down on either side of the head of 
 the fifth figure. As far as visible, all these men are smooth-shaven. This is certainly 
 the case with the fifth and sixth figures and probably with the seventh. This lack of 
 beard evidently indicates the youth of at least three of these four persons (see p. 88). 
 All four wear purple undergarments over which hangs a white overgarment from the 
 shoulders to the lower calves. This white overgarment, not unlike a draped toga or 
 himation, is edged down the front with purple on both sides in such a way as at first 
 
84 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 sight to look deceivingly like purple breeches, and we at first thought they were the 
 well-known Persian breeches, but this is probably not the case. This white over- 
 garment exposes the sleeves of a purple shirt and hangs over the left arm, just freeing 
 the hand bearing the green branch. AIl these men wear low white shoes like the first 
 figure but with white cords for tying. These figures are all painted with the same 
 technique and in the same manner as the third and fourth. Their extraordinary 
 lack of correlation with the floor and the architecture behind them, so that they float 
 like phantoms, has already been noticed above. 
 
 NINTH FIGURE (PLATES VIII AND XVII) 
 
 A boy standing before the line of eight adults in a plane well in advance. The 
 feet are gone, but the right hand is raised, palm outward, as in the case of the five 
 persons above (fourth to eighth figures). The left hangs down and carries a small 
 pitcher. The lad is bareheaded and wears a long white sleeved overgarment edged 
 with red. The figure is very well painted. 
 
 TENTH FIGURE (PLATES VIII AND XVIII) 
 
 A girl standing before the eight adults in the same advanced plane as the ninth 
 figure. The feet are gone, but her right hand is raised, palm outward, in the same 
 posture as the ninth figure. The left arm hangs down and bears an object which was 
 possibly a purse. Her cap, of the same form as that of the lady above, is covered 
 with a full pink drape which falls down to mingle with her full garment, which is 
 likewise pink. Under this pink outer garment is a black bodice. She wears earrings, 
 and below a collar of four rings, like that of the lady above, is a similar elaborate 
 necklace with two medallions and two pendants. 
 
 ELEVENTH FIGURE (PLATES VIII AND pa Se) 
 
 A standing boy in the same advanced plane as the ninth and tenth figures. 
 His feet also are gone, but his right hand is raised to the level of his breast and grasps 
 the two ends of a necklace exactly like that which hangs straight from the hand of the 
 first figure. The boy holds the necklace as it would be when the two ends are gathered 
 together about the neck of a wearer, his hand grasping the tight cords between the 
 two lotus flowers, which appear on either side of his hand. The left hand hangs down 
 and is seemingly empty. The head is bare, the garment is like that of the ninth 
 figure. General Cunningham stated to the writer that this was the best-painted face 
 on the entire wall; but before we saw it some iconoclastic native, possibly a prowling 
 Arab, had barbarously scratched the face. 
 
 It is evident, as we have before stated, that the scene depicts a religious ceremony 
 with two ministrants at the left (second and third figures). These two are evidently 
 priests. Each wears the pointed white cap, a white garment without any colored 
 
 
 
 : 
 7 
 : 
 4 
 ; 
 
 
 
THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 85 
 
 border, has the feet bare, and is engaged in a cultus act, with much the same imple- 
 ments. The other nine figures are also sharing in the religious observances, for all 
 the remaining figures except the eleventh (that is, fourth to tenth) stand in a posture 
 of worship, with the right hand raised, palm outward. As is well known, this is a 
 characteristic posture of Babylonian worship from the earliest times—a posture which 
 we often find on early Babylonian seals. The act of thrusting the branch into the 
 water also suggests old Babylonian ritual in the ceremony of the “tree of life,” origi- 
 nally a palm branch thrust into a vase of water. We evidently have here one of the 
 eclectic cults of Syria, about which we know so little, and in which we expect to find 
 
 
 
 Fic. 51.—GErnNERAL CUNNINGHAM WITH OFFICERS AND PART OF OUR EXPEDITION AT THE SOUTHWEST 
 GATE OF THE DuRA FORTRESS 
 
 General Cunningham stands at the left. To his interest in the wall paintings in the ancient 
 fortress it was largely due that we had the opportunity of studying and recording them. The 
 British East Indian troops have established a battery on the top of the fortified gate. 
 
 not only old Babylonian influences like those already mentioned, but later also a 
 plentiful infusion of neo-Persian elements. 
 
 In this connection it is important to notice that the artist has inserted on the 
 lower portions of seven figures the names of the persons depicted (Fig. 58, p. 101). 
 These names reveal to us the fact that seven of the people depicted in the painting 
 belong to one family, only the two priests and two of the three children being omitted. 
 It is probable that the two children belong to the same family, and their names may 
 
86 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 have been lost at the bottom of the painting where much has disappeared. We may 
 therefore reconstruct the following family tree: 
 
 NIKOSTRATOS (whose portrait does not appear) 
 KONON I (first figure, head of the family) 
 | 
 
 | | 
 BITHNANAIA DIOGENES LYSIAS PATROKLOS 
 
 par v3 II 
 ? 
 —N———§ 
 
 In this family group Konon I, the first figure standing at the extreme left separated 
 from the others by the two priests, is head of the family. Beyond the two priests 
 (that is toward the right, their left) are Konon I’s four children, a daughter and three 
 sons. The third son, Patroklos, has his son Konon II (grandson of Konon I) standing 
 next to him (eighth figure). Of the three children in front, one whose name is not 
 clearly preserved, was a son of Konon, perhaps Konon II, and in that case the group 
 includes four generations. . 
 In the names of these people, as in the art of the paintings surviving in this chapel, 
 we note the obvious presence of elements both Greek and oriental. Although her 
 father bore the Greek name Konon, the lady’s name is evidently oriental. The resto- 
 ration of the | in Bithnanaia, favored by the small size of the lacuna and the surviving 
 traces, is certain; and it is obvious that we must recognize in the second part of this 
 compound name the name of the well-known goddess Navata (II Mace. 1:13). With 
 his customary sagacity Clermont-Ganneau has discussed early Syrian names of this 
 form having Na (here Bu§) as their first element.! To his list of names (op. cit., p. 416) 
 we should now add that of “Bithnanaia, daughter of Konon.”” To Clermont-Ganneau’s 
 discussion of the family relationships involved in names of this form we will not attempt 
 to add anything further, but for the sake of readers who may not have followed the 
 literature of the question we may recall Néldeke’s notice of the fact that in 
 De Vogiié’s Syrie centrale? the statue dedication inscription No. 29 bears the name of 
 Zenobia in the form "2TM2 N20DO=Septimia Bathzabbai, meaning “Septimia, 
 daughter of Zabbaeus.”’ Oberdick*® has contended that filiation is indicated in Pal- 
 myrenian Aramaic by M72 (feminine of 72) and not by the Hebraizing form M2, 
 . so that when M2 is found in Palmyrenian it is used as “filia doni,”’ meaning ‘‘a woman 
 possessing.’’ He therefore renders Zenobia’s Semitic name as ‘‘ Daughter (= possessor) 
 
 1 See Clermont-Ganneau, ‘‘Odeinat et Vaballat,” Revue Biblique, XXIX (1920), pp. 382-419. 
 
 2 ZDMG, XXIV (1870), p. 96. 
 3 Zeitschr. frir die oesterr. Gymnasien, XLI (1890), 699-703. 
 
 
 
THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 87 
 
 of Brilliance ("2X M3).” It cannot be said that this new name Bithnanaia favors 
 Oberdick’s contention, for the lady can hardly have been designated as possessor 
 of the goddess Nanaia. On the other hand she might be called ‘‘ daughter of Nanaia” 
 without in any sense conflicting with the addition of her father’s name in the genitive. 
 Clermont-Ganneau’s essay already cited should also be consulted on this point. 
 
 In contemplating this group of eleven people who look down upon the shrine in 
 the middle of this transverse sanctuary hall, one cannot but feel the probability that 
 the lady who occupies so prominent a position in the scene at the head of all the children 
 of Konon the elder was a woman of rank and power, an Aramaean princess, reminding 
 us, even in the form of her Semitic name, very much of Zenobia of Palmyra, which 
 was but 140 miles distant. May she not have been a local ruler of this region, or 
 the wife of one of the “duces,’’ such as Zenobia was in the beginning ? 
 
