Greek Texts and Armenian Traditions an Interdisciplinary Approach

www.ebook3000.com

Greek Texts and Armenian Traditions

www.ebook3000.com

Trends in Classics – Supplementary Volumes

Edited by Franco Montanari and Antonios Rengakos Scientiﬁc Committee Alberto Bernabé · Margarethe Billerbeck Claude Calame · Philip R. Hardie · Stephen J. Harrison Stephen Hinds · Richard Hunter · Christina Kraus Giuseppe Mastromarco · Gregory Nagy Theodore D. Papanghelis · Giusto Picone Kurt Raaﬂaub · Bernhard Zimmermann

Volume 39

www.ebook3000.com

Greek Texts and Armenian Traditions An Interdisciplinary Approach

Edited by Francesca Gazzano, Lara Pagani, and Giusto Traina

www.ebook3000.com

ISBN 978-3-11-047911-9 e-ISBN (PDF) 978-3-11-048994-1 e-ISBN (EPUP) 978-3-11-048866-1 ISSN 1868-4785 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2016 Walter de Gruyter GmbH, Berlin/Boston Logo: Christopher Schneider, Laufen Satz: Dörlemann Satz GmbH & Co. KG, Lemförde Printing and binding: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

www.ebook3000.com

Table of Contents Preface IX Armenian Transliteration Table

XI

I. Problems of Method Moreno Morani Ancient Armenian Translation from Greek Texts: Questions of Method Giusto Traina Some Observations on the Armenian Pseudo-Callisthenes Alessandro Orengo Eznik of Kołb as a Translator of Methodius of Olympus

3

23

31

Valentina Calzolari The Transmission and Reception of the Greek Cultural Heritage in Late Antique Armenia: The Armenian Translations of the Greek Neoplatonic Works 47

II. History and Historiography Gianfranco Gaggero Armenians in Xenophon

73

Francesca Gazzano Croesus’ Story in the History of Armenia of Movsēs Xorenac‘i

83

Francesco Mari Cyrus the Great in Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘: Telescoping the King 115 Anahide Kéfélian The Roman Army and the Transmission of Latin Loan Words in Old Armenian 143 Federico Frasson Armenia and Armenians in Asinius Quadratus’ Παρθικά

www.ebook3000.com

163

VI

Table of Contents

III. Biblical, Christian and Byzantine Literature Theo Maarten van Lint Among Others: Greek in Context in the Letters of Grigor Magistros Pahlawuni (eleventh century) 197 Armenuhi Drost-Abgaryan The Reception of Eusebius of Caesarea (ca. 264–339) in Armenia

215

Alessandro Capone Pseudo-Athanasius, De incarnatione Christi contra Apollinarium: Some Critical Remarks 231 Lia Raffaella Cresci The Hexaemeron of George of Pisidia and the Armenian Version: Textual Investigations 241

IV. Philological, Linguistic and Lexical Problems Giulia D’Alessandro and Lara Pagani Homer in the Armenian Tradition 257 Chiara Aimi Appendix – Codex M 437 272 Irene Tinti Grecisms in the Ancient Armenian Timaeus

277

Andrea Scala Greek, Syriac and Armenian in Contact: Lexical and Textual Outcomes

299

Rosa Ronzitti Etymological Hypotheses Between Armenian and Indo-Iranian: New Proposals and Remarks 311

Index of passages Greek texts Latin texts

329 337

www.ebook3000.com

Table of Contents

Armenian texts 338 Babylonian texts 341 Egyptian texts 341 Syriac texts 341 Epigraphic texts 341 List of Contributors

343

www.ebook3000.com

VII

www.ebook3000.com

Preface We are delighted to present the Proceedings of the Genoa conference of 21 and 22 October 2013¹, which was held with the aim of resuming a tradition of scholarship that focused on the Classical heritage in the Eastern traditions. Organized in Italy in the years 1980/1990, these early seminars were held under the pivotal guidance of the renowned linguist and Armenologist Giancarlo Bolognesi (1923– 2005). Various language areas of the Christian and Muslim East were taken into consideration, thereby involving Classical philologists and Orientalists in a fruitful scientific dialogue and often marking significant progress in our knowledge of authors such as Aristotle, Aelius Theon, or the Pseudo-Callisthenes. Several years later, two participants who had contributed to these events – Moreno Morani, himself a pupil of Bolognesi, and Giusto Traina – proposed a revitalisation of the spirit of the meetings, focusing on Classical, Christian, and Byzantine Greek texts in the Armenian tradition. Their suggestion was sponsored and supported by the Doctoral Schools of Genoa (“Scienze dell’Antichità”) and Paris-Sorbonne (ED 1, “Mondes antiques et médiévaux”), by the Genoese Departments DAFIST and DIRAAS, and by the Parisian UMR 8167 (“Orient et Méditerranée”). Finally, a contribution was provided by the Association Internationale des Études arméniennes. The scientific board was composed of Professors Paul Demont, Francesca Gazzano, Charles de Lamberterie, Franco Montanari, Moreno Morani, and Giusto Traina. Classical philologists, with the support of ancient historians, benefited from the opportunity to engage in fruitful dialogue with linguists and Armenologists, jointly aspiring to “total philology”. In spite of the “natural” rejection of Oriental texts by the average Classical scholar, who is typically uneasy when faced with the Unknown and haunted by the Demon of Over-Specialization, we succeeded in introducing a spirit of mutual and interdisciplinary cooperation. The methodological problems are addressed in Part I, introduced by Moreno Morani, whose brief history of the studies is followed by a series of specific examples. Case studies (Pseudo-Callisthenes, Eznik of Kołb) are offered by Giusto Traina and Alessandro Orengo; Valentina Calzolari presents an important survey of the Armenian translation of the Neoplatonic works. The studies of Part II mark important progress in the historical and historiographical disciplines. After the remarks on the “Armenian chapters” of Xenophon’s Anabasis by Gianfranco Gaggero, a new file in the studies on Movsēs Xorenac’i is opened. This heralds an innovative perspective on the subject, in

1 A survey of the conference has been drawn up by M. Donato in Maia 66/1, 2014, 200–205.

www.ebook3000.com

X

Preface

contrast to the dismissive appoach often adopted by scholars of ancient history, who (influenced by Robert Thomson’s minimalist approach) frequently reject any assumption that the History of Armenia can constitute a useful source. Francesca Gazzano’s thorough examination of the chapters on king Croesus shows that Xorenac‘i is dependent here on a Greek alternative tradition. In the same spirit, Francesco Mari comments on the chapters regarding Cyrus the Great. An interesting approach to Armenian military vocabulary, with the use of epigraphic and archaeological sources, is offered by Anahide Kéfélian. Finally, Federico Frasson gives a useful contribution on the fragments of Asinius Quadratus as a source of Armenian history. Finally, Part IV gathers together a series of carefully argued studies on philological, linguistic and lexical problems. Lara Pagani and Giulia D’Alessandro offer some observations on an unusual topic in the field of Homeric studies, investigating the presence of Homer’s poems and their contents in the Armenian tradition; the paper is supplemented by an appendix containing a thorough description of ms. Erevan, Matenadaran, 437 by Chiara Aimi. Irene Tinti studies the lexical aspects of the Armenian translation of the Platonic Timaeus, and Andrea Scala presents a cross-consideration of Armenian and Syriac texts, showing the plurilinguistic character of early medieval Armenian culture. Finally, in a critical examination of the Etymological Dictionary of the Armenian Inherited Lexicon by Hrach K. Martirosyan (2010), Rosa Ronzitti highlights the vitality of the etymological interest of Classical Armenian. Given the relevance, first and foremost from a methodological point of view, of events of this kind, we are confident that the Genoa conference will herald the advent of a new series, which the Doctoral Schools involved will hopefully be able to plan in the near future. Francesca Gazzano Lara Pagani Giusto Traina This volume is a special issue (Conference Proceedings): peer review procedures have been entrusted to the conference organizers and scientific committee. The English translation of the contributions by L. R. Cresci, G. D’Alessandro & L. Pagani, F. Frasson, F. Gazzano, G. Gaggero, F. Mari, M. Morani, R. Ronzitti are by Rachel Barritt Costa, as well the English language revision of the contributions by C. Aimi, V. Calzolari, A. Drost-Abgaryan, A. Kéfélian and G. Traina.

Armenian Transliteration Table Ա Բ Գ Դ Ե Զ Է Ը Թ Ժ Ի Լ Խ Ծ Կ Հ Ձ Ղ Ճ Մ Յ Ն Շ Ո Չ Պ Ջ Ռ Ս Վ Տ Ր Ց Ւ Փ Ք Օ Ֆ

ա բ գ դ ե զ է ը թ ժ ի լ խ ծ կ հ ձ ղ ճ մ յ ն շ ո չ պ ջ ռ ս վ տ ր ց ւ փ ք օ ֆ

a b g d e z ē ǝ t‘ ž i l x c k h j ł č m y n š o č‘ p ǰ ṙ s v t r c‘ w p‘ k‘ ō f

I. Problems of Method

Moreno Morani

Ancient Armenian Translation from Greek Texts: Questions of Method Abstract: The aim of the paper is to discuss the most relevant problems the researcher has to solve when investigating the ancient Armenian versions of the so-called Hellenistic School (Yunaban dprocʽ) from an essentially philological point of view. The versions of this school have special features: they are so strictly literal that their technique seems to be not even verbum e verbo, but rather morphema e morphemate. This characteristic often makes it impossible for the Armenian reader to gain a clear understanding of the original text, but it is conceivable (as some scholars have supposed) that their aim was to give standard and unchangeable versions which would represent the Greek text in all details. Their importance is basically philological, because their contribution leads us to assume a Greek text that in several cases is some centuries older than the most ancient manuscripts of the direct tradition. Thus some methodological assumptions are necessary: first of all we must compare Armenian and Greek, in order to establish the text used by Armenian translators and determine the reasons underlying the respective differences. When we conclude that there is a divergence between the original and the translation, we need to ascertain the relationship between the two texts and reach a decision on whether the Greek text presumed from the Armenian version is preferable than the Greek transmitted text or not. Only after these steps can the researcher exactly evaluate the philological relevance of the Armenian version and insert it in the Greek stemma codicum.

The first part of this article focuses attention on the name of a great scholar, the late Giancarlo Bolognesi. The memory of this prestigious figure is still vivid at the University of Genoa, where he taught German Philology as a Lecturer in the early years of his professorial career, and while the greater part of his scientific and teaching activity took place in Milan, at the Università Cattolica del Sacro Cuore, his presence in Genoa has left a powerful impression that still reverberates today. Even now: at a distance of almost half a century, many of his former students, some of whom have themselves entered an academic career, look back gratefully on his teachings. The reference to the name of Giancarlo Bolognesi calls forth memories the author of this paper holds particularly dear. I first met Bolognesi when I attended

4

Moreno Morani

a course he was teaching on Historical Linguistics during the academic year 1965–66, and I maintained contact with his activity throughout over forty years, until his death in December 2005. The present article is not intended to be a commemoration, and I would like to refrain from any kind of academic celebration; but I cannot omit to say from the start that a large part of the ideas and arguments I will put forward here are the outcome of the my Master’s teachings, which I had the benefit of learning not only through reading his works but also by listening to his own voice. The paper also bears a relation to another important experience, in which Bolognesi was directly involved and played a major role: his participation in a series of seminars devoted to the Oriental translations of Greek texts. This was an experience of study and research that extended over almost twenty years and gave rise to a number of meetings held in various cities of Italy, with the participation of renowned scholars from many continents having expertise and specialization in many different fields. The seminars, partially financed by funds from the MIUR – Ministero dell’Università e della Ricerca Tecnologica (the funds which at that time were called “Fondi 40 %”: the name has changed since then, as has that of the Ministry itself) gave rise to substantial volumes of Proceedings. This prolonged activity was initiated thirty years ago, in March 1983, in a little room of the Istituto dell’Enciclopedia Italiana, in Rome, when a small group of scholars sitting around a table began to put forward their ideas and debated the possibility of cooperation among specialists in classical philology and oriental languages. A second seminar was held the year after, and was followed by other meetings in later years, up to the first years of the current century. Those present at that first meeting included distinguished scholars who carried out this research activity within a wide variety of contexts. I would like to recall the names of some who are no longer with us today, in order to pay tribute to their memory among those who had the pleasure of working with them and those who regret not having had the opportunity to meet them in person. In addition – of course – to Bolognesi himself, I cannot neglect a mention of Massimiliano Pavan and Umberto Cozzoli, the ancient historians who organized the meetings, and the multifaceted and extremely active Orientalist, Michel van Esbroeck, as well as Arabic specialists such as Father Yusuf Habbi and specialists in other areas whom it would be impossible, here, to name one by one. Initially, the project was set up by ancient history scholars on the basis of a hope that proved to be illusory, namely the prospect of finding a manuscript containing a complete Armenian translation of Diodorus Siculus. This expectation was prompted by a passage from the book of Suk‘as Somalean, Quadro delle opere di vari autori anticamente tradotti in Armeno (Venezia 1825), which reads as follows:

Ancient Armenian Translation from Greek Texts: Questions of Method

5

Biblioteca storica di Diodoro di Sicilia. Altro tesoro preziosissimo è questo lavoro, la cui versione in Armeno fatta nel quinto secolo, come ne siamo certi, potrebbe ben consolare la repubblica letteraria già afflitta per la perdita di parecchie parti dell’originale¹.

This certainty expressed by Monsignor Somalean was not substantiated by the facts, but his assertion gave rise to a long-lasting and interesting international scientific adventure. The study of ancient translations of Greek texts into other languages represented one of the major aspects on which Bolognesi’s scientific activity focused. But it was not an exclusive interest. Bolognesi was an Indoeuropeanist in the full sense of the term, and he has left us a notable quantity of works that touch on innumerable aspects of Indoeuropean languages. While Armenian and Iranian hold a predominant position in his scientific production, one should not overlook the many important works devoted to the Germanic area, as well as Greek, Latin, Lithuanian and numerous other languages. This broad spectrum of interests was a function of a central idea that was frequently at the forefront of his approach throughout his prolonged activity as a scholar and professor: Bolognesi believed that linguistics played a central role in understanding history and in particular the history of civilizations. The study of languages enabled him to enquire into the history of peoples and explore the flows of civilization. In what could be defined as his Statement of Intent, namely the introductory speech to his course on Historical Linguistics (Glottologia), which he delivered at the Catholic University of Milan in May 1963 – the very title of which reveals his conception of language as an essential element of the culture of a civilization – he asserted his frustration with “un certo modo di concepire la storia esclusivamente come un incessante scontro di popoli”². Rather, it would be desirable, he argued, to move towards a vision of history “fatta non solo di scontri che possono annientare una delle sue parti in lotta, ma anche di incontri che stabiliscono fecondi rapporti tra uomini appartenenti alle più diverse tradizioni culturali, linguistiche e religiose³.

1 Somalean 1825, 28: “Historical library of Diodorus of Sicily. Another priceless treasure is this work, of which the translation into Armenian carried out in the fifth century, as we know for certain, could be of great consolation for the literary republic, already severely afflicted by the loss of numerous parts of the original”. 2 Bolognesi 1964, 147 (“a certain manner of viewing history exclusively as an incessant series of clashes among peoples”). 3 Bolognesi 1964, 147 (“composed not only of clashes that can wipe out one of the battling parties, but also of encounters that establish fertile relations among men speaking the most disparate languages and belonging to the most disparate cultural and religious traditions”).

6

Moreno Morani

Study of the translations of Greek texts into Armenian plays a prominent role in this vision, since nothing is more significant than a translation in forging a dialogue between cultures. The ancient Armenian translations of Greek texts display certain curious characteristics by virtue of which, in a sense, they can be considered as mysterious objects, because they raise many questions to which we can give no answer: who were they for, and what purpose did they serve? What tools did the translators have at their disposal? Given the slavish character of these translated texts, in which the technique of verbum e verbo is often pushed to the limits – so much so that it is replaced by a most singular technique of morphema e morphemate – to the point of making the translated text difficult to grasp for those unable to resort to the aid of the Greek text, they seem to defy the most elementary principles of the theory of translation: for what useful purpose can a translation serve if it fails to enable a foreign reader to gain a proper understanding of a work written in a different language? Allow me to call to mind a piece I wrote many years ago: in an article included in a Festschrift dedicated to the great linguist Eugenio Coseriu, in memoriam, I used the phrase “traduzioni per difetto” (“translations rounded down well below the average”), stating: Giungiamo così al livello più basso dell’attività del traduttore (o dell’interpres, dell’Überträger): siamo al di sotto di quell’ideale minimale di traduzione che permette di cogliere nella lingua d’arrivo il senso complessivo dell’originale pur nel dettato spoglio e disadorno di una lingua e di una sintassi pedissequamente esemplata su quella dell’originale⁴.

However, even this statement must be taken cautiously. Admittedly, these ancient versions must have been strewn with a number of stumbling blocks for the Armenian reader, but it is also true that this difficulty did not preclude their utilization or the possibility for the translated work to be fully incorporated in the cultural horizon of Armenian literature. As stated by Marc Nichanian, these translations never intended to convey the sense, but the Letter, because the entire sense was already contained in the Letter. Thus it was not the translations, but rather the commentaries, that were entrusted with the task of providing an account of the sense. The translations supplied the indispensable basis for the unceasingly renewed work of drawing up the commentaries, and the commentary was the

4 Morani 2003b, 327 (“We thus reach the lowest level of the activity of a translator [or of the interpres, of the Überträger]; here we are below that minimal ideal of translation which makes it possible to take a look at the text in the target language and grasp the overall meaning of the original despite the impoverished and unadorned wording of a language and syntax apishly modeled on the original”).


7

only manner in which, at that time, the life of the spirit was given expression⁵. One need only cite the example of Pseudo-Dionysius: a translation not always devoid of mistakes, in a complex and at times barely intelligible text, gave rise, even in its original form, to a thriving production of scholia and commentaries, which continued through the centuries. The high number of manuscripts that have come down to us (as many as fifty or so are held in the Matenadaran) shows the interest aroused by the translation. The decline and fading interest in this translation dates from many centuries later, as shown by the fact that it was no longer printed (a critical edition of the ancient translation was published by Thomson in 1987) and was replaced in the seventeenth century by a new translation, drawn up by a Polish Armenian scholar, Step‘anos Lehac‘i (“Stephen from Poland”), who produced a translation that was perhaps more easily readable, but was not a first-hand translation since it was created on the basis of a Latin translation. Overall, however, poor readability is a feeling that creates greater frustration in modern Armenian readers, who are easily irritated and bothered by having to cope with a fairly unintelligible text without being able to achieve a full or at least satisfactory understanding of a work which, on the face of it, is written in their own language. Bolognesi always emphasized that in order to study these translations, a good knowledge of Greek is just as important as a good knowledge of Armenian, if not more so: Potrà forse sembrare un paradosso, ma queste traduzioni armene richiedono una maggiore conoscenza del greco che dell’armeno, perché la loro lingua è un artificioso e superficiale travestimento armeno di costrutti e vocaboli propriamente greci. È cioè un “armeno” che deve essere ritradotto in armeno perché diventi comprensibile agli stessi armeni⁶.

This is a judgment shared and confirmed by many other scholars⁷.

5 Nichanian 1989, 235. 6 Bolognesi 1982, 13 (“It may perhaps seem paradoxical, but these Armenian translations call for a greater knowledge of Greek than of Armenian, because their language is a contrived and superficial Armenian camouflage of constructions and words that are actually Greek. That is to say, it is an ‘Armenian’ that has to be retranslated into Armenian in order for it to be understandable to the Armenians themselves”). 7 E.g. Fermeglia 1964–65, 231: “Questi ellenismi danno un aspetto artificioso alla lingua e la sfigurano in modo tale che, se non si potesse attingere alla fonte originale, essa sarebbe difficilmente intelligibile” (“These Hellenisms endow the language with an artificial look and alter it to the point that it would be hardly intellegible, if one were not able to draw on the original source”); Abełian 1936, 4*, quoted from Bolognesi 1982, 13: “Diese Schriften sind oft dermassen sklavisch, Wort für Wort, ja Silbe für Silbe ihrem griechischen Urtext nachgeahmt, dass sie nicht

www.ebook3000.com

8

Moreno Morani

I would like to summarise here a few problematic aspects faced by scholars in seeking to utilise these translations. Those who have read Bolognesi’s works will be well acquainted with his rigorous mode of proceeding, and the fascination arising from an analysis that reaches the point of drawing objective conclusions only when all the data of the problem have been taken into consideration and verified. In the light of this severe discipline, Bolognesi would always warn that it was necessary to exercise caution both in considering the data and in the procedural method adopted. As Bolognesi stated: La scoperta di una traduzione può non solo evidenziare, ma anche risolvere problemi non avvertiti neppure dai più esperti filologi nell’edizione critica di un testo. Da qui l’opportunità che a un’edizione critica collabori anche un linguista, specialmente se esperto di tecniche e processi traduttivi⁸.

einmal der beste Kenner des Armenischen ohne die Kenntnis des Griechischen und des behandelten Stoffes verstehen kann. Nicht nur der gesamthe Satzbau, sondern auch die Formenlehre, zum beträchtlichen Teil auch der Wortschatz vieler Bücher sind eher griechisch als armenisch zu nennen” (“These texts often duplicate their Greek original so slavishly, word by word, even syllable by syllable, that even the best expert in Armenian would be unable to understand them without knowing Greek and the subject matter. Not only the general phrase construction, but also morphology and, to a large extent, even the vocabulary of many books are to be defined as Greek rather than Armenian”). Further, Teza 1893, 6: “l’armeno può dare un ditino per guidare sulla buona strada il greco […]; ma questo darà a quello una mano intiera” (“Armenian may provide a prompt or two so as to lead the Greek in the right direction […]; but the latter will take the reader all the way home”); Mercier 1978–79, 74: “Les traductions de l’E. H. supposent en effet qu’on connaît assez le grec pour procéder à une rétroversion. Même, dans plus d’un cas, cette connaissance, si poussée qu’on l’imagine, ne suffit pas. Il faut avoir le texte grec sous les yeux pour comprendre l’arménien” (“As a matter of fact, the translations of E. H. assume that the reader knows Greek well enough to be able to back-translate the text. However, in more than one case, this knowledge, first-rate though it may be, does not suffice. It is necessary to have the Greek text at one’s disposal to be able to understand the Armenian”); Ter-Petrossian 1984, 12: “Les représentants de l’école grécophile se soumettaient au principe de la traduction littérale de l’original, c’est pourquoi leurs traductions se caractérisent par des tournures stylistiques artificielles et des constructions syntaxiques étrangères à la langue arménienne. Néanmoins, la culture arménienne, ayant assimilé les résultats positifs de l’activité des traducteurs de l’école grécophile, a pu se débarrasser par la suite de toutes les exagérations” (“The representatives of the Hellenizing school were prone to adhere to the principle of literal translation of the original: this is the reason why their translations are characterised by artificial stylistic expressions and syntactic constructions which are alien to the Armenian language. Nevertheless, Armenian culture assimilated the good results of the translators of the Hellenizing school and was subsequently able to overcome any exaggerations”). 8 Bolognesi 1997, 228 (“The discovery of a translation may not only highlight, but even solve problems not detected even by the most expert philologists who have dealt with the critical edi-


9

1 Comparison between the text and the translation a. The work involved in collation. The first task to be undertaken by a researcher is to set up a comparison between the Greek text and the Armenian translation, in order to establish which Greek text underlies the translation. Even though it is said that these ancient translations can be utilized codicis instar, it should always be borne in mind that the texts forming the object of comparison are written in different languages, and consequently there always remains some margin of indeterminacy. Collation of a codex presents an almost absolute degree of objectivity: unless the materials have severely deteriorated, doubtful lessons for which more than one reading is possible are a rarity. In the case of a translation, on the other hand, determination of the original reading is the outcome of a process of deduction and reasoning, and the majority of the language texts are open to several different readings. Despite the attempt by the Armenian translators to achieve a rendering that would reproduce the Greek text as literally as possible, it should not be overlooked that Armenian and Greek have rather different language systems, and these divergences cannot be masked even by a rendering that seeks to be as literal and blindly imitative as possible. Armenian has two verb stems as opposed to the three of Greek, it has no grammatical gender whereas Greek distinguishes three genders, Armenian does not have the distinction between genitive and dative that characterizes Greek, and so forth. This means that in many cases the Armenian translation cannot provide an answer, or cannot provide an absolutely precise answer, to our question concerning which was the reading known to the translators. For instance, whereas a Greek codex allows us to distinguish between ἄνθρωπος and ἄνθρωπον without a shadow of doubt, the mard of Armenian does not distinguish between the two possibilities. Thus on the one hand we have the evidence, and on the other a moral certainty, which we can rely on, but it is always vital to check the data and the outcome. Even when we have no reason to believe that the translator was working with a different text, the possibility that the original available to the translator had a different reading, but compatible with the proposed rendering, is destined to remain unresolved. b. The critical apparatus. The observations put forward so far do not so much concern the direct work on the text as, rather, the consequences for the critical apparatus of critical editions. The reader of a critical edition generally assumes

tion of a text. Hence the need for a critical edition to be drawn up with the aid of a linguist as well, especially if the linguist is an expert in translation techniques and processes”).

10

Moreno Morani

that silence of the critical apparatus implies the lack of variants in a codex. If we find, in a tradition with two codices AB, the note ἄνθρωπος] ἄνθρωπον B, we can reasonably conclude that codex A presents the reading accepted in the text: if the critical apparatus shows no note, we must conclude that both codices are jointly associated with the reading given in the text. On the other hand, if there exists an Armenian translation of this same tradition, the lack of the note Arm. or some such annotation does not authorize us to draw any conclusion. A critical note of the type ἔγραψε] γέγραφε Β tells us nothing with regard to the text underlying the Armenian translation, because the latter would in any case have been unable to distinguish between aorist and perfect.

2 When the text and the translation diverge If one can hypothesize a deviation of the Armenian translation as compared to the underlying text, the task facing the researcher is delicate and involves several different steps. The question becomes even more problematic when the text to which a reading of the Armenian translation apparently leads can cast doubt on the text of the original (and even more so when there is no longer a Greek text for reference because the Greek original is lost). a. The Armenian manuscript tradition. Editors of translations enjoy some advantage as far as evaluation of the witnesses is concerned, because they have the solid basis of a reference point that can provide guidance in evaluating the witnesses. When the witnesses of the Armenian manuscript tradition are divergent, whereas the Greek text is known to be certain and free from manuscript variants, the guiding criterion will naturally be that of preferring the reading that is closest to the Greek text. Clearly, it would be desirable to have reliable critical editions of the Armenian translations, but it has to be said, in all honesty, that this desire does not always correspond to reality. For many translations, it is necessary to have recourse to old texts, often published in the nineteenth century by scholars who were full of good intentions but lacking a solid background in critical editions, and who restricted themselves to the manuscript holdings of western libraries, in particular those of the Mekhitarist Fathers of Vienna and Venice. These were pioneering editions in which the indication of variants was limited to brief notes mentioning the presence of a varia lectio “in a witness (i.e. a codex)” (yōrinaki) or “in some codices”; furthemore they may, here and there, conceal tacit rationalisations of the translation designed to make it more easily readable (rationalisations that come to


11

light only when one gains access to the Armenian manuscripts of the work). For many of the translations the task of making a recensio of the manuscripts is still in its infancy, and even when critical editions drawn up by authoritative modern scholars are available, these are still in an initial phase. To take an example, the edition of Pseudo-Dionysius drawn up by Thomson, excellent though it is, rests on a limited manuscript base: six codices, chosen exclusively with the criterion of greater antiquity as compared to the great quantity that still remain to be examined. Many treasures of the Matenadaran are still to be uncovered. The first task facing the researcher, if a discrepancy between the Greek text and the Armenian translation is noted, is to assess whether the divergence can be eliminated without resorting to excessively disruptive interventions. If a small adjustment suffices to restore full correspondence between the Greek text and its translation, it is good practice to carry out the modification. For instance, in the manuscripts one often finds fluctuations between verbs such as šaradrem and šaragrem: the former corresponds to συντίθημι, the latter to συγγράφειν: the confusion is partly due to the fact that the signs for and are very similar in the Armenian alphabet. Thus whether variants are present in the Armenian manuscript tradition, or whether, instead, one of the manuscripts is handed down unitarily, the editor or scholar must be guided by the Greek text, and choose the Armenian verb corresponding to the Greek original. In the following example the emendation is rather more heavy, but equally evident. The beginning of Pindar’s IX Paean is transmitted by Dionysius of Halicarnassus (Dem. 7) and by Philo of Alexandria’s De Providentia, the Greek original of which is lost; P.Oxy. 5, 841 (frr. 126–128), which contained Pindar’s paeans, is lacunose and devoid of this passage. In ll. 3–4 we read: <τί δ᾿> ἔθηκας ἀμάχανον ἰσχύν <τ᾿> ἀνδράσι καὶ σοφίας ὁδόν ἐπίσκοτον ἀτραπὸν ἐσσυμένα; to which there correspond, in the Armenian translation, the following words: zor edir t‘agawor zawrut‘iwn, zaynr zor ew imastut‘ean čanaparhi xawar anhet ełeal. The interpretation of this passage of Pindar is extremely complicated and has severely tested the skills of modern exegetes: thus it must have been even more taxing for the poor Armenian translator. With regard to t‘agawor “king”, modern critics have given free rein to flights of fancy: quod regem costituisti, Virtutem eius, quo et sapientiae viae evanescunt tenebrae (Aucher); den da setzte der König der Marcht, dessen Weisheitsweg Finsternis nicht kennt (Neumann); quod (vel quam)

12

Moreno Morani

ἔθηκας rex (vel βασιλικήν) ἰσχύν illius quod (vel quam) καὶ σοφίας ὁδόν ἐπὶ σκότον ἀτραπὸν (invium Arm.) existens (Conybeare). Without claiming to endow the Armenian translation with meaning, one can at least improve the text with a few small adjustments. Firstly, the reference to the text of Dionysius, which reads ἀμάχανον ἰσχὺν πτανόν (the expunction of ἰσχύν, proposed by Blass, is not even mentioned in the text of modern editions, and the reading found in the codices is recorded only in the apparatus criticus) could easily suggest us to replace t‘agawor “king” of the Armenian translation by t‘ewewor “winged”; for zaynr one can obtain a better text, closer to the Greek original, by the very simple correction zayrn, which, however, leads us to assume a mistaken word division (roughly speaking, something like ἄνδρα ὅν or some such term for ἀνδράσι); similarly, čanaparhi xawar could easily have been corrected to čanaparh i xawar, a correction that already seems to be implicit (albeit not declared) in the translation by Conybeare and which leads us to presume a reading ὁδὸν ἐπὶ σκότος. These corrections do not restore a completely perfect passage: there are other problematic aspects that will test the mettle both of Armenian scholars and Pindar specialists, but it should be noted that at the beginning of the citation, the Armenian translation has a relative pronoun at the position corresponding to the <τί δ᾿> integrated by Diehl; the relative pronoun does not correspond to the integration by the modern philologist: rather, it suggests a different text and perhaps a less advanced phase of the distortion. Knowledge and study of the habits (both graphic and phonetic) of the Armenian copyists as well as of the phonetic processes of medieval Armenian can provide guidance for the conjectural solution of distortions that have arisen in the Armenian translation. For example, Bolognesi⁹ solves the following passage of the translation of Theon by making reference to an Armenian phonetic custom, the confusion of /o/ and /u/ in the vicinity of a liquid: spananēr i Koṙnt‘osi, or šurǰēr Yason (102, 25–26) ἀπέκτεινεν ἐν Κορίνθῳ, ὅπου διέτριβεν Ἰάσων Here the relative pronoun or should certainly be corrected to owr “where”. But even more interesting is a variant occurring in manuscript A: ənd asown instead of Iason. As pointed out by Bolognesi, here we have the same o ~ u alternation, facilitated also by the possible confusion with asun λογικός, exacerbated, furthermore, by the “mistaken interpretation of y- as a preposition, which has been substituted by the preposition ənd which in certain cases can be the equivalent

9 Bolognesi 2000, 73.


13

of y-”¹⁰. This is an eloquent demonstration of the fact that in the area of Armenian, as in Greek, the worst copyist is the one who claims or presumes he can understand, and who therefore adapts the text instead of restricting himself to transcribing it! In actual fact, this phase of the investigation does not always lead to important results. In the case of the translations of Nemesius and of the works of Pseudo-Dionysius, to mention a couple of cases I am familiar with, the Armenian manuscript tradition is compact and the variants of the manuscripts are qualitatively of little interest: in the overwhelming majority of cases they consist of omission or addition of the article, or of small variations due to phonetic reasons or other similar details. It is quite likely that even the collation of a greater number of manuscripts would not substantially change the picture we can already delineate on the basis of the manuscripts we know. Not all the problems assailing a philologist lurk in this phase of research. b. Mistakes by the translator. Many of the differences between the Greek text and the Armenian translations are to be ascribed to mistakes that arose during the actual translation process. Errors can be observed in all aspects of language: errors in syntactic constructions, partly due to the frequent technique of translating word by word without following the overall meaning of the sentence and the context; errors in morphology, when the translator misunderstands a form and produces the wrong translation. Many errors are generated by itacistic readings, other errors are due to incorrect reading of the text, as in the frequently wrong word divisions (partly resulting from the use of codices written in uncial, in which correct segmentation of the lexemes was less straightforward). As I have pointed out elsewhere, various errors can be explained by taking into account the phonetic habits of medieval Greek and in particular the low articulatory energy of word-final /s/ and /n/: if, as I have hypothesized, the translation was performed with the aid of a dictation procedure, then the fact that forms such as πόλις, πόλιν, πόλεις are easily mistaken for one another provides insight into the reasons underlying many of the imperfections in the translations¹¹. In short, one can understand why the translators may have confused καλῶς with καλῷ, but also potentially with καλός and καλόν.

10 Bolognesi 1990, 313 (“l’erronea interpretazione di y- come preposizione, cui è stata sostituita la preposizione ənd che in certi casi può essere l’equivalente di y-”). 11 Morani 2003a.

14

Moreno Morani

Naturally, not everything can be explained in this manner, and in various cases the responsibility for the mistake should be attributed to a genuine language mistake by the translator. Let us now take a look at some of the most common errors taken from a variety of Armenian translations studied by Bolognesi or by the author of this paper (N = Nemesius, D = Pseudo-Dionysius the Areopagite, T = Theon: as this list has the character of a general overview, a precise indication of the passages involved will be unnecessary). Mistake in reading: Greek text

Armenian translation

Reason

μὲν οὖσα

mnac‘eal

mistake in word division (μένουσα) Ν

διαμονήν

vasn tewout‘ean

mistake in word division (διαμονήν) Ν

εἰ τὰ δοθέντα

ew apa tueal

mistake in word division (εἶτα δοθέντα) Τ

διάφορα

tarberowt‘iwn

wrong accent (διαφορά)

ἑταῖρος

ayl

itacism (ἕτερος) N

παιδιάν

xratu

itacism and wrong accent (παιδείαν) N

κλίσις

koč‘ut‘iwn

itacism (κλῆσις) D

ἔτη

takawin

itacism (ἔτι) D

Χριστοειδές

k‘ałcratesakn

itacism (χρηστο-) D

ποίαν

cparantut‘iwn

itacism (πίαν) T

ἱέρειαι, ὧν

k‘rmic‘

id. and mistake in word division (ἱειρειαίων vel sim.) T

εἰ τύχοι

kam zormoy

itacism (ἢ τοίχου) N

ἡ διάθεσις

kam tramadrut‘iwn

mistake in spiritus (ἢ instead of ἡ) N

ἡ … ἔχουσα

kam unelov

id. Τ

εἰς διόρθωσιν

mi yułłut‘iwn

spiritus and accent (εἷς instead of εἰς) Ν

ὡς … ὡς

orpēs … or

id. (ὡς … ὅς) N


15

Wrong interpretation of a form: Greek text


Reason

οὐσία

ewt‘anē

iota subscriptum neglected or hypercorrected N

οὐσίᾳ

ewt‘iwn

id. N

ἀνάγκη

i harkē

(as if it were ἀνάγκῃ) N

οἷόν τε

karcen

word division and itacism (οἴονται) NT

εἰδότες

tesanelov or teseal

confusion of forms (ἰδόντες) N

εἰδῶμεν

tesc‘uk‘

id. (ἴδωμεν) T

θέωσις

drowt‘iwn

confusion with θέσις D

ἀριστός ἐστιν

hačeli ē

confusion of forms (ἀρεστέος ἐστίν) N

τὰ νοερά

Zawd

mistake in word division and confusion of forms (τὸν ἀέρα) D

εἴπομεν ὡς

hetewabar

id. (ἑπομένως) DT

ἀπορρήτων

tarakuseleac‘n

confusion with ἀπορέω D

αἱρεσθείσης

asac‘eloy

confusion with the forms of ἐρρήθη D

ἐνόντα

miaworeal

confusion with ἑνοῦντα D

συνιόντων

ənd nmayn en

confusion with συνόντων N

Failure to understand the value of the term: Greek text


Reason

πείσεις

hawanumn

connected to πείθω rather than to πάσχω N

δοξάζειν

paṙaworel

“glorify” instead of “believe”, “think” N

συγκρίνειν

bałdatel

“compare” instead of “believe”, “think” N

Other common mistakes include uncertainty between singular and plural, omission of καί especially if the following word starts with /k/, omission and addition of ē, and other problems. But how much of all this is to be ascribed to misunderstandings and inadvertence by the translators and to what extent does it depend on the Greek manuscript tradition? Let us look at an example that illustrates the

16

Moreno Morani

meaning of this question. In Nemesius 8, 16 the syntagm δι᾿ εὐαισθησίαν is rendered by barwok‘ zgestuc‘: evidently there has been a mix-up between a derivative of αἴσθησις and a possible (though unattested) derivative of ἐσθής. This mistake is certainly due to the translators, because a compound made up of εὖ and ἐσθής does not appear to be attested. But could it not be the case that this mistake was encouraged by an itacistic type of spelling in the source codex? Unfortunately, codices dating back to the era in which the Armenian translators were working on their translations are rare; however, the tradition of Dionysius the Areopagite, for example, does provide us with an extremely ancient codex, the precious codex known as the Louvre manuscript (MR 416). This is the codex that presents the closest relationship with the Armenian translation, but it also shows a quantity of errors in writing and careless mistakes by the translator that are not found in codices dating from a more recent era. Thus it would appear that the tradition became progressively more refined over the centuries, eliminating a series of mistakes and impossible readings which are indeed present in two of the earliest witnesses, namely the Paris codex and the Armenian translation. c.1. Armenian translations and the Greek manuscript tradition. Once these two preliminary phases of the work have been completed, one can turn to an investigation of the reasons underlying the difference between the Greek text and its Armenian translation. In cases where the Greek manuscript tradition presents one or more variants, and the Armenian translation is in agreement with one of these variants, then there is little difficulty in forming a judgment. On the other hand, whenever the Armenian translation stands in opposition to a unitary Greek tradition, the value of an Armenian variant must be assessed on the basis of the normal and current techniques of the edition. In this case the Armenian translation can be treated codicis instar, and generally this will be a privileged codex because it will often date back to several centuries earlier than the most ancient codices of the Greek direct tradition. But this should not be taken as implying that the reading found in the Armenian translation must a priori be considered more likely or superior to that of such codices. Merely applying mechanically a criterion of vetustiores ergo meliores would be illogical: a reading must be judged not only on the basis of the age of the witness, but above all on the basis of the extrinsic value both of the witness and of the reading. It could be objected that the Armenian translation, by virtue of its greater antiquity, should be placed high up in the stemma: the problem is that trust in stemmata, and above all their heuristic value, has by now been eroded. The automatism of the stemmata has been replaced in modern philology by a far more fascinating and historically realistic vision of working processes and of the activities involved in study, leading to a gradual refinement of


17

texts through research and comparison among manuscripts. The contamination displayed by codices is not a pathology but the consequence of a lively intellectual approach: to be sure, it makes our work more difficult and the classification of witnesses becomes more fluid; however, it offers a more concrete historical perspective that highlights the human aspect of the copyist (or, in our case, of the translators). Thus without wishing to overestimate the importance of the Armenian translations, they do photograph an ancient situation of the tradition. For example, they can reveal the existence of very early variants, or sometimes of ancient glosses, unwittingly documented inasmuch as the translator often preferred, for the sake of convenience, to translate the gloss he read in the margins rather than the original reading he found in the body of the text. The passage from Pindar examined earlier shows that at the time when the Armenian translation was being composed, the reading πτανόν was alternative to ἀμάχανον, and perhaps it was a gloss that had ended up replacing the original reading, whereas in the Greek witness it appears as a double reading. When the solution lies in a position intermediate between the presupposition of a Greek variant and the possibility of a mistake in the Armenian manuscript tradition, the most economical possibility must be chosen. In the Hexaemeron by George of Pisidia (l. 1625), Gr. θεωρῶν is rendered with gteal, a participle of the verb gtanem “I find”: but, as Bolognesi demonstrated¹², correcting the Armenian reading to diteal, a participle of ditel “to look”, “to observe” is far more plausible. The misunderstanding was brought about by the similarity of the signs for and in the Armenian script. There are cases, however, where it is genuinely impossible to distinguish between a mistaken reading of the Armenian and a variant in the Greek tradition. If an itacistic variant is found in the Greek manuscript tradition, it is impossible to establish whether the genesis of the divergence arose from a false reading by the translator or a variant in the Greek tradition. Philological study of translation necessarily also involves study of the translation techniques adopted when the texts were translated into Armenian. For instance, there was a widespread tendency to double the words, using two Armenian words where the Greek texts shows only one term. This technique (not unknown in other oriental translations of Greek texts) was intended to ensure that no semantic nuance of the original term would be overlooked, in cases where the translator felt that the semantic correspondence between the Greek word and the chosen Armenian equivalent was not fully achieved. The work of comparison between Greek and Armenian will become all the more fruitful, the more the

12 Bolognesi 2000, 144.

18

Moreno Morani

tools we have available are reliable. By the same token, it is equally vital to have complete and trustworthy critical editions for the Greek texts as well. For many authors of classical Greek literature the work of making recensions of the codices and producing editions has achieved an extremely high level. For various authors of Christian Greek literature, on the other hand, we are in a far less fortunate position, and in certain cases scholars are given no possibility of checking the validity of the readings because of unwise editorial choices. It may seem paradoxical, but at times modern editors tend to censure what is not to their liking. One need only consider the following example, taken from the Armenian translation of De divinis nominibus by Pseudo-Dionysius. For this author, we ought to be in a fairly favourable situation: as far as the Armenian translation is concerned we have at our disposal a good critical edition drawn up with modern criteria (albeit established on a rather limited manuscript base), and for the Greek text we have an edition that is the outcome of an intense philological work, carried out with unsparing use of all possible means. This would appear to be the ideal situation for proceeding to study the Armenian translation without undue concern, safe in the knowledge that the material to be worked on is free from uncertainties. But in actual fact this is not always the case. The editor of Dionysius, Beata Regina Suchla, has eliminated from the apparatus criticus all the lectiones singulares which, in her view, were of little significance. The critical edition opens with a list of the Singulärlesarten, including the orthographische Quisquilien, of the first chapter. Now in at least one case, the trifling item in question does have some interest for the Armenian text: in 111.7 the reading ἀνατεινόμεθα is rendered by the subjunctive yarjakesc‘uk‘. Now, codex Rc, the Roman Vallicellianus E.29 with which the Armenian translation often shows close affinity (Armenian + Louvre codex + Vallicellianus often coincide with regard to numerous important variants), has the reading ἀνατεινώμεθα, of which no mention is made in the apparatus criticus. Cases of confusion between and <ω> occur frequently in the Armenian translation, but in this particular case the Armenian reading corresponds to a variant in the Greek manuscript tradition which could turn out to be an ancient variant. As a result of the editor’s silence, scholars engaged in study of the Armenian translation have been deprived of an important criterion for assessment, which could have made a significant contribution even to the very definition of the relationship between the Armenian translation and the Greek codices. A few additional remarks c.2. The Armenian translation and contamination. One might be led to think that since the language in which the translations were drawn up was Armenian, i.e. a language different from Greek, the translation, differently from what happens


19

with the Greek codices, would be free from subsequent processes of contamination. This statement has a fair degree of probability, but, as always, caution must be exercised. Both the Armenian tradition of Nemesius and that of pseudo-Dionysius suggest a transmission of double readings which must have been present in the common ancestor of the Armenian tradition. Of these two readings, one should be regarded as a mistaken reading of a Greek word, the other as the correction. An example, in the second chap. of Nemesius (p. 115.3), διέπειν of the text is rendered by xawsil. This is clearly a mistaken translation: the interpreter misinterpreted this form, confusing it with διειπεῖν, and consequently, he rendered it with the corresponding Armenian verb; at some later stage someone (conceivably the interpreter himself or one of the first readers) realized that the Armenian verb was inappropriate in the given context and replaced it with bašxel “to administrate”, which is the correct rendering of the Greek verb. Evidently, whoever it was who noticed a possible misunderstanding in the translation went to the effort of going to look for a Greek codex and inserted the correction. The two readings appear alternately in the text and the margin of the Armenian manuscripts I have collated. Furthermore, as I have already pointed out in my study on the manuscript tradition of Nemesius¹³, on two occasions the codices of the Armenian translation present variants that someone ferreted out from the lost Syriac translation of the work (on the first occasion [p. 70.7], the absence of a short passage that may well have been an interpolation; on the second, a variant also known to the Greek manuscript tradition). We will not be concerned, here, with discussing the value of the variants, on which I have put forward my opinion elsewhere¹⁴; rather, what is important to note is that some unknown reader took the trouble to compare the Armenian translation with a translation into another language in order to check the validity of a passage that raised doubts in his mind. Naturally, this was hardly likely to have been a systematic and exhaustive collation of the work; rather, in all probability the sequence of events was more or less what we ourselves would do: we read a translated work and, when some important passage in the translation strikes us as unconvincing and we have the possibility of checking it against the

13 Morani 2003a, 45. 14 That 70.7 was superfluous was already pointed out by the previous editor of Nemesius, namely Mathhäi (Halle 1802). The indication that this segment is missing in the Syriac version is given by some Armenian codices, which have a note in the margin that records the divergence of the Syriac translation from the Armenian. Since it is unrealistic to assume that the Armenian readers systematically collated the whole of the Armenian translation with the Syriac (there are only two marginal notes, in total, that make a reference to Syr.), it seems likely that they proceeded to check the Syriac version to assess the reliability of a particular passage that had aroused some doubt.