 
 
 Fig. 52.—Our Camp In THE ANCIENT FortRESS OF DURA 
 
 At the right, the west end of the massive castellum; at the left in the background, the tents of 
 Major C. T. Wright-Warren, who was commandant of the post. Our camp was in front of his tents. 
 In the foreground is the white limestone tomb of a young British officer killed with all his party by 
 the Arabs a short time before our arrival. 
 
 In any case there can be little doubt that the family of Konon, depicted here so 
 prominently in the sanctuary of the temple, was the leading family of Dura. The 
 names of Konon I and of his son Patroklos are mentioned in inscriptions found in the 
 assembly building in the town by Cumont.! Lysias? (probably one of the unnamed 
 
 1 Comptes rendus (1923), p. 22. 
 
 2 The restoration AYCIAC for the sixth name, first suggested by Torrey, exactly suits the 
 surviving traces and is now confirmed by the inscription of Lysias in the court (p. 88). 
 It may further be mentioned in this connection that the lower part of this painting is covered with 
 numerous graffiti in such rapid cursive that they are difficult to read. In the few hours at our disposal 
 it was quite impossible to spend any time on them. They have been collected since our visit by 
 Cumont. 
 
88 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 two children), son of Konon II and grandson of Patroklos, was the one who built a 
 dwelling for the use of this temple alongside the court, and an inscription found in the 
 court commemorates this pious act.! An interesting light is thrown upon the custom 
 of granting to such influential families places of honor on the temple walls, by an 
 inscription or memorandum on a portion of the wall of Hall I.2 It designates an 
 empty space as not yet assigned to anyone, and therefore still to be regarded as avail- 
 able for assignment to some personage of sufficient importance to claim it. The 
 prominent place assigned to the family of Konon I is conclusive evidence of their 
 power and importance. 
 
 The data furnished by this painting, combined with the other inscriptions, are 
 quite conclusive evidence of the date to be assigned to the building and to the paintings. 
 Konon I is mentioned in an inscription in the town which is dated in the year a.p. 61.3 
 Lysias, son of Konon II and grandson of Patroklos, built his dwelling in the temple in 
 A.D. 114.4 It is evident that the painting was executed at a time when the children 
 of Konon I were still young, for all of his sons are depicted as beardless, although in 
 the case of Patroklos we must conclude that he was probably about forty years old. 
 The chronology of the family may be reconstructed as follows: 
 
 Nixostratos. Probably born just before the beginning of the Christian Era, and already 
 dead at the time of the great painting (about a.p. 80). 
 
 Konon. Probably born about a.p. 15; known to be living in a.p. 61; shown in painting 
 (about A.D. 80) as about sixty-five years old. 
 
 PatrokiLos. Probably born about a.p. 35; shown in painting beardless although over 
 forty years old (a.p. 80). 
 
 Konon IJ. Probably born about a.p. 55; shown beardless in painting at over twenty years 
 of age (A.D. 80); built dwelling in Zeus temple in a.p. 114, when about fifty-nine 
 years old. 
 
 UnnaMED Son. Probably born about A.p. 70; shown as a lad in painting in A.p. 80. 
 
 The fixed points in the foregoing scheme will not permit much shifting of the sug- 
 gested dates. We may conclude with certainty that the painting was executed in 
 the last quarter of the first century. The building of the temple may reach back to 
 the middle decades of the same century, when Konon I was in the prime of life. 
 
 HALL II, WEST WALL, D-E (PLATE XX, 1) 
 
 The west or rear wall of Hall II is preserved only at the left (south) end (D-H on 
 plan, Fig. 56). On this wall there was a large painting (Plate XX, 1) with figures on a 
 considerably larger scale than on the adjoining south wall just described. In the 
 foreground is a remarkable attempt to represent a stony landscape with scanty 
 vegetation in the crannies. Crowning a hill is an architecturally developed rectangular 
 base, perhaps an altar; but the upper part, with the right-hand corner, is gone. At 
 
 1 Comptes rendus (1923), pp. 21 f. 2 Tbid., pp. 30-31. 3 Tbid., p. 22. 4 Ibid., pp. 21-22. 
 
 
 
THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 89 
 
 the left of the altar stood two men. Three legs only survive. The two feet which 
 show are shod with shoes which turn up at the toes. These figures wear short, elabo- 
 rately patterned kilts with rich gold-colored border. ‘The figure at the left carries 
 a tall shield, occupying the space at D directly next to the south wall at the left end 
 of this west wall. The shield is decorated with the painting of a lady holding up 
 her skirts with her left hand, her right hand raised in front but lost by damage to 
 the wall. The only other trace of armor on these figures is what are probably greaves 
 on the front of the legs. At the right of the so-called altar is a horse running off to 
 
 
 
 Fig. 53.—Ovur First GLIMPSE OF THE PAINTINGS 
 
 The semicircular shrine (see plan, Fig. 56) covers the lower part of the great painting on the wall 
 of Bithnanaia (Pl. VIII). This is the wall marked A in PI. V, 2. 
 
 the right. Only his hind legs show. <A curious decorated object, coming down, 
 seemingly, from the back of the horse toward the feet of the man, ends in an elaborately 
 bordered volute, which pushes into the landscape. Below, at the extreme right, in a 
 part not included in the photograph (Plate XX, 1), are traces of a wheel, perhaps 
 belonging to a chariot. 
 
 HALL II, EAST WALL, B—-C (PLATE VIII) 
 Facing the wall just described and at right angles to the wall of Bithnanaia, the 
 
 east wall of Hall II is plastered but bears no paintings, as may be seen in Plate VIII. 
 On the thickness of this east wall (A—B) on the south side of the entrance to Hall II 
 
90 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 and facing north is the figure of a priest (see Plate VIII, extreme left) practically 
 identical in posture and equipment with the second figure on the wall of Bithnanaia. 
 The plant which he is thrusting into the water is not, however, visible and his vase 
 and standard are much less elaborate. 
 
 This completes the series of paintings in Hall II. It is evident that such subjects 
 as the customs, the cultus equipment and the ceremonies being carried on, the identity, 
 racial connection, origin, and so on, of these religious practices and of the people 
 engaged in them, might be much more fully discussed; but they must be left to future 
 study, the chief purpose of this report being to make the paintings accessible. 
 
 HALL I, SOUTH WALL 
 
 When the waning light forced us to cease work on the Zeus-Baal temple, we had 
 been able to clear only the north and east walls of Hall I, and these only partially. 
 The hall was still filled with rubbish, above which we could discern projecting traces 
 of paintings on the south wall. It may be that these are the traces which were seen by 
 Sarre (p. 57). It was with great reluctance that we were obliged to leave without 
 investigating this hall farther. When it was cleared by Cumont, he found a door in 
 the south wall. This of course reduced the amount of space available for decoration, 
 and the painting which his clearance disclosed on this wall is not as ambitious a compo- 
 sition as that on the corresponding wall of Hall II; but it has furnished some very 
 important and valuable data,! especially the name of the artist. In an architectural 
 setting between truncated columns of pink hue, three standing figures, all men, are 
 ranged across the observer’s line of vision, as on the wall of Bithnanaia. Each figure 
 is clothed in a sleeved tunic and a white mantle crossed by bands of various shades. 
 All three are engaged in the same ritual act, dipping a leafy branch into a tall vase 
 with the right hand, and with the left extending a blue plate as if for a liba- 
 tion. Cumont notes that these blue plates are evidently of the same blue enamel 
 ware of which the débris in the town has furnished a number of examples. It is the 
 same act of worship which we have also observed in Hall II in the great painting 
 (Plate VIII). As in the great painting also, these three figures of Hall I face directly 
 toward the observer. Beside the third personage we find leaning against one of the 
 columns a child, of whom only the feet have survived. Similar paintings covered the 
 wall at the right and above, but very little of them now survives. 
 
 It is evident that these three figures are intended to be portraits, for all three 
 have their names appended, written in black paint as on the south wall in Hall II. 
 The first two are of the same form asin Hall II: Avotas ’Axteta and Avotas Bapyarovs. 
 In the third, however, we find far more than the name: ’Aro\Nodarnp |’ APnvodapou 
 rod | Znvodorov kal Zn | vodorov Tov €£dded | Hov adrod INacapcos | éypade (“Tlasamsos has 
 painted Apollophanes, the son of Athenodoros, the son of Zenodotos, and his first 
 
 1 Cumont, Les Travaux Archéologiques, pp. 65-66. 
 
 
 

 
 
 
 THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 91 
 
 cousin Zenodotos”).1_ This signature by the artist himself is of far-reaching importance 
 in the history of art. His name, meaning as it does “the Sun is god,” is purely 
 Semitic and reveals to us the author of these paintings in the Zeus-Baal temple as an 
 oriental practicing the art of portraiture and composition. In view of the discovery 
 of unmistakable efforts to paint the light by Egyptian artists at least as far back as 
 the Nineteenth Dynasty in Egypt,? and probably already in the Eighteenth, these 
 
 
 
 Fig. 54.—Britisp East Inptan Troops at Our DisPosiITION FOR EXCAVATING THE GROUND PLAN 
 OF THE TEMPLE IN THE ANCIENT FortTRESS OF DURA 
 
 The ancient floor level of the temple is at the feet of the workmen on the left. The high mound 
 above and beyond them is rubbish covering the temple court and its first hall. It has since been 
 removed by Cumont. 
 
 paintings of ancient Dura may fairly raise the question whether the Greek “shadow- 
 painters” of the Hellenistic Age really originated the earliest chiaroscuro technique. 
 It is not unlikely that Greek civilization borrowed this accomplishment, as it borrowed 
 
 1Cumont, ibid., p. 66. ‘‘His first cousin Zenodotos”’ is of course the child standing beside 
 Apollophanes, as Cumont has noted. The correction from ’Amo\\édwpov to ’ArodAopavny was kindly 
 furnished by Cumont. 
 
 2In the Bulletin of the Metropolitan Museum of Art for December, 1922 (Part 2, p. 52), Mr. N. 
 de Garis Davies calls attention to various instances of true shading in the Nineteenth Dynasty tomb of 
 Queen Nefretiri at Thebes. This phenomenon here and also in the tomb of the official Userhet, who 
 lived in the preceding reign of Seti I, had been already mentioned in Schifer’s second edition of his 
 Von dgyptischer Kunst (Leipzig, 1922, pp. 57-58) as the result of a letter from Davies elaborating a 
 paragraph in his review of Schafer’s first edition (1919) in the Journal of Egyptian Archaeology, VII 
 
92 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 so many others, from the culture of the Orient, and then, as they so often did, the 
 Greeks raised the new acquisition to a degree of excellence and refinement which it 
 had never known before. It is not a little significant that the “shadow-painters”’ 
 arose in Greece after the penetration of the Orient by the conquests of Alexander the 
 Great, conquests which carried the Greeks directly across the Syrian communities 
 where the art of Ilasamsos flourished. Compositions of this style, as Cumont has 
 noted,! displaying human figures separated by columns of pink color, were still 
 produced in Syria in the ninth century of our era and were found by Musil at 
 Kuseir ‘Amra.2 These later examples suggest the persistence of an ancient oriental 
 school long at home on this soil. Such artists as Ilasamsos were available at Palmyra 
 in the age of her splendor, and these paintings of Dura have furnished us the best 
 surviving examples of the art of Palmyra. 
 
 THE LATER GROUP OF PAINTINGS 
 
 When I first published the Dura paintings in the pages of Syria, it was impossible 
 to establish their date with precision. The inscriptions since then recovered by 
 Cumont have made it clear that the large paintings in the southern portions of both 
 Hall I and Hall Il comprise a group which we must date in the last quarter of the first 
 century; while the small paintings from the northern portion of Hall I, especially 
 the painting of the Roman tribune engaged in worship with his troops, are to be dated 
 about a century and a half later, that is, about the middle of the first half of the third 
 century. They likewise form a group by themselves, distinguished most obviously 
 by their small size, but also by a style less clear and defined, displaying far less ability 
 to treat the light. They resemble hastily thrown together sketches, blocking out 
 blurred compositions which were never later given any precision of detail. They are 
 also much more limited in their color resources. On the whole they represent a 
 decline of art from the high standard of the age of Ilasamsos. 
 
 (1921), p. 225. Yet Davies in his review feels that “it is pretty clear that it is only the deepened 
 colour that has been observed, and not its origin in form as a cast shadow.”’ He adds: ‘‘ Although I 
 know no instance of it under Akhenaten [Ikhnaton, XVIII. Dynasty], I think his artists must have 
 introduced it, or at least have pointed the way.”’ In the same journal (p. 4) Davies has denied a use 
 of orpiment for the high lights in an Eighteenth Dynasty painting of the Tell el-Amarna age which 
 Petrie (p. 221) is equally positive shows this characteristic. It has even been suggested in the Metro- 
 politan Museum’s bookiet on The Tomb of Perneb (1916) that “‘in ceiling paintings . . . . found by 
 the Museum’s Expedition [meaning paintings of the reign preceding Ikhnaton’s], the rotundity of 
 the bodies of flying ducks and pigeons . . . . is indicated by this means [shading with pigments].”’ 
 But these Eighteenth Dynasty examples will require further study before the nature of the facts they 
 illustrate is clearly understood. Acknowledgment is due Dr. Caroline Ransom Williams and 
 Dr. T. George Allen for their kindness in collecting the foregoing references to the published 
 evidence. 
 
 1 Les Travaux Archéologiques, p. 66. 
 *Musil, Kuseyr <Amra, Plates XV, XVII, XVIII. 
 
 
 

 
 
 
 THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 
 
 Fia. 55.—AtrRPLANE VIEW OF THE ANCIENT ForTRESS OF DuRA-SALIB{YAH 
 
 By kindness of M. Cumont, General Gouraud, and the French Air Force 
 
 93 
 
 
 
94 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 HALL I, NORTH WALL (PLATE XXI) 
 The Wall of the Roman Tribune 
 
 While the north wall of Hall II has entirely fallen down into the deep wadz on the 
 north and west (Plate VI, 1), the north wall of Hall I (in front of Hall II) is preserved 
 to a height of some 5 or 6 feet from the floor. This north wall, facing south, is entirely 
 covered with paintings from the floor to the top, as far as preserved. As we turned 
 to work on this wall the sun was low in the west, and our observations were interrupted 
 by approaching dusk. ‘There are other very interesting scenes on this wall, but we 
 were able to record only the important one at the right end. Professor Luckenbill, 
 using the last 1318 cm. film plate we had, secured the best negative possible in 
 the waning light. The original is much faded and not easy to discern, especially at 
 the top. The uncleared rubbish prevented moving the camera far enough back to 
 include on the plate the left end of the scene, so that the camera lost at this 
 point one of the three standing gods and also one of the two sitting goddesses. 
 Important details, owing to the weakness of the fading daylight, were likewise 
 lacking in places on our plate. The scene was, however, so important that it seemed 
 desirable to include it in those published in Syrza (Vol. III, Plate XLVIII). Since 
 then Cumont’s clearance has made fuller study of the painting possible, and I am 
 indebted to him for the detailed drawing published herewith (Plate X XI), which 
 he has already published in his preliminary report.! 
 