20

Moreno Morani

original text, we proceed to search for the original passage to determine whether there might perhaps be a mistake in the translation. c.3. Armenian translations and Greek critical editions. Finally, a note on the presence of the Armenian translations in the apparatus criticus of Greek authors. Usually little attention is paid to the translations in the critical apparatus of Greek works: in some cases they were completely neglected, but the situation has improved in recent years, and the world of philology has seen a gradual rise in awareness of the importance of non-Greek witnesses. The Paris edition of Theon, in which Bolognesi also took part, is an example that will be difficult to replicate: here the variants are recorded in the apparatus in their original text. But this is a special example, and it is not always possible to proceed with such a degree of completeness. The differences between the text and the Armenian translation are too pronounced for all the differences to be recorded, as this would lead to an extremely cumbersome apparatus which would be largely pointless, because many variants are probably internal to the Armenian manuscript tradition. In my edition of De divinis nominibus¹⁵, I recorded in the final apparatus many (but not all) of the variants arising from the collation of the Armenian translation, but without any claim to completeness. I (reluctantly) used the apparatus of Suchla and I added many variants resulting from the Armenian translation, inserting here and there the occasional fortasse or ut videtur in cases where the reading was not certain and supplying, if it seemed important, the original Armenian reading in transcription in order to enable the reader to check the proposed solutions more closely.

Bibliography Abełian, A. (1936), Neuarmenische Grammatik, Berlin-Leipzig. Bolognesi, G. (1964), “Incontri di culture nell’antico Oriente indo-europeo”, in: Annuario dell’Università Cattolica del Sacro Cuore, a.a. 1963–19643, 143–165. Bolognesi, G. (1966), “La tradizione culturale armena nelle sue relazioni col mondo persiano e col mondo greco-romano”, in: Problemi attuali di scienze e di cultura 76 (Atti del Convegno Internazionale sul tema “La Persia e il mondo greco-romano”), 569–603 (= Bolognesi 1990, 271–318). Bolognesi, G. (1982), “Traduzioni tardo-antiche ed alto-medievali in Medio Oriente”, in: Processi traduttivi: teorie ed applicazioni: atti del Seminario su “La traduzione”, Brescia, 19–20 novembre 1981, Brescia, 11–38 (= Bolognesi 2000, 1–28)

15 Morani 2010.


21

Bolognesi, G. (1990), Studi glottologici, filologici, orientali, Brescia. Bolognesi, G. (1997), “Dalla traduzione al testo”, in: R. Ambrosini et al. (eds.), Scríbthair a ainm n-ogaim. Scritti in memoria di Enrico Campanile, Pisa, I, 227–234 (= Bolognesi 2000, 217–224). Bolognesi, G. (2000), Studi e ricerche sulle antiche traduzioni armene di testi greci, Alessandria. Fermeglia, G. (1964–65), “Studi sul testo delle due versioni (slava ed armena) dello Hexaemeron di Giorgio Pisida”, in: Memorie dell’Istituto lombardo, Accademia di Scienze e Lettere. Classe di Lettere, Scienze Morali e Storiche, 28, 227–332. Mercier, C. (1978–79), “L’école hellénistique dans la littérature arménienne”, in: Revue des Études Arméniennes 13, 59–74. Morani, M. (1981), La tradizione manoscritta del De natura hominis di Nemesio di Emesa, Milano. Morani, M. (2003a), “Premessa sul valore della traduzione dello Pseudo-Dionigi l’Areopagita”, in: A. Sirinian / S. Mancini Lombardi / L. D. Nocetti (eds.), Le scienze e le ‘arti’ nell’Armenia medievale. Atti del Seminario internazionale, 29–30 ottobre 2001, Bologna, 37–47. Morani, M. (2003b), “Sensum de sensu, verbum e verbo. Riflessioni su teoria e storia della traduzione in margine a uno scritto di Eugenio Coseriu”, in: V. Orioles (ed.), Studi in memoria di Eugenio Coseriu, Udine, 317–336. Morani, M. (ed.) (2010), Dionigi, I nomi divini, introduzione e testo critico; traduzione e note di G. Regoliosi; commento di G. Barzaghi, Roma-Bologna. Nichanian, M. (1989), Âges et usages de la langue arménienne, Paris. Somalean [Somal], P. S. (1825), Quadro delle opere di vari autori anticamente tradotte in armeno, Venezia. Ter-Petrossian, L. (1984), La littérature arménienne ancienne de traduction, Erevan. Teza, E. (1893), “Nemesiana. Sopra alcuni luoghi della natura dell’uomo in armeno”, in: Rendiconti della Reale Accademia dei Lincei. Classe di Scienze Morali, Storiche e Filologiche (ser. 5) 2, 3–16.

Giusto Traina

Some Observations on the Armenian Pseudo-Callisthenes Abstract: A new edition and commentary of Pseudo-Callisthenes’ recensio vetusta is obviously welcome. Richard Stoneman and Tristano Gargiulo have published the first two books, in the “Lorenzo Valla” collection: the third is forthcoming. However, although Gargiulo’s critical text is useful, his treatment of the Armenian text raises some doubts as Gargiulo, instead of seeking external expertise, preferred to understate the Armenian text in order to bypass the problem. In the present paper, three examples (Ps. Callisth. 1.13.4, 2.12.1, 2.21.12) are examined, in order to understand the relationship of Arm. with the ancient recensions α and β. As a result, a general reconsideration is needed: the Armenian recension should no longer be considered a mere source of the critical apparatus of the Greek, but as an independent recensio.

For almost ninety years, we have relied on Wilhelm Kroll’s praiseworthy edition of Pseudo Callisthenes’s recensio vetusta¹. Today, a new edition is being published in the collection “Scrittori greci e latini” of the Lorenzo Valla Foundation². This very welcome edition has been entrusted to Richard Stoneman, author of the general introduction, and to Tristano Gargiulo, now deputy director of the “Lorenzo Valla” collection, who prepared the critical edition and made the Italian translation³. In addition to the recensio vetusta, based on Parisinus Gr. 1711, the Latin Iulius Valerius has been edited, as well as two other Greek versions of PseudoCallisthenes: the β, usually dated to the fifth century, and, oddly enough, γ, actually a later compilation based on β and on the Middle Byzantine ε⁴. Volume I, edited by Stoneman, is more reader-friendly but less attentive to philological

1 Kroll 1926. 2 Book I: Stoneman 2007; book II: Stoneman/ Gargiulo 2012 (see Franco 2013); book III is forthcoming. 3 In the future, an English version of this edition will be published by Oxford University Press. See also the excellent Italian translation by Franco 2001. 4 β (and sub-recension λ): Bounoure/Serret 1992; γ: Jouanno 2009, 5–30; ε: Sempéré 2006. For a general survey of the tradition of the Pseudo-Callisthenes, see Jouanno 2002. Some useful observations can also be found in Franco 1999. On the other hand the late Byzantine Alexandergedicht (Marcianus Gr. 408, ed. Aerts 2014) is a very important tradition for the reconstruction of α: see the case study in Aerts 2011.

24

Giusto Traina

problems, with the exception of some additional notes by Gargiulo, embedded in Stoneman’s commentary. Volume II, provided with a richer commentary, is jointly edited with Gargiulo, who explains his textual choices in a long preliminary note⁵. A crucial problem arises in the treatment of the Armenian materials by the new edition⁶. The Armenian version (= Arm. A), dated to the second half of the fifth century, is usually considered as a main witness of the recensio vetusta⁷. Kroll gave due consideration to this version for the constitution of his critical text. As he did not read grabar (Classical Armenian), he used Raabe’s Greek retroversion (1896), based on the Mekhitarist edition by T‘reanc‘ (1842). In his introduction, Stoneman rightly warned against uncritical use of the Greek retroversion, but he was no less uncritical, as he relied on the English translation by Albert M. Wolohojian (1969)⁸. Gargiulo, on the other hand, while duly considering the Armenian lectiones in his critical apparatus, nonetheless did not examine the original text. In vol. I he followed Kroll’s example, relying on Raabe’s retroversion⁹; in vol. II he made some use of my own Italian translation (2003), although this “diplomatic” translation is based exclusively on V 424, where several folia are missing. An additional problem is the new text published by Simonyan (= Arm. B): this incomplete version, preserved by a codex unicus, apparently likewise dates from the fifth century, but seems, on the other hand, to be rather independent¹⁰. Of course, it would be unfair to criticize Gargiulo’s critical edition. However, the following rather peremptory statement arouses some doubts: … when A and the Armenian version coincide, the text of A may be accepted, but we hardly advise to correct a lectio on the sole basis of the Armenian text, no matter how it seems to be a safe one. As a matter of fact, in not a few passages, the Armenian version evidently omitted, added, misunderstood or changed [the original text]¹¹.

5 Gargiulo, “Nota al testo”, in: Stoneman/Gargiulo 2012, XXXIII–LVIV. 6 The recent collective volume on Alexander in the Oriental traditions (Stoneman/Erickson/Netton 2012) does not consider the Armenian texts. 7 Ed. Simonyan 1989. See also Simonyan 1998; Traina 1996; Id. 1998. 8 Wolohojian was aware of the most important Armenian philological studies, such as Tašean 1892. See Stoneman 2007, LXXV. 9 For the abbreviations of Armenian manuscripts see Coulie 1992. 10 Simonyan 1998; Mancini Lombardi/Uluhogian 1998. These contributions are listed in the bibliography of Stoneman 2007 and Stoneman/Gargiulo 2012, and Stoneman (2007, LXXV n. 1) is well aware of the new Armenian version, complaining of the lack of a translation. But there is no hint of these philological problems, either in the apparatus or in the commentary. 11 Gargiulo, in: Stoneman/Gargiulo 2012, XLI.

Some Observations on the Armenian Pseudo-Callisthenes

25

Unlike Kroll, Gargiulo awards a definitely marginal role to the Armenian text in the tradition of the Greek recensio vetusta, deliberately ignoring not only the Armenian text(s), but also the rich and important literature on the Armenian techniques of translation, finely elaborated by the Armenian scholars, and, in the West, by Giancarlo Bolognesi and his students¹². Gargiulo is certainly right to adopt a more complex approach in considering the Armenian version than Kroll did: for instance, there are important differences between the Greek and Armenian version, namely in the epistolary sections. But Gargiulo’s methodological approach could not rely on the support of direct examination of the original text. Working without the assistance of an Armenian scholar, Gargiulo had no choice but to marginalize the Armenian version whenever it was possible. Therefore, his edition did not provide a constitutio textus based on all the available evidence¹³. It will be useful to consider some passages in order to check Gargiulo’s hyper-conservative positions¹⁴. As a first example, I will consider the answer of the Indian king Poros to Darius’ request for help (Ps. Callisth. 2.12.1): λαβὼν δὲ ὁ Πῶρος γράφει· “Πῶρος βασιλεὺς Ἰνδῶν Δαρείῳ βασιλεῖ Περσῶν χαίρειν. ἀναγνοὺς [δὲ] τὴν ἐπιστολὴν καὶ λυπηθεὶς τὴν ὁρμὴν πρός σε εἶχον, ὡς ἂν σκεψώμεθα … ἀλλ’ ἐμὲ κωλύει ἡ συνεχής μου νόσος”. Kroll supplevit ex B’ Arm. (Byz.) περὶ τῶν συμφερόντων (ὑμῖν) Gargiulo ὡς ἂν σκεψώμεθα, ἀλλ’ ἐμὲ κτλ.

Arm. A, § 170 (p. 207 Simonyan) reads: Ew ǝnt‘erc‘eal zgrealsn Povros ew p‘oxanak nma greac‘ namak ayspisi awrinakav: “Povros t‘agawor Parsic‘ Darehi xndal. Ǝnt‘erc‘eal mer ew zgrealsl jer, mek‘ yoyž trtmec‘ak‘. Zi t‘ēpēt ew kamēi patahel k‘ez ew xorhel vasn awgti jeroy, sakayn argeleal linim anǝndhat hanapazord hiwandut‘enē marmnoy …”. Poros read the letter he (sc. Darius) had written, and wrote him in return a letter of this sort: “The king Poros greets Darius, king of the Persians. We read your letter, and we were strongly afflicted. Even if I wanted to come to you and medi-

12 See the methodological considerations, and the bibliographical references of M. Morani’s contributions in this volume. See also Shirinian 2001; Calzolari 2014. 13 For a fitting example, see the edition of Socrates Scholasticus (Hansen 1995). 14 He explains his approach in Gargiulo 2009, 203, and confirms it in Stoneman/Gargiulo 2012, XXXV–XXVI.

26

Giusto Traina

tate something about helping you, on the other hand I am prevented (from doing so) by a chronic and stubborn bodily infirmity …”. On the basis of the Armenian version, Kroll conjectured a lacuna after σκεψώμεθα, suggesting that the gap should be integrated with the complement περὶ τῶν συμφερόντων (ὑμῖν). Yet in this case the Armenian version does not merely give a more verbose text, but an actual variant deviating from α (and also from β). Two solutions are possible: either to admit a lacuna, as Kroll did, or to consider, with Gargiulo, the verb σκέπτομαι in an absolute sense, citing two examples from Aristophanes and Thucydides¹⁵. In fact, not only is Kroll’s hypothesis more reasonable, but there is no reason to reject the Armenian tradition. On the other hand, both Kroll and Gargiulo were misled by Raabe, who tried to find a common meaning both in the Greek and in the Armenian text. Another example, from book 2, is Alexander’s answer to Rhodogune and Stateira, Darius’s wife and mother (Ps. Callisth. 2.22.12). The text established by Kroll reads: “Παραιτοῦμαι τὰς ἰσοθέους τιμάς· ἐγὼ γὰρ ἄνθρωπος φθαρτὸς γεγένημαι καὶ εὐλαβοῦμαι τὸ τοιοῦτον· κίνδυνον γὰρ φέρει τὸν περὶ ψυχῆς. ἐπαινῶ δὲ καὶ δέχομαι ὑμῶν τὸ φρόνημα· πειράσομαι οὖν ἄξιον ὑμῶν τοῦ εὐγενοῦς … περὶ ὑμῶν. ἔρρωσθε”. Kroll reconstituted the text according to Raabe’s retroversion of Arm. A, § 206 (p. 249 Simonyan): “Hražarim i hangitastuacakan patwoyn, k‘anzi es mahkanac‘u mard ełeal, erknč‘el part ew aržan ē ayspisums. K‘anzi vtang bereal matuc‘anē yałags ogwoy. Bayc‘ govem ew šnorhakal em ew ǝndunim zjer imasdut‘iwnd, ew p‘orjec‘ayc‘ aržanis cnołac‘n jeroc‘ hog tanel vasn jer. Ołǰ leruk‘!”¹⁶. “I renounce the honour of being equal to the gods¹⁷, for I am a mortal man and must fear this sort of thing, for it presents danger to the soul. But I laud and thank you and acknowledge your thoughtfulness. And I shall try to be worthy of your forefathers by taking care of you. Farewell”.

15 Gargiulo, in: Stoneman/Gargiulo 2012, XLVI. 16 This chapter is also found in Arm. B (p. 410 Simonyan), but in a strongly reworked and abridged form. 17 Raabe rightly renders the adjective hangitastuacakan with τὰς ἰσοθέους τιμάς.


27

Apparently, the text is corrupted. Gargiulo rejects the Armenian tradition, integrating the lacuna identified by Kroll by means of β¹⁸: Παραιτοῦμαι τῶν σῶν θεῶν τιμάς· ἐγὼ γὰρ ἄνθρωπος φθαρτὸς γεγένημαι καὶ εὐλαβοῦμαι τὸ τοιοῦτον· κίνδυνον γὰρ φέρει τὸν περὶ ψυχῆς. ἐπαινῶ δὲ καὶ ἔγνων ὑμῶν τὸ φρόνημα· πειράσομαι οὖν ἄξιον ὑμῶν τοῦ γένους <φροντίσαι> περὶ ὑμῶν. ἔρρωσθε. Relying on Raabe’s retroversion, Kroll preferred δέχομαι to ἔγνων, restored by Gargiulo who claims that a conjecture based on the Armenian version is arbitrary; moreover, this tradition is missing in β and γ (!)¹⁹. As a matter of fact, the Armenian text presents a more complex phrase, which suggests a simplification by the copyist. Finally, a particularly important case is a passage containing the list of young Alexander’s teachers (1.13.4). Ἐπεὶ δὲ τῆς παιδικῆς τάξεως <ἐξέβη>, παιδαγωγὸς ἦν αὐτοῦ † Λακρητητις Μέλανος †, τροφεὺς Λευκονίδης, διδάσκαλος δὲ γραμμάτων Πελλαῖος Πολυνείκης, μουσικῆς δὲ Λήμνιος Ἄλκιππος, γεωμετρίας Πελοποννήσιος Μένιππος, ῥητορικοῦ δὲ λόγου Ἀθηναῖος Ἀριστομένης, φιλοσοφίας δὲ Μιλήσιος Ἀριστοτέλης, ὁπλοκτυπίας δὲ † Λαμψάκης ὁ σευρωτάς †. The corresponding Arm. A, § 29 (p. 93 Simonyan) helps to improve a visibly corrupted text²⁰: Ew ēr nora dayeak: kinn Lakrinē, Keltac‘woc‘ Melanosi k‘oyrn. Ew mankavarž ew dastiarak ew tayeak: ayr Lewondēs Lakonac‘i; ew vardapet ergoc‘: Pawłinikos, ew eražštakanut‘ean (Ł)ewkios Lesnac‘i; erkrač‘ap‘ut‘eann: Menek‘mos Pelēponesac‘i; ew hṙetorakan banic‘: Anak‘simēnos Aristotełeay Ampsakenac‘i. isk imastasirut‘eann: Aristotelēs Nikomitak‘ay Stagirac‘i i Meleteay k‘ałak‘ē. Bayc‘ yišeac‘ ew

18 Ὁ δὲ Ἀλέξανδρος δεξάμενος τὰ γράμματα αὐτῶν ἀντιγράφει αὐταῖς οὕτως· “Ἐπαινῶ ὑμῶν τὸ φρόνημα. πειράσομαι οὖν ἄξια τοῦ γένους ὑμῶν φροντίσαι. κἀγὼ γὰρ φθαρτὸς ἄνθρωπος γεγένημαι. ἔρρωσθε”. 19 Gargiulo, in: Stoneman/Gargiulo 2012, LIX. The Byzantine Alexandergedicht (v. 4103) has ἔγνωκα. I am grateful to Corinne Jouanno for this observation. 20 For the version Arm. B, p. 366 Simonyan presents only two variants. The expression mankavarž ew tastiarak dayeak is replaced by a simpler mankakal; the term vardapet, “teacher”, “doctor”, referring to the music teacher, is replaced by usuc‘ič‘ “teacher”.

28

Giusto Traina

Pap‘ovranos yałags aysocik i č‘orordum girs iwrum yamenazan hnaragēt patmuteanc‘. (Favorinus, fr. 61a Amato). His governess was Lakrine, the Celt, sister of Melanos; and his pedagogue and tutor was Leuondides, the Lakonian. The teacher of songs, Polinikos; and of Music, Leucios de Lesnos (sc. Leucippos, the Lemnian); of geometry, Menechmos, the Peloponnesian; of rhetoric, Anaximenos of Aristocles, the ampsakenian; and of philosophy, Aristoteles of Nikomitakes, the Stagirite from the city of Meliteos. But Paphovranos [sc. Favorinus] mentioned these matters in the fourth book of his all-encompassing learned histories²¹. Gargiulo rightly notes that both the Greek and Armenian version depend on an earlier corrupted Greek text. The confusion between names and place names increased the problems in this passage. The Armenian tradition preserved the right name of the master of geometry, the Peloponnesian Menaechmus (attested by Stob. 2.31.115), in contrast to Menippus (α) and Melemnus (β). An important semantic nuance is the use of Armenian dayeak, a lemma with the main meaning of “nurse”, which also corresponds to Gr. τροφεύς²². Given the gender ambiguity of the term, the Armenian text specifies that Alexander had firstly a dayeak woman – Lacrine, sister of the Celt Melanos – and then a male one, Leontides, with the triple function of “teacher, tutor and dayeak”. For the final part, the Armenian version seems to agree in part with β: ῥητορικῶν δὲ λόγων <Ἀν>αξιμένης Ἀριστοκλέους ὁ Λαμψακηνός, φιλοσοφίας δὲ Ἀριστοτέλης Νικομάχου ὁ Σταγειρίτης. The quotation from Favorinus (absent in Arm. B) has been invoked to attest the superiority of the Armenian version²³. Stoneman considered this passage one “of the strongest arguments to fix the date of composition of the novel (after 140 AD)”²⁴. As Eugenio Amato remarked in his accurate commentary of Favorinus, the list of Alexander’s teachers is also found in the Syriac Romance of Alexander (which omits the name of Favorinus)²⁵. The differences between the Armenian version and Iulius Valerius seem to suggest that both texts separately rely on an earlier version of α.

21 A parallel text is Iulius Valerius 1.13: si quid inquirere curiosius voles, sat tibi habeto Graecum Favorini librum, qui Omnigenae historiae superscribitur (Favorinus, fr. 61 Amato). 22 See Traina 2004. 23 Tašean 1892, 225. 24 Stoneman 2007, 498. 25 Amato 2010, 297–299 n. 271.


29

The examples cited above confirm the importance of the Armenian version for the text of Pseudo-Callisthenes: Gargiulo’s conservative approach shows its evident limitations. On the other hand, we can no longer maintain Kroll’s reductive approach. In fact, there is no need to consider the Armenian version(s) as a mere support of the Greek text: as a matter of fact, the Armenian version should be considered on the same level of the Greek versions. Not differently from the Latin Iulius Valerius, the Armenian Pseudo-Callisthenes is, rather, an authoritative testimonium, extremely useful for the constitutio textus of the recensio vetusta. Its position in the stemma codicum²⁶ should necessarily be revised.

Bibliography Aerts, W. J. (2011), “The Ismenias Passage in the Byzantine Alexander Poem”, in: M. P. Futre Pinheiro / S. J. Harrison (eds.), Fictional Traces: Receptions of the Ancient Novel, 1, Groningen, 69–82. Aerts, W. J. (2014), The Byzantine Alexander Poem, Berlin-New York. Amato, E. (ed.) (2010), Favorinos d’Arles, Œuvres. Tome III. Fragments, Paris. Bounoure, G. / Serret, B. (eds.) (1992), Pseudo-Callisthene, Le Roman d’Alexandre. La vie et les hauts faits d’Alexandre de Macédoine, Paris (reprint 2004). Calzolari, V. (ed.), (2014), Armenian Philology in the Modern Era: from Manuscript to Digital Text (Handbook of Oriental Studies/Handbuch der Orientalistik), Leiden-Boston. Coulie, B. (1992), Répertoire des bibliothèques et des catalogues de manuscrits arméniens, Turnhout. Cowe, S. P. (1996), “Aspects of the Translation and Redaction Process of the Alexander Romance in Armenian”, in: D. Sakayan (ed.), Proceedings of the Fifth International Conference on Armenian Linguistics, Delmar (N. Y.), 245–260. Finazzi, R. B. / Valvo, A. (eds.) (1998), La diffusione dell’eredità classica nell’età tardoantica e medievale. Il romanzo di Alessandro e altri scritti, Alessandria. Franco, C. (1999), “Il Romanzo di Alessandro”, in: Quaderni di storia 49, 45–102. Franco, C. (ed.) (2001), Vita di Alessandro il Macedone, Palermo. Franco, C. (2013), Review of Stoneman/Gargiulo 2012, in: Lexis 31, 455–457. Gargiulo, T. (2009), “Un caso particolare di intertestualità nel Romanzo di Alessandro: i fraintendimenti di Β e Γ rispetto ad Α”, in: P. Mureddu / G. F. Nieddu / S. Novelli (eds.), Tragico e comico nel dramma attico e oltre: intersezioni e sviluppi parateatrali, Sassari, 201–221. Gaullier-Bougassas, C. (ed.) (2014), La fascination pour Alexandre le Grand dans les littératures européennes (Xe–XVIe siècle), Turnhout. Hansen, G. C. (ed.) (1995), Sokrates, Kirchengeschichte. Mit Beiträgen von Manja Širinjan, Berlin. Jouanno, C. (2002), Naissance et métamorphoses du Roman d’Alexandre. Domaine grec, Paris.

26 Stoneman 2007, LXXIV–LXXV.

30

Giusto Traina

Jouanno, C. (ed.) (2009), Histoire merveilleuse du roi Alexandre, maître du monde, Toulouse. Kroll, W. (ed.) (1926), Historia Alexandri Magni. I, Recensio vetusta, Berlin (reprint 1958). Mancini Lombardi, S. / Uluhogian, G. (1998), “Due redazioni per il Romanzo di Alessandro armeno: tessere di un mosaico perduto?”, in: Finazzi/Valvo 1998, 157–174. Merkelbach, R. (1977), Die Quellen des griechischen Alexanderromans, München2. Raabe, R. (ed.) (1896), Ιστορία Ἀλεξάνδρου: die armenische Übersetzung der sagenhaften Alexanderbiographie (Pseudo-Callisthenes) auf ihre mutmassliche Grundlage zurückgeführt, Leipzig. Sémpéré, C. (2006), “La récension epsilon du Roman d’Alexandre: l’écriture infinie, ou le ‘Roman’ d’un mythe”, in: L’information littéraire 58, 35–39. Shirinian, E. M. (2001), “Translations from Greek in Armenian Literature”, in: Eikasmos 12, 229–240. Simonyan, H. A. (1989), Patmut‘iwn Ałek‘sandri Makedonac‘woy [History of Alexander of Macedonia], Erevan. Simonyan, H. A. (1998), “La versione armena del Romanzo di Alessandro e i principi ispiratori del testo”, in: Finazzi/Valvo 1998, 281–287. Stoneman, R. (ed.) (2007), Il Romanzo di Alessandro, I, Milano. Stoneman, R. / Erickson, K. / Netton, I. (eds.) (2012), The Alexander Romance in Persia and the East, Groningen. Stoneman, R. / Gargiulo, T. (eds.) (2012), Il Romanzo di Alessandro, II, Milano. Tašean, Y. (1892), Usumnasirut‘iwnk‘ Sdoyn-Kalist‘eneay varuc‘ Ałek‘sandri [Studies on Pseudo-Callistene’s Life of Alexander], Wien. Topchyan [T’op‘č‘yan], A. (2011), “The Alexander Romance in Medieval Armenian Historiography”, in: C. Gaullier-Bougassas (ed.), L’historiographie médiévale d’Alexandre le Grand, Turnhout, 85–101. Traina, G. (1996), “Lo Pseudo-Callistene armeno. Nota introduttiva”, in: L. Zurli / C. Santini (eds.), Ars narrandi. Studi di narrativa antica in memoria di Luigi Pepe, Napoli, 133–150. Traina, G. (1998), “La recensio α e i suoi paralleli orientali: osservazioni sull’edizione di Kroll”, in: Finazzi/Valvo 1998, 311–322. Traina, G. (2003), La Storia di Alessandro il Macedone. Codice armeno miniato del XIV secolo (Venezia, San Lazzaro, 424), Padova. Traina, G. (2004), “Un dayeak armeno nell’Iberia precristiana”, in: V. Calzolari / A. Sirinian / B. L. Zekiyan (eds.), Bnagirk‘ yišatakac‘/ Documenta memoriae. Dall’Italia e dall’Armenia. Studi in onore di Gabriella Uluhogian, Bologna, 255–262. Traina, G. (2014a), “Alexandre dans la tradition médiévale arménienne” in: Gaullier-Bougassas 2014, 639–647. Traina, G. (2014b), “Alexandre en Arménie: du Moyen Age à la christianisation”, in: GaullierBougassas 2014, 1209–1221. T‘reanc‘, Ṙ. (1842), Patmut‘iwn Ałek‘sandri Makedonac‘woy [History of Alexander of Macedonia], Venezia. Wolohojian, A. M. (ed.) (1969), The Romance of Alexander the Great by Pseudo-Callisthenes, New York.

Alessandro Orengo

Eznik of Kołb as a Translator of Methodius of Olympus Abstract: As is well known, the de Autexusio by Methodius is one of the most relevant sources used by Eznik of Kołb in composing his treatise, conventionally known as Ełc Ałandoc‘ (Refutation of the Sects). The latter included an almost integral translation or reworking of the former, divided into two parts. This was highlighted for the first time by Father Grigoris Galēmk‘earean, who devoted to the subject the greater part of his study on Eznik’s sources, published as a monograph in 1919. In 1924–1925, Louis Mariès addressed the same topic in his work on Eznik’s text. However, in later years, the subject seems to have ceased to draw the attention of Armenologists. In this contribution, after a short summary of the contents of the works both of Methodius and Eznik, we address Eznik’s manner of selecting the sects that are the object of his criticism. We then highlight how Eznik more or less freely translates Methodius’ Greek text, providing some relevant examples.

Eznik of Kołb, bishop of Bagrewand, is one of the most prominent Armenian writers of the fifth century. His treatise ‒ which reached us without a title, but is conventionally known as Ełc Ałandoc‘ (Refutation of the Sects) ‒ is among the first original works ever written in Armenian, contending for absolute primacy with Koriwn’s Vark‘ Maštoc‘i (Life of Maštoc‘). Although it is impossible to establish which is older, they were both composed in the fifth decade of the fifth century, that is, just a few decades after the invention of the Armenian script, when Armenia ‒ no longer an independent country since 428 ‒ ran the risk of being culturally assimilated by Sassanid Persia. Such an assimilation would have chiefly involved a forced conversion of the Armenian people to Mazdeism (in the Zurvanite variety prevailing in Persia at that time). Since the newly-invented script acted as a bulwark for the preservation of Armenian identity, it is easy to understand the import of Koriwn’s enterprise which focused on commemorating the men who invented it and the means through which they achieved their goal. The reasons behind Eznik’s literary activity are also easy to comprehend, since he aimed at refuting certain religious or philosophical beliefs spread in the territories inhabited by Armenian-speaking people. In order to realise his project, Eznik does not hesitate to use Greek and Syriac texts: chief among them is the De

32

Alessandro Orengo

Autexusio by Methodius, an almost integral translation or reworking of which is included (in two parts) in his treatise. This was highlighted for the first time by Viennese Mechitarist Father Grigoris Galēmk‘earean, who – following up on a suggestion by another member of the same congregation, Father K‘erovbē Spenean – devoted the largest part of his study on Eznik’s sources to the subject. This piece of research was published initially in the journal Handēs Amsōreay (years 1893–1894 and 1896), and later as a monograph (Galēmk‘earean 1919).

1 Methodius As for Methodius, not much is presently known about him, save that he was a Christian teacher living in Lycia toward the end of the third or the beginning of the fourth century AD. According to an old tradition, related by Jerome (De vir. ill. 83) and Socrates Scholasticus (Hist. eccl. 6.13) but challenged by modern scholarship, he was bishop of Olympia, in Lycia, and later on of Tyrus. He was possibly martyrized in Chalcis (Euboea), around 311 AD¹. Among his works, the one that is relevant for the present article, the De Autexusio (Περὶ τοῦ αὐτεξουσίου), has reached us in a nearly complete Old Slavonic translation, titled On God, Matter and Free Will. However, in this translation, a few pages are missing, and there are other minimal omissions as well. In addition, the Greek text of the initial section of the treatise is preserved in a manuscript of the Biblioteca Medicea Laurenziana in Florence (Plut. IX, 23), and large excerpts are quoted in the Praeparatio evangelica by Eusebius of Caesarea, in the Philocalia by Origen, as well as in other texts. Finally, large excerpts are also preserved in a dialogue ascribed to one Adamantius² ‒ probably written in the first half of the fourth century AD ‒ which was also translated into Latin by Rufinus of Aquileia. The fragments of the Greek text of the De Autexusio, alongside a German translation of the Old Slavonic version, were published twice, in 1891 and 1917, by Gottlieb Nathanael Bonwetsch (Bonwetsch 1891; Bonwetsch 1917). In the second edition, the editor also took into account Eznik’s treatise. A critical edition of the

1 On Methodius’ biography as documented in ancient literature and evaluated by modern scholarship, see Moreschini/Norelli 1995, 445. 2 A section of his De recta fide ad Deum, or Περὶ τῆς εἰς θεὸν ὀρθῆς πίστεως, is based on excerpts from Methodius’ De Autexusio. On the so-called Adamantius and his work, see Moreschini/ Norelli 1995, 449.

Eznik of Kołb as a Translator of Methodius of Olympus

33

Slavonic and Greek texts, with reference to Eznik’s work, was prepared by André Vaillant and published in the Patrologia Orientalis (Vaillant 1930). As for the De Autexusio’s contents, three characters take part in the dialogue: two heterodox ones and an orthodox one, to be identified with the Author himself. The heterodox speakers postulate that beside God, Matter (ὕλη) also exists, as eternal as the former, and the origin of all evil; the orthodox refutes such opinions and counters them with his own (i.e. the opinions of a Christian). After a short introduction, the real dialogue begins: the first heterodox points out that the universe’s workings testify to the existence of a divine economy, so that it can either come from something as eternal as God, or from God himself. On the other hand (and here Eznik’s first excerpt begins), evil exists in the world as well, and men commit it. Consequently, it is necessary to accept that some Matter must have existed, co-eternal to God, from which the latter created the universe, and which is also the origin of evil. However, the last statement is refuted by the orthodox, according to whom it is impossible to assume the existence of two uncreated beings. Besides, even if such Matter should exist, that would not exonerate God from responsibility for the existence of evil: indeed, evil would still be created by Him, albeit indirectly, because He would necessarily be the one to give quality to Matter, which is devoid of it. Therefore, in such a case, postulating the existence of Matter is absolutely useless. Now it is the second heterodox’s turn to speak, starting from a different assumption. In his opinion, Matter is endowed with quality from the beginning: consequently, it can be the origin of evil. This statement is refuted by the orthodox, who argues that, if that were true, God would be useless. After silencing his second opponent this way, the orthodox continues his discussion with the first. They speak again of evil: the orthodox distinguishes between real evil and divine punishments, which are not actually evil, but merely acts of justice. As for true evil, this is an accident, not a substance; furthermore, no act can be considered evil by nature, because the same act can be both good and bad, according to the purpose of the performer. To sum up, it is men who perform evil. The heterodox asks whether evil can be considered a human invention, or rather whether men have been created by God already inclined to perform it, or, finally, whether they act this way on someone else’s instigation. In his answer, the orthodox completely refutes the second hypothesis: man has been created as a free being, and consequently, he can obey whomever he wants. Evil began when man, endowed with free will, refused to obey God. At this point, the orthodox’s speech becomes fully Christian in tone, as he explicitly refers to Adam’s sin, although it is clear that his opponents do not share his beliefs. The orthodox continues his argument, stating that God did not ignore what the snake, that is the devil, would do, but nonetheless cannot be considered

34

Alessandro Orengo

responsible for the evil that ensued: the devil acted freely, testing man’s free will, and he continues to exist as an instrument for verifying man’s ability to resist temptation. At this point Eznik’s first excerpt from Methodius ends. The second one, containing what remains of the Greek treatise, is quite distant from the first in the Armenian text. It concerns the reason why God created the world: according to the orthodox, the reason is that God wished to implement his creative faculty, in order to demonstrate his goodness and divine essence. Consequently, He cannot be considered as a mere handicraftsman, compelled to work on pre-existing Matter, but rather as the real creator that He is. These are, in short, the dialogue’s contents: later on, we will try to determine to which school of thought the two heterodox characters adhere.

2 Eznik We will now address Eznik’s text, beginning with a short résumé of it. After detailing the features of the divinity and those of rational beings, the Author faces his first opponents, the pagan Greek dualists, according to whom evil exists because, at the moment of creation, there was another being beside God, the Hylē (hiwł), that is, Matter. At this point the first and larger excerpt from Methodius is inserted, which occupies paragraphs 4–56 in Mariès and Mercier’s edition (Mariès/Mercier 1959)³. After this section, the discussion continues, dealing with the relativity of evil among animals and plants, the causes of sickness and death, diabolic possessions, and, finally, the nonexistence of some alleged divinities of the pagan Armenian pantheon. The treatise then continues with sections respectively devoted to Persian dualism of the Zurvanite variety, to astrology, and to the theories conceived by Greek philosophers. This last part ends with the second and shorter excerpt from Methodius, occupying paragraphs 353–354 in the aforementioned edition of Eznik⁴. In the Greek text, this excerpt contains the opinion of the orthodox, with whom Eznik agrees, so much so that he inserted the passage in a different part of his treatise than the longer one. After confuting various sects external to the Christian faith, the Author deals with one internal heresy, namely, Marcionism, and with this the treatise comes to a close.

3 The first excerpt occupies sections 1.4.2 to 1.14.12 in the edition by Martiros Minasean as reprinted in Minasean 2003. 4 The second excerpt occupies sections 3.17.3 to 3.17.18 in Minasean 2003.


35

Beside the Bible, the De Autexusio is the most important source of the Ełc Ałandoc‘: it has been translated almost in its entirety and inserted in the Armenian text. The translation, partly literal, partly very free⁵, occupies more than one seventh of the Armenian text. Methodius might have appeared to Eznik as a very peculiar writer, partly because, as mentioned above, he was supposed to have died a martyr’s death. This particular aura surrounding the Greek writer could explain why, beside translating and inserting Methodius’ treatise on Free Will into his own work, Eznik also briefly quoted two other books of his, namely, De Resurrectione and De Sanguisuga⁶. In any case, there is another possible explanation for the origin of the excerpts from the De Autexusio in Eznik’s treatise: he could have drawn them, at least in part, from the aforementioned dialogue by the so-called Adamantius. In some cases, Methodius and Adamantius have different readings, and Eznik seems to follow sometimes the former, sometimes the latter⁷: therefore, it is possible that the Armenian Author knew Adamantius. Alternatively, it could be argued that Eznik read the De Autexusio in a text slightly different from the one that has reached us and more similar to Adamantius’; however, we are not going to discuss the matter any further in the present contribution.

3 The Valentinians Resuming our discussion on the De Autexusio, we are now going to address specifically the matter of the doctrine to which Methodius’ two heterodox opponents might have adhered. They have often been identified as Valentinians, both on the basis of the Laurenziana codex, in which the first of them is indicated as Οὐ(α)⁸, and, more importantly, of Adamantius’ dialogue, where the two heterodox speakers are explicitly presented as followers of Valentine. In ancient times, other writers, while mentioning Methodius’ dialogue, suggested that the Author’s

5 This imperfect parallelism between Methodius’ text and its Armenian translation can be explained as a consequence of Eznik’s reworking of his model, but it made Paul Peeters suppose that Eznik could have translated his text not directly from Greek, but rather from a Syriac version of it (about which, however, we have no information). See Peeters 1926, 173–174. 6 For these texts as a source of Eznik, see Mariès 1924, 38–39 of the offprint. 7 Cf. Mariès 1924, 64–67 of the offprint and Mariès 1925, 100–104 of the offprint. 8 Also in Dunderberg 2008, 67–74 (and, more generally, in chapter 4) this first heterodox character at least is considered a Valentinian.