 The painting is a small one, about 1.75 meters wide and nearly a meter high. It 
 depicts a scene of worship with statues of five divinities at the left and a group of 
 worshipers at the right. The latter are separated from the statues by a small burning 
 altar or censer and a Roman signifer holding his standard upright. The worshipers 
 are standing in two groups of eight each, with five additional heads indistinctly visible 
 in the background behind these sixteen. The nearer group are in two planes with four 
 in each plane, the heads of each four being on the same level; the more distant group 
 are either in a plane much farther back, or they stand on a raised step behind the 
 front group, for their figures project, from the waist up, above the heads of the nearer 
 group. ‘The second group are also ranged in two planes of four each, with all eight 
 heads on the same level. ‘The five additional heads visible behind the second group 
 are very summarily done and are about on the same level. In so far as the figures are 
 visible, they all stand with the weight on the right leg and the left swung backward. 
 The left hand rests on the girdle in front and grasps some object which is sometimes 
 the sword hilt, while the right is raised and held palm outward in the posture of 
 worship which we found also in the great painting (Plate VIII). The only exception 
 
 1Comptes rendus (January, 1923), p. 27. While the foregoing reasons made it inadvisable to 
 include in this publication the color plate of the wall of the tribune published in Syria, some readers 
 
 may find the colors in that plate of value; but for details only the drawing furnished by Cumont 
 should be consulted. 
 
 
 
95 
 
 quouny "WW Aq peysrainy ATpury suonippe yA 
 VUO( dO SSaUL4O AHL NI AIANA], IVVG-SATZ THL 40 NVIG—9g ‘DL 
 
 
 
 
 
 
 
 
 
 
 
 
 QOS OOO KOO ROO OT 
 : | SEER NNN 
 KSC ‘s Sos SKK 
 : ee a 
 2 panecoeenaten as SENG G] peace 
 4 OL, XO QSOS o> — ar GS am q 
 SSN KKK 
 oye OCOD > & 
 2 Ne 1 < eceeseseseseseseeees ! ios ae 
 a N ea eetceoestctcececeee a II ali VH 
 P SSR” «SLB L 1 a 
 N a 5 iil & ® G91 
 = | : ance ove 4] ! cae 
 1 8 SONS ! See 
 ; a WV TEESE ESSE LaaeWy Fs 
 aes , VOCOOCCC OO Ce 26.0 pa teens 
 ° ae SSSR SOR NOR FD 0 
 [ee <2 SRR SSS NON ofl fae wear 
 ip OX KKK OS oo oe a 
 nn mee SKS : 
 i OOOO ' 
 Z ! s ee seetcacasentareceretctetsteaeteichebenen ! eget 
 2 $ or) 
 : bere ates jm 
 = 2 par PRN ND Eset a 
 5 ae ee Fs | | 
 ae Sipe eres Cea usehay 7 
 < a 
 ES Chorf--o1e- £9 ae oe kles een. ---085 -==4 ] } --009-- 
 = Ze Xr x NM A 
 = 
 
 
 
 
 
 
 
 
 
 
 
96 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 is the tribune himself. All the faces are depicted en face, so that these twenty-two 
 worshipers (including the szgnzfer) all face the observer and not the divinities they are 
 worshiping. The details of costume are sufficiently clear in the sketch, but it should 
 be noted that the kilts worn by the nearer group of eight are all edged along the 
 bottom with pink, which may have faded from the red or purple clavus. This portion 
 of the rear group of eight cannot be seen. 
 
 The nearer group are headed by a Roman officer, who has written before his 
 figure the inscription: 
 
 IVL-TEREN 
 TIVS: TRIB. 
 
 painted in black letters. As we should expect, the inscription is in Latin uncials, not 
 in the Greek uncials which we have had in the great painting. The tribune carries 
 in his left hand what seems to be a roll and may contain the ritual of the service which 
 he is evidently leading. With his right hand he is dropping incense into the burning 
 censer before him, like the priest (third figure) in the great painting (Plate VIII), 
 who also uses his right hand for this office. An examination of available lists of Roman 
 tribunes has not revealed the name of Julius Terentius elsewhere, but it cannot be 
 doubted that he was the commander of the Roman garrison, which included a cohort 
 of Palmyrenians, occupying the place in a.p. 229 or 230 according to a Latin inscrip- 
 tion on a rebuilt block found by Cumont in front of the shrine. It is badly mutilated, 
 but as restored by Cagnat it is dedicated by the Palmyrenians of the X Xth Cohort 
 to Alexander Severus in A.D. 229 or 280.1 
 
 The further group is led by a priest who is probably to be understood as standing 
 beside the tribune a little farther in the background. The inscription before his figure, 
 in Greek uncials, is as follows: 
 
 OEMHCY Oéuns 
 MOKIMGY Mokiyolv] 
 IEPEYC iepels 
 
 Cumont has noted Oatuns (for Oéuns) and Mox:uos as names common enough in the 
 surviving Palmyrenian records. We therefore find here one “Themes, son of Mokimos, 
 priest,’’ doubtless a Palmyrenian priest, assisting the Roman commandant of Dura 
 in the worship of the city gods. 
 
 The service is evidently the official worship of the military garrison, and as such 
 is headed not only by the tribune, who is the Roman commandant of the garrison, 
 but also by the Roman standard. This standard is the well-known vexillum? consisting 
 as usual of a flagstaff, possibly a lance shaft, held vertically, with a transverse stick 
 at the top from which is suspended a banner of some red textile edged with yellow 
 
 1 Comptes rendus (January, 1923), p. 18. 
 
 2 Compare Domaszewski, Die Fahnen im rémischen Heere, p. 77 (Fig. 94). 
 
 
 
THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 97 
 
 or gold and fringed below. It is surmounted by a yellow ring, possibly a wreath, 
 but no traces of the hand sometimes visible at this point are discernible. Unfortu- 
 nately the red field of the banner contains no inscription or insignia by which the body 
 of troops to which the vexillum belonged may be identified, nor the name of the emperor, 
 which was sometimes inserted here. The inscription just cited, however, is probably 
 sufficient for this identification. It is of interest to notice that the field of the banner 
 is red, in accordance with the familiar passage in Plutarch’s Life of Fabius, regarding 
 the red (kéxkwos) battle flag floating over the tent of Varro before the battle of Cannae, 
 and several other references.! In so far as I know, our painting furnishes the first 
 example of such a vexillum in which the color is preserved. The costume of the 
 signifer is not sufficiently clear to discover whether he wears an animal’s skin as on 
 the arch of Constantine. 
 
 At the left of the standard are two rows of divinities. The top row consists of 
 three statues, each standing on a circular base. They are figures of men in military 
 costume and accoutrement, and the new drawing, together with the inscriptions found 
 since our visit, must essentially alter our conclusion that they were to be considered 
 as statues (¢magines) of the deified Roman emperors. On learning that the graffiti 
 show the temple to have been sacred to the three chief gods of Palmyra, Baalsamin 
 (“Lord of the sky,” that is, Zeus-Baal), Yarhibol, and Aglibol, Clermont-Ganneau at 
 once suggested that the three statues depicted in this scene of worship must be the three 
 Palmyrenian gods to whom the temple was dedicated. This suggestion, highly 
 probable in itself, is rendered certain by the detai! of the insignia and equipment 
 borne by the figures and first clearly disclosed by the drawing. The figure in the 
 middle, which is the tallest in the group, bears in the left hand a celestial globe, showing 
 us that we have here the “Lord of the sky.”” The two military figures standing on 
 his either hand must therefore be Yarhibol and Aglibol. It is interesting to observe 
 that all three display a bright golden aureole behind the head, showing that, like the 
 deified emperors later, they have already been solarized. It was from such figures 
 as these that the aureole passed over into Christian painting. 
 
 Even more instructive are the figures of the two goddesses sitting below the three 
 Palmyrenian gods. They are two figures of Tyche, or goddesses of Fortune such as 
 became very common in Roman times, whether as bronze figures, on coins, or in wall 
 paintings, depicted bearing a horn of plenty and sometimes with a rudder to indicate 
 guidance of the fortunes of men. Our two Fortunas, however, are quite clearly 
 derived from that form of the goddess which was sculptured by the most gifted of the 
 several artists who bore the name of Eutychides. He was the pupil of Lysippos and 
 in the year B.c. 300 he wrought a statue of the patron goddess of Antioch which was 
 widely celebrated (PAUSANIAS vi. 2. 7). In a relaxed, almost reclining posture, the 
 goddess sat upon a rock. Upon her head she wore a crown like a fortified tower; 
 
 1 See Saglio-Pottier, s.v. “Signa,” p. 1310 (note 6). 
 
98 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 she bore a sheaf or a stalk of grain in her hand, and beneath her feet emerged the 
 river god of the Orontes. This figure has survived to us on the coins of Antioch and 
 in a marble copy now in the Vatican. 
 