36

Alessandro Orengo

target were the Gnostics in general, or even Origen. André Vaillant⁹, following an opinion already expressed by others, noted that the whole discussion about the origin of evil from Matter has to do with Platonic or Neoplatonic theories, whereas the discussion about the devil as an uncreated being, or as a being created as the Evil One, could originate from Valentinian or, more generally, Gnostic theories. Identifying the philosophical school or sect to which the two heterodox characters in Methodius’ work adhere is obviously relevant for Eznik as well. Although it is theoretically possible that the Armenian author used a work refuting a certain school of thought in order to contend against another ‒ more or less similar ‒ one, it is nonetheless also possible that his target was the same as Methodius’. In the introduction to our Italian translation of Eznik’s text, published back in 1996, we argued that, while reworking Methodius’ text, Eznik was more likely set on refuting Platonic or Neoplatonic theories¹⁰. At present, we are no longer so convinced of that explanation. It is true that the Armenian author does not mention the Valentinians, but the latter are possibly the target of Methodius’ criticism, as mentioned above. Furthermore, Valentinian theories are certainly expressed by the two characters in the aforementioned dialogue by Adamantius, whose work Eznik possibly knew, and which was probably written in the first half of the fourth century AD (cf. supra). Consequently, roughly a century before Eznik wrote his own treatise, such theories were clearly labeled as Valentinian. Last but not least, it is worth noting that both Emperor Julian and Ambrose of Milan¹¹ speak of the presence of Valentinians, in the second half of the fourth century, in Osrhoene, and more precisely in Edessa and the area of Callinicum¹² (the latter being at that time a relevant fortress and trade center, as stated by Ammianus Marcellinus, 23.3.7). On the basis of this evidence, dating back to just a few decades before the Ełc Ałandoc‘ was written, it might be assumed that Valentinians were also present in territories inhabited by Armenian-speaking people at the time of the treatise’s composition.

9 Vaillant 1930, XI-XIX. 10 Orengo 1996, 17–18. 11 Julian, Ep. 115 (ed. J. Bidez) to the Edessene, written at the end of 362 AD or at the beginning of the following year; Ambrose, Ep. 74 (ed. M. Zelzer = Maur. 40) to Emperor Theodosius, written in 388 AD (also reproduced, with a slightly different text, as extra collectionem 1a, Zelzer). On the same topic see also another letter by Ambrose, extra collectionem 1 (Zelzer = Maur. 41), to his sister Marcellina. 12 Καλλίνικος in Greek, also known as Nicephorium.


37

4 Eznik’s modus operandi If indeed the Valentinians ‒ which would then be considered as a non-Christian sect ‒ were the target of the first part of his treatise, we could find here a confirmation of Eznik’s method as a polemicist. As is well known, his work was composed in a peculiar moment in Armenian history, when Armenians felt threatened in their cultural independence, and the opponents with whom the writer contends represented a real menace in the territories inhabited by his people. Unsurprisingly, among these opponents are the “historical” adversaries, the Zurvanite Mazdeists, to whose refutation, however, not much room is devoted (although the section is not so reduced as to suggest that an opportunistic censorship might have intervened). The Marcionists are also addressed: as detailed in the eighth chapter of von Harnack’s monograph on Marcion¹³, their presence was widespread in this area of the East ‒ particularly in the countryside ‒ during the fifth century AD. On the other hand, no specific section is devoted to Manichaeism, and this deserves some attention. As recently pointed out by Desmond Durkin-Meisterernst (Durkin-Meisterernst 2012), all references to Mani contained in the Ełc Ałandoc‘ just aim to prove the similarity between the latter’s doctrines and those of the Mazdeans or the Marcionites. Therefore, it might be assumed that, when Eznik composed his work, Manichaeism was no real threat for the Armenians, or at least it was no longer perceived as such¹⁴. We could also take a step forward, and wonder whether the treatise’s concern with contemporary sects and heresies could also be considered as its greatest limit, and have caused the progressive disappearance of any reference to it in the Armenian literature of the immediately following centuries. Actually, we believe that it is important to distinguish between the solid reputation enjoyed by Eznik as a protagonist of religious and cultural Armenian life in the fifth century AD, and the renown of his treatise. We can assume that his work was read immediately after its composition, both because its Author was a well-known and celebrated ecclesiastic, and because works written in Armenian were relatively few at that time. However, later on such a work probably drew less attention, possibly because it was considered obsolete, since some of the sects refuted by the Author were no longer perceived as a threat by the Armenians, or had disappeared alto-

13 Harnack 19242, particularly 153–160. See also Fiey 1970, and Jullien/Jullien 2001, 38–40. 14 The situation was likely different in the third century AD, when, partially as a consequence of Mani’s travels to Armenia, a Manichaean church did exist in the country. On the topic, see Yevadian, 2007, 169–190, 235–238, 249–250, Yevadian 2011, and also Russell 1998.

www.ebook3000.com

38

Alessandro Orengo

gether¹⁵. Be it as it may, it appears that no sure reference to the Ełc Ałandoc‘ can be detected in Armenian literature immediately after the fifth century. Another piece of evidence ‒ admittedly not very strong ‒ that may suggest a lack of interest in Eznik’s treatise could also be seen in its being extant in just one manuscript (M 1097, copied in 1280)¹⁶. Actually, it has been suggested that a second manuscript might have been the basis of the editio princeps (Smyrna, 1762–1763), but there is no consensus on the matter. In any case, we are well aware that this is an argument to be considered with the greatest caution, because in ancient Armenian literature there are other instances of undoubtedly relevant works that have nonetheless survived in a single manuscript.

5 Eznik as a translator We are now going to discuss how Eznik translates or reworks Methodius’ text. As an example, we have chosen the short section at the beginning of the first excerpt from the De Autexusio in the Armenian text. While quoting from Eznik’s text, we refer both to Mariès/Mercier 1959 and Minasean 2003, with some minimal orthographical alterations. While quoting from the Greek text, we took into account both Bonwetch’s and Vaillant’s editions (Bonwetsch 1917 and Vaillant 1930), giving preference to the variant readings that are closer to Eznik’s text. Readings that have no relevance for our comparison are not quoted. Here are the original texts¹⁷:

15 For references to Eznik’s text in the most ancient Armenian literature, see Orengo 2014. On the topic, see also the brief remarks by Blanchard/Young 1998, 11–12, 30–31, and Thomson 1992, 307. 16 In this manuscript, the text of the Ełc Ałandoc‘ is present together with other works. A recent description of the manuscript can be found in MC’HJ 2008, 295–298. 17 For the Greek text: Bonwetsch 1917, 150–152 (3.1.5); Vaillant 1930, 12–17. For the Armenian text: par. 4 in Mariès/Mercier 1959; 1.4.2–8 in Minasean 2003. See also Galēmk‘earean 1919, 22‒24.


39

METHODIUS

EZNIK

Ἑώρων δύο τινὰς ὁμογενεῖς, ἀνθρώπους δὲ λέγω, διαπληκτιζομένους καὶ διαλοιδορουμένους ἀλλήλοις, ἕτερον δ᾽ αὖ πάλιν ἀμφαιματώσαι πειρώμενον τὸν πλησίον.

Tesanemk‘ erkows ars miazgis ǝnd mimeans grgṙeal, mimeanc‘ mahow ew areamb carawi en.

Ἤδη δέ τινες καὶ δεινότερα τολμᾶν ἤρχοντο. Ὃς μὲν γάρ ἐσκύλευε νεκρὸν καὶ τὸ κρυφθὲν ἤδη σῶμα τῇ γῇ πάλιν ἐδείκνυεν ἡλίῳ, καὶ τῆν ὁμοίαν αὐτῷ ὕβριζεν εἰκόνα, βορὰν κυσὶ καταλιπὼν τὸν νεκρόν.

Ew aylk‘ zgerezmans krkten ew zt‘ałeal marmins ǝnd hołov merkac‘eal, xaytaṙakeal, arewow c‘owc‘anen, ew zmecareal din aṙ p‘owt‘i č‘cackeloy t‘erews ew šanc‘ ews kowr arkanen.

Ὁ δὲ ξίφος ἐγύμνου καὶ ἐπὶ τὸν ὅμοιον ἄνθρωπον […] ἐχώρει· καὶ ὁ μὲν φυγῇ τὴν σωτηρίαν πορίζεσθαι ἤθελεν, ὁ δὲ διώκειν οὐκ ἐπαύετο οὐδὲ τοῦ θυμοῦ κρατεῖν ἤθελε. Καὶ τί δεῖ πλείονα λέγειν, ἀλλ᾽ὅτι χωρήσας ἐπ᾽αὐτὸν εὐθέως ἐπάϊσσε τῷ ξίφει· ὁ δ᾽ ἱκέτης τῷ πλησίον ἐγίνετο καὶ χεῖρας ἱκεσίας ὤρεγεν, καὶ τὴν μὲν ἐσθῆτα διδόναι ἤθελεν, μόνον δὲ τὸ ζῆν ἔχειν ἠξίου· ὁ δ᾽ οὐκ ἔθραυέ τι τὸν θυμὸν οὐδ᾽ ἠλέει τὸν ὁμογενῆ, οὐδὲ ἑαυτὸν διὰ τῆς εἰκόνος ἐν ἐκείνῳ βλέπειν ἤθελεν, ἀλλ᾽ ὡς ἄγριος θὴρ τῷ ξίφει τῆς βορᾶς ἤρχετο, ἤδη δὲ καὶ τὸ στόμα τῷ ὁμοίῳ προσέφερε σώματι, τοσοῦτος γὰρ ἦν τῷ θυμῷ· καὶ ἦν ἰδεῖν τὸν μὲν ἤδη κείμενον, τὸν δὲ λοιπὸν σκυλεύοντα καὶ μηδὲ γῇ σκεπάζοντα τὸ σῶμα οὗ τῆς ἐσθῆτος ἐγύμνωσεν.

Ew ē erbek‘, zi min p‘axowc‘eal ert‘ayc‘ē anjnapowr owrek‘, zkeans anjinn šahel, ew miwsn c‘asmamb borbok‘eal, srov zkni ǝnt‘ac‘eal, oč‘ dadarē, minč‘ew yagec‘owc‘anē zc‘asowmnn: owsti ē ayn anyag c‘asowmn?

Πρὸς δὲ τούτοις ἕτερος προσῄει, ὃς τοῦ πλησίον γυναῖκα παίζειν [alternative reading: βιάζειν]¹⁸ ἤθελε, λῃστεύων γάμον ἀλλότριον καὶ ἐπὶ παράνομον κοίτην τραπῆναι παρορμῶν, τὸν γεγαμηκότα γνήσιον πατέρα γίνεσθαι μὴ θέλων.

Ew owrowmn i mti edeal gołanal zayloy amowsnowt‘iwn ew eleal anawrēnowt‘eamb yawtar mahičs, čtay linel hayr ordwoc‘ aynm, or awrinawk‘n amowsnac‘eal ic‘ē.

Ew omn merkanay zhanderjs ǝnkerin, ew t‘e ǝnddēm daṙnayc‘ē, yarewē ews arkanē.

//In this situation one is allowed to believe the myths about Thiestes and Oenomaus//. … καὶ τῇ τῶν ἀδελφῶν διὰ ξίφους φιλονεικίᾳ οὐκ ἀπιστῶ.

Ew erbek‘ erbek‘ paterazmownk‘ šaržin, owr meławor ew ardar aṙ hasarak kotorin. Ew mahk‘ taražamk‘, ew axtk‘ č‘aračark‘!

18 The reading βιάζειν, not attested in Greek, is suggested on the basis of the Old Slavonic translation.

40

Alessandro Orengo

Here are their translations: METHODIUS

EZNIK

I saw two persons of the same race, I mean men, who were fighting and insulting each other, one trying to spill his neighbor’s blood.

We saw two men of the same race disputing between themselves, eager for each other’s death and blood.

And others have already begun to dare to do more terrifying acts as well. One was unclothing a dead man, and showing once again to the sun his corpse already concealed in the earth, making violence to the image that was similar to his own, leaving the dead as food for dogs.

Others are digging in the graves and showing to the sun, naked and insulted, the corpses buried into the earth, and, acting in a hurry, they don’t conceal the honored body, but possibly they also leave it to the dogs as food.

Another one was unsheathing his sword and advancing […] toward a man, a similar being to himself. The latter wanted to escape and save himself, but the former continued to chase him and did not want to suppress his anger. What can we add, if not that the former was urging the latter, immediately dashing against him with the sword. The latter was begging his neighbor, extending his suppliant hands and saying that he agreed to give him his garments, provided that the adversary would let him live. But his enemy did not suppress his anger, nor had compassion for a member of his own race, nor wanted to see himself in his neighbor’s image, but, like a wild beast, began to chop his prey with the sword, already thrusting the latter’s tip (στόμα)¹⁹ toward a body similar to his own: so great was indeed his anger. And you could see one of them lying on the ground and the other unclothing him and then not covering with earth his corpse, which he himself had stripped of the dress.

And sometimes it happens that someone escapes and seeks refuge somewhere, in order to save his own life, and his adversary, burning with anger, chases him with the sword in his hand, and doesn’t stop until he has given vent to his fury: whence comes such an insatiable rage?

Then another one came forward, wishing to seduce [alternative reading: to rape] his neighbor’s wife, stealing the other’s marriage and forcing the woman to enter an illegitimate bed, thus preventing the bridegroom to be the real father of his sons.

And another one, resolving to steal the marriage of another person, entered, without any right, into a bed which is not his own, preventing the one who legally married the woman to be the father of his sons.

And someone steals his neighbor’s garment and if the latter remonstrates, he kills him too

19 There is a pun in the text, because στόμα means both “tip (of the sword)” and, more usually, “mouth”, “jaws”.


41

//In this situation one is allowed to believe the myths about Thiestes and Oenomaus.// … and I thought not unworthy of belief the rivalry between the brothers, settled with the sword.

And sometimes wars break out, during which both the sinner and the righteous are killed together. And what about premature deaths? And awful sufferings?

Eznik’s working method, as can be seen in the excerpt quoted above, is clear enough. Methodius seems to focus on the following four situations: 1. two men fighting to death; 2. graves being desecrated; 3. most probably, a brigand assaulting his victim; alternatively, an enemy being killed; 4. an adulterer. These are followed by a reference to the credibility of the atrocities described in old myths. As far as Eznik is concerned, we obviously do not know whether he was reading Methodius’ text as we know it. Moreover, here and elsewhere, the Armenian text might contain lacunae due to the process of manuscript transmission. Nonetheless, if we accept this passage as it has reached us, we can see that the Author follows his model ‒ albeit without providing a truly literal translation ‒ in the first, the second and the fourth section. On the other hand, the third section has been heavily shortened and divided into two parts, respectively describing the killing of an enemy and the assault of a brigand. As for the mythological references that were present in the source text, Eznik did not deem it appropriate to transfer them into Armenian, but the reference to the rivalry between brothers, to be identified with Eteocles and Polynices, might have given him the opportunity to speak about wars, to which he also adds the mention of premature deaths and sufferings. Before moving on to another topic, it is worth mentioning that the Armenian word miazgis, “of the same race”, used by Eznik in the first section, is certainly a calque of the Greek word ὁμογενεῖς, even if the whole sentence is modified. The Armenian term might even have been created by Eznik himself, if we consider that this is the only occurrence to be quoted in the Nor Baṙgirk‘ Haykazean Lezowi²⁰ (admittedly not a very complete thesaurus). In any case, it should be

20 NBHL 1837, 264, s.v. miazgi.

42

Alessandro Orengo

pointed out that this calque is actually an interpretation rather than a mechanical rendering of its model, because, otherwise, Greek ὁμο- would have been rendered with Armenian ham-. As a matter of fact, Armenian hamazgi is indeed attested, but only in later Authors²¹. Although, in the passage analyzed above, Eznik significantly alters the text of Methodius, in other instances he renders his source quite faithfully. In the De Autexusio, at one point, the orthodox says that man has received a marvelous gift by God, being endowed with free will whereas all the other beings are compelled to obey divine orders, like slaves. We find the same argumentation in Eznik: as in Greek, the statement is followed by an example, which is partially reproduced below²².

METHODIUS

EZNIK

Ἐάν τε γὰρ οὐρανὸν εἴπῃς, ἕστηκε φέρων τὸν δεσπότην, οὐ μετακινούμενος τοῦ ὡρισμένου τόπου·

Et‘e zerknic‘ asic‘es, kay hastateal ew oč‘ šarži i sahmaneloy nma telwoyn.

καὶ ἐάν τε περὶ ἡλίου τὸν λόγον ποιεῖσθαι θέλῃς, ἐκτελεῖ οὗτος τὴν ὡρισμένην κίνησιν, οὐ παραιτούμενος τὸν δρόμον, ἀλλὰ ἀνάγκῃ τινὶ δουλεύων τῷ δεσπότῃ.

Ew et‘e zaregakanē kamic‘is asel, sakayn ew na katarē zzatowc‘eal nma zšaržowmnn ew č‘išxē xowsap‘el yǝnt‘ac‘ic‘n, ayl i harkē caṙayē terownakan hramanin.

If you want to speak about the sky, it stands still carrying its Lord, and doesn’t move from the place assigned to it;

If you want to speak about the sky, it stands still, and it doesn’t move from the place assigned to it;

if, on the contrary, you prefer to speak about the sun, it fulfills the movement assigned to it, not refusing to follow its way, but serving its Lord, driven by some necessity.

if, on the contrary, you prefer to speak about the sun, it too fulfills the movement assigned to it, and does not dare to escape from its way, but by necessity fulfills its Lord’s orders.

In this passage, as can be easily seen, Eznik translates his model quite faithfully, and just omits the reference to the sky carrying its Lord, a statement that, if indeed present in the Greek source he was using, would probably have sounded weird to him. Alternatively, we can assume that Eznik actually translated that sentence,

21 NBHL 1837, 14, s.v. hamazgi. 22 For the Greek text: Bonwetsch 1917, 186–187 (16.3); Vaillant 1930, 70–71. For the Armenian text: par. 43 in Mariès/Mercier 1959; 1.11.4–5 in Minasean 2003. See also Galēmk‘earean 1919, 60.


43

which would have later been lost in the manuscript tradition of the Ełc Ałandoc‘. Be it as it may, we believe that, even considering similar opposite explanations (either a dependence on a Greek text different from the one that has reached us, or the presence of a lacuna in the Armenian text), studying possible omissions operated by Eznik, particularly in passages otherwise literally translated, would be of some interest. As mentioned above, Eznik’s translation is faithful, but not slavishly so. This can be showcased, for instance, by considering the different ways in which Greek participles are rendered²³. They can be translated either with a coordinate clause (οὐ μετακινούμενος = ew oč‘ šarži, “and it does not move”; οὐ παραιτούμενος = ew č‘išxē xowsap‘el, “and it does not dare to escape”; δουλεύων = caṙayē, “it slavishly fulfills”) or with another participle (τοῦ ὡρισμένου = i sahmaneloy, “from the assigned [place]”; τὴν ὡρισμένην = zzatowc‘eal, “assigned”), followed by an indirect object which was not present in the Greek text (namely, nma, “to it”). It is true that, while translating τὸν λόγον ποιεῖσθαι, Eznik prefers a trivial asel, “to say”, which he had already used before, but we will not blame him for it: it is not always true that variatio delectat. This approach is likely based on the methods adopted while translating the Holy Books, and possibly, on the previous training derived from the practice of translating the Scriptures orally in front of a congregation, before the invention of the Armenian alphabet made a written and more fixed translation possible, as is documented in Armenian sources. The passage at hand prompts a further remark. Eznik uses here a second person singular (“if you want … if you prefer”) in order to render the same person in the original text. It might be assumed that, in this case, “you” (singular) has an impersonal value in Armenian; nevertheless, it is significant that the same feature is present in Methodius’ text, where “you” is used as part of a dialogue. According to Louis Mariès, Eznik might have failed to adapt the dialogic structure of the source text to his own narrative one: this could be explained by assuming that Eznik himself would translate the text orally, while his assistants put it into writing²⁴. As the French scholar points out, a similar process is documented for Jerome’s Latin version of the book of Tobias, according to the information provided in the preface by the translator himself. We might add the case of Dawit’ hiwpatos and Step‘anos Siwnec‘i, who, during the eighth century AD, translated into Armenian some works by the Pseudo-Dionysius the Areopagite, as well as

23 For the rendering of Greek participles into Armenian in fifth-century texts, see Bănățeanu 1937. 24 Mariès 1924, 89 of the offprint.

44

Alessandro Orengo

the De Natura Hominis by Nemesius of Emesa. For these enterprises, Moreno Morani²⁵ suggested a partly similar translation process: one of the authors would read the Greek text aloud, while the other translated it into Armenian. Once more, comparing Methodius’ text with Eznik’s Armenian translation seems to offer, albeit indirectly, some indication on the methods followed by the vardapet of Kołb in composing his work. In so doing, it clarifies a process that would find parallels among Armenian authors of his time.

Bibliography Bănățeanu, V. (1937), La traduction arménienne des tours participiaux grecs, Bucarest. Blanchard, M. J. / Young, R. D. (eds.) (1998), A Treatise on God Written in Armenian by Eznik of Kołb (floruit c. 430-c. 450), an English Translation, with Introduction and Notes, Leuven. Bonwetsch, G. N. (ed.) (1891), Methodius von Olympus. 1, Schriften, Erlangen-Leipzig. Bonwetsch, G. N. (ed.) (1917), Methodius, Leipzig. Dunderberg, I. (2008), Beyond Gnosticism. Myth, Lifestyle, and Society in the School of Valentinus, New York. Durkin-Meisterernst, D. (2012), “Eznik on Manichaeism”, in: Iran and the Caucasus 16, 1–11. Minasean, M. (ed.) (2003), Eznik Kołbac‘i, Ełc Ałandoc‘, in: Matenagirk‘ Hayoc‘ / Armenian Classical Authors [director of publication Zaven Yegavian], vol. 1, Ant‘ilias, 429–512. Fiey, J.-M. (1970), “Les Marcionites dans les textes historiques de l’Église de Perse”, in: Le Muséon 83, 183–188. Galēmk‘earean, G. (1919), Noragoyn alberk‘ Eznkay Kołbac‘woy Эnddēm alandoc‘ matenin, Vienna. Harnack, A. von (19242), Marcion: das Evangelium vom fremden Gott. Eine Monographie zur Geschichte der Grundlegung der katholischen Kirche, zweite, verbesserte und vermehrte Auflage (1st ed. 1921), Leipzig. Jullien, C. / Jullien, F. (eds.) (2001), Les Actes de Mar Mari. L’apôtre de la Mésopotamie, introduction, traduction et notes, Turnhout. Mariès, L. (1924–1925), “Le De Deo d’Eznik de Kołb connu sous le nom de ‘Contre les sectes’. Études de critique littéraire et textuelle”, in: Révue des Études Arméniennes 4, 113–205 (= offprint 1–93), 5, 12–130 (= offprint 94–212). Mariès, L. / Mercier, C. (eds.) (1959), Eznik de Kołb, De Deo, édition critique du texte arménien (Patrologia Orientalis XXVIII/3), Paris (reprint Turnhout 2004). MC‘HJ (2008), Mayr c‘owc‘ak hayerēn jeṙagrac‘ Maštoc‘i anowan matenadaranin, 4, Erevan. Morani, M. (2003), “Sul valore della traduzione dello Pseudo-Dionigi l’Areopagita”, in: A. Sirinian / S. Mancini Lombardi / L. D. Nocetti (eds.), Le scienze e le ‘arti’ nell’Armenia medievale. Seminario internazionale (29–30 ottobre 2001), Bologna, 37–47. Moreschini, C. / Norelli, E. (1995), Storia della letteratura cristiana antica greca e latina. 1, Da Paolo all’età costantiniana, Brescia.

25 Morani 2003, 44–45.


45

NBHL (1836–1837), G. Awetik‘ean / X. Siwrmēlean / M. Awgerean (eds.), Nor Baṙgirk‘ Haykazean Lezowi, Venetik (reprint Erevan 1979–1981). Orengo, A. (ed.) (1996), Eznik di Kołb, Confutazione delle sette (Ełc Ałandoc‘), introduzione, traduzione e note, Pisa. Orengo, A. (2014), “Lo Ełc ałandoc‘ di Eznik e la sua ricezione nella letteratura armena dal V all’VIII secolo”, in: La teologia dal V all’VIII secolo fra sviluppo e crisi. XLI Incontro di Studiosi dell’Antichità Cristiana, Roma, 9–11 maggio 2013, Roma, 803–823. Peeters, P. (1926), review of: “L. Mariès, Le De Deo d’Eznik de Kołb connu sous le nom de ‘Contre les sectes’. Études de critique littéraire et textuelle. Paris, 1924”, in: Analecta Bollandiana 44, 171–174. Russell, J. R. (1998), “A Manichaean Apostolic Mission to Armenia?”, in: N. Sims-Williams (ed.), Proceedings of the Third European Conference of Iranian Studies, 1, Wiesbaden, 21–26 (reprinted in: Id., Armenian and Iranian Studies, Cambridge [MA], 2004, 893–898). Thomson, R. W. (1992), “‘Let Now the Astrologers Stand Up’: The Armenian Christian Reaction to Astrology and Divination”, in: Dumbarton Oaks Papers 46, 305–312 (reprinted in: Id., Studies in Armenian Literature and Christianity, Aldershot-Brookfield, 1994 [article XI]). Vaillant, A. (ed.) (1930), Le De Autexusio de Methode d’Olympe, version slave et texte grec édités et traduits en français (Patrologia Orientalis XXII/5), Paris (reprint Turnhout 1974). Yevadian, M. K. (2007), Christianisation de l’Arménie. Retour aux sources. 1, La genèse de l’Église arménienne, Lyon. Yevadian, M. (2011), “Mani et l’Arménie”, in: Haykazean Hayagitakan Handēs 31, 405–412.

Valentina Calzolari

The Transmission and Reception of the Greek Cultural Heritage in Late Antique Armenia: The Armenian Translations of the Greek Neoplatonic Works* Abstract: This paper deals with the ancient Armenian translations of Greek philosophical texts (esp. the Neoplatonic corpus). Before focusing on this topic, a broad overview on the presence of Armenian students in the Greek schools in Late Antiquity will be presented, followed by a survey of the Armenian translations issued from the so-called “Hellenizing school”, through which the works necessary for learning the disciplines (artes) of the trivium (grammar, rhetoric, and dialectic) were introduced into Armenia. The core of this article will deal with the Neoplatonist heritage and the translatio studiorum in Armenia, with special emphasis on the Armenian translations of the Greek commentaries of the Neoplatonic philosopher David, who was called the “Invincible” in the Armenian tradition. The second part of the paper will pay particular attention to some methodological criteria that should be followed while studying such works in Greek and in Armenian. After some information on the approach of Classical philologists to the Armenian translations from Greek, the importance of cross-comparison and of an interdisciplinary investigation will be stressed in conclusion.

Over the centuries, Armenia played an important role in the preservation and transmission of the Greek cultural heritage. The old centres of translation of Late Antique Armenia as well as the scriptoria and libraries attached to the monastic schools of medieval Armenia contributed significantly to the diffusion of the Greek texts. These texts were first studied directly in Greek at the schools where they were used as textbooks, and later translated into Armenian, thus contributing to

* This paper was presented as a webcast lecture to the Genoa workshop from the Department of Near Eastern Languages and Civilizations at Harvard, which hosted me during the academic year 2013–2014 as a visiting scholar. I wish to express warm thanks to professor David Elmer and professor Gregory Nagy, who both helped me to organise this webcast lecture. Special thanks also to Ms Elizabeth DeLozier (Harvard University Network Services Group) for her perfect technical assistance and to the IT staff at the University of Genoa.

48

Valentina Calzolari

the formation of a distinctive Armenian cursus studiorum. While the first translations of Greek writings, in the first half of the fifth century, were concerned with religious works, by the end of the fifth century Armenian translators turned their interests towards secular works as well. These latter works were called artak‘in in Armenian, i.e. “external”, a noun equivalent to a Greek expression (οἱ ἔξωθεν, “those of outside”, ἡ ἔξωθεν, “the [paideusis] of outside”) already used by Paul¹ and by the Church Fathers to refer to pagan authors or texts. Familiarity with Greek authors “of outside”, however, goes back to a more ancient time. As early as the fourth century, some Greek sources provide information on the presence of Armenian students in the most prestigious schools of Antiquity: Athens, Antioch, and later Alexandria, Byzantium, and Trebizond. Before introducing and discussing the ancient Armenian translations of Greek philosophical texts, which is the topic of this paper, a broad overview of the presence of Armenian students in the Greek schools will be first given. This will be followed by a survey of the Armenian translations composed by the so-called “Hellenizing school”, with emphasis on the philosophical works. The second part of the paper will pay particular attention to methodological criteria that should be followed while studying such works, in Greek and in Armenian, focusing above all on the Armenian translations of the Greek commentaries by the Neoplatonic philosopher David, who was called the “Invincible” in the Armenian tradition². The importance of a multilateral approach and an interdisciplinary investigation will be stressed in the conclusion of the paper.

1 Armenian students in the Greek schools of rhetoric of Athens and Antioch 1.1 Athens In the fourth century, the school of rhetoric of Athens was an international centre. In this respect, Paul Gallay wrote:

1 1 Tim. 3.7. 2 On this topic, see also my previous articles Calzolari 2005; Calzolari 2007; Calzolari 2009a; Calzolari 2009b; Calzolari 2012; Calzolari 2014, from which I drew some information and considerations incorporated in this paper.

The Armenian Translations of the Greek Neoplatonic Works

49

Au point de vue de l’enseignement, Athènes gardait un prestige incontesté ; par ses écoles, elle était encore la demeure de l’éloquence. Pour appartenir à l’élite des gens cultivés, il fallait presque necessairement avoir passé par Athènes³.

Among the learned people who received an education at the desks of the School of Athens, one may recall the Armenian Prohaeresius (276–369), an important figure often mentioned in the context of the cultural contacts between the Greek and Armenian world, and concerning whom we are informed in particular by Eunapius of Sardis⁴. In his Lives of the Sophists (Βίοι φιλοσόφων καὶ σοφιστῶν)⁵ Eunapius relates that Prohaeresius was of Armenian origin; after studying first at Antioch under the professor of rhetoric Ulpian, Prohaeresius distinguished himself as one of the most brilliant students at the Athens school of rhetoric, where he later became a professor. As he came from the East, the students from Pontus and Asia Minor were assigned to his class. In Athens, Prohaeresius’ most famous pupils included Gregory of Nazianzus and Basil of Caesarea, as well as Eunapius himself⁶. Basil mentioned Prohaeresius as a professor⁷, while Gregory of Nazianzus wrote an epitaphium in honour of his master⁸. Eunapius also mentions a journey by Prohaeresius to Rome, where he was honoured with a statue bearing an inscription that read Η ΒΑΣΙΛΕΥΟΥΣΑ ΡΩΜΗ ΤΟΝ ΒΑΣΙΛΕΥΟΝΤΑ ΤΩΝ ΛΟΓΩΝ “Rome, the Queen, to the King of Eloquence”⁹. Several other sources contain a reference to the activity of Prohaeresius; some are interesting not only for the reconstruction of his biography, but also as evidence concerning the presence of Armenian students in Athens. Additional information on Prohaeresius and more generally on Armenian students attending Greek schools in the fourth century is also provided by Socrates, Sozomenus¹⁰, and other Greek authors. In an encomium in honour of Basil, written after his death, Gregory stressed that Basil was a brilliant student at the time of his attendance at the School of Athens, far outshining the others in his class, to the point of raising the jealousy

3 Gallay 1943, 37–38. 4 On Prohaeresius, see Goulet 2000; Watts 2006; Di Branco 2011, especially on the question of his faith, Christian or pagan, and of his banishment from Athens for a while. 5 LCL 134, 476–514. 6 See Socrates, Hist. eccl. 4.26, Sozomenus, Hist. eccl. 6.17; see also the following footnotes. 7 PG 29.23. 8 Gregory of Nazianzus, Epit. 5 (PG 38.15). 9 See also Gregory of Nazianzus, Epit. 5 13A (PG 38.15); Libanius, Ep. 278. 10 See supra, n. 6.

50

Valentina Calzolari

of some classmates, and especially of some Armenian students¹¹. In a Letter¹², Gregory relates that at the beginning of his stay in Athens (350–355/6), some Armenians, former pupils of the father of Basil, Basil the Older, in Neocaesarea, approached the future Father of the Church and began a discussion aiming at testing Basil’s talent. Their real aim was to make him fail, but they were shamefully defeated by his expertise in using the weapon of syllogisms and Homeric quotations. But regardless of this negative judgment, which may have been real or exaggerated¹³, the mention of Armenian students at the School of Athens is an interesting piece of evidence for our purposes. The renown of the school of Athens in Armenia was so great that the Armenian medieval sources established a close connection between two founding figures of Armenian thought, namely David the Invincible and the historian Moses of Khoren – both considered as fifth century authors¹⁴ –, attributing to them a common sojourn in this Greek school, where, moreover, they are depicted classmates of Gregory and Basil (fourth century). This information cannot be possible for obvious chronological reasons.

1.2 Antioch Still in the fourth century, other Armenians attended the school of rhetoric directed by Libanius in Antioch, which was a deeply Hellenized Syrian city. In his numerous letters, Libanius mentions several pupils of Armenian origin. Paul Petit underlined this statement with the following words: De toutes les provinces orientales, c’est l’Arménie qui fournit à Libanius son plus important contingent avec ses 20 ressortissants, et c’est vraiment étonnant. D’autre part, s’il y a parmi eux quelques chrétiens [4/5], ils appartiennent tous à des familles curiales, ceux du moins – ils sont 11 – sur lesquels nous sommes suffisamment renseignés¹⁵.

11 Gregory of Nazianzus, Or. 43.17 (PG 36.517 = SC 384.43 17) (Funerary encomium of Basil). See also n. 13. 12 Gregory of Nazianzus, epist. 43.17. 13 Some possible personal reasons explaining this sharp judgment were given by Bernardi 1984, 352–359. 14 While the chronology of Moses is still debated (see for instance Traina 1991, for a dating in the fifth century, and Garsoïan 2003–2004 for a later chronology), it is accepted that David, a supposed author of the first half of the fifth century according the medieval tradition, was a later author, living in the sixth century (see above). 15 Petit 1956, 132–135.


51

The students probably came from Armenia Minor, i.e. Western Armenia, which was under Roman influence, and not from Greater Armenia. Nevertheless, this information is noteworthy.

1.3 Alexandria, Trebizond and Constantinople Later, around the sixth century, Armenian students attended the Neoplatonic school of Alexandria, on which we will focus below. In the seventh century, Anania Širakac‘i, who introduced the arts of the quadrivium to Armenia, was a pupil of the Greek Tychikos at the school in Trebizond, to which Anania had travelled in order to study above all mathematics and philosophy. As he explained in his autobiography¹⁶, on his way “to the country of the Greeks”, he met several other Armenians coming from Byzantium, who dissuaded him from going to the capital, and recommended he should rather go to Trebizond, in order to attend the classes of Tychikos, a Byzantine doctor who lived in Armenia and knew Armenian¹⁷. With regard to the Armenian presence in Byzantium, we may also mention that in the eighth century, Step‘annos Siwnec‘i went to the capital, where he met a number of Armenians who introduced him to the library of the Hagia Sophia and worked together with him on the translation of pseudo-Dionysus the Areopagite¹⁸.

2 The influence of Greek thought in Armenian native literature: the example of Eznik By attending Greek schools, Armenians gained familiarity with the textbooks used in these centres, and began to translate them. This period of Armenian literature is generally called “Hellenizing School” (in Armenian, Yunaban Dproc‘)¹⁹, where the use of the term “school” aims at indicating the main common – or supposed common – translation technique and vocabulary of the translations dating from this epoch, more than a localisation or the name of some teachers, which remains mostly unknown.

16 Berbérian 1964. 17 On Tychikos, see Lemerle 1964. 18 On an Armenian presence in Byzantium, see also Terian 1982. 19 On the Hellenizing School, see Manandean 1928; Mercier 1978–1979; Terian 1982; Calzolari 1989; Zekiyan 1997; Muradyan 2014.

52

Valentina Calzolari

Before focusing attention on this period of Armenian literature, and above all on the translations of the philosophical works, certain aspects should be underlined. Philosophical literature in the Armenian language includes both works in translation and works written directly in Armenian²⁰. A survey of the development of philosophical literature in Armenia and generally speaking of the influence of Greek thought should also take into consideration Armenian works that are not specifically philosophical, but which may presuppose traces of the influence of Greek philosophy. Too often the question of the knowledge of Greek sources by the Armenian authors is viewed only in terms of the existence or absence of Armenian translations. This is an argument ex silentio, which in itself is not determinant; moreover, the Armenian writers could read Greek sources directly in Greek. Consequently, even in the absence of Armenian translations, or with regard to the period before some of the sources were actually translated, it is legitimate to ask whether any of the Greek texts were actually known and whether they influenced Armenian literature. The corpus of Armenian texts is more difficult to define, but it is nevertheless important for an understanding of the conditions in which Greek philosophical thought spread in Armenia. One such example is the theological treatise by Eznik Against Sects [Ǝnd or Ełc ałandoc‘], from the classical age of Armenian literature (first half of the fifth century)²¹. One section of this treatise is directed against “outside” (artak‘in) philosophers, that is to say, the pagan philosophers, and more specifically the Pythagoreans, Peripatetics, Stoics, and Epicureans. Eznik’s refutation is aimed particularly against Plato who, in Eznik’s assessment, “appeared to the Greeks as the most pious of all philosophers,” but who, more than any other, fell into impiety (§ 357). Eznik contended that Greek philosophers were to be reproached for failing to distinguish the Creator from the creature; they also wrongly supposed, like the Persian dualists and the heretic Marcion, the existence of a source of evil independent from God, the Hylē. Although Eznik provides information on the philosophical currents mentioned above, excerpts from his treatise on the Greek philosophers appear to rely heavily on Greek patristic sources such as the Anakephalaiosis, included in the Panarion by Epiphanius of Salamis († 403)²² and the De Universo by Hyppolitus of Rome (third century). It cannot be established with certainty whether the Armenian author had direct knowledge of

20 This chapter does not seek to give exhaustive bibliographical references on the subject, but rather to provide a critical overview of the history of research. For additional bibliographic information, consult Thomson 1995 and Thomson 2007. On the works from the Hellenizing School, see also Zuckerman 2001. 21 See also the contribution by A. Orengo, in this volume. 22 Orengo 1996, 17–19.


53

the philosophical works that he refutes. Nevertheless, it is clear that Eznik’s mode of development of arguments in his treatise seems to be inspired largely by Greek philosophy, if not directly by Aristotelian logic; this is one area of research that deserves to be examined further²³. In his works, we find for the first time in Armenian literature certain ideas that were widespread in the ancient world, such as the theory of the four elements (earth, air, fire, water), or lamps that revolve around fixed paths, returning each day to mark the passage of time. The reception of philosophy and more generally of Greek thought in ancient Armenian historiography is thus an area for further research²⁴. The influence of Neo-Platonism on medieval poetry, for example, on Gregory of Narek²⁵ and on later poets, is also a major field that demands investigation.

3 Translations of Greek works at the time of the Hellenizing School The diffusion of Greek philosophical literature in ancient Armenia was closely linked to the activity of translators from the Hellenizing School, which developed from the late fifth and early sixth century to the eighth. Through the activity of these translators, the works necessary for learning the disciplines (artes) of the first cycle of the encyclopaedic cursus – the trivium (grammar, rhetoric, and dialectics) – were introduced into Armenia. Among the first works translated into Armenian during this epoch is the Grammar attributed to Dionysios Thrax (second century BC)²⁶. Another noteworthy translation is the Armenian version of the first century AD rhetorician Aelius Theon’s Progymnasmata (preliminary exercises for the study of rhetoric)²⁷, as well as the Book of Chreias (in Armenian Girk‘ Pitoyic‘), which is a Christianized edition of Aphthonius of Antioch’s Progymnasmata (fourth–fifth centuries AD)²⁸.

23 Contin 2005. 24 Stepanyan 2009. 25 Tamrazyan 2004; see also Calzolari 2014, 388 and n. 2. Attention should likewise be devoted to the influence of philosophical thought on the Church Fathers or theologians translated into Armenian, such as Pseudo-Dionysius the Areopagite and Gregory of Nyssa. See among others, La Porta 2007; La Porta 2008; Thomson 1987; Vardanyan 2008. 26 Adontz 1970; Sgarbi 1991; Clackson 1995; Weitenberg 2001; Sgarbi 2004. In the following centuries, the Grammar became the object of several Armenian commentaries: Adontz 1970; Ervine 1995. 27 Patillon 1997, with an introduction to the Armenian translation by G. Bolognesi. 28 Muradyan 1993.

54

Valentina Calzolari

The introduction of the third liberal art, namely dialectics and more generally, philosophy, was largely due to translations of the writings of Aristotle and Porphyry, as well as the commentaries by David on Aristotle and Porphyry. In addition to these works, the philosophical corpus that was created thanks to the translations of the Hellenizing School also included pseudo-Aristotle’s De Mundo and De Virtutibus, some writings of the Corpus Hermeticum, a collection of the Definitions, the treatise De Natura attributed to Zeno²⁹, and several works of Philo of Alexandria or of the pseudo-Philo³⁰. Still open is the question of the dating of different versions of Plato’s dialogues (Euthyphro, Apology of Socrates, Minos, Laws, Timaeus). According to some scholars, they were probably written by the translators of the Hellenizing School³¹, whereas other philologists believe they were the result of the work of Grigor Magistros, in the tenth and eleventh centuries³². The corpus of Greek philosophical writings translated into Armenian also included Proclus’ Elements of Theology. However, this work was not translated until in the Middle Ages, by Simēon Gaṙnec‘i, from an older Georgian version, rather than from the Greek text.

4 The Neoplatonist heritage in Armenia and the translatio studiorum A large part of the corpus of philosophical texts reveals the debt that Armenian philosophical speculation owes in particular to the Neoplatonic schools of Alexandria. This is an important feature from the point of view of the transmission of the Neoplatonist Greek heritage outside the centres of production. The question of the translatio studiorum is usually invoked to describe the transfer of the scholarly centres that followed the closing of the last philosophical school directed by pagans – the Neoplatonic school of Athens –, under the order

29 A non-exhaustive list. On the translation of the philosophical works into Armenian, see Zuckerman 2001. See also Mahé 1998; Stone/Shirinian 2000, 7–15. There are also several manuscripts preserved in the Library of Ancient Manuscripts in Erevan (Matenadaran) that contain works still unedited and unidentified: see Stone/Shirinian 2000, 11–12, and nn. 40–42. Knowledge on the Armenian ancient philosophical corpus thus remains incomplete. 30 Quaestiones et Solutiones in Genesim et in Exodum, De Providentia I-II, De Animalibus, Legum Allegoriae I-II, De Abrahamo, De Vita Contemplativa; Ps.-Philo, De Jona, De Sampsone: bibliography by Terian 2001; see also Mancini Lombardi/Pontani 2010. 31 Arevšatyan 1971. 32 Leroy 1935. Recent studies in this direction have been carried out by Dr. Irene Tinti: see also her contribution, in this volume.


55

of the emperor Justinian, in 529. The translatio then took the direction of the Sassanid Empire at the time of the king Husrō I Anuširvān (531–579). According to the testimony of the Byzantine historian Agathias (536–582), the Iranian king is said to have offered hospitality to the philosophers banished from Athens, among whom one may mention Simplicius and Damascius³³. The translatio took various paths. One well-known route is the Syro-Arab path, which led from Alexandria to Baghdad, after the closing of the Alexandrian school in the seventh century. This path took a Western direction, crossing through Cordoba and Toledo. At the same time, the “Libraries of the Neoplatonists”³⁴ also circulated in Armenia. As of the sixth and the seventh century, the activity of the translators shows a close link with the curricula of the ancient Neoplatonic schools of Athens and Alexandria, and especially with the Aristotelian curriculum (see also infra).

5 Similarities between the Aristotelian cursus of the Neoplatonic schools and the corpus of Armenian translations of the philosophical works The syllabus of the Neoplatonic schools is known to have been organized around the two main lines of the Aristotelian and Platonic cursus, where the first was considered as an introductory phase necessary to approach the second. Both the curricula were structured in a precise order proceeding from the supposed easiest to the most difficult texts. Each commentary was preceded by an introduction organized around a pre-defined series of discussion points, which had the function of orienting the reading of the texts³⁵. Study of Aristotle began with a general introduction to philosophy, in which four questions of Aristotelian inspiration³⁶ were raised and solved on the grounds of the definitions given by the ancient philosophers (Pythagoras, Plato, and Aristotle himself):

33 According to a hypothesis put forward by Michel Tardieu, these philosophers settled in the city of Hârran (unless these two transfers constitute a single transfer), where they apparently opened a new school: Tardieu 1986, 1–44. But this hypothesis is not unanimously shared. On this question, see also Hadot 2014. 34 See D’Ancona 2007. 35 Hadot 1997, 173; see also Wildberg 1990; Mansfeld 1994; Mansfeld 1998. 36 See Arist. An. post. 2.89b 23; Barnes 19942, 203–204; Hadot 1987, 100 n. 1.

56

Valentina Calzolari

Does philosophy exist? (εἰ ἔστι) What is philosophy? (τί ἐστι) What is it like? (ὁποῖόν τί ἐστι) Why does it exist? (διὰ τί ἐστι) The introduction also addressed the question of the different parts of philosophy – theoretical and practical –, with their subdivisions³⁷. This is the structure of the Prolegomena to Philosophy, i.e. the first work of the curriculum. Study of Aristotle then continued with the introduction par excellence to Aristotelian logic, i.e. Porphyry’s Isagoge (third century), along with a commentary on the same text. After students had learned the Isagoge, they were considered ready to face the reading of the works of Aristotle, and in particular the Categories, the De Interpretatione, and the Prior Analytics (in this order). All these works were read and explained pericope after pericope, following the procedure of the so-called “continuous commentary” genre, which was the most common literary form of teaching philosophy, in Late Antiquity, and which survived the decline of the Neoplatonic schools³⁸. A comparison between the Aristotelian syllabus and the corpus of Armenian translations made by the Hellenizing School allows many possible connections to be identified³⁹. Thus the philosophical works translated into Armenian include Aristotle’s Categories and De Interpretatione⁴⁰, Porphyry’s Isagoge⁴¹, David’s commentaries on Isagoge, Categories, and Prior Analytics, as well as his Prolegomena to Philosophy. The same corpus also comprised two anonymous Armenian commentaries on the Categories and De Interpretatione attributed to David, mentioned above⁴², as well as a third Commentary on the Categories which is fragmentary and anonymous⁴³. These parallels may be considered as evidence of the contacts between Armenians and the Neoplatonic Greek schools. If the compar-

37 Hadot 1979. 38 Hoffmann 2007; Hadot 1997, 169. 39 A different case is that of the Platonic syllabus, which is distinct from the Platonic corpus of the Armenian translations. The only parallel is offered by the Timaeus, which was very popular in the Late Antiquity. 40 On the Armenian translations of the writings of Aristotle, see Tessier 1979; Bodéüs 2001, CLVII-CLXI; Tessier 2001; mention should also be made, despite its early date of publication, of Conybeare 1892. For other bibliographical references, see Thomson 1995, 35–36; Zuckerman 2001, 427–428; Thomson 2007, 170. 41 Sgarbi 1972. 42 Attributed in the manuscripts to a certain “Amelawxoy”, or “Amelaxos”, a noun in which some scholars have read a deformation of Greek Ἰάμβλιχος. On this topic, see Mahé 1989, who challenges this hypothesis. 43 Shirinian 2009, 101–102.


57

ison is extended to encompass the Syriac philosophical collection created in the same centuries, then further interesting considerations arise: for instance, the Syriac corpus also features the translations of Porphyry’s Isagoge, Aristotle’s De Interpretatione and the Prior Analytics, as well as of some treatises attributed to Aristotle: Aristotle’s Categories (three translations, sixth–seventh centuries) Aristotle’s De Interpretatione (two translations, sixth–seventh centuries) Aristotle’s Prior Analytics (two translations, seventh century) Porphyry’s Isagoge (two translations, sixth–seventh centuries) See also: Ps.-Aristotle’s De Mundo (sixth century) Ps.-Aristotle’s De Virtutibus (sixth century) Beside the translations, the Syriac corpus contains several commentaries on Aristotle’s Categories, De Interpretatione, and Prior Analytics as well as on Porphyry’s Isagoge⁴⁴. The comparison with the Armenian renderings is interesting in various respects. The many correspondences that can be detected between the Armenian and Syriac corpora can be regarded as evidence for the circulation of an ensemble of works in common between the two oriental areas in the same epoch (sixth–eighth centuries)⁴⁵. In the Syriac tradition, the names and personalities of many translators and commentators are known. In this respect, mention should be made at least of Sergius of Reshaina, Paul the Persian, Jacob of Edessa, Probus, Athanasius of Balad, Severus Sebokt, and George, bishop of the Arabs. In contrast, the transmission of Neoplatonic philosophy in Armenia is closely connected to only one figure, David the Invincible.

6 David the Invincible and his works in Greek and in Armenian According to the majority of modern scholars, David studied at the Neoplatonic school of Alexandria at the time of Olympiodorus (still alive in 565) and Elias (sixth century). His Armenian origin, claimed by the Armenian medieval sources, is

44 On the Syriac corpus, see Hugonnard-Roche 2007; Hugonnard-Roche 2009. The Syriac corpus does not include, for that period, any translation of Plato. 45 On the circulation of the Neoplatonic libraries, see D’Ancona 2007.

58

Valentina Calzolari

not unlikely, even if it cannot be proved⁴⁶. David may have been one of the numerous Armenian students who studied in the Greek schools in Late Antiquity. Like the Armenian sophist Prohaeresius did about two centuries before, David obtained a chair and pursued his commentatorial activity at the same school where he had formerly been a student. It is important to bear in mind that it is only on the basis of intertextual parallels and similarities concerning the content and the structure, shared by the Greek commentaries attributed to David and some other commentaries of the Alexandrian school, that David may be presumed to have belonged to the Ammonius – Olympiodorus – Elias line⁴⁷. No other information on David’s life can be found in the Greek tradition⁴⁸, whereas the Armenian tradition, especially in the Middle Ages, is overflowing with biographical details, very often contradictory or legendary⁴⁹. Among many other details, in addition to the translation of his own commentaries on Aristotelian logic (see infra) the tradition further attributes to David the translation of Dionysius Thrax’ Grammar, and of the works of Aristotle and Porphyry. By virtue of this attribution, David can be seen as one of the central figures of the Hellenizing School, and was instrumental introducing the arts of trivium to Armenia. Today, a corpus of three works is generally attributed to David, in Greek and Armenian: the Prolegomena, the Commentary on the Isagoge, and the Commentary on the Categories, although with reservations by some scholars regarding the authorship of the latter⁵⁰. To these is added a fourth work, the Commentary on the Prior Analytics, attested only in Armenian, unless we recognize its model in the fragmentary Greek text of the Commentary on Prior Analytics by Elias, identified by Leendert Westerink in the 1960s⁵¹. As an epigone of the school of Alexandria, in the sixth century, David stands on the shoulders of a long tradition. An in-depth understanding of his works not only in Greek, but also in Armenian, can be achieved only by taking into due consideration the links with the prior Greek tradition. This is an aspect often neglected by Armenian scholars in the past. Thanks to the investigations of

46 On David’s life, see Thomson 1983; Mahé 1986; Calzolari 2009a. 47 Westerink 1990, XXXVII. 48 On David in the Greek tradition, see Barnes 2009. 49 This topic has already been discussed in Calzolari 2009a. See also Arevšatyan 1981; Kendall/ Thomson 1983, XI-XXI; Mahé 1990; Mahé 1997. 50 The question is summarized in Calzolari 2009a, 29–33; see also Mahé 1990, which summarizes the main points of Arevšatyan 1969. 51 Westerink 1961. The question is debated: Papazian 2009; Sweeting 2009; Topchyan 2009; Topchyan 2010.


59

modern historians of ancient Greek philosophy, today we know the main characteristics of the Neoplatonic commentaries and in particular of the Prolegomena⁵². We know, for instance, that these commentaries were highly stereotyped in structure and content, as well as in the choice of questions, examples, and style of argument. David’s Greek works followed the same patterns⁵³. If David inherited and continued a long tradition in Greek, the Armenian translator, or translators, whoever he or they may have been, undertook a pioneering task in Armenia, where the study of Aristotelian logic was only at its beginning. This situation, specific to the Armenian tradition, as well as the distance of the centres of translation from the centres of Greek production, allowed the translator, or translators, to enjoy a certain freedom from the more stereotypical aspects of the Greek tradition. The differences between the Greek and the Armenian texts represent important features to understand how the Greek heritage was transmitted outside its milieu of production. All these considerations lead to a general question: what is the correct methodology for evaluating the divergences of the Armenian texts from their Greek models? That is to say, what is the right approach to follow while studying the Armenian translations of the Greek Neoplatonic commentaries?

7 Methodological considerations concerning the study of the Armenian and Greek works Several avenues of investigation may be suggested. First of all, a correct enquiry into the differences between the Armenian and the Greek texts presupposes a general study of the method of translation that was adopted. In this regard, it can be stressed that unlike a number of other works of the Hellenizing School, the translations of the Prolegomena and of the Commentary on Porphyry’s Isagoge, for example, are less verbatim⁵⁴. A number of reasons may be put forward for the differences, which can be summarized in the following main points⁵⁵:

52 See for instance Barnes 19942; Hadot 1987; Hadot 1990; Hadot 1991; Hadot 1997; Mansfeld 1994; Mansfeld 1998; Mueller-Jourdan 2007; Westerink 1990. 53 Hoffmann/Luna 2001, 863. 54 See Calzolari 2009b; Muradyan 2009; Muradyan 2014. The case of the translation of the Commentary on the Categories appears to be rather different; a study on this work has been carried out by Manea E. Shirinian in the framework of the Commentaria Aristotelem Armeniaca (see infra). 55 In this respect, it is worth mentioning that Giancarlo Bolognesi and his school, in addition to other researchers in Italy and elsewhere, paved the way to correct methodological research on

60

Valentina Calzolari

a) In some cases, they may be explained as the result of an adaptation of the source language (Greek) to the target language (Armenian)⁵⁶. We are aware that in current research on translation such an argument could appear fairly obsolete. Nevertheless, this is an issue that does have relevance when studying ancient translations – not only in Armenian, but also for instance in Syriac⁵⁷ –, which calls for both a linguistic and philological approach. b) In certain cases, the differences may be explained as the result of a misreading by the translators or of a corruption attested in the manuscripts (Greek or Armenian). The latter point can be further investigated as follows in point c) below: c) In some cases, the divergences may suggest a Greek model different from that which has been handed down by the Greek manuscripts at our disposal today. A close examination of the variants of both the Armenian and Greek versions is therefore essential, in order to determine whether the Armenian translations have preserved passages now lost or corrupted in the Greek tradition that has come down to us. This is a very plausible hypothesis. In effect, the Greek manuscripts employed by the Armenian translator(s) in Late Antiquity were more ancient than the medieval manuscripts that modern editors of the Greek commentaries may read today. Quite often, the testimony of the Armenian is fundamental for restoring the Greek. For instance the Prolegomena and the Commentary on Porphyry’s Isagoge have many examples of lacunae by homoteleuton which can be filled through the Armenian translations⁵⁸. The importance of the Armenian translations as witnesses for the critical restitution of the Greek text deserves greater attention from specialists in classical philology. The above-mentioned observations are primarily philological and linguistic. Armed with dual training in Armenian studies and classical studies, scholars should focus on the question of the relationship between the Armenian versions and the underlying Greek models, in order to evaluate the contribution of Armenian as a “tool” for restoring the Greek originals. In certain cases, the Armenian evidence is essential, especially when the Greek original is lost, as is the case with some of the writings of Philo, Pseudo-Philo, and David’s Commentary on

the Armenian translations resulting from the Hellenizing School, and especially on the correct interpretation of the differences between the Greek and the Armenian. See for instance his classical work on the Progymnasmata of Aelius Theon: Bolognesi 1962; on the scholarly research on the Armenian translations from Greek, see Finazzi 2012. 56 See examples in Calzolari 2009b, 41–48. 57 See the important works of Henri Hugonnard-Roche (for instance Hugonnard-Roche 2004) and of Sebastian Brock (for example Brock 1993 and Brock 2003). 58 See Calzolari 2009b, 55–65; Muradyan 2009.


61

Prior Analytics⁵⁹. Therefore, Armenian philology can and must be of service to classical philology. It is nevertheless important to stress that the Armenian translations should also be studied for the sake of their own specificity. In order to understand the manner of reception of Greek thought in Armenia, the differences compared to the original text are notably more interesting than the fidelity verbum de verbo. For instance, as we have pointed out elsewhere, in the Armenian version of David’s Prolegomena to Philosophy the omission of a passage dealing with Aristotelian teaching on the nature of the sky and of the First Mobile seems to reveal the desire to hedge a theme – that of quintessence – which provoked polemical reactions by both Christian and pagan ancient commentators alike⁶⁰. d) Finally, it is therefore crucial for a correct analysis of the Armenian translations of the Greek texts to devote attention to differences that may be the result of intentional modifications by the translator. This is a difficult part of the inquiry, which should be conducted by adopting an enlarged and broader approach, based not only on an intra-Armenian perspective, but also taking into account the results of the most recent investigation on Neoplatonic literature in Greek, as already mentioned above. Only thus can insight be gained into the mode of transmission and reception of Greek philosophical thought outside its sphere of production. Such research is thus a task of relevance not only to Armenologists, but also to specialists in the history of Greek philosophy and, more generally, to scholars interested in the mode of transmission of Greek thought in the East. In this respect, it is crucial to promote a multilateral appraisal of the philosophical literature in translation, and to give a short overview of the history of the interdisciplinary research already conducted in this field.

8 Towards closer collaboration between specialists in ancient Greek philosophy and Armenologists The importance of the Oriental translations of the Greek texts was already clear to scholars in the first half of the nineteenth century. For instance, the academics of the Royal Society of Science, meeting in Göttingen in 1830, expressed the

59 But see above, n. 51. 60 Calzolari, 2009b, 49–54.

62

Valentina Calzolari

wish to create a list of ancient translations of Greek works into Eastern languages, including Armenian, while recalling the value of these versions as evidence for reconstituting the original Greek⁶¹. This project was then initiated by scholars such as Johann Georg Wenrich, who, in his De auctorum Graecorum versionibus et commentariis Syriacis, Arabicis, Armeniacis, Persicisque commentatio, collected a great deal of information about the ancient Armenian translations of Aristotle, Porphyry, Plato, and David the Invincible, thanks to the Catalogue of the Oriental Manuscripts of the Bibliothèque royale de France, as well as the historical and philological works of the Mekhitarist Fathers of Venice and of Jean Saint-Martin⁶². The heritage of ancient Armenian translations also drew the interest of Oxford scholars such as Frederic C. Conybeare, who conducted many philological studies on the Armenian translations of Philo, Plato, Aristotle, Porphyry, and David, in which he sought to identify the critical value of Armenian in relation to the Greek⁶³. Of course, the work of the nineteenth century scholars needs to be completed or corrected. However, these studies are important evidence of an enduring interest in the literature involving Armenian translations. They also contributed to paving the way to a new avenue of research in textual criticism of the Greek texts. In this respect, it is worth mentioning the edition of the De Mundo by William L. Lorimer, who included Armenian variants in the apparatus of his edition based on the work of Conybeare⁶⁴. The results of collating the Armenian text with the Greek text of the Categories and De Interpretatione⁶⁵ carried out by Conybeare later served as the basis of a new edition of the Greek by Lucio Minio-Paluello in 1949⁶⁶. The turn of the millennium saw the beginning of a new impetus in this field. In recent works devoted to Greek philosophical texts, evidence from the Armenian versions has been analysed. This was the case with the edition of De Interpretatione by Elio Montanari and the Categories by Richard Bodéüs as well as the translation of Porphyry’s Isagoge, with commentary by Jonathan Barnes⁶⁷. An interest in the corpus of Armenian works by David can also be seen in the work of

61 Wenrich 1842, IV: “Ut colligantur notitiae de versionibus auctorum Graecorum Syriacis, Arabicis, Armeniacis, Persicis, quarum versionum historia accurata adhuc caremus”. 62 Saint-Martin 1818; Sevin 1739. 63 Conybeare 1889; Conybeare 1891; Conybeare 1892; Conybeare 1893; Conybeare 1895. 64 Lorimer 1933; on the limitations of this edition, see Tessier 1979, 46–49; new critical apparatus useful for restituting the Greek can be found in Tessier 1979, 53–122. 65 According to Waitz 1844. 66 Minio-Paluello 1949. 67 Montanari 1984; Bodéüs 2001; Barnes 2003.


63

Ilsetraut Hadot, director of a research project on the Greek work of Simplicius⁶⁸, or of Richard Goulet, the editor of the Dictionnaire des philosophes antiques⁶⁹. Another recent article, by Fr. Henri D. Saffrey, seeks to reconstruct the itinerary of the famous Greek manuscript A of Plato’s works (Parisinus gr. 1807) from Alexandria and Byzantium to the Italy of Petrarch, taking into account a possible stopover in Armenia at the time of Grigor Magistros⁷⁰. Closer cooperation among experts from both subject areas constitutes a strong desideratum for the future. In this respect, a pioneering role has been played by the team focusing on the treatise by the Pseudo-Zeno, led by Michael E. Stone and Jaap Mansfeld, with the collaboration of David T. Runia and Manea E. Shirinian from the Matenadaran in Erevan⁷¹. More recently, research has begun in the context of an extensive project on the works of David the Invincible, set up at the University of Geneva under the direction of the chair of Armenian Studies (held by the author of this paper), in cooperation with two experts in the history of ancient philosophy, i.e. Jonathan Barnes (University of Geneva and subsequently University of Paris-Sorbonne) and Dominic O’Meara (University of Fribourg, Switzerland). This collaborative study has been extended to include scholars of the Institute of ancient Armenian manuscripts, known as the “Matenadaran” (Manea E. Shirinian, Gohar Muradyan, Aram Topchyan, Sen Arevšatyan and Arminē Melkonyan), and in the first phase of the project, the Erevan State University (Albert Stepanyan). The project has resulted in the creation of an editorial program called Commentaria in Aristotelem Armeniaca, which was incorporated in the collection Philosophia Antiqua⁷². It attempts to take into consideration all the above-mentioned main criteria necessary for a correct and fruitful interpretation of the Armenian translation of Greek philosophical literature.

68 Hadot 1990. 69 Goulet 2000. With regard to the issue of the authorship of the Commentary on the Categories, he mentions the need to resort to “la prise en compte systématique de la tradition arménienne” (p. 65). 70 Saffrey 2007; but see Tinti 2012. 71 It is also important to mention the collaboration between Manea E. Shirinian and Doug Hutchinson on the Greek and Armenian texts of the Pseudo-Aristotelian treatise, De Virtutibus. See Shirinian [Širinjan] 2002. 72 The project of the Commentaria in Aristotelem Armeniaca is presented in Calzolari/Barnes 2009, XI-XIV, in Topchyan 2010, IX-X and in Muradyan 2015, XVIII-IX.

64

Valentina Calzolari

9 The “Commentaria in Aristotelem Armeniaca. Davidis Opera” project The series Commentaria in Aristotelem Armeniaca. Davidis Opera will consist of five volumes dedicated to the commentaries on Aristotelian logic which the tradition attributes to David the Invincible. The first collective volume outlines the questions to be dealt with and serves the purpose of introducing the series. Each of the following four volumes will present a separate work: Prolegomena to the philosophy, Commentary on Porphyry’s Isagoge, Commentary on Aristotle’s Categories, and Commentary on Aristotle’s First Analytics. Each volume will include the revision of the critical edition, a new commented translation of the Armenian text, as well as a comparison of the Armenian with the underlying Greek model, whenever the latter can be identified. This projects has five chief goals: 1. To identify the position of the Armenian indirect witnesses within the Greek direct tradition of David the Invincible’s works. 2. To assess the value of the Armenian translations for an editor working on textual recovery of the Greek original. 3. To analyse the differences between the Armenian versions and the Greek originals so as to investigate the methodology adopted by the Armenian translators when facing a new audience. 4. To pay particular attention to examination of the language and the translation technique of the Armenian version. 5. To inquire, more generally, into the ways Greek thought was transmitted to the Armenian tradition. Three volumes have now appeared⁷³. We hope that the project will be able to revive scholars’ awareness of this heritage and its importance for insight into the circulation of ideas and cultural exchange between East and West in Late Antiquity⁷⁴. As a final consideration, I will conclude this article by insisting, once again, on the fact that in order to appreciate the real value of Armenian philosophical writings, the only proper methodological approach is an interdisciplinary study. In this respect, the meeting organised in Genoa offered a major opportunity to renew an important tradition created some decades ago in Italy, and to develop stronger cooperative research in this direction.

73 Calzolari/Barnes 2009; Topchyan 2010; Muradyan 2015. 74 On this respect, see for instance Lemerle 1971; Cavallo 2001, 189–199.


65

Bibliography Adontz, N. (ed.) (1970), Denys de Thrace et les commentateurs arméniens, Louvain (French translation of the Russian original edition, published in Saint Petersburg in 1915). Akinean, N. (1959), Dawit‘ Hark‘ac‘i Anyałt‘ P‘ilisop‘ay. Keank‘n ew gorcerǝ/David Harkazi Unbesiegter Philosoph. Leben und Wirken, Vienna. Arevšatyan, S. S. (1969), “Dawt‘i Anhałt‘i žaṙangut‘yunǝ nor lusabanut‘eamb” [The legacy of David the Invincible in a new light], in: Banber Matenadarani 9, 7–22. Arevšatyan, S. S. (1971), “Platoni erkeri hayeren t‘argmanut‘yan žamanakǝ” [The epoch of the Armenian translation of Plato’s works], in: Banber Matenadarani 10, 7–20. Arevšatyan, S. S. (1981), “David l’Invincible et sa doctrine philosophique”, in: Revue des Études Arméniennes 15, 33–43. Barnes, J. (ed.) (19942), Aristotle, Posterior Analytics, Oxford (1st ed. 1975). Barnes, J. (ed.) (2003), Porphyry, Introduction, Oxford. Barnes, J. (2009), “David and the Greek Tradition”, in: Calzolari/Barnes (eds.) 2009, 3–14. Barnes, J. / Bobzien, S. / Flannery, K., s.j. / Ierodiakonou, K. (eds.) (1991), Alexander of Aphrodisia, On Aristotle Prior Analytics 1.1–7, London. Berbérian, H. (1964), “Autobiographie”, in: Revue des Études Arméniennes 1, 189–194. Bernardi, J. (1984), “Nouvelles perspectives sur la famille de Grégoire de Nazianze”, in: Vigiliae Christianae 38/4, 352–359. Bodéüs, R. (ed.) (2001), Aristote, Catégories, Paris. Bolognesi, G. (1962), “La traduzione armena dei Progymnasmata di Elio Teone”, in: Accademia Nazionale dei Lincei. Rendiconti della Classe di Scienze morali, storiche e filologiche (ser. 8) 17, 86–125; 211–257. Brock, S. (1993), “The Syriac Commentary Tradition”, in: C. Burnett (ed.), Glosses and Commentaries on Aristotelian Logical Texts. The Syriac, Arabic and Medieval Latin Traditions, London, 3–10. Brock, S. (2003), “Syriac Translations of Greek Popular Philosophy”, in: P. Bruns (éd.), Von Athen nach Bagdad. Zur Rezeption griechischer Philosophie von der Spätantike bis zum Islam, Bonn, 9–28. Calzolari, V. (1989), “L’école héllenisante: faits et circonstances”, in: M. Nichanian (ed.), Ages et usages de la langue arménienne, Paris, 110–130. Calzolari, V. (2005), “Du pouvoir de la musique dans la version arménienne des Prolegomena de David le platonicien (Orphée et Alexandre le Grand)”, in: A. Kolde / A. Lukinovich / A.-L. Rey (eds.), Κορυφαίῳ ἀνδρί. Mélanges en l’honneur d’André Hurst, Genève, 417–431. Calzolari, V. (2007), “Aux origines de la formation du corpus philosophique en Arménie: quelques remarques sur les versions arméniennes des commentaires grecs de David”, in: D’Ancona (ed.) 2007, 259–278. Calzolari, V. (2009a), “David dans la tradition arménienne”, in: Calzolari/Barnes (eds.) 2009, 15–36. Calzolari, V. (2009b), “La version arménienne des Prolegomena philosophiae de David et son rapport avec le texte grec”, in: Calzolari/Barnes (eds.) 2009, 39–65. Calzolari, V. (2012), “The Ancient Armenian Translations of Greek Philosophical Texts: the Works of David the Invincible”, in: Scripta & e-Scripta. The Journal of Interdisciplinary Medieval Studies, 10–11, 131–146.

66

Valentina Calzolari

Calzolari, V. (2014), “Philosophical Literature in Ancient and Medieval Armenia”, in: Calzolari (ed.) 2014, 349–376. Calzolari, V. (ed.) / with the collaboration of M. E. Stone (2014), Armenian Philology in the Modern Era. From Manuscript to Digital Text, Leiden-Boston. Calzolari, V. / Barnes, J. (eds.) (2009), L’oeuvre de David l’Invincible et la transmission de la pensée grecque dans la tradition arménienne et syriaque, Leiden-Boston. Cavallo, G. (2001), “Qualche riflessione sulla continuità della cultura greca in Oriente nei secoli VII e VIII”, in: Fiaccadori (ed.) 2001, 189–199. Clackson, J. P. (1995), “The Techne in Armenian”, in: V. Law / I. Sluiter (eds.), Dionysius Thrax and the Techne Grammatike, Münster, 121–133. Contin, B. (2005), “Alle origini del pensiero filosofico armeno. Eznik di Koghb e Dawith l’Invincibile”, in: Annali di Ca’ Foscari 14/3, 69–78. Contin, B. (ed.) (2014), Davide l’Invincibile, Definizioni e divisioni della filosofia, introduzione, traduzione e note, Palermo. Conybeare, F. C. (1889), “Note on the Ancient Armenian Versions of Plato”, in: The Classical Review 3, 340–343. Conybeare, F. C. (1891), “On the Ancient Armenian Versions of Plato”, in: The American Journal of Philology 12, 193–210. Conybeare, F. C. (1892), A Collation with the Ancient Armenian Versions of the Greek Text of Aristotle’s Categories, De Interpretatione, De Mundo, De Virtutibus et Vitiis and of Porphyry’s Introduction, Oxford. Conybeare, F. C. (1893), “Aristotēli yałags ǝndōrinakut‘ean ‘Yałags Astucoy’” [A manuscript of Aristotle’s De Deo], in: Handēs Amsōreay 7, 227–232. Conybeare, F. C. (1895), Philo about the Contemplative Life, or the Fourth Book of the Treatise Concerning Virtues, Oxford. Criscuolo, U. (1994), “Sull’‘epitafio’ di Gregorio di Nazianzo per il retore Proeresio”, in: C. Curti / C. Crimi (eds.), Scritti classici e cristiani offerti a F. Corsaro, Catania, 189–195. D’Ancona, C. (ed.) (2007), The Libraries of the Neoplatonists. Proceedings of the Meeting of the European Science Foundation Network ‘Late Antiquity and Arabic Thought. Patterns in the Constitution of European Culture’, Strasbourg, March 12–14, 2004, Leiden-Boston. Di Branco, M. (2011), “Politica e religione in Atene nel IV secolo d.C. Il caso del sofista Proeresio”, in: La parola del passato 76, 31–46. Ervine, R. (1995), “Yovhannēs Erznkac‘i Pluz’s Compilation of Commentary on Grammar as a Starting Point for the Study of Medieval Grammars”, in: J. J. S. Weitenberg (ed.), New Approaches to Medieval Armenian Language and Literature, Amsterdam, 149–165. Fiaccadori, G. (ed.) (2001), Autori classici in lingue del Vicino e Medio Oriente, Roma. Finazzi, R. B. (2012), “Cinquant’anni di ricerche sulle antiche traduzioni armene di testi greci”, in: C. Baffioni / R. B. Finazzi / A. Passoni Dell’Acqua / E. Vergani (eds.), Gli studi orientalistici in Ambrosiana nella cornice del IV centenario (1609–2009). Primo Dies Academicus 8–10 novembre 2010, Milano, 144–170. Frede, M. (2005), “Les Catégories d’Aristote et les Pères de l’Eglise grecs”, in: O. Bruun / L. Corti (eds.), Les Catégories et leur histoire, Paris, 135–173. Gallay, P. (1943), La vie de saint Grégoire de Nazianze, Lyon-Paris. Garsoïan, N. (2003–2004), “L’Histoire attribuée à Movsēs Xorenac‘i: que reste-t-il à en dire?”, in: Revue des Études Arméniennes 29, 29–48.


67

Goulet, R. (2000), “Prohérésius le païen et quelques remarques sur la chronologie d’Eunape de Sardes”, in: Antiquité Tardive 8, 209–222. Hadot, I. (1987), “Les introductions aux commentaires exégétiques chez les auteurs néoplatoniciens et les auteurs chrétiens”, in: M. Tardieu (ed.), Les règles de l’interprétation, Paris, 99–122. Hadot, I. (dir.) (1990), Simplicius, Commentaire sur les Catégories, 1 and 3, Leiden-New York-København-Köln. Hadot, I. (1991), “The Role of the Commentaries on Aristotle in the Teaching of Philosophy according to the Prefaces of the Neoplatonic Commentaries on the Categories”, in: H. Blumenthal / H. Robinson (eds.), Aristotle and the Later Tradition, Oxford, 175–189. Hadot, I. (1997), “Le commentaire philosophique continu dans l’Antiquité”, in: Antiquité Tardive 5, 169–176. Hadot, I. (2014), Le néoplatonicien Simplicius à la lumière des recherches contemporaines. Un bilan critique, Sankt Augustin. Hadot, P. (1979), “Les divisions des parties de la philosophie dans l’Antiquité”, in: Museum Helveticum 36, 201–223. Hadot, P. (1990), “La logique, partie ou instrument de la philosophie?”, in: I. Hadot (dir.), Simplicius, Commentaire sur les Catégories, 1, Leiden-New York-København-Köln, 183–188. Hoffmann, P. (2007), “What was Commentary in Late Antiquity? The Example of the Neoplatonic Commentators”, in: M. L. Gill / P. Pellegrin (eds.), The Cambridge Companion to Ancient Philosophy, Oxford, 597–622. Hoffmann, P. / Luna, C. (eds.) (2001), Simplicius, Commentaire sur les Catégories d’Aristote: chapitres 2–4, Paris. Hugonnard-Roche, H. (2004), La logique d’Aristote du grec au syriaque. Études sur la transmission des textes de l’Organon et leur interprétation philosophique, Paris. Hugonnard-Roche, H. (2007), “Le corpus philosophique syriaque aux VIe-VIIe siècles”, in: D’Ancona (ed.) 2007, 279–291. Hugonnard-Roche, H. (2009), “La tradition gréco-syriaque des commentaires d’Aristote”, in: Calzolari/Barnes (eds.) 2009, 153–173. Kendall, B. Thomson, R. W. (eds.) (1983), David the Invincible Philosopher, Definitions and Divisions of Philosophy, English translation of the Old Armenian version with introduction and notes, Chico, CA. La Porta, S. (2007), Two Anonymous Sets of Scholia on Dionysius the Areopagite’s Heaveny Hierarchy, Leuven. La Porta, S. (2008), The Armenian Scholia on Dionysius the Areopagite. Studies on their Literary and Philological Tradition, Leuven. Lemerle, P. (1964), “Note sur les données historiques de l’autobiographie d’Anania de Shirak”, in: Revue des Études Arméniennes 1, 195–202. Lemerle, P. (1971), Le premier humanisme byzantin, Paris. Leroy, M. (1935), “Grégoire Magistros et les traduction arméniennes d’auteurs grecs”, in: Annuaire de l’Institut de Philologie et d’Histoire Orientales 3 (= Mélanges Jean Capart), 263–294. Lorimer, W. L. (ed.) (1933), Aristotelis qui fertur libellus De mundo, Paris. Lössl, J. / Watt, J. W. (eds.) (2011), Interpreting the Bible and Aristotle in Late Antiquity. The Alexandrian Commentary Tradition between Rome and Baghdad, Farnham.

www.ebook3000.com

68

Valentina Calzolari

Mahé, J.-P. (1987), “Quadrivium et cursus d’études au VIIe siècle en Arménie et dans le monde byzantin”, in: Travaux et Mémoires 10, 159–206. Mahé, J.-P. (1989), “Amelaxos”, in: R. Goulet (ed.), DPhA, 1, Paris, 160. Mahé, J.-P. (1990), “David l’Invincible dans la tradition arménienne”, in: I. Hadot (dir.), Simplicius, Commentaire sur les Catégories, 1, Leiden-New York-København-Köln, 189–207. Mahé, J.-P. (1995), “L’Arménie et les Pères de l’Eglise: histoire et mode d’emploi (Ve–XIIe siècle)”, in: J.-C. Fredouille / R.-M. Roberge (eds.), La documentation patristique, Laval-Paris, 157–179. Mahé, J.-P. (1998), “Du grec à l’arménien”, in: Encyclopédie philosophique universelle. 4, Le discours philosophique, Paris, no. 64. Manandean, H. (1928), Yunaban dproc‘ǝ ew nra zargac‘man šrǰrannerǝ [The Hellenizing School and the phases of its development], Vienna. Mancini Lombardi, S. / Pontani, P. (eds.) (2010), Studies on the Ancient Armenian Version of Philo’s Works, Leiden-Boston. Mansfeld, J. (1994), Prolegomena. Questions to be settled before the Study of an Author, or a Text, Leiden-New York-Köln. Mansfeld, J. (1998), Prolegomena Mathematica. From Apollonius of Perga to Late Neoplatonism, Leiden-New York-Köln. Mercier, C. (1978–1979), “L’Ecole hellénistique dans la littérature arménienne”, in: Revue des Études Arméniennes 13, 59–75. Minio Paluello, L. (ed.) (1949), Aristotelis Categoriae et liber de interpretatione, Oxford. Montanari, E. (1984 and 1988), La sezione linguistica del Peri hermeneias di Aristotele. 1, Il testo; 2, Il commento, Firenze. Mueller-Jourdan, P. (2007), Une initiation à la philosophie de l’Antiquité tardive. Les leçons du Pseudo-Elias, Fribourg. Muradyan, G. (1993), Gik‘ pitoyic‘ [Book of Chreias], Erevan. Muradyan, G. (2009), in: Calzolari/Barnes (eds.) 2009, 67–88. Muradyan, G. (2014), “The Hellenizing School”, in: Calzolari (ed.) 2014, 321–348. Muradyan, G. (2015), David the Invincible’s Commentary on Porphyry’s Isagoge, Leiden-Boston. Orengo, A. (ed.) (1996), Eznik di Kołb, Confutazione delle sette (Ełc Ałandoc'), introduzione, traduzione e note, Pisa. Papazian, M. B. (2009), “The Authorship of an Armenian Commentary on Aristotle’s Prior Analytics”, in: Calzolari/Barnes (eds.) 2009, 105–118 (First edition in: Journal of the Society for Armenian Studies 10 [1998–1999], 55–62). Patillon, M. (ed.) (1997), Aélius Théon, Progymnasmata, Paris. Petit, P. (1956), Les étudiants de Libanius, Paris. Rapava, M. (1982), “Traditions et innovations dans l’École néo-platonicienne d’Alexandrie (Ammonius Hermias et David l’Invincible)”, in: Bedi Kartlisa 40, 216–227. Richard, M. (1950), “’Απὸ φωνῆς”, in: Byzantion 20, 191–222 (= Opera Minora, 3, Turnhout-Leuven 1977) Saffrey, H.-D. (2007), “Retour sur le Parisinus Graecus 1807, le manuscrit A de Platon”, in: D’Ancona (ed.) 2007, 3–28. Saint-Martin, A.-J. (1818), Mémoires historiques et géographiques sur l’Arménie, 2 vols., Paris. Sanjian, A. K. (ed.) (1986), David Anhagt’, the ‘Invincible’ Philosopher, Atlanta (GA). Sevin, F. (dir.) (1739), Catalogus Codicum Manuscriptorum Bibliothecae Regiae, 1, Paris.


69

Sgarbi, R. (1972), “Osservazioni sul testo e sulla lingua della versione armena dell’Isagoge di Porfirio”, in: Memorie dell’Istituto Lombardo. Accademia di Scienze e Lettere. Classe di Lettere, Scienze morali e storiche 31/5, 365–477. Sgarbi, R. (1991), Studio contrastivo sull’adattamento strutturale armeno della Téchne dionisiana, Milano. Sgarbi, R. (2004), “Problemi lessicali legati alla terminologia della versione armena dell’Ars dionisiana”, in: V. Calzolari / A. Sirinian / B. L. Zekiyan (eds.), Bnagirk‘ yišatakac‘ / Documenta memoriae. Dall’Italia e dall’Armenia studi in onore di Gabriella Uluhogian, Bologna, 349–357. Shirinian [Širinjan], M. E. (2002), “Armenian Translation of [Aristotle’s] De Vitiis et Virtutibus”, in: Verbum: Aristotel i srednevekovaja metafizika, vypusk 6, Sankt-Peterburg, 166–171. Shirinian, M. E. (2009), “The Armenian Version of David the Invincible’s Commentary on Aristotle’s Categories”, in: Calzolari/Barnes (eds.) 2009, 89–102. Sorabji, R. R. (ed.) (1990), Aristotle Transformed. The Ancient Commentators and their Influence, Ithaca (NY). Stepanyan, A. (2009), “On the Basic Idea of the History of the Armenians by Moses Khorenatsi” in: Calzolari/Barnes (eds.) 2009, 181–196. Stone, M. E. / Shirinian, M. E. (eds.) (2000), Pseudo-Zeno, Anonymous Philosophical Treatise, Leiden-Boston-Köln. Sweeting, Clive (2009), “The Relationship between the Armenian Translation of the Commentary on Aristotle’s Analytics of David and the Greek Text of the Commentary on Aristotle’s Analytics of Elias”, in: Calzolari/Barnes (eds.) 2009, 137–150. T‘amrazyan, H. (2004), Grigor Narekac‘in ev Norplatonakanut‘yunǝ [Grigor Narekac‘i and the Neoplatonism], Erevan. Tardieu, M. (1986), “Sabiens coraniques et Sabiens de Harran”, in: Journal Asiatique 274, 1–44. Tarrant, H. (1998), “Introduction”, in: Olympiodorus, Commentary on Plato’s Gorgias, Leiden-Boston-Köln, 1–52. Terian, A. (1982), “The Hellenizing School. Its Time, Place and Scope of Activities Reconsidered”, in: N. G. Garsoïan / T. F. Mathews / R. W. Thomson (eds.), East of Byzantium. Syria and Armenia in the Formative Period, Washington (DC), 175–186 (= St. Nersess Theological Review 13 [2008]: Opera Selecta Teriana. A Scholarly Retrospective, 25–44). Terian, A. (2001), “Appendix: The Armenian Translation of Philo”, in: Fiaccadori (ed.) 2001, 425–448 (1st ed. in Constantin Zuckerman [ed.], A Repertory of Published Armenian Translations of Classical Texts, Jerusalem 1995, 36–44). Tessier, A. (1979), Il testo di Aristotele e le traduzioni armene, Padova. Tessier, A. (2001), “Frederick Cornwallis Conybeare e le versioni armene da Aristotele”, in: Fiaccadori (ed.) 2001, 133–138. Thomson, R. W. (1987), The Armenian Version of the Works attributed to Dionysius the Areopagite, Leuven. Thomson, R. W. (1995), A Bibliography of Classical Armenian Literature to 1500 AD (Corpus Christianorum), Turnhout. Thomson, R. W. (2007), “Supplement to A Bibliography of Classical Armenian Literature to 1500 AD: Publications 1993–2005”, in: Le Muséon 120, 163–223. Tinti, I. (2012), “On the Chronology and Attribution of the Old Armenian Timaeus: A Status Quaestionis and New Perspectives”, in: EVO. Egitto e Vicino Oriente 35, 219–282.

70

Valentina Calzolari

Topchyan, A. (2009), “Remarks on David the Invincible’s Commentary on Aristotle’s Prior Analytics, its Structure, Contents, Language, and the Problems of Translator and Authorship”, in: Calzolari/Barnes (eds.) 2009, 119–135. Topchyan, A. (ed.) (2010), Commentary on Aristotle’s Prior Analytics: Old Armenian Text with an English Translation, Introduction and Notes, Leiden-Boston. Traina, G. (1991), Il complesso di Trimalcione. Movsēs Xorenac‘i e le origini del pensiero storico armeno, Venezia. Vardanyan, S. (2008), S. Grigor Niwsac‘i, Tesut‘iwn i mardoyn kazmut‘iwn (Yałags kazmut‘ean mardoy) / St. Gregory of Nyssa (On the making of man), Erevan. Waitz, T. (ed.) (1844), Aristotelis Organon Graece, Lipsiae. Watts, E. J. (2006), City and School in Late Antique Athens and Alexandria, Berkeley-Los Angeles-London, 48–78. Wenrich, J. G. (1842), De auctorum Graecorum versionibus et commentariis Syriacis, Arabicis, Armeniacis, Persicisque commentatio, Leipzig. Westerink, L. G. (1961), “Elias on the Prior Analytics”, in: Mnemosyne 14, 126–139 (= Texts and Studies in Neoplatonism and Byzantine Literature, Amsterdam 1980, 59–72). Westerink, L. G. (1990), Prolégomènes à la philosophie de Platon, Paris. Wildberg, C. (1990), “Three Neoplatonic Introductions to Philosophy. Ammonius, David and Elias”, in: Hermathena 149, 33–51. Zekiyan, B. L. (1997), “Quelques observations critiques sur le Corpus Elisaeanum”, in: R. F. Taft (ed.), The Armenian Christian Tradition, Roma, 71–123. Zuckerman, C. (2001), A Repertory of Published Armenian Translations of Classical Texts, in: Fiaccadori (ed.) 2001, 425–448.