 Fortunately for us, our two goddesses bear inscriptions designating them as ‘‘ Tyche 
 of Dura” (on the right) and “Tyche of Palmyra” (on the left). Each goddess 
 displays an aureole and wears on her head a castle showing that she is the personifica- 
 tion of a strong city, a symbol often seen on the heads of Syrian goddesses and already 
 employed by Eutychides. Below each figure there are additional symbols of impor- 
 tance, which appear for the first time in the drawing. For here our photographic 
 plate did not cover the left-hand goddess nor include the lower portion of the scene 
 under the other goddess—a loss which seriously interfered with a proper understanding 
 of the two goddesses. As Cumont has recognized, below the rocks on which the right- 
 hand goddess sits there is the figure of a man immersed above his middle in water in 
 which he is swimming. This is evidently a personification of the Euphrates, above 
 which Dura lies, while the child at the right rising from the waters is Dura itself, 
 which owes its origin to the river and enjoys the favor of the goddess, who lays her 
 hand protectingly upon the child’s head. Similarly under the rocks on which the other 
 goddess sits appears the figure of a woman placing her hand upon her breast, who 
 personifies the springs flowing from beneath Palmyra, which are the sources of its 
 life and sustenance. Above the personified waters of Palmyra sits a lion, the symbol 
 of the city, on the head of which the goddess lays her protecting hand. 
 
 Such Fortunas as these had in Roman times become the personification of good luck 
 commonly worshiped by the legionaries, and were localized much as we find the Virgin 
 to be in later times, so that each city or even each legion might have its own Fortuna 
 or Tyche. That the Roman troops levied in Palmyra and serving in the garrison of 
 Dura in the first quarter of the third century should continue their worship of their 
 Palmyrenian Fortuna, would be most natural. It cannot be doubted that the other 
 Fortuna, “Tyche of Dura,”’ is the patroness of the city in which our temple is situated.! 
 This painting has thus preserved to us the ancient name of the city with which we are 
 dealing—a name otherwise unknown in the documents recovered at Sdlihiyah. The 
 identification is rendered certain by the new Greek or Macedonian name given to 
 the place by Nikanor, compared with the reference found by Cumont in one of the 
 parchments recovered during his excavations. The inhabitants of our city are called 
 in this passage avw évres Eipwraton, ‘citizens of Huropos by ancestry,’’? and as we have 
 seen (p. 88), Europos is affirmed by Isidor of Charax to have been the Greek name of 
 
 1 As I suggested in Syria, III, 203, and as Cumont has made still more clear by the disclosure 
 of the situation of the city within the fortress of Sdlihtyah. In view of this evidence the suggestion 
 of G. Hoffmann (Sarre and Herzfeld, op. cit., pp. 394-95) that the ancient name of Dura was 
 TAAATAAA is untenable. 
 
 2 Cumont, Les Travaux Archéologiques, pp. 69-70. 
 
 
 
THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 99 
 
 Dura. The fertile plain of Khana-Mari on the east side of the river would obviously 
 
 be closely connected with its ruling city on the west side. 
 
 The references in the ancient 
 
 sources, which place our city on the left (east) bank of the Euphrates, doubtless refer to 
 
 a stronghold which would naturally protect the river 
 crossing at the other end, like the bridgeheads of later 
 times, and might easily acquire the same name as 
 the city proper on the right (west) bank of the river. 
 
 This painting raises some interesting questions 
 concerning the composition as a whole. Attention 
 has already been called to the fact that the worship- 
 ers all face the observer and not the gods and 
 goddesses they worship. It will be noticed also that 
 the divinities likewise face the observer rather than 
 their worshipers. It is obvious that all the figures 
 in the composition, human and divine, must be 
 turned around 90 degrees: the worshipers toward 
 the left and the divinities toward the right. Such 
 a change, however, at once involves us in difficulties 
 with the ground plan and the planes in which the 
 figures now appear. It is evident that we are view- 
 ing the divinities from in front, so that the chief god 
 would be in the middle, not behind the god nearest 
 to the worshipers, as he now seems to be. In other 
 words, we are viewing the three gods exactly as they 
 would be seen by the worshipers in front of them. 
 If we turn their bodies around 90 degrees to the 
 right we shall have ranged them single file with the 
 right-hand god at the head of the file toward the 
 worshipers. It is evident that such was not their 
 position as conceived by the artist. To place them 
 as they must have been, and as the position of the 
 niches in the shrine in the temple (p. 71) shows 
 they certainly were, we must not only turn the 
 three figures around 90 degrees to the right, but 
 we must also shift the position of the plane in 
 which they stand and make it nearly parallel with 
 
 
 
 
 
 ecco | 
 
 oS 
 
 Fie. 57.—Coiuar, NECKLACE, AND 
 PECTORAL ORNAMENTS OF THE 
 Lapy BITHNANAIA 
 
 See description, pp. 82-83 
 
 the observer’s line of vision. That is to say, the middle god (Baalsamin) when 
 turned around will remain on the same spot, but the other two gods on his left and 
 right must continue to stand on his left and right after he has been turned around. 
 
 1See Cumont, Comptes rendus (January, 1923), p. 38. 
 
100 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 They will then be standing in a plane at right angles to that which they now occupy. 
 The position of the worshipers is simpler. If we leave the ground plan of their feet 
 on the stage, as it were, just as they now are, merely turning them around 90 degrees 
 to the left, we shall have replaced them very nearly as they ought to appear. It 
 should be noted that in so doing we should not be changing the direction in which they 
 extend their right arms in the posture of worship, that is, toward the divinities. The 
 artist did not venture to be consistent and represent their right arms as extended 
 toward the observer in their present position, as he should have done. He has made 
 an exception of their right arms and represented them extended toward the divinities 
 as they in real fact were. 
 
 A consideration of similar questions in connection with the great painting (Plate 
 VIII) discloses an essentially identical disregard of the actual spatial relations and 
 at the same time furnishes interesting confirmation of our conclusions regarding the 
 painting of the tribune. In the great painting, however, this disregard is less evident, 
 owing to the fact that the artist has not depicted the divinities on the wall. This is 
 doubtless due to the fact that the artist (Ilasamsos) was painting in the same room 
 where the shrine containing the images of the three great gods also stood. He conceives 
 the images as actually present and therefore does not paint them. That this is in 
 fact the conception of the artist in the great painting is shown by the fact 
 that the right arms of the worshipers are extended straight out toward the observer, 
 that is, toward the shrine and the images of the gods in the middle of the hall. A 
 comparison of Plate VIII and the plan of Hall II (p. 95) will make this fact evident. 
 Why should the arms of the worshipers in the scene of the tribune be extended in 
 worship toward the left, and in the great painting (Plate VIII) directly toward the 
 observer? Because in both cases the direction depicted is that of the images of the 
 gods: in the first case on the wall, in the second in the shrine in the same room. This 
 analysis of the artist’s treatment of spatial relations is of importance in the proper 
 appraisement of this ancient Palmyrenian art. Such disregard of spatial relations is 
 essentially oriental and goes far back in the history of art in the Orient. It is quite 
 in keeping with the fact that the only artist at Dura whose name we know was an 
 oriental. There can be little doubt that he was practicing his art in accordance with 
 inherited tradition transmitted from far earlier oriental culture. 
 