II. History and Historiography

Gianfranco Gaggero

Armenians in Xenophon Abstract: Herodotus and Ctesias, the first Greek authors who mentioned Armenia, seldom quoted the peoples of that region, particularly the individual inhabitants. Different, although curious, is the case of Xenophon: even though Armenia is one of the well-known backgrounds of the renowned retreat described in the Anabasis, the Armenian people remain in the shadows, with the mention only of two or three satraps who, however, were ethnically Persians. By contrast, in the Cyropaedia the author uses the Armenians in order to describe a special, but fictitious, relationship between the latter and the emergent Achaemenid empire; additionally, within the Armenian nation, there stands out a figure, likewise fictitious, portrayed as their king Tigran, one of the leading allies of the Xenophontean Cyrus, and overall a major personality of the whole work.

As is known, the earliest information on Armenia in the classical sources dates back to Herodotus, who mentions the region as the thirteenth among the Achaemenid satrapies of the time of Darius and includes the indication that the tribute paid by this region was 400 talents. So, through Herodotus’ account, a new place was assured in the annals of history for a land that until not long before had been the home of the Urartian people, even if – it should be mentioned en passant – whether there was any relationship between that ancient population and the Indo-European new arrivals is still a debated question¹. Herodotus also cites Armenian contingents among Xerxes’ troops in 480 BC, “armed like the Phrygians, of whom they were descendants”, and led by the Persian Artochmes, a son-in-law of Darius². After Herodotus, the next reference to Armenia is found in Ctesias of Cnidus, who lived a few decades later than the historian of Halicarnassus and could potentially have been a precious source on account of his prolonged period at the Persian court as the royal physician. Nevertheless, in terms of the information he

1 For Herodotus see 3.93.1. With regard to the previous period of Urartu, and the possible relations between the ancient population of that land, related to the Hurrites, and the Indo-European newcomers, see the still classical Piotrovskij 1966, and then Zimansky 1985; Haas 1986; Salvini 1995, and above all Diakonoff 1984. 2 See Hdt. 7.73. Additional Herodotean details on the physical and human geography of Armenia, although providing information of lesser interest, are found in 1.72, 1.180.1, 1.194.2 and 1.194.4–5; 5.49.6, 5.52.3–5.

74

Gianfranco Gaggero

provides, he proves to be rather disappointing. Although he devotes a number of pages both to Cyrus the Great and to the revolt of Cyrus the Younger against Artaxerxes II, thus describing the same historical background as Xenophon’s two “oriental” works, his fragments – with a single exception, which bears little importance, as we will see shortly – feature no ethnically defined Armenian. Furthermore, the cases that do contain a reference to Armenia are irrelevant for our purposes and amount to no more than two or three. In one instance, Armenia is portrayed as the land of exile for Artoxares, a Paphlagonian eunuch, who had been banished for making a statement in favour of Xerxes’ son-in-law Megabyzos, after which Artoxares himself had moved away from the region when, having become King Darius II, he joined the sovereign³. Elsewhere two fragments, of uncertain classification⁴, speak of a spring from which there emerged highly poisonous black fishes. Thus, the only Armenian cited by name proves to be a certain Barzanes, who appears in Diodorus Siculus’ account of the conquests achieved by the legendary Assyrian sovereign Ninus: on the occasion of the brutal campaign waged by the latter against Armenia, Barzanes was – Diodorus claims – generously spared after surrendering⁵. Xenophon is therefore the only author who, in the early decades of the fourth century BC, displays more than a mere passing interest in the land of Armenia and its inhabitants. It is known, above all, that Armenia is frequently mentioned among the countries through which he himself travelled, together with the Greek mercenaries who had survived after Cunaxa and were following a route to reach the southern coast of the Pontus Euxinus. The Xenophontean citations have been analyzed in depth by modern criticism, together with those pertaining to the other regions of eastern Anatolia, with the aim of reconstructing the

3 See respectively F 14 = Phot. Bibl. 72.41a, par. 43 and p. 133 Lenfant; F 15 = Phot. Bibl. 72.42a, par. 50 and p. 137 Lenfant. 4 See F 61a from Antigonus of Carystus (mirab. 165) and F 61b from Pliny the Elder (NH 31.19.25): both on pp. 216 and 217 Lenfant. 5 This piece of information, obviously devoid of any historical truth, and which in any event concerns a character bearing a Persian name (of the same type as the better known Ariobarzanes and Satibarzanes, as appropriately underlined by Lenfant 2004, 23 n. 89), can be found in F 1b = Diod. Sic. 2.1.8, p. 23 Lenfant. An additional Diodorean reference to Armenia (2.11.4), albeit apparently not deriving from Ctesias, relates that Semiramis, the widow and successor of Ninus, dragged an enormous stone from the mountains of Armenia in order to use it for erecting an obelisk, one of the seven wonders of the world. On Semiramis and the Armenian Ara, whom she loved but fought against – an element that possibly recalls ancient Assyrian-Urartian relations – see however the testimony of Moses of Khoren (1.15 and partly also the subsequent chapters 16 and 18), the appraisal by Russell 1984, 477–485 and Traina 1995, 287–288; 297–302, and my considerations in Gaggero 1997, 18–21.

Armenians in Xenophon

75

details of the itinerary of the retreating troops. However, the critical appraisals have not always yielded uniform results; furthermore, since the testimony of the Anabasis is predominantly of a geographic-topographic nature and is somewhat scanty apart from occasional episodes, it is of limited interest for us. Among the characters appearing in that work, a few – not explicitly identified as Armenians – are described as having military and political-administrative functions in the region. Two of these men, referred to by the name of Orontas, were relatives of Artaxerxes II; one of them is indicated as the commander of the citadel of Sardis, the other as the satrap of eastern Armenia. But the most significant figure is that of Tiribazos, satrap of western Armenia and organizer of ambushes against the Greek fugitives, and later notorious as satrap of Lydia at the time of the war of Corinth⁶. Here, in any case, it will be more illuminating to focus in particular on the picture emerging in the Cyropaedia, where, in contrast to the information given in the author’s other text, the analysis of Armenians at the time of Cyrus the Great seems, at first sight, to be carried out in decidedly greater depth, at least with regard to their political organization and their condition – now as enemies, now as allies – of the rising Persian empire. But first impressions are deceptive. In actual fact the testimony of the Cyropaedia, fairly extensive though it is, and accompanied by a considerable wealth of details, is hardly of any greater use for a proper historical investigation. Moreover, like most of the rest of the Xenophontean writings, it presents a quantity of incongruent elements and deliberate distortions, or even veritable falsifications, which have made Xenophon’s work into a most singular form of documentation⁷.

6 For the first Orontas see Xen. An. 1.6.1–11; for the second Xen. An. 3.5.17, 4.3.4 and 2.4.8–9, 2.5.40, 3.4.13 (whereas neither of these seems to correspond to the Orondes condemned to death after being unjustly accused by the eunuch Satibarzanes of committing adultery with Parysatis, the mother of Artaxerxes II and Cyrus the Younger, cited by Ctesias in F 16 = Phot. Bibl. 72.43b, par. 60 and p. 142 Lenfant; on Satibarzanes cf. Lenfant 2004, 276 n. 644). On Tiribazos (also a satrap in Plut. Artax. 7.3) see Xen. An. 4.4.4 and then 4.4.18, 4.4.21, 7.8.25. On all these passages see also the still excellent comment by Lendle 1995, above all 50–52 (first Orontas); 108–110; 209 (second Orontas); 225–238, but also passim 230–268 (Tiribazos). Various citations of Armenia can also be found in Xen. An. 3.5.17, 4.1.3, 4.3.1, 4.3.3, 4.3.20, 4.4.1, 4.5.33–34. 7 In recent decades the Cyropaedia has aroused great interest among scholars, and has been the subject of a number of valuable works. Cf., among others, Due 1989; Tatum 1989; Mueller-Goldingen 1995; Nadon 2001; Gray 2011, and most recently the important contribution by Tuplin 2013, 67–90. There is also a well written and carefully balanced Introduction by Marcel Bizos to the Xenophontean work in the Belles Lettres edition (Bizos 1971, IV-LIV). In particular, considerable attention to a comparison between Xenophon and his predecessors Herodotus and Ctesias, and to the figure of Cyrus the Great in the three different authors, is devoted by Due 1989, 117–139 and Mueller-Goldingen 1995, 1–24.

76

Gianfranco Gaggero

To be sure, we may momentarily set aside the general problem of the meaning of the book, and overlook such aspects as the difficulty of making use of the fanciful – or indeed, far-fetched – data that are abundantly scattered throughout the text and frustrate a comparison with other sources. Similarly, we may disregard the evident apriorism displayed by the attempt to enrich the story by spinning yarns or tall tales depicting an ideal sovereign, of unblemished virtue, superior human talents and confident political and military skills, or rather the marked divergence between the way the Armenians are portrayed and the treatment reserved to the other peoples, whether friends or foes of the Achaemenid chief. For present purposes we will restrict our attention to the Armenians only. After a period during which they were openly hostile to the Persians, they are indicated as new and loyal allies of the latter when the aged former sovereign, facing severe pressure from the victorious Cyrus, who had already captured the king’s offspring and wife, is saved from disaster by his eldest son Tigran. From this point on, Tigran becomes one of the Iranian leader’s closest and most faithful associates, even engaging in a sort of “Socratic” philosophical dialogue with Cyrus, to which we will return below. Tigran reappears a number of times in the text, last but not least as a happy bridegroom and thus as the continuator of the dynastic line of his country⁸. The Xenophontean narration on the Armenians has been often examined by modern scholars, not so much as a work in its own right, but rather in relation to similar excursuses on peoples with whom Cyrus is said to have come into contact. The overwhelming idea is that it contains very little genuine historical material, or indeed none at all⁹. This is in contrast to the opinion put forward by some critics, above all in a less recent period, who contended that Xenophon did effectively draw on some factual information, and that at least the figure of Tigran has some basis in historical reality. But in actual fact this seems quite unlikely. The way in which the Armenian episode is organized and described differs little from other similar episodes. A case in point is the preposterous invention of the figure of Cyaxares, portrayed not – as he really was – the father of Astyages, but as the son of the latter, and therefore as Cyrus’ uncle, a figure whose very existence is totally unknown to the other sources, first and foremost to Herodotus. In the Xenophontean version, on the other hand, for a long period of time Cyaxares

8 On Tigran’s father, see below. The numerous passages that mention Tigran are the following: 3.1.7, 3.1.14–15, 3.1.20, 3.1.23, 3.1.27–28, 3.1.36–41, 3.2.1–3, 3.2.11, 3.3.5, 4.2.9, 4.2.18, 4.2.43, 4.5.4, 4.5.35, 5.1.27, 5.3.42, 6.1.21, 8.3.25, 8.4.1, 8.4.24. 9 Cf. in this regard at least Due 1989, 76–79; Tatum 1989, 134–145; Mueller-Goldingen 1995, 150– 164; Nadon 2001, 77–86. A connection that has sometimes been suggested, but which seems decidedly implausible, is that with the brief passage of Strab. 11.13.5.


77

plays the role of a sort of negative counter-melody, whose prestige and ability both in the political arena and in battle are in any case inferior to the skills attributed to his nephew Cyrus, whom Cyaxares preceded on the throne. Cyrus himself was to rise rapidly to power and, even before he formally replaced Cyaxares, he became the real leader at first of Media and then of Persia¹⁰. Let us now take a closer look at the figures mentioned above. While the anonymity of Tigran’s bride and of his other relatives can in some sense be regarded as plausible and need be no cause for suspicion¹¹, the situation is rather different in the case of Tigran’s father, who is regularly and without exception designated as “king of Armenia”¹². Now, this form of, shall we say, systematic anonymity is attributable not so much to the difficulty of ascertaining the precise physiognomy of a man who may have been difficult to identify, amidst a scanty or practically absent documentation, as, rather, to a constant habit of Xenophon’s, the underlying rationale of which remains unclear. For instance, Xenophon reserves indeed the same treatment to the “king of Assyria”, one of Cyrus’ greatest enemies¹³, and to a few minor cases, among which one notes an unlikely “king of India”¹⁴. In short, the “king of Armenia” is not a person in the flesh, brought on stage with the aim of reconstructing some story or describing some specific trait, but, instead, an almost abstract figure, whose significance is entirely contained within the literary and historiographic function of being the precursor of a vastly superior prince.

10 On this fictitious character, who is to all extents and purposes a second protagonist of the work, cf. Tatum 1989, 115–133; Nadon 2001, 87–100. 11 However, what is curious is the way Tigran’s wife, indicated simply as the monarch’s consort in 3.1.37, 3.1.43, and 8.4.24, where she receives a gift from Cyrus, is defined very rudely by Tigran himself (in 3.1.41) with the term “Armenian woman”. In addition to the bride of Tigran there are in the Cyropaedia, equally unidentified, also the wife and the daughters of the “king of Armenia” (3.1.2). 12 See the relevant passages in 2.4.12, 2.4.18, 2.4.22, 2.4.31, 3.1.1 (but more generally see all the paragraphs 1–43); 3.2.14–24, 3.2.28–31, 3.3.1–4. 13 In actual fact there are two “kings of Assyria” in the Cyropaedia, father and son, both rigorously anonymous, but only the son is portrayed as evil (see 1.4.16–24, 4.6.2–5, 5.2,27–28, and more generally throughout the whole of the fifth book; 7.5.30–32), whereas his father is a basically positive figure (see 1.5.2, 2.1.5, 2.4.7–8, 3.3.43, 3.3.50, 4.1.8, 4.6.2–7, 4.5.40, 5.4.12). But it is worth noting that at the time of Cyrus, or even earlier, at the time of Astyages, there no longer existed any Assyrian kingdom, as this realm had been overthrown in 612 BC by the coalition between the Neo-Babylonians of Nabopolassar and the Medes of the real, historical Cyaxares, the destroyer of Nineveh. 14 On this king see 2.4.1–9, 3.2.25–30, 6.2.1–3, 6.2.9–11.

78

Gianfranco Gaggero

The fact is, however, that Tigran likewise lives in an abstract dimension. Although in this case he is characterized by a clearly distinguishable name¹⁵ and is effectively among the most important figures of the entire work, he too remains a shadowy entity, entirely constructed by Xenophon. Far from finding in him any precise correspondence with a known prince, a number of critics have seen him as an alter ego of Xenophon himself, at least in the episode of the dialogue between Cyrus and Tigran, where it appears a “Sophist”, said to be a friend and teacher of Tigran himself, and who, having been unjustly condemned to death at an earlier time by Tigran’s father, is depicted as the protagonist of a noble, “philosophical” death. Such a context actually seems to evoke the image of the real Xenophon, who, engaged in discussions with the “Sophist” (a term used in this case with non negative implications, and obviously concealing the historical Socrates), deals with philosophical and ethical problems of a recent past and brings memories of a reality which, though totally foreign to the condition of Armenia and Persia of the time, was still painfully and deeply felt by the writer¹⁶. Apart from this, Tigran is a man who has only one string to his fiddle, and who is no less uninspiring than the equally monotonous and artificially constructed Xenophontean Cyrus, albeit more cynical and amoral than the Achaemenid. However, the impression is conveyed that Tigran’s association with Cyrus is of long-standing date, with a bond of reciprocal and well-established friendship that goes back to their childhood and has endured over time; it is depicted as a relationship that also involves an implicit agreement of close military alliance should the need arise, as was the case for example against the Neo-Babylonians, and in numerous other circumstances. Faced with this insubstantiality of the historical reality of the two Armenian princes, it is, of course, difficult to determine what conditions prevailed in Armenia at the time of Cyrus and, earlier, during the Median predominance. One has to look back to the era characterized by the ascendancy of the first, genuine Cyaxares – the Median sovereign who had so efficiently contributed to the collapse of the Neo-Assyrian power – to find any last trace of the Urartian presence,

15 Rather, to be precise, Tigran is the only “relevant” name of an Armenian that Xenophon explicitly mentions. Besides him, but much more important, and quoted only once, it is recorded indeed his younger brother, Sabaris, captured by Cyrus together with their father and relatives. In many other cases, Armenia and the Armenians are cited generically (see for ex. 2.1.6, 2.4.16, 2.4.21, 2.4.30, 3.2.4–17, 5.4.13). 16 The “Sophist” appears in 3.1.14 and then in 3.1.38–40; cf. the commentary by Bizos in the Introduction to the entire Cyropaedia (Bizos 1971, XLII-XLIII), where emphasis is placed on the marked difference between Cyrus’ superior and humane wisdom and Tigran’s empty sophisms. Cf. also Due 1989, 40–41.


79

which was swept away by the advance of the Medes towards central Anatolia. But with regard to the period encompassing the middle of the following century, i.e. the time when Cyrus embarked on his amazing adventure, nothing is known about the ethnical and political setup of Armenia. This region, lying to the north of the new and decisive advance of the Iranians towards Lydia and the whole of the western area of Anatolia that was shortly conquered, must surely have been taken into consideration by the Persian authorities. However, we can do no more than to speculate as to whether any move by the Persians would have been designed merely to forestall unrest in a hostile environment that was still churning with primeval forces, or was already intended to garner the support of a few contingents among the Indo-European newcomers in order to enter into a more lasting relation of cooperation with some local prince or other. None of these hypotheses can be excluded a priori, although it seems unlikely that at the time in question Armenia had anything resembling a structured and well developed political form, or even a state worthy of this name¹⁷. To all this should be added the tradition of the Orontid dynasty, which long supplied the Achaemenids with satraps and survived beyond their era¹⁸. The above mentioned Orontids, and the others who came in their wake, clearly cannot be included in the present investigation, partly for chronological reasons, but also because the two satraps who were contemporaries of Xenophon’s are explicitly stated to be Persian. Instead, it is far less outlandish, in the absence of Greek evidence, to draw on other sources, namely to Moses of Khoren, an author known as an invaluable source of information, but at the same time highly controversial and disputed on a number of points¹⁹. As far as the characters of the Cyropaedia are concerned, Moses certainly owes a strong debt to this work with regard to Tigran, portrayed as a brotherly friend of Cyrus and a faithful follower of the latter, just as in Xenophon. However, Moses draws on other material in casting the Armenian prince in the role of the adversary and, in the end, of the material victor who defeated Astyages, Cyrus’ grandfather, whereas in Xenophon – as

17 With regard to Armenia in this first period, cf. the different considerations put forward by Diakonoff 1984, 101–130 and Chahin 1987, 107–115; 209–210. See also the short but interesting observations by Chaumont 1982, 86–89 on the possible presence, in the passages of the Cyropaedia that deal with Armenia, of echoes of the last wars of the Medes against the Urartians. For a recent reconsideration of the “Median” period, see Rollinger 2010. 18 On the Orontids, and their complex evolution from probable Persian provincial governors to an autonomous and increasingly independent dynasty, cf. Toumanoff 1959, 1–36; Id. 1960, 73–106; Chahin 1987, 211–220; Traina, forthcoming. 19 See MX 1.24–31. For an overall evaluation of Moses and his testimony, cf. also the rich and complex observations by Traina 1991.

80

Gianfranco Gaggero

opposed to the mainstream tradition – Astyages by no means adopted a hostile attitude towards his grandson. Another aspect in which Moses diverges from the Greek historian consists in having massively integrated the Xenophontean story with a vast array of major and minor characters and with a mass of details whose truth cannot now possibly be tested (to cite just one example, Tigran’s act of bestowing his sister Tigranuhi on Astyages, as a bride). Among all this there also appears the name of Tigran’s father, Ervand Sakawakeac‘, whose surname, translated as “Short-life”, probably points to the fact that he reigned for only a few years²⁰. Is he the one we met earlier, designated as the “king of Armenia”, or is Moses making use of a native tradition which, as far as this detail was concerned, had nothing at all to do with the story narrated by Xenophon? This would seem to be by far the most logical hypothesis; on the contrary, one may additionally note in this context that it appears plausible to detect in the Armenian form Ervand (which, in turn, is of Iranian origin)²¹ a version of the Greek form Orontes, thereby justifying the later fortune of this name. A final problem concerns specifically the name of Tigran himself. If Moses of Khoren derived this name from Xenophon, where did Xenophon himself find the name? Although it is often said that the name appears for the first time in the literary works of the Athenian writer²², there is, however, a previous Tigran in Herodotus. On three or four occasions in his work, the historian from Halicarnassus, who – as noted earlier – calls attention to the presence of Armenians among Xerxes’ auxiliaries, also mentions a Tigran described as a commander of Median troops and designated as Achaemenid²³. Thus one can rule out the suggestion that the Tigran mentioned in Herodotus could be Armenian, and in any case the etymology of the name seems to be Iranian²⁴. And even if one allows the possibility that some other ancient testimony, now lost (Ctesias possibly, now

20 On Ervand see MX 1.22, 1.24. On the chronological confusion between these figures and the first century BC characters bearing the same name, cf. the commentary by Robert Thomson on the translation of that passage from Moses (Thomson 1978, 114 n. 8 and 178 ff.). 21 Cf. Bartholomae 1904, s.v. aurvant-, col. 201. 22 Xenophon quotes in his Hellenica (4.8.21) another Tigran, who is however a Persian one, namely the son-in-law of the satrap of Ionia Strouthas, taken prisoner in 396 BC by the Spartan Diphridas and then released with a ransom. One may remember, at any rate, that the dating of both the Hellenica and the Cyropaedia is uncertain, and then it is not surmountable the problem of the priority of one or the other work. 23 See Hdt. 7.62.1, 9.96.2, 9.102.4, and perhaps also 8.26.2, according to an alternative reading of the text, in which what is read is the name of a Tritantaichmes, described as a son of an Artabanos. 24 Cf. the related entries (see especially the name of the Scythians Tigrakhauda = “with the pointed caps”) in Bartholomae 1904, s.v. tigra- (and tigra-xauda-), col. 651.


81

fragmentary?), may have included a different Tigran, more significant as compared to the Herodotean character, the latter seems for the moment to be the only figure from whom Xenophon could have taken the name of “his” Tigran. That Herodotus’ Tigran was not even Median, but actually Achaemenid, is hardly likely to have been a circumstance that would have deterred the younger historian from including the Armenian prince and later king in a tale full of names construable as Median or pseudo-Median, Persian or pseudo-Persian, i.e. names for all seasons, assuredly also functional for the purpose of impersonating men who were officially foreigners. In short, Xenophon’s reutilization of a non-Armenian name for the figure who would thereafter personify, in the eyes of classical culture, the noblest and most praiseworthy exemplar of the Armenian stock was responsible for the most amazing fortune of the name, not only in a native tradition that eventually culminated, disproportionately expanded, in the work of Moses of Khoren, but also in the dynastic customs of the sovereigns of the future independent Armenia: as a matter of fact, right up to Tigran the Great and some of his successors, the Armenian kings repeatedly used an appellation that was as foreign in its etymology as it was in the reality of the princes who materially bore that name.

Bibliography Bartholomae, C. (1904), Altiranisches Wörterbuch, Strassburg. Bizos, M. (ed.) (1971), Xénophon, Cyropédie, 1, Paris. Chahin, M. (1987), The Kingdom of Armenia, London. Chaumont, M.-L. (1982), “Tentations de l’Iran et du monde gréco-romain (Ve siècle av. J. C. – vers 300 ap. J. C.)”, in: G. Dédéyan (ed.), Histoire des Arméniens, Toulouse, 73–114. Diakonoff [D’jakonov], I. M. (1984), The Pre-History of the Armenian People, transl. by L. Jennings, with revisions by the author, Delmar (N. Y.). Due, B. (1989), The Cyropaedia. Xenophon’s Aims and Methods, Aarhus. Gaggero, G. (1997), “Conquistatori ai confini del mondo. Le imprese di Sesostri, Semiramide, Tearco e Nabucodonosor tra realtà storica e deformazione leggendaria”, in: Columbeis 6, 7–37. Gray, V. J. (2011), Xenophon’s Mirror of Princes: Reading the Reflections, Oxford. Haas, V. (ed.) (1986), Das Reich Urartu. Ein altorientalischer Staat im 1. Jahrtausend v. Chr., Konstanz. Lendle, O. (1995), Kommentar zu Xenophons Anabasis (Bücher 1–7), Darmstadt. Lenfant, D. (ed.) (2004), Ctésias de Cnide, La Perse. L’Inde. Autres fragments, Paris. Mueller-Goldingen, C. (1995), Untersuchungen zu Xenophons Kyropädie, Stuttgart-Leipzig. Nadon, C. (2001), Xenophon’s Prince. Republic and Empire in the Cyropaedia, Berkeley. Piotrovskij, B. B. (1966), Il regno di Van Urartu, transl. by M. Salvini, Roma.

82

Gianfranco Gaggero

Rollinger, R. (2010), “Das medische Königtum und die medische Suprematie im sechsten Jahrhundert v. Chr.”, in: G. Lanfranchi / R. Rollinger (eds.), Concepts of Kingship in Antiquity. Proceedings of the European Science Foundation exploratory workshop held in Padova, November 28-December 1, 2007, Padova, 63–85. Russell, J. (1984), “The Platonic Myth of Er, Armenian Ara, and Iranian Ardāy Wīrāz”, in: Revue des Études Arméniennes n.s. 18, 477–485. Salvini, M. (1995), Geschichte und Kultur der Urartäer, Darmstadt. Tatum, J. (1989), Xenophon’s Imperial Fiction. On the Education of Cyrus, Princeton. Thomson, R. W. (ed.) (1978), Moses Khorenats‘i, History of the Armenians, Cambridge (MA)-London. Toumanoff, C. (1959–1960), “A Note on the Orontids”, in: Le Muséon 72, 1–36 and 73, 73–106. Traina, G. (1991), Il complesso di Trimalcione. Movsēs Xorenac‘i e le origini del pensiero storico armeno, Venezia. Traina, G. (1995), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. I”, in: Le Muséon 108, 279–333. Traina, G. (2001), “Note sulla montagna nell’Armenia antica”, in S. Giorcelli Bersani (ed.), Gli antichi e la montagna: ecologia, religione, economia e politica del territorio / Les anciens et la montagne: écologie, religion, économie et aménagement du territoire. Atti del Convegno, Aosta, 21–23 settembre 1999, Torino, 213–221. Traina, G. (forthcoming), “L'Arménie entre les Orontides et les Artaxiades”, in: Ch. Feyel / L. Graslin-Thomè (eds.), Antiochos III et l’Orient, Nancy. Tuplin, C. (2013), “Xenophon’s Cyropaedia: Fictive History, Political Analysis, and Thinking with Iranian Kings”, in: L. Mitchell / C. Melville (eds.), Every Inch a King: Comparative Studies on Kings and Kingship in Ancient and Medieval Worlds, Leiden-Boston, 67–90. Zimansky, P. (1985), Ecology and Empire. The Structure of the Urartian State, Chicago.

Francesca Gazzano

Croesus’ Story in the History of Armenia of Movsēs Xorenac‘i Abstract: In a brief and puzzling section (2.12–13) of his monumental History of Armenia, Moses of Khoren assigns to the Armenian king Artašēs I many military deeds, which include the capture of the Lydian king Croesus and the conquest of the Lydian kingdom. Moreover, this peculiar tradition, with its internal variants, is followed by Moses against the authority of Eusebius’ Chronicon, and is attributed to four, otherwise unknown, “Greek historians”. This paper aims to investigate these interesting chapters from a historical and a historiographical point of view, on the one hand because they preserve a tradition which is completely different both from the Greek (mostly Herodotean) and Persian vulgata about the fate of Croesus, and on the other because Moses’ text is interwoven with echoes, phrases and episodes taken from the classical historiographical models.

The second book of the History of Armenia (Patmut‘iwn Hayoc‘)¹ by Moses of Khoren (Movsēs Xorenac‘i, henceforth MX) includes a dossier, no less remarkable than it is laconic, concerning the ancient sovereign Croesus of Lydia and the historical background of his life and deeds². The aim of this paper is to offer a detailed reading of the text of the dossier, in a framework that is first and foremost Greek, and from an essentially historical viewpoint. Naturally, it cannot be overlooked that work on this author raises a host of problems, many of which have given rise to discussion and even controversy, with regard both to his date and his reliability³. These are certainly thorny problems, but they will not

1 For the text, the Abełean/Yarut‘iwnean 1981 edition is used, as well as Thomson 1978’s English translation and the French translation by Mahé/Mahé 1993. I owe the Italian translation, with critical notes on the chapters examined in this paper, to the willing assistance of Giusto Traina, to whom I wish to express my thanks: for having devoted time and patience to discussing with me numerous aspects of the text and of my paper, I owe him a debt of gratitude beyond words. I follow here his transcription of personal names. Also, I am grateful to Maddalena Luisa Zunino for her careful reading of the original Italian version. 2 MX 2.11–13. 3 Both the chronology and the reliability of MX are still the object of lively debate, and inevitably the overall assessment of the historian is rather negative among those who propose a fairly late date (eighth–ninth centuries AD), as opposed to the “traditional” dating (fifth century AD). Those who prefer the late dating regard him as unreliable, untruthful and a forger, see e.g. Car-

84

Francesca Gazzano

affect the present investigation, as MX’s documentation on Croesus is rich in thought-provoking reflections, from several different perspectives. For instance, among the witnesses of what has come down to us of the literary tradition on the downfall of Croesus, the Armenian historian is one of the very few to present a viewpoint that is completely heterodox in comparison to the far more widespread Herodotean vulgata⁴. Furthermore, the picture conveyed in MX has no link to the stories related by other authors such as Ctesias, Xenophon, Diodorus and Nicolaus of Damascus⁵, who follow Herodotus, but diverge from his account in some respects. And on closer inspection, certain peculiarities can also be noted in the information MX gives on Croesus, reported with great insistence – for reasons that are by no means clear – and with a decidedly non mainstream approach. What is more, the overall corpus of information provided in the chapters on the king of Armenia Artašēs I is built up through a dense and almost uninterrupted web of cross-references and revisitations of passages from Greek writers, citing pieces taken from a variety of sources which, however, actually refer to several different Achaemenid sovereigns. The present paper was prompted by the conjecture that identification of MX’s remote “originals” may contribute to a better understanding of the general extent, if not the direct medium, of his debt to the Greek tradition relating in particular to the most ancient history of the kingdom of Armenia, and may also help to clarify the reasons for his insistence in attributing precisely to Artašēs I the conquest of Lydia and the capture of Croesus.

rière 1893–1894; Thomson 1978 (whose judgment has exerted considerable influence even outside of the sphere of Armenian studies); more recently, Garsoïan 2003–2004. This “hypercritical” line of interpretation is rejected by scholars who on numerous occasions have underlined the role and responsibility of the sources utilized by MX and have more positively reassessed his historical value: see in partic. Traina 1991; Id. 1995; Id. 1998; for a status quaestionis, see Topchyan 2006 and now Traina 2015, 154–157. 4 Hdt. 1.86–92. On the reception of the Herodotean story, cf. Talamo 1985, 150–161; Gazzano 2012, with previous bibliography. 5 Ctes. FGrHist 688 F 9 § 5 (= Phot. Bibl. 72.36b Henry; a similar version preserves Just. Epit. 1.7.9); Xen. Cyr. 7.2.2–29; Diod. Sic. 9.2 (= F 4 Cohen-Skalli = Const. Porph. Exc. de sent. 36 Boissevain); 9.27 (= F 38 Cohen-Skalli = Const. Porph. Exc. de sent. 56 Boissevain); 9.31 (= FF 41–45 CohenSkalli = Const. Porph. Exc. de sent. 59–62 Boissevain); 9.33 (= F 50–51 Cohen-Skalli = Const. Porph. Exc. de sent. 66 Boissevain [F 50]; Const. Porph. Exc. de virt. 58.220 Büttner/Wobst); Nic. Dam. FGrHist 90 F 68 (= Const. Porph. Exc. de virt. 345 Büttner/Wobst); see infra, 87–88.

Croesus’ Story in the History of Armenia of Movsēs Xorenac‘i

85

1 MX and the fate of Croesus Within his impressive work, which reconstructs the history of the Armenian people in three fairly lengthy books, MX devotes three chapters (11–13) of the book II to a profile of the figure of Artašēs I⁶, the founder of the dynasty of the Artaxiads and of an independent kingdom of Armenia. The first of the chapters in question (11), which we will not examine in detail in this paper, concerns the ascent of Artašēs I to the throne and the internal organisation of the kingdom. Here, on the other hand, we will focus on his conquests, surveying the res gestae the historian assigns to this sovereign, whose reign extended from roughly 190 BC to 160 BC and who, it should be recalled, in all probability corresponds to the Artaxias I mentioned by Strabo⁷. As narrated by MX in chapter 12, the exploits of Artašēs I were memorable for their enormous – indeed epoch-making – impact: Artašēs led a gigantic expedition to Asia Minor, he brought to Armenia statues of gods and heroes purloined from temples in Asia and Greece, he even captured and took prisoner Croesus the king of Lydia. But while he was planning to subjugate the entire West, taking advantage of the fact that Rome had by then been weakened by tumultuous revolts, he was unexpectedly killed by his own soldiers, after a twenty-five year reign, on account of a conflict within the army⁸. As can be gathered even from a first glance, the story is not credible, either in its substance or in the details. And at least on one point, the capture of Croesus, MX appears fully aware of the existence of completely different traditions, which he outlines, only to reject them, in the next chapter (13). In the latter, he adds no other events to the bios of Artašēs I, but devotes himself entirely and with determination to confirming the reliability of the tradition of Artašēs I’s relations with the king of Lydia, basing his assertions on the citation, quoted verbatim, of four presumed “historiographers of the Greeks” who had addressed the issue. This notwithstanding, the fragments also provide additional information on the greatness and deeds of king Artašēs, whom the cited authors judged to have been superior to Alexander the Great and all the Achaemenid kings⁹. Naturally, it is hardly necessary to enquire into the historical reliability of the events related by MX (and of the Greek sources he quotes) concerning Artašēs I

6 Artašēs I had already been mentioned, but en passant, by MX in 1.12 e 2.8. See also Traina forthcoming. 7 Strab. 11.14.15: on Artaxias I, see Chaumont 1986, 418–438, esp. 419–420; Garsoïan 1997, 47–52; Traina 1999–2000. 8 See the text in the Appendix, no. 1. 9 See the text in the Appendix, no. 2. On these chapters see also Sargsyan 2012–2013, and the contribution by F. Mari, in this volume.

86

Francesca Gazzano

and his achievements¹⁰, and in particular the capture of Croesus, since there can be no doubt whatsoever that this event took place in the sixth century BC, more precisely in 547/6 BC¹¹. It definitely involved a genuine protagonist, known to have emerged victorious from the event in question, namely the “founder” of the Persian empire, Cyrus II “the Great”¹². MX himself takes note of this chronology, albeit without wholeheartedly endorsing it, as he explicitly states at the opening of Chapter 13: “we heard from some histories that Cyrus (Kiwros) had killed Croesus (Kriwsos) and had destroyed the Lydian kingdom”¹³. Here the wording of the statement, as well as the detail that Cyrus killed Croesus¹⁴, indicate that MX’s source for this tradition, which is chronologically more appropriate, can without a doubt be identified as Eusebius, in the Armenian version of the Chronicon, which makes use of exactly the same expression¹⁵. Although MX actually prefers not to follow Eusebius, the testimony in question is certainly of considerable significance, especially for scholars of archaic history, as these are the only two witnesses, both in Armenian, of a variant of the story according to which king Croesus did not survive the Persian conquest¹⁶. It is worth going into this point in slightly greater detail, as the real fate of Croesus after the fall of Sardis was already disputed in antiquity, and continues to

10 See Sargsyan, 2012–2013, with status quaestionis. 11 The chronology of the downfall of Croesus and the capture of Sardis by Cyrus is based in primis on the testimony of the so-called Nabonidus Chronicle (2.16), and despite the doubts expressed by some scholars (e.g. Cargill 1977; Burkert 2004, 46–47; Rollinger 2008), the most recent re-reading of the document seems to confirm the dating of the event to 547/6 BC: see Van der Spek 2014, 256 n. 184. 12 For the events, see above all Hdt. 1.71–94, with the commentary by Asheri 19974. On Cyrus, see not only Briant 2002; Kuhrt 2007, but now also Daryaee 2013; Van der Spek 2014; cf. also Mari, in this volume. 13 Transl. by Thomson. Cf. the translations by Mahé/Mahé 1993 (“nous avions eu vent par certaines histoires que Cyrus aurait tué Crésus et aboli le royaume de Lydie”) and by G. Traina (“abbiamo appreso da certi storici che Cyrus ha ucciso Croesus e gli ha tolto il regno dei Lidi”). 14 MX 2.13: Kiwrosi spaneal zKriwsosi ew barjeal zt‘agaworut‘iunn Liwdac‘woc‘. 15 Eus. Chron. 1.103 Aucher: zKrisos spaneal Kiwrosi, ebarj zlidac‘woc‘ išxanuin – Croesum Cyrus perimens, Lydorum imperium sustulit. 16 We imply here a variant as compared to the “canonical” Herodotean version because, among the oriental sources prior to Herodotus, the above-cited Nabonidus Chronicle (2.16–18) seems to allude, according to the most recent reconstruction, to the murder of Croesus by Cyrus: “… ina ITI GU4 ana KUR Lu-˹ú˺-[du GI]N 17LUGAL-šú GAZ bu-šá-a-šú il-qí šu-lit šá ram-ni-šú ina ŠÀ! (text: LU) ú-še-li […] 18EGIR šu-lit-su ù <É> šar-ri ina ŠÀ GÁL-ši (ušabši), “in the month Iyyar (Cyrus) [mar] ched to Ly[dia]. He killed its king, he took its valuables (and) a garrison of his own he stationed in it! Afterwards he had his garrison and the royal treasury! (bīt šarri) in it”. Note that the verb GAZ = dâku can either mean “to kill” or “to defeat,” but in the context of an individual the translation “to kill” is to be preferred: thus Van der Spek 2014, 256 n. 184.


87

be an object of debate today¹⁷. In the account given by Herodotus¹⁸, certainly the best known and the most widespread version in antiquity¹⁹, Croesus was indeed defeated and taken prisoner by Cyrus, and was condemned to be burnt at the stake (1.86.2). But he was saved by two episodes that occurred in succession: first, by the curiosity awakened in the Persian king on hearing the Lydian invoke the name of the Athenian Solon (1.86.3–6), and secondly by the divine intervention of Apollo, who sent a thunderstorm to put out the flames of the pyre (1.87). Having thus escaped death, and finally “illuminated”, Croesus even became a counsellor of Cyrus and, subsequently, of Cyrus’ son Cambyses²⁰. Whatever the reasons were that prompted the great Greek historian to structure his account of the events in this detailed manner²¹, his superb description, in which Croesus is saved by the Greek gods and by the Persian king, is to be considered as a sort of milestone of the tradition. All the later sources – from Xenophon in the Cyropaedia, to Ctesias, Diodorus, Nicolaus of Damascus, Pompeius Trogus / Justinus, Boethius, right up to the Byzantine chroniclers²² – were unanimous on a specific point, even though their tales show some modifications here and there: Croesus survived the end of his reign²³. Although this question will not be dealt with in greater detail here, it is worth noting that Eusebius’ “minority” version, quoted word for word by MX, is the one most generally accepted by scholars today, if for no other reason than in the sense that the Lydian king is assumed not to have survived the capture of

17 See La Bua 1976; Evans 1978; Burkert 1985; Duplouy 2000; West 2003, 418–428; Gazzano 2012, 100–103. 18 Hdt. 1.84–91, with the commentary by Asheri 19974, 319–322. 19 For a re-examination of the ancient tradition see Talamo 1985; Gazzano 2012; on its subsequent fortune, cf. Law 1948. 20 Hdt. 1.88–89, 1.130.3, 1.155–156, 1.207–208, 3.14.11, 3.14.34, 3.14.36. 21 The most important reference is to Lombardo 1990; see also the Introduzione by Asheri 19974, XCIX-CXIV; Gazzano 2012, 84–91, esp. 84 n. 47, for the previous bibliography. 22 Xen. Cyr. 7.2.2–29, on wh. Lefèvre 2010, 401–417; Ctes. FGrHist 688 F 9 § 5 (= Phot. Bibl. 72, 36b Henry); Diod. Sic. 9.33 = F 50 (= Const. Porph. Exc. de sent. 66 Boissevain) and F 51 Cohen-Skalli (= Const. Porph. Exc. de virt. 220 Büttner/Wobst); Nic. Dam. FGrHist 90 F 68 (= Const. Porph. Exc. de virt. 345 Büttner/Wobst), on wh. Parmentier 2011, 144–147; Just. Epit. 1.7.9; Boeth. Cons. 2.2; Malal. 157–158 Dindorf = 6.10.122 Thurn. 23 It may incidentally be recalled that even in very late sources, such as Georgius Monachus (1.12, 18–20 De Boor/Wirth) or the so-called First Vatican Mythographer (1.196 Zorzetti/Berlioz), according to which Croesus was killed by Cyrus, this did not take place at the fall of Sardis, but at a later time and/or in a different place. According to Georgius Monachus, after the triumph celebrated by Cyrus, Croesus was transported to Persia and was killed there; in contrast, in the story told by the Mythographus Vaticanus, considerable confusion can be noted between the various Herodotean passages concerning the fate of Croesus and the even more tragic fate of the tyrant of Samos Polycrates (Hdt. 3.124–125), who ended up impaled by the satrap Oretes.

88

Francesca Gazzano

Sardis²⁴. Moreover, the other evidence, albeit very scanty and not explicit, on the end of Croesus, stemming from a period earlier than Herodotus, seem to point towards the same conclusion²⁵. In the two most ancient Greek attestations – both of an iconographic nature, and of which the most reliable, the amphora of Myson, is particularly famous²⁶ – Croesus is depicted on the pyre, pouring a libation while an attendant lights sets the pyre alight, or already surrounded by the flames. The absence of any detail that could allude to salvation in extremis, through human or divine intervention, could be a hint of the intention of the artist (or artists) to represent his real immolation. Furthermore, in the first known literary version of the episode, the poetic rendering of Bacchylides²⁷, Zeus and Apollo do intervene to save Croesus, but it is an undeniably ambiguous intervention. That is to say, while the overall content of the ode reflects a “Delphic” version of the episode, aiming to absolve the oracle of any responsibility in the death of the king, Croesus is indeed saved from the flames through the action of Zeus, but he is miraculously removed from the view of those present thanks to Apollo, who transports him to the land of the Hyperboreans²⁸. Thus it is not impossible, as underlined by Stephanie West²⁹, that this removal of the king to a region of mythic imagination hinted, in a poetic and euphemistic form, at his “real” disappearance into the flames. Set in this perspective, and brusquely interrupting the literary predominance of the “canonical” Herodotean version, the account offered by Eusebius (and MX) therefore deserves careful consideration. Less important on the historical plane, although equally interesting from a historiographical point of view – though proper investigation of this issue calls for a detailed examination that is beyond the scope of this paper – is the second historical/chronological hypothesis mentioned above, and rejected by MX (2.13): namely, that Croesus was a contemporary of Nectanebo. The latter is almost certainly to be recognized as Nectanebo II, the last Pharaoh of the XXX dynasty and of independent Egypt (roughly 360–343 BC)³⁰. Above and beyond the question

24 See the bibliography cited in n. 16; contra, Briant 2002, 36 accepts Ctesias’ version (FGrHist 688 F 9 § 5). 25 On this see esp. Duplouy 2000, 21–37; Gazzano 2012, 81–84. For the parallel testimony of the Nabonidus Chronicle (2.16–18), see supra n. 16. 26 Amphora of Myson: Louvre G 197 (ca. 490–480 BC) = Beazley 19632, 238 n. 1; Hydria of the Painter of Leningrad (?): Corinth T 1144 (first half of the fifth century BC) = Beazley 19632, 571 n. 74. For the interpretation of these vases see also Duplouy 2000, 21–37. 27 Epin. 3.23–62: see. e.g. Crane 1996; Kurke 1999, 130–171; Duplouy 2000, 26–37, with additional bibliography. 28 See in this regard Gazzano 2012, 82–84. 29 West 2003, 419. 30 The last Pharaohs of the XXX dynasty (Nectanebo I, Taos/Tachos, Nectanebo II) were well


89

of the historical facts, Nectanebo II was a popular figure in Oriental and Greek literature of the Hellenistic age³¹, perhaps because he too – not unlike Croesus – represented the last king of a people, the victim of the bellicose intentions of a foreign invader³². Appearing as a character in numerous stories, he is even credited, in one strand of the tradition, with being the real father of Alexander the Great³³. As far as MX’s portrayal of the events is concerned, this clearly has no foundation and is in all probability the outcome of multiple confusions whereby historical sources (Manetho³⁴) and romanticized versions (Pseudo-Callisthenes) became intermingled and blurred, as a result of the parabiographical tradition pertaining to Aesop³⁵. For instance, in the Vita, the fabulist – who reproduces many of the traits of the protagonist of the oriental Tale of Ahiqar³⁶ – is connected not only to Croesus but also to a Pharaoh Nectanebo³⁷, identifiable according to

known to the Greek historical sources (e.g. Xen. Hell. 5.14; Ages. 2.28–31; Diod. Sic. 15.90–93; 16.40–46; Plut. Ages. 36–40) due to the Greeks’ military involvement in Egypt, at times even on a personal basis (the Athenian Chabrias, the Spartan king Agesilaus), for anti-Persian purposes in this period; however, the available evidence is not devoid of contradictions and misunderstandings. In general, see Salmon 1985; Lloyd 1994; Perdu 2010, 153–157; esp. on Nectanebo II see now the in-depth and well documented dissertation by Matthey 2012, esp. 86–156, with extensive bibliography. 31 See e.g. the so-called Demotic Chronicle (Bibliothèque Nationale, Paris, Pap. dem. no. 215 r, third century BC: Spiegelberg 1914; cf. Matthey 2012; on the relationship with Manetho see Moyer 2011, 128–134), the “Aretalogy” of Imhotep–Asclepius (P. Oxy. XI 1381, second century AD), the Dream of Nectanebos (P.Leiden I 396, second century BC: Koenen 1985, in Greek, with the addition of P.Carlsberg 562, first/second century AD, in demotic, on wh. Ryholt 1998) or The Prophecy of Petesis (P.Carlsberg 424, 499 559; PSI inv. D 60 v, first/second century AD: Ryholt 2002), as well as the Alexander Romance and the Vita Aesopi: on all this see Matthey 2012, with discussion of the entire tradition. 32 In this sense see above all Eddy 1961; cf. Lloyd 1982. Greater prudence in Matthey 2012, esp. 34–36, with status quaestionis. 33 [Callisth.] Hist. Alex. 1.1–14, with the commentary in Stoneman/Gargiulo 2007, 468–502, and the analysis of Matthey 2012, passim. 34 Manetho (on wh. see Dillery 1999; Moyer 2011, 84–140) concluded his Aegyptiaca with the reign of Nectanebo II: Man. Hist. FGrHist 609 T 8e (= Sync. 486.17 Mosshammer); F 2 (= Sync. 99.18 Mosshammer = F 6 Waddell); F 3 (= Sync. 144.16 ff. Mosshammer = F 74 Waddell). 35 The Vita Aesopi raises complex issues, which cannot be discussed here: see in general Perry 1952; Ferrari/Bonelli/Sandrolini 1997; Jouanno 2006. More specifically, on the relation between the Vita Aesopi and the Tale of Ahiqar see Konstantakos 2008; Id. 2009; Id. 2013; Matthey 2012. 36 Specifically, the so-called “Babylonian section” (Vit. Aes. 101–123). In general, on the Tale of Ahiqar, see not only the above-cited Konstantakos 2008; Id. 2013, but also the useful works by Fales 1993; Contini/Grottanelli 2005; Toloni 2013. 37 Aesop is Croesus’s addressee in the “Lydian section” of the Vita (92–100), immediately prior to the “Babylonian section” (101–123), which features the presence of the pharaoh Nectanebo.

90

Francesca Gazzano

modern critical investigation as none other than Nectanebo II, the latter being held by the author of the Vita Aesopi to be the (anonymous) pharaoh who was the addressee of Ahiqar³⁸. We have no information as to the source from which MX derived this “variant” that held Croesus to be a contemporary of Nectanebo, but it represents an alternative to the view that sees him as a contemporary not only of Cyrus (Eusebius) but also to the claim, endorsed by MX, that he was a contemporary of Artašēs I (the four “Greek historians”). This notwithstanding, MX’s correct determination of the chronological interval among the various sovereigns could suggest that the tradition, as well as the calculations made by MX himself³⁹, were based on an underlying Greek chronographical source, possibly – as has been hypothesized for other reasons as well⁴⁰ – Julius Africanus⁴¹. The problem, as mentioned, deserves further investigation, but for the purposes of our analysis it is sufficient to underline the special literary fortune of Croesus, a character who often takes on a symbolic/paradigmatic nature, and above all the great care taken by MX in tracing the different traditions concerning this character – as MX himself states (“we made many researches”: 13) – and in selecting the one which in his view was most “worthy of belief”.

38 On the identification of the pharaoh Nectanebo of the Vita see above all Konstantakos 2009; cf. Matthey 2012, esp. 7–14. 39 MX 2.13: “But we have found the period of Croesus to be two hundred years before that of Nectanebo, while Nectanebo’s period is more than another two hundred years before that of Artašēs the First, king of Armenia” (transl. Thomson). 40 Topchyan 2001; Id. 2006, 65–84. 41 In effect, MX specifically asserts, in the introductory chapter of the section on Artašēs, that he will begin his description by starting “from the fifth book of Africanus the Chronographer” (2.10, transl. Thomson); Thomson (1978, 12–13; 146 n. 1) rules out the possibility that MX may have had direct knowledge of the Chronography, and believes that he cited it through Eusebius; an indirect mention, via Eusebius, is also suggested by Annie and Jean-Pierre Mahé (1993, 349 n. 2): but on the evidence from MX with regard to Africanus and from the Edessa archives, see Wallraff 2006b, 49–50, and more recently Roberto 2011, 45–57. Be that as it may, the passage discussed here seems to exclude the view that MX based himself exclusively on Eusebius, as appears from the mention of the detail – non Eusebian but nonetheless of “chronographic” interest – concerning the presumed Croesus/Nectanebo synchronism. Hypothetically, one could perhaps imagine that the Armenian historian derived the various traditions on Croesus from a source that was particularly interested in synchronisms, or better, a source which aimed to correctly position the kingdom of Lydia and the fall of Croesus (conceivably in a perspective of translatio imperii); in this case, the alternative hypotheses (era of Nectanebo, era of Artašēs I) would thus have been mentioned in order to reject them. On the possibility that this tradition could ultimately date back to Julius Africanus, as suggested on the basis of different reasons by Topchyan (2006, 79–84), see infra; cf. the contribution by F. Mari, in this volume.


91

2 Artašēs and the Achaemenids Whatever the truth of the matter, the historical end of Croesus is neither the only nor the main reason why these chapters of the History of Armenia devoted to Artašēs I are extremely fascinating. First and foremost, it can be observed that the res gestae of this king described by MX in chap. 12 and above all by the “Greek historians” mentioned in chap. 13 are none other than a rather stratified (confused?) jumble of images, literary reminiscences, veritable citations of Greek “classical” writers, from Herodotus onwards, the originals of which actually refer to the feats and deeds of several Achaemenid sovereigns. I will limit myself to citing a few significant examples. At the outset, MX (2.12) reports that Artašēs issued an order to muster a force, the size of which proved to be such that its extent could not be known. Artašēs therefore apparently resorted to a stratagem: “on the roads and resting places [he ordered] each man to leave a stone to form a cairn as an indication of the multitude”. The image – a sovereign who has no knowledge of the scale of his troops on account of their vastness – is in actual fact a literary topos, with special reference to the Persian world: it is interpretable either as a malicious allusion to the Great King’s ignorance of things concerning the dominated populations and indeed his own military force⁴², or, in a more frequent rendering, as a means of underlining not only the immensity but also the diverse provenance of the massed contingents, thereby unequivocally flaunting the size (and the power) of the empire⁴³. Even the expedient excogitated by Artašēs to solve the impasse has numerous literary precedents, and it strongly recalls the ruse Herodotus attributed to the Persian king Darius, who, without revealing that his aim was really to count his troops, resorted to a similar expedient as he moved across Thracia, when the army that was proceeding against the Scythians reached the banks of the river Arteskos⁴⁴. This kind of empirical head count is in effect well attested in the sources, in various forms. The same system as employed by Darius and Artašēs is attributed by the Ps. Faustus of Byzantium (P‘awstos Buzand better known as the anonymous Buzandaran Patmut‘iwnk‘) to the king of the Massagetae Sanesan who invaded Armenia⁴⁵, while the Herodotean king Xerxes, at Doriscus, is said to have adopted a different and more convoluted procedure for

42 Thus in Curt. 3.2.8–9; 4.12.9. 43 See esp. Hdt. 7.44 (Xerxes’ round-up at Abidus) and 7.59–101 (round-up and head count of the soldiers at Doriscus). Cf. also Hdt. 4.85–88 (Darius at Chalcedon); Xen. An. 1.2.14–18 (round-up by Cyrus the Younger in Cilicia). On this see Briant 2002, 207–211, and now Vannicelli 2013, 50–65. 44 Hdt. 4.92: see Corcella 20034, 306; Boteva 2011, on the localisation of the Arteskos and on the symbolic function of the piles of stones. 45 Buz. 3.7, on wh. Garsoïan 1989, ad loc.

92

Francesca Gazzano

calculating the number of his troops⁴⁶, namely the same method which, according to Q. Curtius Rufus⁴⁷, was also exploited by Darius III at Issus. Furthermore – again, according to Herodotus – the Scythians took a sort of head count of their troops by using arrow heads⁴⁸, while Procopius attributed to the Parthians the habit⁴⁹ – which he defines as very ancient⁵⁰ – of counting the soldiers before and after military campaigns by means of an arrow deposit, in particular to count up the losses. Be that as it may, let us now turn back to the text of MX (2.12). Among the feats accomplished by Artašēs, in addition to the capture of Croesus – to which MX later returns, and to which we will also return – there is a mention of the appropriation of a series of statues of Greek gods and heroes deriving not only from Asia, to be taken as meaning Anatolia, but also from Greece itself. The historian mentions first of all the “Asian” statues of Artemis, then Heracles (explicitly identified with the Armenian warrior deity Vahagn⁵¹) and Apollo, gold-plated; at the end of the chapter, at this point he adds the “Greek” statues of Dios/Zeus, Artemis, Athena, Hephaistos and Aphrodite. MX is careful to underline that the statue of Heracles was the work of the Cretan sculptors Scyllis and Dipoenus, but it is impossible to determine where he obtained this information: such was their fame as artists – they were even believed to be disciples of Daedalus, and according to the judgment of Pliny they were the first to have achieved fame in the art of sculpting⁵² – that they were mentioned innumerable times in connection with their masterpieces and the pupils assigned to them⁵³. While Thomson points to a possible parallel in Clement of Alexandria⁵⁴, it should also be borne in mind that these two sculptors, regarded as veritable symbols of the most “ancient” art,

46 Hdt. 7.60 (assembling 10,000 “counted” men in a circle, building a fence round the circle, calculating the number of troops in multiples of 10,000): obviously there are doubts (How/Wells 1912, II 151, ad loc.; Vannicelli 2013, 64) on the historical truth of the episode. 47 Curt. 3.2.2, explicit in recalling the model of Xerxes (Xerxes exemplo numerum copiarum iniit): Briant 2002, 693, 796–797. 48 Hdt. 4.81.3, with the commentary by Corcella 20034, 299–300. 49 Procop. Bell. 1.18.52–53: see Börm 2007, 97. 50 Procop. Bell. 1.18.54: οὕτω μὲν οὖν Πέρσαις ὁ νόμος ἐκ παλαιοῦ ἔχει. 51 The Armenian warrior deity Vahagn has some points of contact not only with the Iranian Vǝrǝθraγna attested by the Avesta, but also with the Greek Heracles: cf. Russell 1989; Mahé 1994, 782–783; Petrosyan 2007; Ivanov 2011; Traina 2013. 52 Plin. HN 36.9, who estimated his floruit as being around the time of the 50th Olympiad (580–577 BC), etiamnum medis imperantibus priusque, quam Cyrus in persis regnare inciperet. Modern scholarship is effectively inclined to believe he was active between the end of the seventh and the beginning of the sixth century BC: see in general Stewart 1990, 107–108. 53 See esp. Paus. 2.15.1; 2.22.5; 2.32.5; 3.17.6; 5.17.1; 6.19.14; on the interest of Pausanias in these sculptors see Arafat 1996, 68–73. 54 Thomson 1978, 148 n. 4, in reference to Clem. Al. Protr. 4.47.8, who does actually cite some


93

were also held to have carved a precious statue of Athena Lindia, in green stone, that was on show among the works of the collection in the Palace of Lausus in Constantinople, which burnt down in 475 AD⁵⁵. Be that as it may, this section could spark a wealth of interesting reflections. Firstly, the Achaemenids occupy a far from secondary role in the range of testimonies concerning the removal of Greek cult statues. For instance, Berossus (an author explicitly cited elsewhere by MX⁵⁶) credited Artaxerxes II with introducing the cult of anthropomorphic images of the gods into Persia, with specific reference to Aphrodite-Anahita⁵⁷. Moreover, although the theft of sacred statues was a widespread practice in antiquity, the most powerfully evocative precedent, in the Greek literary perspective, appears once again to have been represented by the “thefts” which, during the Persian wars, were (wrongly, in all probability) blamed on the Achaemenid kings. The finger is pointed both at Darius – though in this case his general Datis is said to have immediately returned the ill-gotten gains⁵⁸ – and above all at Xerxes, described as having looted a vast array of cult images both in Asia Minor, such as the Apollo of Didyma, and also in Greece, as in the case of the Artemis of Brauron⁵⁹; the works he had made off with are then said to have been found in Persia and restored to the Greeks by Alexander or his successors⁶⁰. Consequently, one should not disregard the hypothesis that in drawing up his account of the res gestae of Artašēs, MX may have made use of one or more sources that had addressed the question of the plundering of statues of gods and heroes among peoples of various different religions and cultural backgrounds. But there is also a second and far more cogent element, namely

works of Dipoenus and Scillis, but in a context of total condemnation of pagan idolatry. Cf. Herrero de Jáuregui, 2008, ad loc. 55 Cedr. 1.564, 7–10 Bekker. See Francis/Vickers 1984. On the collection of Lausus, praepositus sacri cubiculi during the reign of Theodosius II (406–450 AD) see esp. Mango 1963. Topchyan 2006, 81–82 also believes that the testimony of Cedrenus, who is said to have utilized Julius Africanus, may indicate that MX’s source for these data as well was none other than the author of the Chronographiae. 56 MX 1.2, on wh. see the extensive commentary by Topchyan 2006, 17–35. 57 Beros. FGrHist 680 F 11 (= Clem. Al. Protr. 5.65.2). This datum is also confirmed by the oriental sources: Boyce/Grenet 1991, 385; Amir-Arjomand 1998; Daryaee 2006, 496–497. It is worth noting that Topchyan 2006, 81–82 makes a reference to the passage in question by Clement of Alexandria, but without realising that it is actually the citation – intended for purposes of putting forward a negative criticism – of a fragment from book III of Berossus’ Babyloniaca. 58 Hdt. 6.118.1–3 (theft of a gold-plated statue of Apollo). 59 In the Greek sources, the Achaemenids are actually notorious as destroyers, rather than thieves, of statues: see Gazzano 2014. 60 Paus. 8.46.3, with the commentary by Moggi 2003, 503–504. On these (and other) alleged thefts by Xerxes see Moggi 1973; Scheer 2003.

94

Francesca Gazzano

the fact that in this portion of the text there are two pieces of linguistic evidence which are helpful in showing that, whatever the extent of MX’s re-elaboration of the materials on Artašēs, he based himself wholly or in part on a source that was indisputably Greek. Of these two features, one concerns the use, extremely rare in MX and already noted by Topchyan⁶¹, of the choronym “Hellas” (Helladà), instead of other terms otherwise used habitually by the Armenian historian for “Greeks” and “Greece” (Yoynk‘ “the Greeks, also in the sense of Greece”, Yunac‘ ergir “the land of the Greeks”, Yunac‘ ašxarh “the country of the Greeks”). The other, which one may consider even more tangible, is represented by the fact that MX mistakenly interprets the original Greek genitive of Zeus, Dios, as a nominative⁶². Finally, the information concerning the death of king Artašēs at the hand of his own troops, an episode with regard to which MX seems to be somewhat skeptical – witness the parenthetical “as they say” – could perhaps be viewed in conjunction with the events related by the Greek sources on the subject of several Achaemenid kings, such as Xerxes I, Xerxes II, Artaxerxes III, assassinated in court plots⁶³, or Darius III, killed by his own generals⁶⁴. However, it is above all in the next chapter, in the quotation of the ipsissima verba of the four unrecognized “Greek historians”⁶⁵, that it inescapably has to be conceded that the deeds attributed to Artašēs – whom these “historians” likened to or even considered greater than Alexander the Great himself – were actually re-elaborated on the basis of materials originally pertaining to the Achaemenids. For example, the statement by Polykrates (2.13) “although he remained in his own country he ruled over Thebes and Babylon; and without crossing the river Halys

61 Topchyan 2006, 81. 62 Credit for this observation (not present either in Thomson 1978, or in Mahé/Mahé 1993, or in Topchyan 2006) should be attributed to G. Traina. Nevertheless, the possibility that the mistake was made by a Greek author cannot be ruled out. 63 Xerxes I: Ctes. FGrHist 688 F 13b, 33; Diod. Sic. 11.69; Ael. VH 13.3; Just. Epit. 3.1 (on wh. Briant 2002, 563–568); Xerxes II: Ctes. FGrHist 688 F 15, 47–50 (with Briant 2002, 588–591); Artaxerxes III: Diod. Sic. 17.5; cf. 16.47–50 on the murderer, the powerful eunuch Bagoas (see in this regard Briant 2002, 269–270, and esp. 769–776). 64 On the death of Darius III see Briant 2002, 864–867; on his figure see esp. Briant 2003. 65 The complex question of the identification (and the very issue of the historical reality) of the four Greek authors (Polykratēs, Evagaros, Skamadros, P‘łegonis) whose citations are given verbatim by MX, cannot be addressed in depth here. With the exception of P‘łegonis, whom scholars generally agree in identifying as Phlegon of Thralles (see infra), the attempts made to single out possible candidates among the authors known from the tradition have been rather unfruitful, for various reasons. Jacoby includes in the FGrHist only the fragment of P‘łegonis/Phlegon (257 F 39), although he classifies it as Unechtes (IID, 1191). Cf. Sargsyan 2012–2013, 82–83 and the contribution by F. Mari, in this volume.


95

(Aliws) he destroyed the Lydian army …” seems to recall the figure of Darius I, who was effectively the lord of Egypt and of Babylonia, but more specifically that of the Darius of The Persians by Aeschylus (ὅσσας δ᾽ εἷλε πόλεις πόρον / οὐ διαβὰς Ἅλυος ποταμοῖο, /οὐδ᾽ ἀφ᾽ ἑστίας συθείς “And what a number of cities he captured! – [865] without crossing the stream of Halys or even stirring from his own hearth”⁶⁶). Furthermore, the subsequent statement “and before arriving in Asia he was announced in the fortress of Attica”⁶⁷ could be an echo of the episode, known in primis from Herodotus⁶⁸, of the heralds sent in 491 BC by Darius to Athens, as in all the Greek cities, to request “earth and water”, thus showing their submission to the Great King⁶⁹. And there is more: beyond all reasonable doubt, the assertions of Evagaros “… for the light of the day was obscured by the dust of the former (scil. Alexander and Darius), but the latter (scil. Artašēs) hid and darkened the sun by his volley of arrows, turning midday into artificial night” mirror, in the first image, the battle of Issus as described by the Ps.-Callisthenes⁷⁰, while in the second image the assertions constitute a reminiscence of a celebrated anecdote narrated by Herodotus⁷¹, possibly filtered through one of the many subsequent

66 Aesch. Pers. 864–866 (transl. Herbert Weir Smyth). 67 Cf. also the rendering in French by Mahé/Mahé 1993, 170: “avant qu’il fût arrivé en Asie, on proclamait déjà son nom dans la citadelle d’Athènes”. 68 Hdt. 6.48, with the commentary by Nenci 1998, 215–216; cf. Hdt. 5.73.2, for an analogous precedent in Athens. Cf. in general Briant 2002, 157–158. 69 As is known, the tradition described by Herodotus (7.32, 7.133; cf. Paus. 3.12.7) held that both the Spartans and the Athenians responded to Darius’ request with the (sacrilegious) killing of the heralds, and that this was the reason why Xerxes, in the imminence of his expedition against Greece, reportedly sent heralds bearing the same request throughout Greece, with the exception of Athens and Sparta. 70 [Callisth.] Hist. Alex. 1.41.7, on wh. see Stoneman/Gargiulo 2007, 559: the story in the Romance refers to the battle of Issus, whereas in the historical sources (Diod. Sic. 17.50.4; 51.1; Curt. 4.15.32) the image of the dust blotting out the sun refers to the battle of Gaugamela. 71 Hdt. 7.226: Λακεδαιμονίων δὲ καὶ Θεσπιέων τοιούτων γενομένων ὅμως λέγεται ἀνὴρ ἄριστος γενέσθαι Σπαρτιήτης Διηνέκης. τὸν τόδε φασὶ εἰπεῖν τὸ ἔπος πρὶν ἢ συμμῖξαι σφέας τοῖσι Μήδοισι, πυθόμενον πρός τευ τῶν Τρηχινίων ὡς ἐπεὰν οἱ βάρβαροι ἀπίωσι τὰ τοξεύματα, τὸν ἥλιον ὑπὸ τοῦ πλήθεος τῶν ὀιστῶν ἀποκρύπτουσι. τοσοῦτο πλῆθος αὐτῶν εἶναι. [2] τὸν δὲ οὐκ ἐκπλαγέντα τούτοισι εἰπεῖν ἐν ἀλογίῃ ποιεύμενον τὸ Μήδων πλῆθος, ὡς πάντα σφι ἀγαθὰ ὁ Τρηχίνιος ξεῖνος ἀγγέλλοι, εἰ ἀποκρυπτόντων τῶν Μήδων τὸν ἥλιον ὑπὸ σκιῇ ἔσοιτο πρὸς αὐτοὺς ἡ μάχη καὶ οὐκ ἐν ἡλίῳ. “This then is how the Lacedaemonians and Thespians conducted themselves, but the Spartan Dieneces is said to have exhibited the greatest courage of all. They say that he made the following speech before they joined battle with the Medes: he had learned from a Trachinian that there were so many of the barbarians that when they shot their missiles, the sun was hidden by the multitude of their arrows”. (transl. A. D. Godley).

96

Francesca Gazzano

reformulations⁷², among which one may mention that of the Ps.-Callisthenes, who in turn attributed the words to the Athenian orator Demades⁷³. The detail of the rivers dried up by the army can likewise be found on more than one occasion in the Herodotean account of the march of the Persian army in Greece⁷⁴, and in this same fragment the recourse to a unit of measure as a means of counting the troops once again recalls the words of Herodotus in his description – mentioned earlier⁷⁵ – of the system excogitated by Xerxes to calculate the extent of his infantry (a demarcated area, designed to contain 10,000 men at a time). As a final addition to this list of similarities, one should not overlook the tears and exclamation of Artašēs at the sight of the army, which appear to be modelled on the tears of Xerxes, and on his sic transit gloria mundi⁷⁶. Turning now to the fragment of Skamadros, which is built up exclusively on the Croesus episode, it would appear to be indisputable that this account derived from the Herodotean story of Cyrus’ capture of the Lydian king⁷⁷ – possibly filtered through some intermediary such as Diodorus⁷⁸ – includ-

72 The innumerable occurrences of this image (e.g. Plut. Apophth. Lac. 225b; Cic. Tusc. 1.42.101; Stob. Flor. 3.7.45) confirm its topic character. 73 [Callisth.] Hist. Alex. 2.2.14, with the commentary by Stoneman/Gargiulo 2013, 380–381. 74 Hdt. 7.43.1: ἀπικομένου δὲ τοῦ στρατοῦ ἐπὶ ποταμὸν Σκάμανδρον, ὃς πρῶτος ποταμῶν, ἐπείτε ἐκ Σαρδίων ὁρμηθέντες ἐπεχείρησαν τῇ ὁδῷ, ἐπέλιπε τὸ ῥέεθρον οὐδ᾽ ἀπέχρησε τῇ στρατιῇ τε καὶ τοῖσι κτήνεσι πινόμενος […]. “When the army had come to the river Scamander, which was the first river after the beginning of their march from Sardis that fell short of their needs and was not sufficient for the army and the cattle to drink” (transl. A. D. Godley); cf. also Hdt. 7.108; 7.127.2; 7.196. 75 Supra, 91–92, also nn. 46 and 47. 76 Hdt. 7.45–46: […] ὡς δὲ ὥρα πάντα μὲν τὸν Ἑλλήσποντον ὑπὸ τῶν νεῶν ἀποκεκρυμμένον, πάσας δὲ τὰς ἀκτὰς καὶ τὰ Ἀβυδηνῶν πεδία ἐπίπλεα ἀνθρώπων, ἐνθαῦτα ὁ Ξέρξης ἑωυτὸν ἐμακάρισε, μετὰ δὲ τοῦτο ἐδάκρυσε. “[…] When he saw the whole Hellespont covered with ships, and all the shores and plains of Abydos full of men, Xerxes first declared himself blessed, and then wept” (transl. A. D. Godley). 77 See supra, 86–87. 78 Diod. Sic. 9.2.3–4 = F 4.3–4 Cohen-Skalli: [3] ὁ δὲ Κροῖσος ὕστερον γενόμενος αἰχμάλωτος ὑπὸ Κύρου καὶ μέλλων ἐπὶ μεγάλῃ πυρᾷ κατακαίεσθαι, τῆς Σόλωνος ἀποφάσεως ἐμνημόνευσεν. διὸ καὶ τοῦ πυρὸς ἤδη περιφλέγοντος ἀνεβόα συνεχῶς τὸ τοῦ Σόλωνος ὄνομα. [4] ὁ δὲ Κῦρος προσπέμψας τοὺς πευσομένους, τίς ἡ συνεχής ἐστι τοῦ Σόλωνος ὀνομασία, μαθὼν τἀληθὲς μετέπεσε τοῖς λογισμοῖς καὶ νομίσας τὴν ἀπόκρισιν τοῦ Σόλωνος ἀληθινὴν εἶναι τῆς μὲν ὑπερηφανείας ἀπανίστατο, τὴν δὲ πυρὰν κατασβέσας ἔσωσε τὸν Κροῖσον καὶ τὸ λοιπὸν ἕνα τῶν φίλων κατηρίθμησεν. “And at a later time, when Croesus had been taken prisoner by Cyrus and was about to be burned upon a great pyre, he recalled the answer Solon had given him. And so, while the fire was already blazing about him, he kept continually calling the name of Solon. [4] And Cyrus sent men to find out the reason for his continual calling of the name of Solon; and on learning the cause Cyrus changed his purpose, and since he believed that Solon’s reply was the truth, he ceased regarding Croesus with contempt, put out the burning pyre, saved the life


97

ing the verbatim citation of the Delphic oracle⁷⁹ and the invocation of the name of Solon, as well as the final salvation⁸⁰. Finally, the fragment of the last author on whom MX draws is a veritable pastiche of material that is both literary and historical at the same time. The author is a certain P‘łegonis whom criticism tends, with a fair degree of reasonableness, to identify with Phlegon of Thralles⁸¹. Here the grandeur of Artašēs’ achievements is revealed at once through the image of an adynaton (sailing over the land, walking on the sea⁸²), an image which, from Herodotus⁸³ to Lysias (2.29), to Isocrates (4.89), to the poets of the Anthologia Palatina (9.304), was unequivocally linked – at times, not without irony⁸⁴ – to the hybris of Xerxes, who had overturned the natural order by digging a canal for his fleet on the slopes of Mount Athos and

of Croesus, and counted him henceforth as one of his friends” (transl. C. H. Oldfather). Diod. Sic. 9.31.1 = F 42 Cohen-Skalli: [1] ὅτι Κροῖσος ἐπὶ Κῦρον τὸν Πέρσην ἐκστρατεύων ἐπύθετο τοῦ μαντείου. ὁ δὲ χρησμός, “Κροῖσος Ἅλυν διαβὰς μεγάλην ἀρχὴν καταλύσει”. ὁ δὲ τὸ ἀμφίβολον τοῦ χρησμοῦ κατὰ τὴν ἑαυτοῦ προαίρεσιν ἐκδεξάμενος ἐδυστύχησεν. “When Croesus was taking the field against Cyrus the Persian, he made inquiry of the oracle. And the answer ran: ‘If Croesus crosses Halys, a mighty realm will he destroy’. He received and interpreted the ambiguous answer of the oracle in the light of his own purpose and so came to grief” (transl. C. H. Oldfather). Cf. Gazzano 2012, 88 ff. 79 The oracle in question is probably the most famous and most frequently mentioned oracle of antiquity: Parke/Wormell 1956, II 24, no. 53; Fontenrose 1978, 113–114, no. 67. In contrast to Hdt. 1.53, where the oracle is reported in an indirect form, the majority of the subsequent citations quote it in verse. In addition to the already cited Diod. Sic. 9.31.1 (= F 42 Cohen-Skalli), see e.g. Arist. Rh. 1407a; Lucian Iupp. Conf. 14; Iupp. Trag. 20 and 43; Oenom. fr. 5 Hammerstaedt (= Eus. Praep. Evang. 5.21.2); Max. Tyr. 5.2a, 13.5e; Schol. Ar. Plut. 8b5. Cf. also Malal. 155–156 Dindorf = 6.8.120–121 Thurn. See Gazzano 2012, 98 ff. 80 Hdt. 1.86. 81 FGrHist 257 F 39. The fragment (classified as uncertain by Jacoby FGrHist IIC 848) is not included in the most recent edition of Phlegon, Stramaglia 2011 (but see Praef. XV-XVI n. 40); for a recent Italian translation see now Braccini/Scorsone 2013, 59–60 F 39. See Traina 1991, 73–74; Topchyan 2006, 84; Sargsyan 2012–2013, 82 n. 13; Mari, in this volume. 82 On the fortune of this image, see Crevatin/Lefons 1991, 411–426. 83 Hdt. 7.22–24, 7.33–36; in Aesch. Pers. 745–750 there is a description of the “yoking” of the Hellespont, but the theme of the incision cut into Mount Athos is absent. On the latter undertaking see Traina 1988, 320–332 (who highlights the ideological aspect of the “condemnation” of Xerxes); for an in-depth analysis of the text of Herodotus see Vannicelli 2013, 24–36. 84 For Lucian (Rh. Pr. 18) this is one of the topoi that were considered an absolutely essential feature for anyone called upon to deliver a speech before a court: κἂν περὶ ὑβριστοῦ τινος ἢ μοιχοῦ λέγῃς Ἀθήνησι, τὰ ἐν Ἰνδοῖς καὶ Ἐκβατάνοις λεγέσθω […] καὶ ἀεὶ ὁ Ἄθως πλείσθω καὶ ὁ Ἑλλήσποντος πεζευέσθω καὶ ὁ ἥλιος ὑπὸ τῶν Μηδικῶν βελῶν σκεπέσθω […]. “If your subject is assault or adultery in Athens, cite the Indians and Medes. […] Hardly less so (scil. indispensable) are a fleet crossing Mount Athos, an army treading the Hellespont, a sun eclipsed by Persian arrows […]” (transl. Fowler).

98

Francesca Gazzano

by leading his army on foot across the Hellespont on bridges made of boats⁸⁵. The lists of conquered populations given immediately afterwards undoubtedly portray feats attributed to the same Persian king, and in the last analysis all these deeds can be traced back to the Herodotean narration of the second Persian war⁸⁶. In contrast, the grievous end of Artašēs is presented through a comparison with the most significant defeats suffered by the Persian kings, this time listed in opposition (Cyrus, Cambyses, Darius and Xerxes), once again following a model supplied in primis by the historian of Halicarnassus⁸⁷. The whole of this chapter of MX, as can readily be imagined, presents problems that offer no easy solution or indeed no conceivable solution: an in-depth discussion of this aspect would be beyond the scope of this paper – though a possible line of interpretation, which I believe can be shared, is put forward in this same volume by Francesco Mari. Here, attention will be limited to pointing out some elements which seem to run counter to the widely held view that MX “invented” these traditions ex nihilo, attributing them to unknown Greek authors⁸⁸. Firstly, consideration should be given to the fact that in all four “Greek historians” mentioned by MX, Artašēs is curiously defined as “Parthian”⁸⁹; secondly, and more importantly, the episodes in question all have a highly literary flavor, in which the Persian kings are subsumed into a single character, who combines within himself events and characteristics of various different Achaemenid

85 Thomson 1978, 19 proposes a comparison with 2 Macc. 5.21 (ὁ γοῦν ᾿Αντίοχος ὀκτακόσια πρὸς τοῖς χιλίοις ἀπενεγκάμενος ἐκ τοῦ ἱεροῦ τάλαντα θᾶττον εἰς ᾿Αντιόχειαν ἐχωρίσθη, οἰόμενος ἀπὸ τῆς ὑπερηφανίας τὴν μὲν γῆν πλωτὴν καὶ τὸ πέλαγος πορευτὸν θέσθαι διὰ τὸν μετεωρισμὸν τῆς καρδίας); however, this represents a further example of the fortune of the expression, since in the fragment cited by MX the parallel with Xerxes is unequivocal due to the references to the Hellespont and Thrace. 86 The parallels are: Hdt. 7.129–130 (Xerxes threatens to transform Thessaly into a lake; cf. 7.172– 173: invasion of Thessaly); 7.132 (the Locrians were among those who accepted the Great King as their overlord; cf. 8.66); 7.222–228 (massacre of the Spartans at the Thermopylae); 8.32–33 (Phocis laid waste). One curious element is the allusion by P‘łegonis/Phlegon to the Bithynians, whom Herodotus mentions only twice with this name, once in relation to Croesus’ conquests (1.28), and once relating to the period of Xerxes’ expedition (7.75). 87 Hdt. 1.204–214 (expedition and death of Cyrus against the Massagetes); 4.122–142 (expedition and failure of Darius in Scythia); 3.17–26 (expedition and failure of Cambyses in Ethiopia). As is known, Xerxes’ expedition against Greece occupies almost the whole of books VII-IX of the Histories: esp. on Xerxes’ retreat see Hdt. 8.97–117. In 8.118–120 the historian reports – without endorsing it – a more dramatic version of the Great King’s escape by sea. 88 In this sense, see above all Thomson 1978; cf. the extensive status quaestionis in Topchyan 2006, with previous bibliography. Contra, see not only Topchyan 2006, but also Traina 1991; Id. 1995; Id. 1998. 89 On the possible reasons for this slip, see esp. Sargsyan 2012–2013; Mari, in this volume.


99

sovereigns⁹⁰. This makes it somewhat improbable that the “forger” MX would have been able, purely on the basis of the Armenian sources, to build up a portrait of Artašēs so richly interwoven with literal references to the historical sources of the classical age, and above all a portrait which, although certainly ahistorical, is free from patently evident fable-like and/or romanticized features. Thus if he did indeed have a base, one could thus surmise that he had found the information on Artašēs I already “structured”, wholly or in part, in a tradition stemming from a Greek background, though how he had come across this tradition is a question which, as stated earlier, goes beyond the scope of this paper and cannot be addressed here. However, the hypothesis that this involved a chronographic source, conceivably Julius Africanus⁹¹, as Topchyan has sought to demonstrate elsewhere using a different line of reasoning, is an idea that can be shared, not only on account of the arguments put forward so far, but also for an additional reason, which I outline briefly below, prior to my concluding remarks.

3 The importance of being Croesus Let us then turn to an aspect which I feel has not yet been sufficiently examined – although it could provide the key to understanding the operation undertaken by MX: namely, the question as to why he engaged in such a project. Why, within the economy of his “historical” reconstruction, did he believe it was fundamental to demonstrate, against other traditions that included the authoritative account by Eusebius (and indeed against historical reality) that the figure who achieved victory over Croesus and the Lydians was Artašēs I, the founder of the kingdom of Armenia, and not Cyrus the king of Persia or some other character? Why single out Croesus, the last sovereign of a kingdom which, after the Persian conquest (547/6 BC), disappeared for ever from the face of History as an independent and autonomous entity? Now, Cyrus’ rise to power and the “birth” of the Persian empire appear as epoch-making events, reflected also in terms of a chronological watershed with the bifurcation of two different literary (and ideological) traditions: the biblical perspective – the Persian conquest of Babylon in 539 BC allowed the Hebrews to escape from “captivity” and eventually led to the founding of Jerusa-

90 See the already cited Sargsyan 2012–2013; Mari, in this volume. 91 The observations that follow – prompted by an examination of MX’s text – owe much to the general interpretation of the work put forward by Roberto 2011. See also the essays contained in Wallraff 2006a (in particular those by Burgess 2006; Roberto 2006; Wallraff 2006b). For the edition of the fragments, see Wallraff/Roberto/Pinggéra 2007.

100

Francesca Gazzano

lem⁹² – and the perspective which led in the opposite direction, towards Greek civilization, given that the fall of Croesus coincided historically with the end of an era of coexistence and contiguity, if not of genuine cultural assimilation, between Greeks and Orientals in Anatolia. The Persian conquest of the poleis of Asia was a veritable choc for the Greeks, and would later be perceived as the first act in a history of clashes – not only of a military nature – between two civilizations destined to experience a protracted period of absence of communication, aloof from each other⁹³. The centrality of Cyrus in these two different traditions – the Hebrew and the Hellenic tradition – is probably the reason why one notes in the later chronographic literature an overlaying – or indeed a crasis between the two different events in which the founder of the Persian empire was a protagonist. In this regard, suffice it to mention the episode narrated in the Chronographia of John Malalas, whose sources unquestionably included a work by an author particularly alert to the synchronisms and the symphonia of events⁹⁴, namely the Chronographiae of Julius Africanus. In Malalas’ account of the events, king Cyrus, wisely counselled by the prophet Daniel, freed the Hebrews from captivity precisely after having defeated and taken prisoner Croesus of Lydia⁹⁵. Given the absence of any reference to the côté of the Hebrew tradition, the importance attributed to the capture of Croesus in MX would thus appear to confirm the choice of a quintessentially Hellenic perspective, but this is not the fundamental point. Rather, there was another aspect that must have struck the Armenian historian as far more significant: in Herodotus, the Lydia of Croesus seems to be awarded the role of the last power to hold out against the irresistible rise of the Persians who were

92 The Hebrew tradition that (correctly) linked the liberation of the Hebrews to Cyrus’ conquest of Babylon, without any reference to the fall of the kingdom of Lydia, is also accepted by Joseph. AJ 11.1–3: see Ackroyd 1990; Grabbe 2004, esp. 143–155 and 209–237. 93 On this see Lombardo 1990; Asheri 19974; Gazzano 2012. 94 Roberto 2011, 67 ff. 95 Malal. 153–156 Dindorf = 6.5–11, 119–122 Thurn. In the passage, the narration of Cyrus’ conquest of Lydia and defeat of Croesus, wherein Cyus was “helped” by the prophet Daniel, is extended by Malalas’ addition of the list of authors (Thallus, Castor, Polybius and, finally, Herodotus) who ταῦτα […] ἱστορήσαν. It should be borne in mind that the same three Greek authors (Castor, Thallus and Polybius) are cited in a significant fragment (F 34 Wallraff/Roberto, with the important observations in Wallraff/Roberto/Pinggéra 2007, 73 ff.) of the Chronographiae by Julius Africanus, namely in a fragment which is designed to set up a synchronism between Hebrew history and Greek history, just as in this passage from Malalas the figure of Cyrus acts as a unifying element between the biblical (Daniel) tradition and that of the Hellenes (Croesus). Thus it could be hypothesized that here Malalas’ source was in fact Africanus himself; on his method, see Roberto 2011, 88–103.