 The presence of a Roman tribune worshiping with his troops in this temple on the 
 outermost oriental frontier of the Roman Empire is, historically speaking, only of 
 casual interest. The fact that he has had himself depicted on the temple walls thus 
 engaged in worship is of far greater importance. He and his comrades are the eastern- 
 most soldiery of Rome ever found depicted on the monuments. That he treated 
 with respect the painted memorials of the noble families of Dura elsewhere preserved 
 on the walls of the temple, like the great painting of Konon’s family executed by 
 Ilasamsos a century and a half earlier, is worthy of note. Julius Terentius as Roman 
 
 
 
THE WALL PAINTINGS OF THE ZEUS-BAAL TEMPLE 101 
 
 commandant of the stronghold of Dura, might easily have chosen for himself more 
 extensive and more prominent space on the walls of the temple halls. The scene in 
 which he appears is less than 2 meters long, while the family of Konon occupy the 
 entire wall in the sanctuary itself with eleven life-size figures. Nevertheless, the fact 
 that the tribune has had himself painted here at all, with his name appended to his 
 figure, shows a self-consciousness which reminds us of the custom, first observable under 
 Marcus Aurelius, of permitting the tribunes to record themselves as having erected 
 the emperor’s statues.!. In these temple paintings the Roman tribune has slipped 
 into the social and religious position once occupied by such noble families as that of 
 
 First Figure vey, N (Ce) N N | KOC ATOY 
 INV MY MWA OWT 
 
 Fourth Figure ! RI y N AND 
 KONWNOZ 
 
 Fifth Figure  AJOPTENHG KON CONOC 
 
 Sith Figure YSY MGC KONUDNOC 
 
 Seventh Figure [| A TPOK AOC KGNWNOC 
 Eighth Figure KON GUN MAT POKA € OYC™ 
 
 Ninth Figure Uy G2 
 
 KONWNOC 
 
 Fig. 58.—INScRIPTIONS ON THE WALL OF BITHNANAIA 
 
 A study of the photographs since our return to America has revealed two short inscriptions above 
 the right shoulder of the second figure. Immediately above the right shoulder is the inscription in 
 two lines: KONG@N NIKOCTPATOY. This must refer to the first figure. At the top, and 
 ending at the corner of the open door behind the head of the second figure, is a 4-line inscription 
 of which prolonged study by a skilled Greek epigrapher might possibly result in a reading. In the 
 course of our hurried single day’s work these inscriptions were not discerned. 
 
 Konon. But he, and perhaps other influential people of Dura in the first quarter of 
 the third century, were no longer able to secure for their commissions any artists 
 comparable in ability with Ilasamsos of four to five generations earlier. 
 
 On the same wall with that of the tribune, as well as on the neighboring space 
 on the front (east) wall of the same hall (I), are other small paintings belonging to 
 the later group with which we are now engaged.” Like the scene of the tribune, these 
 
 1 Domaszewski, Die Religion des romischen Heeres, p. 69. 
 
 2 We saw these smaller paintings around and near that, of the tribune, but there was no time even 
 to note their subjects. The following description is taken entirely from Cumont, Les Travaux Archéo- 
 logiques, pp. 68-69, and Comptes rendus (January, 1923). 
 
102 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 smaller paintings are far inferior in execution to the work of Ilasamsos. In subject- 
 matter they are largely obscure. 
 
 At the left of the painting of the tribune are several small scenes inclosed in rec- 
 tangular, painted frames. The upper register, of which the top is fragmentary, 
 displays a reclining woman, to whom a servant crowned with roses carries a dish of 
 food—a scene which reminds Cumont of the ‘funerary banquet,’’ so common at 
 Palmyra. Farther along four ministrants are presenting offerings on as many altars. 
 The lower register is subdivided into several small scenes: Hercules supported on his 
 club; a young man, the right hand lifted in the posture of worship; a she-goat, a ram, 
 a woman with an aureole about her head standing with face toward the front; and three 
 other persons wearing draped costumes. Below these pictures the twenty-four 
 letters of the Greek alphabet are inscribed in large characters. These letters, the 
 symbols of the elements of the world and the celestial bodies, which the same name 
 arotxeia designates, possessed for the ancients a sacred character, and they are fre- 
 quently found employed in magic on the phylacteries and in astrology as substitutes 
 for the twelve signs of the zodiac. In our temple also they certainly have a religious 
 significance, which it is difficult to define exactly. 
 
 On the neighboring pier weapons are depicted. The bow, the quiver, and arrows 
 recall the fact that the Palmyrenians were archers of repute and served as sagittarii 
 in the Roman army. These sagzttarzi are furthermore mentioned in a Latin graffito 
 traced on the wall of one of the towers of the inclosure, on which is also painted in 
 black and red a curious talisman against the evil eye: a poniard and a harpoon are 
 directed against the pupil of a large eye, which is likewise assailed by a bird and some 
 serpents. 
 
 Finally the east wall, between the entrance door and the south wall, is decorated 
 with two superimposed scenes which are closely related. The lower one of the two 
 is the only one entirely preserved. In a landscape where several gnarled shrubs 
 appear, a man is seated on a rock, with his head somberly resting on his left hand; 
 three companions advance extending their arms toward him, and as a characteristic 
 detail the third carries a long cudgel or crutch and appears to be a cripple. These 
 same personages differently grouped reappear in the upper register, which is seriously 
 damaged. ‘The interpretation of this scene is still to be found. 
 
 
 
 q 
 

 
 INDEX 
 
 Abd-Hadad, king of Hierapolis, 16 
 
 Africa, 7, 22 
 
 Aglibol, 4, 5, 71, 72, 73, 97 
 
 >Ahle Haqq, 17 
 
 Akhlami, 34 
 
 Albu Kamil, 2, 24, 42, 53, 54, 55, 58, 59, 60, 64, 65 
 
 Aleppo, 1, 6, 16, 22, 28, 27, 37, 55, 65, 76, 77 
 
 Alexander the Great, 1, 5, 21, 27, 28, 37, 38, 
 39, 92 
 
 Alexandria, 39 
 
 Ammi-Bail, 29 
 
 Ammonios, 43 
 
 Amorite Empire, 34 
 
 cAnah, 21, 22, 23, 27, 31, 32, 42, 55, 58 
 
 Anazah Arabs, 23 
 
 Antioch, 97, 98 
 
 Antiochus, 39 
 
 Apollophanes, 48, 90 
 
 Arabia Petraea, 18 
 
 Arabian Desert, 21 
 
 Arabs, 2, 15, 18, 49, 54, 58, 61, 75, 84 
 
 Aramaeans, 47 
 
 Ardashir, 46 
 
 Armenia, 26 
 
 Asia, 1, 6, 7, 22, 73 
 
 Asia Minor, 21, 26, 39, 41 
 
 Asicha, 66 
 
 Assur, 34 
 
 Assurnagirpal III, 37 
 
 Assyria, 21, 26, 34, 37, 62 
 
 Assyrians, 36, 40, 50 
 
 Athenodoros, 90 
 
 Athens, 43 
 
 Aurelian, 1, 47, 48 
 
 Baal, 5. See also Baalsamin and Zeus-Baal 
 
 Baalsamin, 43, 71, 97, 99. See also Baal and 
 Zeus-Baal 
 
 Babylon, 17, 21, 68 
 
 Babylon, First Dynasty of, 29, 33 
 Babylonia, 21, 26, 31, 33, 34, 56, 62, 73 
 Babylonians, 31, 40, 50 
 
 Baghdad, 1, 6, 23, 49, 52, 54, 55 
 Beduin, 4, 19 
 
 Beirfit, 3, 81 
 
 Bel, 16 
 
 Bithnanaia, 19, 45, 57, 64, 86, 99, 101 
 Bithnanaia, Wall of, 3-5, 18, 58-60, 76, 89, 90 
 Book of the Dead, 6, 7 
 
 British, 51 
 
 Byzantium, 8 
 
 Caesars, 18, 19, 47 
 Cairo, 7 
 
 Cannae, Battle of, 58, 97 
 Carchemish, 21 
 Carthage, 41, 42 
 Chaldean Empire, 37 
 Chaldeans, 17 
 
 Cilicia, 32 
 
 Ciliza, 16 
 
 Circesium, 16, 20, 24, 27, 29, 36, 48 
 Coffin Texts, 7 
 Constantine, 48, 97 
 Cyrus, 33, 34 
 
 Damascus, 15, 19 
 
 Darius, 34 
 
 Deinocrates, 39 
 
 Deir el-Far, 29 
 
 Deir ez-Zér, 23, 24, 30, 35, 38-45, 61, 62, 69, 73 
 
 Diocletian, 16, 24, 47, 65 
 
 Diogenes, 86, 101 
 
 Dionysus, 43 
 
 Dog River, 82 
 
 Dur, 37 
 
 Dura, 1-5, 8, 15, 18-20, 30, 31, 34, 36-48, 50-52, 
 62, 65, 66, 67, 71-73, 79, 85, 87, 91-98, 96, 
 98-101. See also Europos and Saélihtyah 
 
 Duri, Land of, 34 
 
 E-mach, temple, 68 
 
 Eannatum, king of Lagash, 31 
 
 Egypt, 3, 17, 18, 22, 26, 33, 34, 91 
 
 Emperor, Cult of, 18 
 
 Euphrates River, 1, 5, 6, 15, 17, 18, 20-26, 28-42, 
 44, 45, 47, 50, 52-56, 58, 62, 63, 65, 66, 71, 
 73, 98 
 
 Europe, 1, 2, 5, 6, 8, 44 
 
 Europos, 4, 38, 39, 46, 50, 66, 98. See also 
 Dura and Salihtyah; also Jerablis 
 
 EKutychides, 97 
 
 103 
 
104 ORIENTAL FORERUNNERS OF BYZANTINE PAINTING 
 
 Fallijah, 54, 55 Lagash, 31 
 
 Florence, 8, 44 Lucian, 16 
 
 Fortunas, 46, 71, 97, 98 Lysias, 45, 72, 86, 87, 88 
 
 France, 51 Lysippos, 97 
 
 Gaios, 16 Macedonians, 21, 37, 38, 65, 67 
 
 Gordian III, 19, 20, 47 Ma’er. See Mari 
 
 Great Mother, 16 Magi, 17 
 
 Great War, 1, 50 : Marcus Aurelius, 101 
 
 Greece, 92 Mari, 28-34, 36 
 
 Greeks, 27, 38, 92 Mediterranean Sea, 1, 6, 21-23, 39-42, 55, 56, 
 58, 81, 82 
 
 Hadithah, 28-30 Merra. See Merrha 
 
 Hadrian, 15, 45 Merrha, 30, 31, 36 
 
 Halabtyah, 35 Meskenah, 23, 37, 76 
 
 Hama, 3 Mesopotamia, 15, 17, 21, 28, 34, 37, 46 
 
 Hammurabi, 32-34 Mesopotamian Desert, 24, 26 
 
 Hana, 28. See also Khana Meyadin, 23, 29, 30, 46, 61, 66 
 
 Hercules, 102 Middle Ages, Art of, 19 
 
 Hierapolis, 16 Mithra. 17 
 
 Hit, 25-27, 56 Aner 
 
 Hittite headdress, 16 Molon, 39 
 
 Homs, 3, 34 Moses, 16 
 
 House of Blood, 20 Moslems, 16, 25, 49 
 
 Ibla, 32 Mosul, 6 
 
 Ikhnaton, 92 
 
 Tlasamsos, 5, 44, 64, 73, 90, 92, 100, 101 Nabagath, 66 
 
 Irzi, 30, 31 NRE 
 
 Theses Fit Nanaia, 86, 87 
 
 Near East, 5, 6, 26, 51 
 Nefretiri, Tomb of, 91 
 Nerab, Stela of, 17 
 
 New Persia, 46, 49 
 Nikanor, 37, 38, 62, 67, 98 
 Nikostratos, 86, 88, 101 
 Nile River, 6, 17, 22 
 
 Isar-lim, king of Khana, 28 
 Ishbi-Irra, 33 
 
 Isidor of Charax, 30, 31, 36, 37, 65, 98 
 Isin, Dynasty of, 33 
 
 Islam, 16 
 
 Izi-Dagan, 33 
 
 Jerablis, 21, 46 Nineveh, 21, 26 
 Jews, 17 Nosairis, 17 
 Judaism, 17 
 Julian, 19, 48, 66 Odenathus, 17, 47 
 Old Testament, 16 
 Kadesh, 3 Orient, 2, 5, 6, 7, 8, 16, 34, 38, 92, 100 
 El-Kaim, 22; 27, 33 Orontes River, 26, 80, 98 
 Kan-kaleh, 20, 66 9 ae 
 Kashtiliashu, 29 Palestine, 6 
 Khabar River, 16, 24, 25, 28, 34, 36, 50, 58, 65,66 Palmyra, 1, 15, 17, 18, 40-48, 56, 71, 73, 87, 92, 
 Khana, 29, 30, 34, 36, 61. See also Hana 97, 98, 102 
 Khana-Mari, Plain of, 25-31, 34, 36, 37, 40, Parthia, 21, 40, 47 
 46-51, 64, 99 Parthians, 15, 39-41, 46, 65, 67 
 Konon I, 16, 18, 44, 57, 64, 72, 86,87, 88,100, 101 Patroklos, 86-88, 101 
 Konon II, 86, 88, 101 Pausanias, 97 
 Kurdistan, 17 Perneb, Tomb of, 92 
 Kurds, 6 Persia, 5, 21, 26, 53 
 
 Kuseir <Amra, 18, 92 Persian Empire, 37 
 
 
 
INDEX 
 
 Persian Gulf, 21, 42 
 Persians, 34 
 
 Petra, 15 
 
 Philip the Arab, 19, 47 
 Philo of Byzantium, 67 
 Phoenicia, 6 
 
 Plutarch, 7, 58, 97 
 Polybius, 39 
 
 Pompey, 40 
 
 Pyramid Texts, 7 
 
 R&ahaba, 29 
 
 Ramadan-Beg ibn Shalldsh, sheikh, 69, 70 
 
 Ramadi, 55, 56 
 
 Rapiku, 34 
 
 Ravenna, 2, 8, 19, 73, 79 
 
 Red Sea, 22 
 
 Roman Empire, 5, 18, 21, 27, 33, 41, 47, 48, 49, 
 65, 100 
 
 Romans, 21, 24, 27, 40, 58 
 
 Rome, 15, 19, 40, 42, 45, 72, 100 
 
 Rummunidi, 36 
 
 Sabkhah, 36, 72 
 
 Sahara Desert, 22 
 
 Saladin, 50 
 
 Salihtyah, 2, 3, 4, 15-20, 31, 40, 42, 48, 50-56, 
 58, 61, 62, 65, 66, 93, 98. See also Dura and 
 Europos 
 
 San Vitale, Church of, 2 
 
 Sargon I, 32 
 
 Sargon II, 32 
 
 Seleucid Empire, 39 
 
 Seleucid Era, 43 
 
 Seleucids, 1, 40, 47 
 
 Seleucus I, 37 
 
 Seleucus, father of Apollophanes, 43 
 Semites, 33, 34 
 
 Sennacherib, 8 
 
 Septimia Bathzabbai. See Zenobia 
 Severus, Alexander, 4, 46, 96 
 Severus, Septimius, 15, 46 
 Shamash, 31 
 
 Shamshi-Adad, 28 
 
 Sirqu, 36 
 
 Subi, 28 
 
 Sulla, 40 
 
 Suwan, sheikh, 73 
 
 Syria, 1, 3, 15, 19, 21, 32-34, 39, 40, 44, 47, 72, 85 
 Syrian Desert, 1, 21, 37 
 
 Syro-Mesopotamian Desert, 21, 22, 25 
 
 Tabernacles, Feast of, 17 
 Tadmur. See Palmyra 
 
 Tartar River, 36 
 
 Tell <Ashara, 29, 30, 61 
 Terentius, Julius, 46, 58, 96, 100 
 Thebes, 6 
 
 Themes, son of Mokimos, 96 
 Tibni, 35, 71 
 
 Tigris River, 6, 21, 26, 34, 36 
 Tirka, 28, 29, 30, 31, 61 
 
 Trajan, 5, 15, 45, 64 
 Tukulti-Inurta I, 34 
 Tukulti-Inurta IT, 36 
 Tukulti-Ninib. See Tukulti-Inurta 
 Tyche, 46, 97, 98 
 
 Ur Dynasty, 33 
 Userhet, Tomb of, 91 
 
 Varro, 58, 97 
 Vatican, 98 
 
 Verus, Lucius, 15, 46 
 Virgin, 98 
 
 El-Werdi, 25 
 
 Yarhibol, 4, 5, 43, 71-73, 97 
 Yarmuti, 32 
 
 Zaitha, 19, 20, 66 
 
 Zautha. See Zaitha 
 
 Zenobia, 17, 44, 45, 47, 56, 57, 73, 86, 87 
 
 Zenodotos, 90, 91 
 
 Zeus, 4 
 
 Zeus-Baal, 43, 71, 72, 73, 97. See also Baal and 
 Baalsamin 
 
 Zeus-Baal, Temple of, 4, 39, 48, 64, 68, 90, 91, 95 
 
 Zeus Megistos. See Zeus-Baal 
 
 Zim{ri], king of Mari, 29 
 
 PRINTED IN THB U.S.A. 
 