101

“running rampant” beyond the natural border of the Halys and were thus well on their way to becoming the lords of the whole of Asia⁹⁶. By virtue of this resistance, Lydia was sometimes included among the “universal powers” of antiquity that formed part of the “list of succession of the empires”⁹⁷, and as such it is presented in a number of authors, already (implicitly) in Herodotus⁹⁸, perhaps in Ctesias⁹⁹, then explicitly in Aristoxenus¹⁰⁰, Nicolaus of Damascus¹⁰¹, and then in Julius Africanus¹⁰² and John Malalas¹⁰³. Thus it cannot be excluded that MX’s interest in the capture of Croesus derived from the possibility of attributing a universal empire (albeit of fleeting duration) to Artašēs I, who was effectively the founder of the kingdom of Armenia. There was indeed a pre-existing tradition represented by the four Greek authors whom MX mentions, which ascribed the deeds of various Achaemenid kings to a single “Parthian” sovereign, whose name could be “superimposed” or confused with that of the Armenian Artašēs¹⁰⁴. That such was in fact MX’s intention may be deduced from the title of the chapter itself, which makes an explicit mention of the “universal character” (tiezerakalut‘iwn) of Artašēs’ kingdom. In this sense, the capture of Croesus, the lord of a universal dominion, represented a key feature which, in line with the interest of the chronographers on whom MX based his account, made it possible to include Armenia in the succession of universal empires of the past, long before the reign of Tigran II. If these conjectures have any foundation, then the charge of “falsification” often directed against MX would not perhaps be entirely effaced, but it would

96 Hdt. 1.95.1–2, 1.103.2, 1.130. Research on these themes is currently being undertaken by myself, in the framework of the Progetti di Ricerca di Ateneo (PRA 2013) of the University of Genoa. 97 As is well known, this theme is extremely vast, and can in no way be the subject of investigation here; for a few overviews, see Alcock/D’Altroy/Morrison/Sinopoli 2001; Foraboschi/Pizzetti 2003; Lanfranchi/Roaf/Rollinger 2003; Millar 2006 (chap. 3, 5); Gnoli/Muccioli 2007; Morris/ Scheidel 2009. 98 See supra, n. 96 99 Ctes. FGrHist 688 (= F 1 b, p δ, F 5, F 8 Lenfant): in what has come down to us from Ctesias there is no genuine “list” of the succession of empires, but the presence of Lydia cannot be ruled out, given the interest in this question shown by the historian. 100 Aristox. F 50 Wehrli (= Ath. 12.545a-546b): see Zecchini 1988. 101 Nic. Dam. FGrHist 90 (= F 22, F 66 Parmentier, with the commentary ad loc.). 102 Afric. Chron. F 73 Wallraff/Roberto (= Exc. Lat. Barbari 314.19–316.18 Frick), on wh. see Wallraff/Roberto/Pinggéra 2007, 223–225. 103 Malal. 153–156 Dindorf = 6.5–11, 119–122 Thurn. 104 MX 2.13 tit. “Vkayut‘iwnk‘ tiezerakalut‘eann Artašisi ew kapeloyn zKriwsos yayloc‘ patmagrac‘”. Transl. Thomson 1978: Testimonies from other historians to the empire of Artashēs and his taking Croesus prisoner; transl. Mahé/Mahé 1993: Témoignages d’autres historiens sur les conquêtes d’Artachês et sur la captivité de Crésus; transl. Traina: Testimonianze dell’universalità di Artašēs, e di come mise Kriwsos in catene, da parte di altri storici.

102

Francesca Gazzano

have to be transferred from the plane of invention to that of manipulation, and it would take on an ideological-patriotic rather than a strictly historical character. But one may go further: in creating a linkage (artificial, to be sure) between the real figure of Croesus and a protagonist of Armenian history, MX was not committing a particularly “scandalous” operation: Herodotus himself had already identified the turning-point in the unfolding historical and human events involving Croesus, and he described the landmark change as arising from the meeting – albeit chronologically rather improbable – between Croesus and Solon, the wise lawgiver of Athens¹⁰⁵. It is only after this crucial moment, in which Croesus scorned the warning voiced by the Greek – who, in the Herodotean logos, becomes the mouthpiece of a typically and quintessentially Greek wisdom – that the Lydian king’s hitherto triumphal progress underwent a dramatic reversal. The result of his actions turned into a veritable tragedy, which affected his loved ones (death of his first-born son), the military campaign against the Persians, the destiny of his kingdom and his own personal fate, up to the final illumination, of which Solon is once again an involuntary protagonist¹⁰⁶. In short, the entire logos of Croesus, as presented by Herodotus (the “historicity” of which is not open to doubt, in general terms) hinges on the centrality of a meeting which, in all probability, never took place at all. It is quite likely that the tradition subsequent to the Halicarnassus historian had more than a few doubts about this account of the course of events, given the particular literary fortune of the episode, which was among the most frequently quoted of the Histories, and Plutarch certainly mused on it, in the Vita devoted to Solon (27.1): “As for his interview with Croesus, some think to prove by chronology that it is fictitious. But when a story is so famous and so well-attested, and, what is more to the point, when it comports so well with the character of Solon, and is so worthy of his magnanimity and wisdom, I do not propose to reject it out of deference to any chronological canons, so called, which thousands are to this day revising, without being able to bring their contradictions into any general agreement”¹⁰⁷. It is impossible to determine whether MX had any opportunity to acquire knowledge on this motive, but Herodotus’ charming contrivance, and Plutarch’s readiness in justifying it, could to some extent contribute to illuminating, albeit “through a glass, darkly” the words of the Armenian historian (and mitigate his

105 Hdt. 1.29–33, with the detailed commentary by Asheri 19974, 281–282: here status quaestionis on the chronological difficulties raised by the meeting. See, in the same sense, also infra, n. 108. 106 For this depiction see again Lombardo, 1990; Gazzano 2012, with previous bibliography. 107 On this see esp. Piccirilli 1980, 268–270, who likewise clearly underlines the chronological difficulties raised by the imaginary conversation; in general see more recently Rhodes 2003, 58–72.


103

disarming nonchalance): “because there are many who say that our Artašēs took Croesus prisoner, and narrate this plausibly, I am so persuaded”, and could lift the veil on the reasons that led MX to establish an association between the founder of the kingdom of Armenia and none other than Croesus, the last king of Lydia.

104

Francesca Gazzano

Appendix Movsēs Xorenac’i, History of Armenia, 2.12–13 (transl. Thomson 1978) 12. Artashēs marches to the east; he takes Chroesus prisoner and sends as spoils to Armenia the images of the idols Then Artashēs ordered an army to be raised from the east and north, such a great one that he did not know its number: but on the roads and resting places [he ordered] each man to leave a stone to form a cairn as an indication of the multitude. He then marched to the west and took as prisoner Chroesus, king of Lydia. Finding in Asia images of Artemis, Heracles and Apollo that were cast in bronze and gilded, he had them brought to our country to be set up in Armavir. The chief priests, who were of the Vahuni family, took those of Apollo and Artemis and set them up in Armavir; but the statue of Heracles, which had been made by Scyllas and Dipenes of Crete, they supposed to be Vahagn their ancestor and so set it up in Tarawn in their own village of Ashtishat after the death of Artashēs. But Artashēs, having subdued the land between the two seas, filled the ocean with the multitude of his ships, wishing to subject the whole west. Because great tumults were occurring in Rome, no one offered him strong resistance. But I cannot say from what cause arose a fearful turmoil, and the innumerable troops slaughtered each other. However, Artashēs fled and was killed, as they say, by his own army. He had reigned for twenty-five years. He also took form Hellas images of Zeus, Artemis, Athena, Hephaistos and Aphrodite, and had them brought to Armenia. But before they had arrived in our land the sad news of Artashēs’ death was heard. [Those bringing them] fled and brought the images to the fortress of Ani. The priests followed and stayed with them. 13. Testimonies from other historians to the empire of Artashēs and his taking Chroesus prisoner Those events are described by the Greek historians, not by one or two but by many. Being doubtful of these same events we made many researches because we heard from some histories that Cyrus had killed Chroesus and had destroyed the Lydian kingdom. And again the proposition of Chroesus to Nectanebo is reported. And this Nectanebo is said by Manetho to have been the last king of Egypt, while some have called him the father of Alexander. But we have found the period of Chroesus to be two hundred years before that of Nectanebo, while Nectanebo’s period is more than another two hundred years before that of Artashēs the First, king of Armenia.


105

But because there are many who say that our Artashēs took Chroesus prisoner and narrate this plausibly, I am so persuaded. Polycrates speaks as follows: “Artashēs the Parthian is for me superior to Alexander the Macedonian, because although he remained in his own country he ruled over Thebes and Babylon; and without crossing the River Halys he destroyed the Lydian army and captured Chroesus; and before arriving in Asia he was announced in the fortress of Attica. Alas for his fate! If only he had died in power and not in flight!”. In agreement with him speaks Evagaros: “The war of Alexander and Darius is minor compared of that of Artashēs. For the light of the day was obscured by the dust of the former, but the latter hid and darkened the sun by his volleys of arrows, turning midday into artificial night. He did not allow the Lydians to flee and bring the news, but even their King Chroesus he ordered to be placed in a cauldron. Because of him the torrents did not swell the river, for by the drinking [of his soldiers] it shrank to its winter level. By the multitude of his army he rendered the use of numbers inadequate so that there was need for measurement rather than counting. At this he was not boastful, but wept, saying: ‘Alas for this transitory glory’”. Scamadros also writes as follows: “The haughty Chroesus of Lydia was deceived by the reply of the Pythian oracle: “After passing the Halys river Chroesus will break the powers.” He thought that it referred to that of others, but he broke himself. For the Parthian Artashēs captured him and ordered him to be thrown into an iron cauldron. But Chroesus, remembering the saying of Sołon the Athenian, said in his own tongue: “O Sołon, Sołon, you said well that one should not call a man’s fate happy until his death.” When those standing nearby heard this, they told Artashēs that Chroesus was invoking some new god. Artashēs had pity and ordered him to be brought. And when he had inquired and learned what it was that he had cried out, he ordered him to be spared torments”. Phlegonios also writes: “The most powerful of all kings was the Parthian Artashēs. Not only did he put the Lydians to flight and capture Chroesus, but in the Hellespont and in Thrace he changed the nature of the elements. On the land he was borne along as if sailing the sea, over the sea he marched on foot. He threatened the Thessalians, and his repute made the Hellenes wonder. He destroyed the Lacedemonians, he put the Phocians to flight; the Locrians surrendered, the Bithynians were a part of his forces; all Hellas held him in awe. A short time later his disasters surpassed all others. Not so unfortunate was Cyrus warring against the Massagetae; not so many misfortunes did Darius suffer in flight from the Scythians, or Cambyses from the Ethiopians; insignificant was Xerxes’ march against Hellas with an army, when he left them his treasures and tents and alone escaped alive by a hair’s breadth. But he [Artashēs], overweening through his splendid victories, was slaughtered by his own army”.

106

Francesca Gazzano

Now I think that these accounts are worthy of belief, and that the Chroesus whom they say lived in the time of Cyrus and Nectanebo either is fictitious, or else there lived many kings with the one name – as is the custom of many. Movsēs Xorenac‘i, Patmut‘iwn Hayoc’ 2.12–13 (Italian translation and notes by G. Traina, 2013) 12. Campagna di Artašēs in Occidente. Prende Kriwsos¹ prigioniero, saccheggia le immagini degli idoli e le consegna all’Armenia Allora Artašēs dà ordine di allestire, (con truppe venute) da Oriente e da Settentrione, un esercito enorme, al punto da non conoscerne il numero. Ordina altresì che lascino una pietra nei luoghi di passaggio e di stazionamento per marcare il numero degli uomini, (formando) un cumulo di pietre a segno di questa moltitudine. Si dirige quindi a Occidente, e prende prigioniero Kriwsos, re dei Lidi². Nell’Asia trova statue di bronzo fuso ricoperte d’oro di Artemid, di Heraklēs e di Apołon, le fa portare nella nostra terra, perché vengano erette ad Armawir; ciò fecero i gran sacerdoti (ch’erano della stirpe dei Vahuni): eressero ad Armawir quella di Artemid e quella di Apołon. Quanto alla statua virile di Heraklēs, ch’era stata fatta da Skiwłos e da Dipinos il Cretese³, poiché pensavano che fosse quella del loro antenato Vahagn⁴, la eressero nel Tarawn, nel villaggio di Aštišat che faceva parte dei loro possedimenti, dopo la morte di Artašēs. Ma Artašēs, che aveva sottomesso la terra che si trovava tra i due mari, riempì l’Oceano con una moltitudine di navi, desiderando asservire l’intero Occidente. Poiché a Roma vi era turbolenza e confusione, nessuno gli si oppose vigorosamente. Tuttavia, non so dire in virtù di quale influsso, vi fu un tremendo tumulto di confusione, e le truppe di questa moltitudine si sterminarono a vicenda. Artašēs, fuggito, muore come dicono per mano dei suoi soldati, dopo aver regnato per venticinque anni.

1 La tradizione manoscritta presenta altre due varianti del nome: zKiwrsos (il ms. L riporta Kewrsos) e zKiwros: quest’ultima è un’evidente lectio facilior, che comunque non può essere tenuta, in quanto in contraddizione con la menzione di Ciro a 2.13. 2 Il gen. plur. Lidac‘woc‘ è emendato da alcuni manoscritti, con diverse varianti, in Libēac‘woc‘ (“dei Libyi, degli Africani”). Anche in questo caso sembra trattarsi di lectio facilior. 3 Varianti: Dipenos, Diponos. 4 Effettivamente la divinità guerriera armena Vahagn ha dei punti di contatto non solo con il Vǝrǝθraγna iranico attestato dall’Avesta, ma anche con l’Eracle greco: cf. Russell 1989; Mahé 1994, 782–783; Petrosyan 2007; Ivanov 2011; Russell 2013; Traina 2013.


107

In ogni caso egli aveva preso delle statue anche in Grecia: quelle di Dios⁵, di Artemid, di At‘ena, di Hep‘estos, di Ap‘roditē, le aveva fatte portare in Armenia. Esse non avevano avuto il tempo di penetrare nel paese, che si udì la triste nuova della morte di Artašēs. Fuggiti, gettarono le statue nel castello di Ani. E i sacerdoti, che si trovavano con esse, rimasero loro accanto. 13. Testimonianze dell’universalità di Artašēs, e di come mise Kriwsos in catene, da parte di altri storici Questo è raccontato da storiografi dei Greci non da uno solo o due, bensì da molti. Poiché abbiamo avuto dei dubbi in merito, abbiamo fatto una ricerca accurata. Giacché abbiamo appreso da certi storici che Kiwros ha ucciso Kriwsos e gli ha tolto il regno dei Lidi. E inoltre si racconta dell’opposizione tra Kriwsos e Nek‘tanebos. Questo Nek‘tanebos è l’ultimo re degli Egizi, come racconta Manet‘os, e alcuni dissero ch’era il padre di Ałek‘sandros. Noi troviamo che i tempi di Kriwsos si sono svolti duecento anni prima di quelli di Nek‘tanebos, e quelli di Nek‘tanebos duecento anni prima di Artašēs, primo re dell’Armenia. Ma giacché sono in molti a dire che Kriwsos fu catturato dal nostro Artašēs, e lo raccontano con un certo stile, io ci credo. Infatti dice così Polikratēs: “Artašēs il Parto mi sembra superiore ad Ałek‘sandr il Macedone. Infatti, pur tenendosi nella propria terra, ebbe il potere su T‘ebē e Babelon. Non aveva ancora attraversato il fiume Aliws, che fece a pezzi l’esercito lidio⁶, e catturò Kriwsos. Prima che arrivasse in Asia, veniva pubblicamente annunciato nella fortezza dell’Attikē⁷. Sorte ria! Fosse almeno morto mentre aveva il potere, e non fuggendo!”. Similmente a lui dice Evagaros: “Piccola è la guerra di Ałek‘sandr e Dareh, in confronto con quella di Artašēs. Infatti, ‘per via della polvere di loro non si vedeva nettamente la luce del giorno’. Ma lui, con le frecce scagliate, coprì il sole e lo mise in ombra, creando in pieno giorno una notte artificiale. Non lascia che i Lidi fuggano per dare l’annuncio, e ordinò anche di porre Kriwsos, il loro re, su una graticola. A causa sua, i torrenti non poterono ingrossare il fiume, poiché ci si abbeverava, scendendo ai livelli della magra invernale. D’altronde fece sì che le cifre sembrassero inadeguate agli eserciti, per via della loro moltitudine, sì che occorse ricorrere alle unità di misura più che ai numeri. Ma non se ne inorgogliva, anzi diceva tra le lacrime: ‘ahi gloria mia passeggera!’”.

5 Il contesto rivela la dipendenza da un testo greco: qui Mosé considera Διός come un’altra divinità, anziché come il genitivo di Ζεύς. 6 L’aggettivo liwdakan è trasformato in libēakan da alcuni mss.: cf. sopra. 7 Variante Ettikē.

108

Francesca Gazzano

Scrive anche così Skamadros⁸: “Si illudevano d’orgoglio i Lidi alla risposta della Pizia al quesito di Kriwsos: ‘Kriwsos, passando il fiume Aliws, distruggerà il dominio’. Pensando si trattasse di quello degli stranieri, se stesso egli distrugge. Poiché, catturato dal Parto Artašēs, questi lo mette in una graticola di ferro. Allora Kriwsos, ricordandosi dei discorsi di Sołon l’Ateniese, dice nella propria lingua: ‘O Sołon Sołon, bellamente dicesti di non felicitarsi per la buona sorte di un uomo, fino alla sua morte’. Ascoltandolo quelli che si trovavano nei pressi, raccontarono ad Artašēs ‘Kriwsos invoca un qualche nuovo dio’. Mosso a compassione, Artašēs lo fece condurre. Interrogatolo, e apprendendo ciò ch’egli gridava, ordinò di graziarlo delle torture. Scrive anche P‘łegonis⁹: “Di tutti i re, il più potente è stato il Parto Artašēs; non solo ha sconfitto i Lidi e messo Kriwsos in catene, ma nell’Ellesponto e in Tracia ha mutato la natura degli elementi; portato sulla terra, navigava, mentre sul mare camminava. Minacciando i Tessali, la sua fama sbalordiva l’Elleno; sgominò i Lacedemoni, mise in fuga i Focesi: i Locresi gli si arresero e i Beoti fecero parte del suo apparato. L’insieme dell’Ellade gli presentava il timore. Poco tempo dopo, le sue sventure superarono quelle di tutti. Non ebbe tali difficoltà Kiwros¹⁰, guerreggiando contro i Massageti, né tali passioni patì Dareh contro gli Sciti, e neanche Kambiwsēs contro gli Etiopi. Piccole sono quelle di K‘sersēs che nella campagna in Ellade abbandonò il tesoro e le tende, e fuggito scampò restando vivo. Ma lui, tronfio per le sue grandiose vittorie, viene massacrato dai suoi propri soldati”. Ora, io ritengo che queste storie siano degne di fede, e quel Kriwsos di cui narrano al tempo di Ciro ovvero di Nectanebo o è fasullo oppure hanno messo insieme vari re con un unico nome, come è abitudine di parecchi.

Bibliography Abełean, M. / Yarut‘iwnean, S. (eds.) (1981), Movsēs Xorenac‘i. Patmut‘iwn Hayoc‘, Tiflis, 1913, reprinted in R. W. Thomson (ed.) (1981), Moses Khorenats‘i, Patmut‘iwn Hayoc‘, Delmar-New York. Ackroyd, P. R. (1990), “The Biblical Portrait of Achaemenid Rulers”, in: H. Sancisi-Weerdenburg / J. W. Drijvers (eds.), Achaemenid History. V, The Roots of European Tradition, Leiden, 1–16. Alcock, S. E. / D’Altroy, T. N. / Morrison, K. D. / Sinopoli C. M. (2001) (eds.), Empires. Perspectives from Archaeology and History, Cambridge.

8 Variante Kamadros. 9 Varianti: P‘łegonios, P‘igonios. 10 Varianti: Kiwrsos, Kriwsos.


109

Amir-Arjomand, S. (1998), “Artaxerxes, Ardašir, and Bahman”, in: Journal of the American Oriental Society, 118/2, 245–248. Arafat, K. W. (1996), Pausanias’ Greece. Ancient Artists and Roman Rulers, Cambridge. Asheri, D. (ed.) (19974), Erodoto, Le Storie. I, La Lidia e la Persia, Milano. Beazley, J. D. (19632), Attic Red Figure Vase Painters, Oxford (online: www.beazley.ox.ac.uk/ index.htm). Börm, H. (2007), Prokop und die Perser. Untersuchungen zu den römisch-sasanidischen Kontakten in der ausgehenden Spätantike, Stuttgart. Boteva, D. (2011), “Re-reading Herodotus on the Persian Campaigns in Thrace”, in: R. Rollinger / B. Truschnegg / R. Bichler (eds.), Herodot und das Persische Weltreich / Herodotus and the Persian Empire. Akten des 3. Internationalen Kolloquiums zum Thema “Vorderasien im Spannungsfeld klassischer und altorientalischer Überlieferungen” (Innsbruck, 24.–28. November 2008) Wiesbaden, 735–759. Boyce, M. / Grenet, F. (1991), A History of Zoroastrianism. III, Zoroastrianism under Macedonian and Roman Rule, Leiden. Braccini, T. / Scorsone, M. (eds.) (2013), Flegonte di Tralle, Il libro delle meraviglie, Torino. Briant, P. (2002), From Cyrus to Alexander. A History of the Persian Empire, Winona Lake (IN) (transl. of Histoire de l’empire perse de Cyrus à Alexandre, Paris 1996). Briant, P. (2003), Darius dans l’ombre d’Alexandre, Paris. Burgess, A. (2006), “Apologetic and Chronography. The Antecedents of Julius Africanus”, in: Wallraff 2006a, 17–42. Burkert, W. (1985), “Das Ende des Kroisos: Vorstufen einer herodoteischen Geschichtserzählung”, in: C. Schäublin (ed.), Catalepton. Festschrift für Bernhard Wyss zum 80. Geburtstag, Basel, 4–15. Burkert, W. (2004), “Gyges to Croesus: Historiography between Herodotus and Cuneiform”, in: A. Panaino / A. Piras (eds.), Melammu Symposia. IV, Proceedings of the Fourth Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project (Ravenna, October 13–17, 2001), Milano, 41–54. Cargill, J. (1977), “The Nabonidus Chronicle and the Fall of Lydia: Consensus with Feet of Clay”, in: American Journal of Ancient History 2, 97–116. Carrière, A. (1893–1894), Nouvelles sources de Moïse de Khoren. Études critiques (1893); Supplément (1894), Wien. Chaumont, M.-L. (1986), “Armenia and Iran. II, The Pre-Islamic Period”, in: Encyclopaedia Iranica, II/4, 418–438. Contini, R. / Grottanelli, C. (eds.) (2005), Il saggio Ahiqar. Fortuna e trasformazioni di uno scritto sapienziale. Il testo più antico e le sue versioni, Brescia. Corcella, A. (ed.) (19992), Erodoto, Le Storie. Libro IV. La Scizia e la Libia, introd. e comm. di A. Corcella; testo critico di S. M. Medaglia; trad. di A. Fraschetti, Milano. Crane, G. (1996), “The Prosperity of Tyrants: Bacchylides, Herodotus, and the Contest for Legitimacy”, in: Arethusa 29/1, 57–85. Crevatin, G. / Lefons, C. (1991), “Veleggiar per monti: storie di viaggi impossibili”, in: G. Camassa / S. Fasce (eds.), Idea e realtà del viaggio, Genova, 411–426. Daryaee, T. (2006), “The Construction of the Past in Late Antique Persia”, in: Historia 55/4, 493–503. Daryaee, T. (2013), Cyrus the Great. An Ancient Iranian King, Santa Monica (CA). Dillery, J. (1999), “The First Egyptian Narrative History: Manetho and Greek Historiography”, in: Zeitschrift für Papyrologie und Epigraphik 127, 93–116.

110

Francesca Gazzano

Duplouy, A. (2000), “Le bûcher de Crésus. Exploitation et transformation d’une tradition iconographique et textuelle à travers le Ve siècle”, in: Rivista di Filologia e di Istruzione Classica, 128, 21–37. Eddy, S. K. (1961), The King is Dead. Studies in the Near Eastern Resistance to Hellenism 334–31 BC, Lincoln. Evans, J. A. S. (1978), “What happened to Croesus?”, in: The Classical Journal, 74, 34–40. Fales, F. M. (1993), “Storia di Ahiqar tra Oriente e Grecia: la prospettiva dell’antico Oriente”, in: Quaderni di Storia, 38, 143–166. Ferrari, F. / Bonelli, G. / Sandrolini, G. (eds.) (1997), Romanzo di Esopo. Milano. Fontenrose, J. E. (1978), The Delphic Oracle: its Responses and Operations with a Catalogue of Responses, Berkeley-Los Angeles-London. Foraboschi, D. / Pizzetti, S. (eds.) (2003), La successione degli imperi e delle egemonie nelle relazioni internazionali, Milano. Francis, E. D. / Vickers, M. (1984), “Green Goddess: A Gift to Lindos from Amasis of Egypt”, in: The American Journal of Archaeology 88/1, 68–69. Garsoïan, N. G. (1989), The Epic Histories attributed to P‘awstos Buzand (Buzandaran Patmoutiunk), Cambridge (Mass.). Garsoïan, N. G. (1996), “The Emergence of Armenia”, in: R. G. Hovhanissian (ed.), The Armenian People from Ancient to Modern Time, I, New-York, 37–62. Garsoïan, N. G. (2003–2004), “L’Histoire attribuée à Movses Xorenac‘i: que reste-t-il à en dire?”, in: Revue des Études Armeniennes 29, 29–48. Gazzano, F. (2013), “L’immagine di Creso nella tradizione post-classica. In margine al silenzio di Xanto”, in: V. Costa (ed.), Tradizione e trasmissione degli storici greci frammentari. II, Atti del Terzo Workshop Internazionale (Roma, 24–26 Febbraio 2011), Tivoli, 73–105. Gazzano, F. (2014), “Sovrani ‘barbari’ e santuari greci: qualche riflessione”, in: L. R. Cresci (ed.), Spazio sacro e potere politico in Grecia e nel Vicino Oriente, Roma, 119–162. Gnoli, T. / Muccioli, F. (eds.) (2007), Incontri fra culture nell’Oriente ellenistico e romano, Milano. Grabbe, L. L. (2004), A History of the Jews and Judaism in the Second Temple Period. I, Yehud: A History of the Persian Province of Judah, London-New York. Herrero de Jáuregui, M. (2008), The Protrepticus of Clement of Alexandria. A Commentary, PhD Diss., Bologna. How, W. W. / Wells, J. (1912), A Commentary on Herodotus, 2 vols., Oxford. Ivanov, V. V. (2011), “A Probable Structure of a Protoform of the Ancient Armenian Song of Vahagn”, in: Aramazd 6/1, 7–23. Jouanno, C. (2006), Vie d’Ésope. Livre du philosophe Xanthos et de son esclave Ésope. Du mode de vie d’Ésope, Paris. Koenen, L. (1985), “The Dream of Nektanebos”, in: The Bulletin of the American Society of Papyrologists 22, 171–194. Konstantakos, I. M. (2008), Ακίχαρος. Η Διήγηση του Αχικάρ στην αρχαία Ελλάδα, 2 vols., Αθήνα. Konstantakos, I. M. (2009), “Nektanebo in the Vita Aesopi and in Other Narratives: A Study in Comparison”, in: Classica et Mediaevalia 60, 99–144 Konstantakos, I. M. (2013), Ακίχαρος. III, Η Διήγηση του Αχικάρ και η Μυθιστορία του Αισώπου, Αθήνα. Kuhrt, A. (2007), “Cyrus the Great of Persia: Images and Realities”, in: M. Heinz / M. H. Feldman (eds.), Representations of Political Power. Case Histories from Times of Change and Dissolving Order in the Ancient Near East, Winona Lake (IN), 169–191.


111

Kurke, L. (1999), Coins, Bodies, Games and Gold: the Politics of Meaning in Archaic Greece, Princeton. La Bua, V. (1976), “Sulla fine di Creso”, in: Studi di storia antica offerti dagli allievi a Eugenio Manni, Roma, 177–192. Lanfranchi, G. B. / Roaf, M. / Rollinger, R. (eds.) (2003), Continuity of Empire (?): Assyria, Media, Persia, Padova. Law, H. H. (1948), “Croesus: From Herodotus to Boccaccio”, in: The Classical Journal, 43, 456–462. Lefèvre, E. (2010), “The Question of the Good Life. The Meeting of Cyrus and Croesus in Xenophon”, in: V. Gray (ed.), Xenophon (Oxford Readings in Classical Studies), Oxford-New York, 401–417 (Engl. transl. of: “Die Frage nach dem βίος εὐδαίμων. Die Begegnung zwischen Kyros und Kroisos bei Xenophon”, in: Hermes 99, 1971, 283–295). Lloyd, A. B. (1982), “Nationalist Propaganda in Ptolemaic Egypt”, in: Historia, 31/1, 33–55. Lloyd, A. B. (1994), “Egypt, 404–332 BC”, in: D. M. Lewis / J. Boardman / S. Hornblower / M. Ostwald (eds.), The Cambridge Ancient History. VI, The Fourth Century BC, 2nd ed., Cambridge, 337–360. Lombardo, M. (1990), “Erodoto storico dei Lidi”, in: G. Nenci / O. Reverdin (eds.), Hérodote et les peuples non grecs (Entretiens Fondation Hardt, XXXV), Genève, 171–203. Mahé, J.-P. / Mahé, A. (eds.) (1993), Moïse de Khorène, Histoire de l’Arménie (introduction, traduction de l’arménien et notes), Paris. Mahé, J.-P. (1994), “Un dieu guerrier à la campagne: l’exemple du Vahagn arménien”, in: Comptes rendus des séances de l’Académie des Inscriptions et Belles-lettres CRAI, 779–804. Mango, C. (1963), “Antique Statuary and the Byzantine Beholders”, in: Dumbarton Oaks Papers, 17, 53–75. Matthey, P. (2012), Pharaon, magicien et filou: Nectanébo II entre l’histoire et la légende, Thèse de doctorat, Univ. Genève. Millar, F. (2006), Rome, the Greek World, and the East. III, The Greek World, the Jews and the East, Chapel Hill. Moggi, M. (1973), “I furti di statue attribuiti a Serse e le relative restituzioni”, in: Annali della Scuola Normale di Pisa 3rd ser., 3, 1–42. Moggi, M. (ed.) (2003), Pausania, Guida della Grecia. VIII, L’Arcadia, testo e trad. a cura di M. Moggi; comm. a cura di M. Moggi e M. Osanna, Milano. Morris, I. / Scheidel, W. (eds.) (2009), The Dynamics of Ancient Empires. State Power from Assyria to Byzantium, Oxford. Moyer, I. S. (2011), Egypt and the Limits of Hellenism, Cambridge. Nenci, G. (ed.) (1998), Erodoto, Le Storie. Libro VI. La battaglia di Maratona, Milano. Parke, H. W. / Wormell, D. E.W. (1956), The Delphic Oracle, II, Oxford. Perdu, O. (2010), “Saites and Persians (664–332)”, in: A. B. Lloyd (ed.), A Companion to Ancient Egypt, I (Blackwell Companions to the Ancient World) Malden (MA), 140–158. Perry, B. E. (ed.) (1952), Aesopica: A Series of Texts Relating to Aesop or Ascribed to him or Closely Connected with the Literary Tradition that Bears his Name, Urbana. Petrosyan, A. (2007), “State Pantheon of Greater Armenia”, in: Aramazd 2, 174–201. Piccirilli, L. (ed.) (1980), Plutarco, La Vita di Solone, a cura di M. Manfredini e L. Piccirilli, Milano. Rhodes, P. J. (2003), “Herodotean Chronology Revisited”, in: P. Derow / R. Parker (eds.), Herodotus and his World: Essays from a Conference in Memory of George Forrest, Oxford, 58–72.

112

Francesca Gazzano

Roberto, U. (2006), “Julius Africanus und die Tradition der hellenistischen Universalgeschichte”, in: Wallraff 2006a, 3–16. Roberto, U. (2011), Le Chronographiae di Sesto Giulio Africano. Storiografia, politica e cristianesimo nell’età dei Severi, Soveria Mannelli. Rollinger, R. (2008), “The Median ‘Empire,’ the End of Urartu and Cyrus the Great’s Campaign in 547 BC (Nabonidus Chronicle II.16)”, in: Ancient East and West 7, 51–65. Russell, J. R. (1989), “Carmina Vahagni”, in: Acta Antiqua, 32/3–4, 317–330 = Id., Armenian and Iranian Studies, Cambridge (MA) 2004, 357–370. Ryholt, K. (1998), “A Demotic Version of Nectanebo’s Dream [P. Carslberg 562]”, in: Zeitschrift für Papyrologie und Epigraphik, 122, 197–200. Ryholt, K. (2002), “Nectanebo’s Dream or the Prophecy of Petesis”, in: A. Blasius / B. U. Schipper (eds.), Apokalyptik und Ägypten. Eine kritische Analyse der relevanten Texte aus dem griechisch-romischen Ägypten (OLA 107), Leuven-Paris-Sterling (VA) 2002, 221–241. Salmon, P. (1985), “Les relations entre la Perse et l’Egypte du VIe au IVe siècle av. J.-C.”, in: E. Lipinski (ed.), The Land of Israel: Cross-roads of Civilizations (OLA 19) Leuven, 147–168. Sargsyan, A. (2012–2013), “The Iranian Interpretation of the Greco-Persian Wars reflected in the History of Armenia by Movses Khorenatsi”, in: Iran-Nameh. Armenian Journal of Oriental Studies, 81–89. Scheer, T. S. (2003), “Die geraubte Artemis. Griechen, Perser und die Kultbilder der Götter”, in: M. Witte / S. Alkier (eds.), Die Griechen und der Vordere Orient. Beiträge zum Kultur- und Religionskontakt zwischen Griechenland und dem Vorderen Orient im 1. Jahrtausend v. Chr. (Orbis biblicus et orientalis 191), Göttingen, 59–86. Spiegelberg, W. (1914), Die sogenannte Demotische Chronik des Pap. 215 der Bibliothèque Nationale zu Paris, Lepizig. Stewart, A. (1990), Greek Sculpture. An Exploration, New Haven-London. Stoneman, R. / Gargiulo, T. (eds.) (2007), Il romanzo di Alessandro, I, Milano. Stoneman, R. / Gargiulo, T. (eds.) (2013), Il romanzo di Alessandro, II, Milano. Stramaglia A. (ed.) (2011), Phlegon Trallianus, Opuscula de rebus mirabilibus; de longaevis, Berlin-New York. Talamo, C. (1985), “Erodoto e le tradizioni sul regno di Lidia”, in: Storia della Storiografia 7, 150–161. Thomson, R. W. (ed.) (1978), Moses Khorenats’i, History of the Armenians, Cambridge (MA)-London. Toloni, G. (2013), “Ahiqar tra leggenda e rielaborazione letteraria. Una tradizione e i suoi riflessi”, in: Sefarad 73/1, 7–30. Topchyan, A. (2001), “Julius Africanus’ Chronicle and Movsēs Xorenac‘i”, in: Le Muséon 114, 153–185. Topchyan, A. (2006), The Problem of the Greek Sources of Movsēs Xorenac‘i’s History of Armenia, Leuven-Paris-Dudley (MA). Traina, G. (1988), “Scienza, tecnica e architettura nel mondo ellenistico-romano”, in: Civiltà classica e cristiana 9, 303–349. Traina, G. (1991), Il complesso di Trimalcione (Movsēs Xorenac‘i e le origini del pensiero storico armeno), Venezia. Traina, G. (1995), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. I”, in: Le Muséon, 108, 279–233.


113

Traina, G. (1998), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. II”, in: Le Muséon, 111, 95–138. Traina, G. (1999–2000), “Épisodes de la rencontre avec Rome (IIe siècle av. J.-C.-IIIe siècle ap. J.-C.)”, in: Iran & the Caucasus, 3/4, 59–78. Traina, G. (2013), “Beobachtungen zur Inschrift von Maris, Casiti filius”, in: Zeitschrift für Papyrologie und Epigraphik, 185, 279–285. Traina, G. (2015), “Tradition et innovation dans la première historiographie arménienne”, in: P. Blaudeau / P. Van Nuffelen (eds.), L’historiographie tardo-antique et la transmission des savoirs, Berlin-New York, 153–164. Traina, G. (forthcoming), “L'Arménie entre les Orontides et les Artaxiades”, in: Ch. Feyel / L. Graslin-Thomè (eds.), Antiochos III et l'Orient, Nancy. Van der Spek, R. J. (2014), Cyrus the Great, Exiles and Foreign Gods. A Comparison of Assyrian and Persian Policies on Subject Nations, in: W. F. M. Henkelman / C. E. Jones / M. Kozuh / C. Woods (eds.), Extraction and Control: Studies in Honor of Matthew W. Stolper (Studies in Ancient Oriental Civilization 68), Chicago, 233–264. Vannicelli, P. (2013), Resistenza e intesa. Studi sulle guerre persiane in Erodoto, Bari. Wallraff, M. (ed.) (2006a), Julius Africanus und die christliche Weltchronistik (Texte und Untersuchungen zur altchristlichen Literatur 157), Berlin-New York. Wallraff, M. (2006b), “Die neue Fragmentensammlung der Chronographie des Julius Africanus. Bemerkungen zur Methodik anhand einiger Dubia vel Spuria”, in: Wallraff 2006a, 45–59. Wallraff, M. / Roberto, U. / Pinggéra, K. (eds.) (2007), Iulius Africanus, Chronographiae. The Extant Fragments, transl. by W. Adler (Die griechischen Christlichen Schriftsteller, Neue Folge 15), Berlin-New York. West, S. (2003), “Croesus’ Second Reprieve and Other Tales of the Persian Court”, in: The Classical Quarterly n.s. 53/2, 416–437. Zecchini, G. (1988), “Una nuova testimonianza sulla translatio imperii (Aristosseno, Vita di Archita, fr. 50 Wehrli)”, in: Klio 70, 362–371.

Francesco Mari

Cyrus the Great in Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘: Telescoping the King* Abstract: Although scanty, the tradition on the Persian king Cyrus II “the Great” found in the Patmut‘iwn Hayoc‘ (History of Armenia) by Movsēs Xorenac‘i provides us with an original narrative, often contrasting with the events as related by Greek and Latin sources. Such a narrative is usually regarded as the result both of bad understanding and voluntary alteration by Movsēs Xorenac‘i of the account in the classical sources, with the main goal of exalting the role of Armenia in ancient Near East history. Despite these difficulties, in this paper we re-examine two episodes of the Patmut‘iwn Hayoc‘ concerning Cyrus (1.24–31, 2.11–13). Making careful use of cross-comparison with the Greek tradition, we hope to show that it is possible to extract from Movsēs Xorenac‘i’s text some pieces of original historical information, if not about events as they took place (but the Patmut‘iwn Hayoc‘ could be helpful to corroborate some historical hypotheses recently made concerning the Median kingdom on other bases), certainly about the formation of Iranic tradition on the events in question throughout the centuries.

In his Patmut‘iwn Hayoc‘, Movsēs Xorenac‘i mentions the Persian king Cyrus II “the Great” (approx. 600–530 BC) on two separate occasions that have no link to one another: 1) in relation to a king Tigran of Armenia, who was a contemporary of Cyrus; the context is that of the fall of the empire of the Medes (1.24–31). In this tale that displays a wealth of elements taken from the world of legend, Cyrus is presented as an ally and friend of the Armenian king, who undertakes a war against the Medes and their king Astyages. For despite having taken Tigran’s sister as his

* I would like to thank Dominique Lenfant, Francesca Gazzano and Giusto Traina for their infinite patience in re-reading and correcting this work, starting from the early phases of its drafting. I also owe a special word of thanks to Prof. Traina for his great help in providing me with the most literal and accurate translation possible of the Armenian original of Movsēs Xorenac‘i. Finally, I am grateful to Lara Pagani for her very concrete help. The reference edition for the Patmut‘iwn Hayoc‘ by Movsēs Xorenac‘i (abbreviation: MX) is that of Thomson 1981. Unless otherwise stated – and, in general, as far as concerns the spelling of the proper names, which I harmonise with the spelling conventions followed in this work – the translations of MX are taken from Thomson 1978 (the square brackets indicate Thomson’s interventions and integrations in the text; my own interventions are in italics).

116

Francesco Mari

bride, Astyages is secretly plotting to have the Armenian king murdered. But the latter, upon finding out about the plot, declares war on his father-in-law, defeats him in battle and conquers the entire empire “of the Medes and the Persians”; 2) in relation to the deeds of king Artašēs I (2.1113), whom Movsēs credits with the victory against Croesus of Lydia. Cyrus is cited in this passage only in order to reject the tradition that had become consolidated in ancient historiography ever since the time of Herodotus, namely the tale that casts Cyrus in the role of a protagonist in the capture of Sardis and the defeat and capture of king Croesus in battle, with the additional indication that Cyrus was instrumental in saving him from the death penalty, taking him into his own service instead. Basing himself on the testimony of four probable Greek Schwindelautoren, Movsēs associates these and other enterprises historically undertaken by Achaemenid sovereigns to the action of the Armenian Artašēs¹. At first glance, the historical material transmitted by Movsēs is fairly scanty. Indeed, the general impression it conveys to those who have a background in Greek texts of the classical period² (from which, in any case, the greater part of Movsēs’ information seems to derive) is that of coming across a rather jumbled tradition, probably the fruit of repeated misunderstandings not only by the author in interpreting his sources, but also by the sources themselves³. For instance, on the one hand Cyrus the Great is reduced to the role of supporting actor who merely assists an otherwise unknown king of Armenia. Movsēs attributes to this sover-

1 This passage is discussed by F. Gazzano, in this volume. The reader will find both passages in the Appendix at the end of this paper. 2 The main classical sources on Cyrus II of Persia have now been collected by Lenfant 2011. A separate issue concerns the Cyropaedia of Xenophon (to which we will shortly return) and the Biblical tradition, taken up by Flavius Josephus in the Jewish Antiquities (11.3–21). On Cyrus cf. also Briant 2002, 13–49, Daryaee 2013 and Waters 2014, 35–53. 3 Over the last two centuries little historical credence has been given to the references to classical antiquity in Patmut‘iwn Hayoc‘; rather, Movsēs Xorenac‘i has been castigated for impreciseness if not indeed branded as a forger (cf. the chapter Topchyan 2006 devotes to studies on Movsēs Xorenac‘i, 5–14: particular attention is paid to the criticism of Movsēs advanced by Gutschmid 1876 and Xalatjanc 1896 and 1903); Traina 1995, 279–233, underlines above all that the assessment in Thomson 1978 was influential in the development of the “hypercritical tendency” set in motion by Carrière 1894); most recently, see Traina 2015, 154–157. Recently, however, reasons have been put forward for greater indulgence towards Movsēs Xorenac‘i: cf. Traina 1995, 280 (the contribution, rich in bibliography and itself an example of an approach that is at one and the same time both rigorous and non aprioristic, underlines in particular the importance of the work of Mahé/Mahé 1993; the second part of the study can be found in Traina 1998); Topchyan 2006, 13 for his part, emphasizes the pioneering role of Sargsyan 1956, and then also cites the two Mahés as well as Traina 1991.