 

 
 1.—Looking Southwest in Northwest Wadi: Extreme Western Corner of Fortress 
 at Left. 
 
 
 
 2—Gate near Middle of Southwest Wall 
 
 
 
 Pl. I. — Fortress of Saélihiyah 
 
 
 
 
 
 
 
4 
 ’ 
 
 ea] 
 
 all 
 
 3 
 
 yt 
 
 
 

 
 
 
 1.—Looking Northwest along Southwest Wall: Southwest Gate near Middle (Pl. I, 2) 
 Seen on Right. Note Sand Drifts Rising to Crest of Wall in Places. 
 
 
 
 2.—View along Inside of Same Wall. Looking Southeast past Inside of Southwest 
 Gate. Note Sand Drifts Rising to Crest of Wall in Places. 
 
 Pl. Il. — Fortress of Saélinfyah 
 

 
 2 ’ 
 a - \ . 
 , 
 2 = - ~ 
 - - = 
 . to = - & 
 - fre : p 
 5 - 
 a 
 = 
 > 
 - 
 ~ “ - 
 ‘ — 
 _ AS 
 ' 
 i 
 - x 
 A > 
 if 
 - - \ 
 — 7 
 ? 
 A “ 
 i t 2 
 h ‘ 4 
 P) X { 
 i 
 ll 
 . 
 t 
 - ~ 
 ~ 
 ‘: 
 - 
 : Fe - 
 J * 
 * 
 

 
 1.—Inside View Looking East along Massive Castellum and North Wall 
 
 
 
 2—Northwest Angle of Castellum, Looking a Little East of North 
 
 Pivilie— Fortress ol salihiyan 
 
 
 
 
 
+ 
 
 
 

 
 1.—Outside View of North End of Castellum, Looking East 
 
 
 
 2.—Northwest Bastion of the Castellum, Looking Northeast to the Euphrates 
 
 Pl. 1V. — Fortress of Salihiyah 
 

 
 > 
 . 
 

 
 1.—Cross-Seetion of Wall, Showing Evidenees of Two Structural Stages 
 
 
 
 2—Outside View of Extreme West Angle Containing Chapsl, Looking Southeast. 
 Shrine below A. 
 
 Pl. V. — Fortress of Salihtyah 
 

 
 ~ 
 ~ 
 

 
 1.—Outside View of Extreme West Angle Containing Chapel, Looking West. Shrine 
 at F; Wall of Bithnanaia at CD; Rubbish from Excavations Lying at the Angle of Rest 
 at Right of Middle. 
 
 
 
 2—Shrine in Hall II, Looking Southwest 
 
 Pl. VI. — Fortress of Sdlihiyah 
 
 
 
a 
 
 t 
 i 
 ’ 
 & 
 i 
 1 
 ‘ 
 ‘ 
 ' 
 Na 
 i“ 
 j 
 
 
 

 
 1.—The Two Columns (G and H in Fig. 836) at West Side of Court 
 
 
 
 2—Steps Leading into North End of Hall II at O (in Fig. B6) 
 
 Pl. VII. — Chapel of SAlihiyah 
 

 

 
 
 
 General View (C-D) 
 
 in Hall II 
 
 VIII. — The Wall of BithnanafYa 
 
 JPA. 
 

 
 / 
 

 
 
 
 RB ccige eee se GEM 
 
 ale =2 
 
 
 
 Pl. 1X. — The Wall of BithnanaYa in Hall II : The three ministrants 
 
 ChieSiesecconcdeancd tinted eng mres) 
 
m4 
 
 
 
mae 
 
 — The 
 
 Wall 
 
 of BithnanayYa 
 
 and her group (second t« 
 
 > 
 
 
 
 in Hall II : BithnanatYa 
 
 sixth figures) 
 
Me 
 
 er 
 
 
 
So pies Sia cg RN inet ay RRM mien in <e: Monws 
 
 ef 
 
 
 
 Pl Xl = The Wall of Bithnanaiavin Fall ll: Head of the Second 
 Figure, Showing Injury to Left Eye (Restored in Pl. X). 
 

 

 
 Pl. XIII. — The Wall of Bithnanaia in Hall Il: The Three Ministrants 
 (First, Second, and Third Figures). 
 

 
sn hin: 
 
 Va mae Ee i ; 
 sepa SEMA TRE Rin a | 
 i ‘ md 
 
 tod 
 
 ene RR Sai 
 ¥ 
 
 
 
 Pl. XIV. — The Wall of Bithnanaia in Hall II: Bithnanaia and Her 
 Group (Second to sixth Figures and Blevyenth Higuire). 
 

 

 
 Plate XV.— The Wall of Bithnanaia in Hall II: Upper Part of 
 Figure of Bithnanala. 
 
“« 
 
 
 

 
 the 
 
 5 drulelat, lraxels 
 
 ay Aeslkswilil Wil 
 
 ithnanaia 
 
 It, 2OWOAL, == Teves WAVeall @ie 15} 
 
 Four Men with Green Br 
 
 aneches. 
 
 2 
 
 
 

 

 
 Pie Lia Lie VWallvor sithnanaia in Fall iis (Ninth igure, 
 the First Boy in the Foreground. 
 
 
 
4 
 
 
 

 
 PI ACV == The Wall-or Bibnanaiain Hall Il: Tenth Figure, 
 was (Ganell may wae I Cyreexeperonuiiavel,. 
 
 
 

 

 
 
 
 
 
 Pl. XIX. — The Wall of Bithnanaia in Hall II: Eleventh Figure, 
 the Second Boy in the Foreground. 
 
‘<= 
 
 whe fi 
 
 a “ol 
 
 e 
 
 “py Ae 
 
 
 

 
 1—Weest Wall of Hall II (D-E in Fig. 86), with Desert West of Fortress on Right 
 
 
 
 2—Partly Execavated North Wall of Court (P-Q in Fig. 56), Looking Northeast along 
 Northwest Angles of the Fortress down the Northwest Wadi to the Plain of Khana-Mari. 
 
 Pl. XX. — Chapel of Salihiyah 
 
 
 
euNqIdT 241 JO 11BM eUL — 
 
 “LXX 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i ‘A - y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J ma M8 SS 
 es iy) aS 
 NEO * 
 Zo 
 
 arf 
 
 i EO name, 
 i a cs 
 GH EH 3 j ’ 
 oa Th ee 
 q Someta 
 Ut 
 
 
 
 2UMOUND "TY ko paysrtstf Cpury Suemvsrg 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
*f 
 
 ie 
 
 Gn ret 
 ae 
 
 a aan 
 
 
 

 
 2.—Empress Theodora and Suite 
 
 Pl. XXII. — Sixth-Century Mosaics in the Basilica of S. Vitale at 
 Ravenna. 
 
. 
 
 7 As oN V 
 
 =a 
 YP? 
 Dut 
 
 
 

 
 1.—Priest Offering Sacrifice to the God Bel. Relief from Ciliza (Killiz) 
 
 
 
 o—Mortuary Relief from Palmyra. Musée du Louvre 
 
 Pip sii 
 
 
 
@ a = A 2? Gia sD 
 b ey 
 ae, 
 ae 2 
 * a Sy) 
 3 ® ‘ PY, 4, 
 < oP 
 , 
 } 
 i] 
 ; 
 
 Ss 
 
 Se Stet Le Sh ed 
 bn 
 
 
 
A os . 
 ape ea 
 
 ok 
 
 oh ie i