Cyrus the Great in Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘

117

eign not only the conquest of the Median empire, but also – or at least, this is the impression one derives from the mention of the “Greeks” – the subsequent westwards expansion of the kingdom. And on the other hand, some of Cyrus’ greatest feats are brought forward to a date of at least four hundred years later and then attributed to a king Artašēs of Armenia. One can, with considerable difficulty, make out in this king the figure of Artaxias I, who had already acted as strategos of the Seleucid king Antiochus III and who was subsequently the founder of the Arsacid kingdom of Armenia, knowledge of which has come down to us from Strabo⁴. But, unless we wish to view Movsēs as the most skilful fabricator of all, the multiple problems affecting this account of the events should be ascribed to the state of the sources the Armenian writer had at his disposal. Examined in this perspective, study of even a small, confused case such as that of the tradition concerning Cyrus of Persia in Movsēs Xorenac‘i can give rise to highly fruitful historical reflection. Naturally, a study of this kind will hardly be enlightening with regard to the figure of Cyrus himself; rather, the analysis presented in this paper may perhaps help to identify among the meanders of the text of Patmut‘iwn Hayoc‘ some narrative element of Iranian provenance that can serve to corroborate the Greek sources of the classical period.

1 Cyrus or Artašēs? Let us therefore start out from the second passage, concerning the deeds of Artašēs I. A first and very evident problematic aspect concerns the fact that Movsēs attributes to an Armenian king actions we know to have been carried out by Achaemenid kings. Setting aside, for the moment, the question of the fusion of numerous Achaemenid sovereigns in the figure of a single monarch, Movsēs’ real mistake is due to the equivalence between the Armenian name Artašēs and the Achaemenid name Artaxerxes. Despite the different spellings adopted by the

4 Strab. 11.14.15: Eἶθ᾽ὑπὸ τῶν Ἀντιόχου τοῦ μεγάλου στρατηγῶν τοῦ πρὸς Ῥωμαίους πολεμήσαντος διῃρέθη δίχα, Ἀρταξίου τε καὶ Ζαριάδριος· καὶ ἦρχον οὗτοι, τοῦ βασιλέως ἐπιτρέψαντος· ἡττηθέντος δ᾽ἐκείνου, προσθέμενοι Ῥωμαίοις καθ᾽αὑτοὺς ἐτάττοντο, βασιλεῖς προσαγορευθέντες. (“Then the country was divided into two parts by Artaxias and Zariadris, the strategoi of Antiochus the Great, who made war against the Romans; and these strategoi ruled the country, since it was turned over to them by the king; but when the king was defeated, they joined the Romans and were ranked as autonomous, with the title of king”. The translation is that of Jones 1928, ad loc., slightly modified). On Artaxias cf. Grainger 1997, 83 and Facella 2006, 192–198. On the formation of the kingdom of Armenia and on the role of the sovereign, see Traina 1999–2000, 61 and Traina, forthcoming.

118

Francesco Mari

Greek sources⁵, this equivalence between the names of the monarchs must have seemed a mere trifling matter to Movsēs. That he was unaware of the possible confusion is corroborated by his mention of some steles bearing the name of Artašēs⁶. Similar steles have effectively been found by archaeologists, and the Aramaic spelling of the name engraved on some of them does genuinely correspond to the form used for Artaxerxes in the Achaemenid chancery⁷. As has been convincingly demonstrated, this utilization of overlapping names no doubt reflected Artašēs’ precise intent to found the new Armenian monarchy in a tradition with a national character that would be clearly distinguishable from the legacy of the recent Seleucid government⁸. Thus in all probability it was the very propaganda set in motion by the king himself that would create the conditions, centuries later⁹, which were to enable an Armenian author to attribute to him the achievements of an Achaemenid namesake. However, this was not the only instance of overlapping names. To buttress his version, Movsēs quoted some Greek authors (who had never, or hardly ever, been

5 Spellings that render the Armenian variant (Strab. 11.14: Ἀρταξίας) and the old-Persian variant (Ἀρταξέρξης, in the fifth and fourth centuries BC authors) of one and the same anthroponym. 6 MX 2.56. Even though Movsēs traces these steles back to an Artašēs (II) distinct from the Artašēs (I) whom he deals with in 2.11–13, it is almost universally accepted that both these figures should be set in relation with Artaxias I. Cf. Sargsyan 2012–2013, 81: “The contemporary science in general either rejects the first of them or identifies both of them with the founder of the Artaxid dynasty Artashes I of Armenia (189–160 BC)”. 7 “A large number of these steles have been found, with inscriptions in Aramaic. The name of the king, ‘RTḤŠS[Y] in the inscriptions, corresponds to the Aramaic form of the name of the Achaemenid king Artaxerxes in an inscription of the first half of the 3rd [sic] century B. C. from Naqš-e Rostam [placed below the Elamitic text of DNb; the name in question is found in l. 20: cf. Frye 1982, 89], and the use of Aramaic may be seen as a survival of the chancellery Aramaic of the Achaemenid administration in Armenia” (Russell 1986). Cf. Perikhanyan 1971 and Khatchadourian 2007 (especially 48–55) and Perikhanyan 1971. 8 Cf. Khatchadourian 2007, 52–53 who sees an affinity between the “act of legitimation by association with Achaemenid grandeur” that can be recognised in the text of the inscriptions and the reminiscence of the Urartian past suggested by the shape of the steles. On Artaxias’ desire to present himself as the descendant of the prestigious Persian lineage of the Orontids, who had long governed the region, cf. Traina 1999–2000, 61–62; Facella 2006, 203. In actual fact, the archaeological and epigraphic data are in contrast with the reconstruction put forward by Movsēs Xorenac‘i, who (with his admittedly rather imaginative chronology) tries to set up a continuity starting from the rise to power of Artašēs, and therefore quotes oral traditions that indicate a dynastic change in Armenia following the war between Artašēs and the Orontids (MX 2.44–46). 9 There is no agreement among scholars on the date of the work by Movsēs Xorenac‘i. Here we choose to follow the opinion of those who, like Topchyan 2006, 12 n. 31 or Traina 1999–2000, 66, put forward credible arguments for maintaining the date of the fifth century AD (contra e.g. Russell 1986–1987, 254).


119

heard of elsewhere) in whose work Artašēs is described as a Parthian sovereign¹⁰. At a distance of roughly seven-hundred years after the facts, a feature of this kind can be explained by noting that: 1) the Arsacids of Armenia were a junior branch of the dynasty that had reigned in Iran from 247 BC to 224 AD; 2) from precisely the year 224 onwards, the Arsacids of Armenia were fully able to consider themselves as the legitimate heirs of the entire structure of the Iranian monarchy, whose throne had been usurped by the Sassanids¹¹. At this point, however, the question becomes rather more complex, since it is evident that it was, in primis, Movsēs’ sources that made the mistake of superimposing the figure of a Parthian sovereign on the Achaemenid kings. As a result of the blurring of distinctions, the feats attributed to Artašēs can be recognized as partially made up of elements the classical sources have transmitted to us as belonging to the biographies of Darius, Xerxes, Darius III and, naturally, Cyrus the Great, who represents perhaps the most striking case¹². It was pointed out above that Movsēs explicitly cites the name of the founder of the Achaemenid empire precisely in order to distance himself from the tradition that regarded Cyrus as an adversary of the Lydian king¹³. Thus rather than using the (Greek) sources that relate the canonical version of the events, Movsēs opted to use other (equally Greek) sources which displayed the particular feature of gathering together under a single name the historiographic material on many Achaemenid sovereigns, a name assimilable to that of Artaxerxes.

10 2.13, citing Polykrates: “Artašēs the Parthian”, and shortly afterwards in Skamadros: “the Parthian Artašēs” and in P‘łegonis: “the Parthian Artašēs”. Cf. Gazzano, in this volume. 11 For the main sources on these events, see Frye 2005, 461–464. Movsēs’ favourable view of the Arsacid dynasty is explicit in MX 1.22; as far as the aversion towards the Sassanid “Persians” is concerned, cf. e.g. the judgments by MX 1.α: “Why do you trouble us for those absurd and incoherent Persian stories, notorious for their imbecility? (…) What need do you have of these false fables; what use are these senseless and stupid compositions?” and β: “The stories and deeds of which we speak, and especially those whose recital offends our ears, (…) are hateful for us, for we did not speak of them in our first book, nor did we deem them worthy of inclusion even at the end of that account, but in a separate and special place” (see also Traina 2007). 12 For the details cf. the contribution by F. Gazzano, in this volume. 13 Cyrus is not, however, the only case of dissociation of the historical figure from his deeds, since the same phenomenon is found in the case of Xerxes, cited in 2.13. One reason for the possible confusion between the names of Xerxes and Artaxerxes concerns the influence of the book of Esther in the version of the LXX, where the Hebrew Ahashweros (corresponding to the Persian Khashayarsa, which the fifth century Greeks had rendered with Ξέρξης) is mistakenly translated as Ἀρταξέρξης.

120

Francesco Mari

What do we know about these sources? It is fairly certain that when he was drawing up this passage – and the same holds true for many other passages as well – Movsēs had at hand the work of Eusebius of Caesarea¹⁴, though he made every effort to dissociate himself from Eusebius’ chronology. We heard from some histories – Movsēs states – that Kiwros (i.e. Cyrus) had killed Kriwsos (i.e. Croesus) and had destroyed the Lidian kingdom. […] But because there are many who say that our Artašēs took Kriwsos prisoner and narrate this plausibly, I am so persuaded¹⁵.

Immediately following this assertion, Movsēs quotes from four Greek authors, whose historicity is, to say the least, uncertain: Polykrates, Evagaros, Skamadros and P‘łegonis¹⁶. It is in these sources – which Movsēs decides to put to use in order to substantiate his argument on the contemporaneousness of Artašēs and Croesus of Lydia – that we find the character of the “composite” Achaemenid sovereign bearing the name Artaxerxes. In this regard, Artyom Sargsyan has recently put forward the hypothesis that the material of the four authors cited by Movsēs does not derive – in contrast to what has sometimes been suggested – from the tradition of Greek rhetoric¹⁷, but rather from re-elaboration of a theme of Parthian monarchic propaganda dating back to the third century BC¹⁸. The Arsacids, Sargsyan argues, favoured the dissemination of the positive image of an archetypal Achaemenid sovereign, such as had already become widespread in earlier Iranian epic. According to this line of reasoning, by promoting this image the Arsacids aimed to establish a dynastic link between their own family and that of the Achaemenids and thus to restore the cultural integrity of the region, which

14 “Eusebius of Caesarea is a guarantee, which our blessed teacher Maštoc‘ had had translated into Armenian” (MX 2.10). On the use of Eusebius by Movsēs Xorenac‘i see Thomson 1978, 32 f. and Topchyan 2006, 33 and 63–64; on this passage in particular, see the reasons adduced by F. Gazzano, in this volume. 15 MX 2.13. 16 See Gazzano, in this volume. 17 Cf. the judgment of Mahé/Mahé 1993, 351 n. 5, who seem to think that Movsēs “fabricated” his sources by reworking some teaching material. 18 Sargsyan 2012–2013, 82–83: “We think the ancient version of Achaemenian origin of the Arsacids, put forward by the latter in proof of their right to the Iranian throne, may cast light on the problem. In general, this version, as far as we know, has been treated by the researchers within the framework of the dynastic policies of various houses of the Hellenistic period that claimed descent from the Achaemenids for the sake of their own prestige and legitimization [cf. supra, n. 8]. […] Hardly is it a mistake to think that for the Parthian Arsacids the idea of their Achaemenid descent was something more than for other contemporary dynasties with their local opportunities” (with previous bibliography).


121

had been shattered by the Graeco-Macedonian conquest¹⁹. A number of points of Sargsyan’s interpretation deserve to be further clarified²⁰. But if it were to be well-founded, his hypothesis would provide a convincing explanation for the formation, in the popular culture of a vast area of the Near and Middle East, of the figure of an Artaxerxes the Parthian similar to the one of whom Movsēs read in his sources²¹. These were composed of material reflecting a tradition that stemmed from an Iranian background, but this vulgata was passed down through the filter of Greek and Hellenistic historiography, as proven by the intermingled Herodotean reminiscences. However, there remains the fact that Polykrates, Evagaros and Skamadros are practically unknown sources for us. Only P‘łegonis has been recognised – though not without a degree of uncertainty – as Phlegon of Tralles²². On the basis of this identification, scholars have come to the conclusion that Movsēs Xorenac‘i probably had no first-hand knowledge of these authors, but that he had read of them

19 Cf. Sargsyan 2012–2013, 85: “The Parthian kings could deliberately maintain the collective character of their alleged epical progenitor, […] hereby emphasizing their concern for Achaemenian values in general. The collective image of Artaxerxes/Arsaces who personified the victories over the Greeks, and the idea of Iran’s integrity broken by the Greeks and now subject to restoration, must have been inseparably connected to each other”. 20 In effect, A. Sargsyan does not always recognise the traces of the classical tradition in the text by Movsēs (on which see Gazzano, in this volume); furthermore, his use of the work by Chariton of Aphrodisias does not seem appropriate when he maintains that “it is tempting to seek for some common roots of the images of Artashes in Movses’ History and, for example, of a Persian king Artaxerxes in Chariton’s novel Chaereas and Callirhoe (I century AD)” (Sargsyan 2012–2013, 89, n. 44): as is in fact known, a crucial element which distinguishes this novel from all the other surviving ones is the historical background of its plot (set in the fifth century Syracuse and involving the Syracusan general Hermocrates and the Persian king Artaxerxes II as characters) and he cannot be compared with the figure of the composite character that interests Sargsyan (cf. Borgogno 2005, 22 and, for the figure of Artaxerxes in the Greek novel, Daude 2001); finally, one should also note the poor understanding of Joseph. AJ 11.184, which is actually more an interpretation of the text of the LXX than a true piece of historical information (cf. infra, n. 35). 21 Sargsyan 2012–2013, 83 adduces numerous explanations as to why the epic crystallised precisely around the name of Artaxerxes. They concern both its frequency in Achaemenid onomastics and also the longevity of the sovereigns who bore the name, as well as ideological and religious reasons. On the basic equivalence of the name Artaxerxes with the Parthian dynastic name Arsaces, cf. Plut. Artax. 1.4 = Ctes. F 15b: ὁ δ’ Ἀρτοξέρξης Ἀρσίκας πρότερον ἐκαλεῖτο. The spelling “Arsica” is attributable to Plutarch, if we compare the passage with Phot. Bibl. 72, 43b 3 = Ctes. F 16, where the name is Ἀρσάκης. But cf. the misgivings of Lenfant 2004, 275 and Schmitt 1977. 22 FGrHist 257. He lived under Hadrian and was the author of a historical work bearing the title Olympiads. He is known above all for his paradoxographic production. Cf. the recent edition by Stramaglia 2011.

122

Francesco Mari

in the Chronographies of Sextus Julius Africanus²³, an author who was active in the Severian period and whom, in a passage appearing shortly earlier²⁴, Movsēs explicitly mentions as forming part of his background framework. Unfortunately, the fragmentary state in which the work of Julius Africanus has come down to us makes it impossible either to identify the specific references in Movsēs’ text, or to use Patmut‘iwn Hayoc‘ as a reliable witness for reconstructing the content of the Chronographies²⁵. The only indication linking the tradition on Artašēs used by Movsēs to Julius Africanus is, in effect, a fragment of the Chronographies. It contains a mention of the Olympiads by Phlegon of Tralles, which fixes the first year of Cyrus the Great at the beginning of the 55th Olympiad²⁶. This is somewhat insufficient to maintain that the work of a writer whom we know to have been interested in a thorough philological study of the texts²⁷ would have endorsed such a highly unorthodox tradition as that which replaced Cyrus the Great of Persia by Artaxerxes the Parthian. However, suggesting that Julius Africanus may have cited Polykrates, Evagaros, Skamadros and P‘łegonis is by no means equivalent to imagining that he pointed to them as accredited sources, or as examples to be followed. A passage from the Byzantine chronicler George Syncellus could help to shed some light on this problem. On the subject of Cyrus, Syncellus points out that “Esdra in his book calls him both Artaxerxes and Artasasthas, but the one who freed Esdra, thereby honouring Jerusalem, was another Artaxerxes, called Longimanus, grandson of Darius, son of Hystaspes”²⁸. This detail is designed to solve the contradiction rising from the fact that Esdra’s books relate that the reconstruction of the temple of Jerusalem was interrupted upon the order of Artaxerxes, and then resumed later during the second year of

23 Cf. Topchyan 2006, 84 and the paper by F. Gazzano, in this volume. The most recent edition of Julius Africanus is that of Wallraff/Roberto/Pinggéra 2007. Among the publications that have addressed this topic, see above all Wallraff 2007 and Roberto 2011. Cf. also Topchyan 2006, 65. 24 MX 2.10, to be read with Topchyan 2006, 77–78. On the implications of Africanus’ stay in Edessa, see Roberto 2011, 50 f. 25 “Unfortunately, the material in Xorenac‘i that seems to come from Africanus, with the exceptions of separate phrases, is not found in the surviving Greek fragments. Furthermore, Movsēs’ mention and knowledge of the source are unclear, and several questions must be answered” (Topchyan 2006, 66). 26 FGrHist 257, F 34, 17. Julius Africanus cited Phlegon on at least one other occasion, in connection with a full solar eclipse under Tiberius (F 39, 14–20). See Roberto 2011, 95 and 203–205 and Topchyan 2006, 84. 27 Roberto 2011, 40. 28 Sync. 282 Mosshammer: Τοῦτον ὁ Ἔσδρας καλεῖ Ἀρταξέρξην ἐν τῇ γραφῇ αὐτοῦ καὶ Ἀρτασασθάς. Ὁ δὲ τὸν Ἔσδρας ἀπολύσας μετὰ τιμῆς εἰς Ἱερουσαλὴμ ἄλλος Ἀρταξέρξης ἐστίν, ὅ λεγόμενος Μακρόχειρ, ἔγγονος Δαρείου τοῦ Ὑστάσπου; cf. also 291.


123

Darius²⁹, whereas Darius actually lived fifty years or so before Artaxerxes. Syncellus continues by citing the possible identification of Cambyses with the Nebuchadnezzar of the book of Judith, and his source for this statement is none other than Julius Africanus³⁰. Fundamentally, Africanus seems to be the best candidate as the source for the entire passage in Syncellus³¹. Thus one may begin to suspect Africanus may have cited Polykrates, Evagaros, Skamadros and P‘łegonis in order to confute them³². We know, for example, that one of the major issues to which the Chronographies devoted great effort was the attempt to reconcile the Graeco-Hellenistic historiographic tradition with the chronology that Christianity – of which Africanus was a follower – drew from the biblical tradition³³. Thus if one admits the existence of a passage of the Chronographies devoted to the bad habit whereby the Septuagint tended to confuse Cyrus with Artaxerxes³⁴, then the mention, in such a fragment, of the four authors cited by Movsēs does not seem impossible. In this case, the mention of Polykrates, Evagaros, Skamadros and P‘łegonis would have been intended as examples of a misunderstanding that could lead precisely to this type of confusion³⁵.

29 LXX, 1 Esd. 2.12–30 and 2 Esd. 4.7–24. No agreement has been reached yet on the real chronology of the events narrated in the books of Esdra (cf. Adler/Tuffin 2002, 343 n. 5). A plausible explanation is that the author did not organize his material according to a chronological approach but rather by proceeding according to thematic clusters (see BJ 544). 30 Sync. 282 Mosshammer. Cf. also Eus. Chron. 2.104. 31 Cf. Topchyan 2006, 83 n. 71. 32 See, as a parallel, the useful observations of Serbat 1973, 41–47 on the usus citandi of Pliny the Elder. 33 See Roberto 2011, 67–106. 34 This may have caused some confusion as early as in Joseph. AJ 11.184: Τελευτήσαντος δὲ Ξέρξου τὴν βασιλείαν εἰς τὸν υἱὸν Κῦρον, ὃν Ἀρταξέρξην Ἕλληνες καλοῦσιν. Here the very reading of the name Artaxerxes which we have presumed to have been given by Julius Africanus seems to be applied to the Book of Esther, although in the latter case the Ἀρταξέρξης of the LXX should be replaced with Xerxes (cf. supra, n. 13). In this case, a possible source of confusion may have arisen from LXX, 2 Esd. 4.6–7, where the Artaxerxes to be identified with Cyrus appears as immediately subsequent to Xerxes. 35 All the more so since there is no lack of evidence suggesting that on the question of the Medes and the Persians, Africanus followed the scheme which dominates the Greek historiography (already in Herodotus and, above all, in Ctesias); such a scheme had not been altered by Jew ancient historians. See Roberto 2011, 118.

124

Francesco Mari

2 Intermezzo: a hypothesis Finally, as a conclusion to this step of the discussion, we will take the liberty of formulating a hypothesis of our own as well. We stated earlier that the material contained in the four authors cited by Movsēs has every appearance of being an Iranian oral tradition re-elaborated in the light of Graeco-Hellenistic historiography. If the story was really so widespread in the Near-Eastern area, then it may not be so far-fetched to suggest that Julius Africanus, a native of Ælia Capitolina (the colony founded on the site of Jerusalem at Hadrian’s behest), had encountered it and had taken it upon himself to rectify its contents³⁶. And as far as Movsēs Xorenac‘i is concerned, could it be that faced with the need to make a choice between two alternative chronologies, on the one hand that proposed by Eusebius and Africanus, and on the other, the alternative narration he actually also found in the Chronographies, he was led to choose the latter precisely by the fact that he too, on the basis of oral tradition³⁷, was familiar with a similar story? Thus the very text of the Chronographies offered him the possibility of replacing the solution Africanus had chosen by another version which not only required no substantial change in the onomastics, but also allowed him to set a legendary story in a context that featured an Armenian sovereign. Thus Movsēs had the opportunity to rationalise, thanks to the support of the Greek sources, what he must have regarded as an Iranian tradition and, as such, rather unreliable³⁸. Naturally, here we are obviously in the field of speculation. However, in our view this explanation of the genesis of the story of Artašēs’ achievements does have some support in other passages of Patmut‘iwn Hayoc‘, as Movsēs’ penchant for euhemerism is a characteristic of his approach. More than once he is quite explicit in this regard and shows himself to be proud of his ability to defuse ancient legends and trace them back to events of Armenian national history³⁹.

36 See the observations by Roberto 2011, 44 on the stimulating context of the Roman Syria-Palestine and its great openness to the cultural wealth of the Near East. 37 On the “oral sources” in Movsēs Xorenac‘i, see Traina 1998, 100–104, who tends to view them as the collections of songs that constituted the manner of setting down in writing the mythological-religious heritage of pre-Christian Armenia. 38 It was Movsēs himself who declared his preference for Greek historiography, which he held in high regard on account of the qualities of rationality and precision that he could not discern in the Middle Eastern sources. The latter, in his view, were either inadequate or the outcome of an unreliable oral tradition. (Cf. MX 1.2–3). 39 On the subject of the Persian stories, cf. MX 1.β: “The stories and deeds of which we speak (…) today I shall set out by my own hand, giving a meaning to their irrationality. And behold I shall reveal their [the Persians’] most ancient events, which are incomprehensible to themselves, provided that you gain some pleasure from them or useful profit”. As far as the Armenian legends


125

One such case is that of the war between the Median king Astyages and Tigran of Armenia, to which we will now turn.

3 Cyrus, Astyages and Tigran: classical sources and Iranian orality As mentioned earlier, the episode of the war between the Median Astyages and the Armenian Tigran, with the Persian Cyrus reduced to a mere bit player, is completely unknown in Greek historiography. But one should also bear in mind that it is far from clear what source Movsēs Xorenac‘i used, although it is very likely to have been the mysterious and wise author of the four rhapsodies, cited by Movsēs shortly before in his work⁴⁰. We would argue that the hypothesis which sees this source as having set down in writing the Iranian epic heritage is corroborated precisely by the absence of a parallel Greek tradition, whose first “codification”, as far as we are aware, is to be found in Herodotus. Thus it is clear that the historian from Halicarnassus selected, from among the many versions he was familiar

are concerned, cf. in particular MX 1.31, where the figure of the god Vahagn is listed by Movsēs among the – mortal – offspring of king Tigran. Vahagn was a divinity of Iranian origin assimilated into the Armenian pantheon (cf. Avestan Vərəθraγna). Sun-like in his radiance, he was also considered equally responsible for the outburst of storms, when, “fishing” a dragon out of Lake Van, he dragged it up to the sky in order to kill it (on Vahagn cf. Mahé 1994 and Russell 1989, 320–330). His feats were celebrated in songs, from which Movsēs even cites a passage (see Ivanov 2011), before proceeding to explain that the legend originated from the achievements of one of Tigran’s sons, who was later deified and identified with the Greek Heracles. By so doing, Movsēs not only euhemerised an Iranian myth, but he also went so far as to dissociate it from the Armenians, even though the latter had included Vahagn among their main divinities (Mahé/ Mahé 1993, 341 n. 3). And in any case, an ancient Armenian cult akin to the contemporary Sassanid beliefs must have appeared fairly incongruous to a Christian historian such as Movsēs. With regard to Movsēs’ reliance on the example of the Greek myths, “which have hidden in themselves allegorically the meaning of the events”, see MX 1.α; cf. also Traina 1998, 103–104. 40 1.21. Cf. also 1.24 and 1.25: “Here what relates the one who narrates” (my emphasis). There has been extensive debate on this source, starting out above all from linguistic considerations. However, whatever the nature of the source, it seems fairly well established that it had an epic character: see Mahé/Mahé 1993, 64 nn. 3–4 and Traina 2001, 215; contra Thomson 1978, ad. loc., who identified it with Mar Abas Catina and translates: “which he [Mar Abas Catina] relates as follows” (idem at 1.26); but a little further on (1.29), Thomson himself ponders on a moot point: “Is Moses implying that he took this story from Mar Abas Catina, or from the “Web of Chries” [i.e. Girk‘ Pitoyic‘, a handbook of rhetoric on which see Muradyan 1998–2000] (I 19 n. 2), or from the “Four Rapsodies” (I 21, n. 2) that specifically deal with the reign of Tigran?” (ibid. 118 n. 1).

126

Francesco Mari

with, the most plausible account of the events concerning the end of the Median empire: From here, our story demands that we inquire further about Cyrus and the Persians: who was this man who destroyed the empire of Croesus, and how did the Persians become leaders of Asia? I shall write this account using as my sources certain Persians who do not intend to magnify the deeds of Cyrus but rather to tell what really happened, although I know of three other ways in which the story of Cyrus is told⁴¹.

Furthermore, one cannot disregard Movsēs’ explicit identification of the king Astyages with Aži Dahaka, the monstrous dragon of Persian mythology created by Angra-Mainyu (i.e. Ahriman) to destroy the material world⁴². This identification enabled Movsēs to explain the legends – those of Armenia, in this case – concerning the descendants of the dragon (višap) who are said to have inhabited the slopes of Azat Masis⁴³. Once Astyages had been defeated, so we read in Patmut‘iwn Hayoc‘, Tigran is said to have had Astyages’ wife and daughters deported

41 Hdt. 1.95: ῾Ως ὦν Περσέων μετεξέτεροι λέγουσι οἱ μὴ βουλόμενοι σεμνοῦν τὰ περὶ Κῦρον, ἀλλὰ τὸν ἐόντα λέγειν λόγον, κατὰ ταῦτα γράψω, ἐπιστάμενος περὶ Κύρου καὶ τριφασίας ἄλλας λόγων ὁδοὺς φῆναι (translation by A. L. Purvis in Strassler 2008, ad. loc.). 42 Mahé/Mahé 1993, 65. 43 Cf. 1.30: “because Aždahak, in our tongue is dragon”. The Armenian name of Astyages (Aždahak) fits with the name of Aži Dahaka (the serpent Dahak): “Dans la mythologie iranienne ce monstre a ordinairement pour adversaire le héros de l’orage Hroudên (Thraētaona). Mais, comme l’a montré Abeḷian ([Erker] 1966, t. I, [Erevan,] p. 134–8), Moïse réussit, grâce à un jeu de mots sur Mar, qui signifie à la fois “Mède” en arménien et “serpent” en iranien, à transformer le mythe en conflit entre le roi des Arméniens, Tigrane, et le roi des Mèdes, Adjahak, et à rattacher son récit aux légendes [arméniennes] sur les rois serpents ou dragons, qui, demeurant sur les pentes de l’Azat Masis, se font la guerre, montent à cheval, capturent ou prennent en otages les sœurs ou les épouses de leurs adversaires, tout comme des rois humains” (In the Iranian mythology the adversary of this monster is the storm hero Hroudên [Thraētaona]. However, as Abeḷian [(Erker) 1966, t. I, (Erevan,) p. 134–8] has shown, Movsēs managed to assimilate such a myth to the conflict between the Armenian king, Tigran, and the king of Medes, Adjahak, thus connecting his own narration to Armenians legends on snake-kings and dragons [Movsēs actually exploits the fact that the word Mar means both “Mede” in Armenian and “snake” in iranian]: Mahé/Mahé 1993, 338 n. 5). For these legends cf. also Traina 2011, 215–216. Movsēs himself gives a version of the Persian myth of Aži Dahaka in his chapter on “Persian fables” (1.β). Cf. also the independent oral tradition of the “rhythmic songs that the inhabitants of the wine-producing district of Gołt’n have lovingly preserved” (MX 1.30): “These include – Movsēs Xorenac‘i continues – songs about Artašēs and his sons, and they recall in allegorical fashion also the descendants of Aždahak, whom they call descendants of the dragon”. On the survival of the theme in the legends of the contemporary Kurdish nation, see Liverani 2003, 11–12.


127

(together with many thousands of prisoners) to none other than the region of that great mountain⁴⁴. Movsēs’ rationalisation has an indisputably nationalist character, which becomes quite evident in mistakes such as that of attributing to this Tigran a range of achievements performed by the far more famous Tigranes the Great, an ally of Mithridates Eupator in the war against the Romans (first century BC), such as the founding of Tigranocerta⁴⁵. This notwithstanding, if the tradition that has been incorporated into this passage of Patmut‘iwn Hayoc‘ could be traced back, in the last instance, to oral traditions, then it would perhaps be possible to draw from Movsēs’ tale some genuinely Iranian historical elements on the Medes and sixth century BC Persia. Such elements could then be compared with the Herodotean account which – while remaining the main source of information on the topic – is not devoid of problematic aspects that have increasingly often been underlined⁴⁶. Moeover, Herodotus himself declares he has made a choice from among four local versions concerning the overthrow of the Median empire at the hands of the Persians, and the Greek tradition is far from being univocal. The other two main Greek sources on these events, namely the fragmentary Assyriaka by Ctesias of Cnidos and Xenophon’s Cyropaedia, give different accounts of the course of events. But if one conjectures that not only these various texts but also the History of Armenia may preserve traces of the material Herodotus decided against including, then it becomes possible to draft a comparative assessment. Naturally, given the almost thousand-year span of time separating Herodotus’ investigations and the redaction of the Patmut‘iwn Hayoc‘, great caution should certainly be exercised in dealing with the material, but this does not imply that it is impossible or illegitimate to make an attempt in this direction. First of all, it has been made clear that the tradition examined here was known to Movsēs in the form of songs, which had in all likelihood also been set down in writing⁴⁷; secondly, both in the case of Herodotus and of Movsēs what we find ourselves dealing with is epic material⁴⁸. Thus possible comparisons will not be made between plots, but rather

44 “So are you not now more amazed at the truth of our history, how we have revealed the secrets of the dragons who live up on Noble Masis?” (MX 1.30). 45 MX 1.30. On Tigran the Great in the work of Movsēs Xorenac‘i see Breglia Pulci Doria 1979 and Geller/Traina 2013. 46 Cf. recently, Rollinger 2010, 64–68. See also Liverani 2003, 1–4. More generally, Bichler 2010, 156–161. 47 Cf. supra, n. 37. 48 We may perhaps take the liberty of benefiting from the metaphor of the large furnished house which, already fifty years ago, Eric Havelock used to describe Homer’s poetic practice: the oral poet “chooses a winding and leisurely route which shall in the course of a day’s recital allow him to touch and handle most of what is in the house. The route that he picks will have its own

128

Francesco Mari

between recurrent elements, that is to say, between some of the structures of these tales, and such structures are known to have a formidable capacity to withstand the flow of time⁴⁹. Indeed, there is no lack of elements of contact between the different versions, and it is these elements that are most likely to represent epic patterns revelatory of the oral origin of the tradition on Cyrus the Great, Astyages and the end of Median power. Furthermore, as shown by the table below, many of the narrative elements are shared by more than one version, although fairly few are common to all versions⁵⁰. For instance, while a Median empire is present throughout all four versions, the motif of Cyrus as Astyages’ young cup-bearer is mentioned only by Ctesias and the Cyropaedia; also, although the prophetic dream is absent from Xenophon’s narration, it clearly constituted a central part of the legends concerning Cyrus, despite being re-interpreted in various ways in the different traditions, which attribute the nocturnal vision now to Astyages, now to Cyrus’ mother. Overall, one has the impression of being faced with different ways of assembling narrative elements that are to some extent independent of one another, although they can all be traced back to what one might call the “legend of Cyrus”. Thus the interwoven web of events found in Herodotus, Ctesias, Xenophon and Movsēs is reminiscent of a procedure of narrative construction akin to what is known with regard to other great epic works, from the saga of Troy to the stories in the Breton Cycle. Moreover, such an interpretation is supported by Herodotus’ statement, quoted above, that he was aware of another three versions of the story of Cyrus in addition to the one he chose (the λόγων ὁδοί)⁵¹. In short, during the fifth century

design. This becomes his story, and represents the nearest that he can approach to sheer invention. This house, these rooms, and the furniture he did not himself fashion” (Havelock 1963, 88–89, but cf. also Assmann 1997, 68 f.). 49 Cf. Levine Gera 1991, 13–22: see in particular the example of the persistence, in the Sassanid epic (Kār-Nāmag ī Ardašīr ī Pābagān) and also in epic of a later date (Šāhnāma) on the childhood of Ardashir I, of modules that already formed part of the story of Cyrus, as we know it from Nicolaus of Damascus (FGrHist 90 F 66), who drew his material from the Persika of Ctesias (18 f.). Among the references, found in ancient texts, to an oral Achaemenid epic, see Hdt. 1.132; Strab. 15.3.18; FGrHist 690 F 9 = Ath. 633d-e. For the traces of oral tradition in medieval Iranian epic, see Davidson 1994, 54–72. 50 However, it should be kept in mind that the Greek sources exerted reciprocal influence on one another. Consequently, some elements that are common to more than one version should certainly be attributed to this phenomenon: cf. e.g. the name of Cyrus’ mother in the Cyropaedia (Cassandane), which Xenophon certainly drew from Herodotus’ Histories. For Xenophon’s sources, cf. Due 1989, 117–146; for those of Ctesias, see Lenfant 2004, xxvii–xxxix. 51 In this regard, cf. also the Italian translation by V. Antelami (nom de plume of Augusto Fra-


129

BC the character of the material on which Herodotus drew for his Medikos logos must have been fairly fluid. But let us now return to Patmut‘iwn Hayoc‘. If we compare the narrations in the classical sources with the version offered by Movsēs Xorenac‘i, we note that the more we become aware of points of contact with Xenophon’s Cyropaedia, the less original Movsēs’ account seems. The most interesting passage for a comparison with Patmut‘iwn Hayoc‘ is the one in which Cyrus has the Armenian king “put on trial” because the latter is deemed guilty of rebellion against the Medes. As is well known, the Cyropaedia is a very difficult work to utilize as a source for Persian history. Indeed, some historians have preferred to ascribe it more to the category of the romance than the historical chronicle⁵². Note, in this regard, that Xenophon’s intention was not that of writing history: on the contrary, he sought to portray for his contemporaries a model of the perfect governor, to which Cyrus merely lent a face⁵³.

schetti) in Asheri 1988, I 95 (“tre tipi diversi di storie”), with the explanation by Asheri himself, ad. loc.: the four “types” of stories known to Herodotus, must have been versions of Median or Persian origin, already elaborated and rationalized by previous Greeks: the “type” that Herodotus chooses, as well as those chosen by Xenophon (popular sagas and songs about Cyrus: Cyr. 1.2.1), by Ctesias (FGrHist 688 F 66), and by the author summarized by Diodorus seem to suggest as much, the latter being himself obliged to try to find his way through the labyrinth of a number of unreconcilable traditions: διαφωνοῦσιν οἱ παλαιότατοι τῶν συγγραφέων περὶ τῆς μεγίστης τῶν Μήδων ἡγεμονίας (2.32). 52 An entire chapter of Schmeling 1996 (581–600) is devoted to the Cyropaedia. Cf. Knauth 1975, 12: “So kommt man zum didaktischen Roman” and Due 1989, 19: “Xenophon has a literary, a dramatic point to make. He is not writing history”. But not all the opinions are so negative: on the contrary, recent more favourable reappraisals of Xenophon’s literary achievements have seen a number of proposals suggesting that by adopting a more attentive and carefully thought-out approach to his work, the Cyropaedia could be a useful source to reconstruct not so much the historical events as, rather, a part of the Achaemenid tradition on the rise of Persian domination in central Asia (Tuplin 1996 and 2013; Azoulay 2004, 148). On the Cyropaedia in general see, in addition to the above cited Due 1989, also Tatum 1989; Nadon 2001. 53 Xen. Cyr. 1.1.3: ἀνθρώπῳ πεφυκότι πάντων τῶν ἄλλων ῥᾷον εἴη ζῴων ἢ ἀνθρώπων ἄρχειν. Ἐπειδὴ δὲ ἐνενοήσαμεν ὅτι Κῦρος ἐγένετο Πέρσης, ὃς παμπόλλους μὲν ἀνθρώπους ἐκτήσατο πειθομένους αὑτῷ, παμπόλλας δὲ πόλεις, πάμπολλα δὲ ἔθνη, ἐκ τούτου δὴ ἠναγκαζόμεθα μετανοεῖν μὴ οὔτε τῶν ἀδυνάτων οὔτε τῶν χαλεπῶν ἔργων ᾖ τὸ ἀνθρώπων ἄρχειν, ἤν τις ἐπισταμένως τοῦτο πράττῃ. (“For man, as he is constituted, it is easier to rule over any and all other creatures than to rule over men. But when we reflected that there was one Cyrus, the Persian, who reduced to obedience a vast number of men and cities and nations, we were then compelled to change our opinion and decide that to rule men might be a task neither impossible nor even difficult, if one should only go about it in an intelligent manner” [transl. by Miller 1925, ad. loc.]). Cf. already Cic. QFr. 1.1.23: “Cyrus ille a Xenophonte non ad historiae fidem scriptus, sed ad effigiem iusti imperii”.

130

Francesco Mari

Table 1: Different traditions on Cyrus the Great. HERODOTUS

CTESIAS (FGrHist 688)

Four Median kings mentioned, tot. 128 years (1.102, 1.130) + 28 (1.106)

After the fall of the Assyrian kingdom, hegemony of the Medes (9 sovereigns for 282 years) (F 5)

Invasion of the Scythians (28 years) (1.106)

War with the Cadusians (War with the Sacae and revolt of the Parthians) (F 5)

Astyages is the son of Cyaxares (1.107)

Astyages is the son of Astibares (F 5)

Cyrus is the husband of Cassandane (2.1)

Cyrus is not the grandson of Astyages, but he marries Amytis, the daughter of Astyages (F 9 § 1)

Cyrus is the son of the Persian Cambyses and of Mandane the daughter of Astyages (1.107)

Cyrus is the son of Argoste (F 8d) and Atradates, and he is subsequently made the satrap of Persia

Prophetic dream attributed to Astyages, who dreams of his daughter Mandane (1.107)

Prophetic dream attributed to Argoste (F 8d)

Cyrus survives the attempt by Astyages to get rid of him. He grows up as a shepherd and returns to his family in Persia when he is already almost a teenager (1.109.122)

Cyrus arrives at the court as a servant and subsequently enters into the service of Astyages while he is still a boy (F 8d)

Cyrus becomes Astyages’ cup-bearer (F 8d)

Cyrus causes a rebellion of the Persians (1.126–127)

War between Astyages and Cyrus’ Persians (F 8d)

Astyages dies of natural causes in Cyrus’ house (1.130)

Astyages dies at the hand of the eunuch Petesakas and of Oibaras (F 9 § 6)

Ninos king of Assyria launches a campaign against the Armenians (F 1b), who become his vassals

Cyrus’ conquests

Cyrus’ conquests

Cyrus dies in the North-East of the empire (vs. Massagetae) (1.214)

Cyrus dies in the North-East of the empire (vs. Derbikes) (F 9)


131

XENOPHON

MOVSĒS XORENAC‘I

The war of the Assyrians against the Medes and the Persians breaks out upon the death of Astyages (1.5.2)

8 kings of the Medes, of whom the last 4 = Hdt. (1.22)

Astyages is the son of a Cyaxares, and the father of a Cyaxares Cyrus marries the daughter of Cyaxares (8.5.17)

Cyrus is the son of Cambyses and Mandane (1.2.1)

Prophetic dream attributed to Astyages (1.26)

Astyages summons to court his daughter Mandane, with Cyrus, in order to get to know him. Cyrus stays at court for many years until his father calls him back (1.3.1–1.4.26) Cyrus says he can be a better cup-bearer than Astyages’ cup-bearer (1.3.9) No war but normal succession to the throne (8.5.17)

Astiages starts to plot against Tigran (1.27)

Astyages dies in his own bed (1.5.2)

Tigran kills Astyages in battle (1.29)

Cyrus has the physical and moral qualities typical of the model Iranian monarch (1.2.2)

Tigran has the physical and moral qualities typical of the model Iranian monarch (1.24)

Armenia is an independent kingdom forced by the Medes to pay tribute (2.4.22). In the era of Cyrus, a rebellion arises in Armenia (2.4.9)

The kingdom of Armenia is independent, although Astyages is married to Tigran’s sister, Tigranuhi (1.24)

Cyrus defeats the king of Armenia and the country is once more forced to pay tribute (3.1.40). Cyrus’ conquests

Tigran conquers the Median empire and “subjugates the Greeks” (1.24), with the aid of Cyrus

Cyrus and Tigran are childhood friends because they used to go hunting together (3.1.38)

Tigran and Cyrus are friends and allies (1.25–26)

Cyrus dies in his own bed (8.7)

132

Francesco Mari

The same fragment that concerns us here is in actual fact an example of Socratic literature, which Xenophon includes within a narrative framework that has a Persian atmosphere: as such, very little credence can be given to his account⁵⁴. However, as long as one is not tempted to use it for purposes of reconstruction of the events, this passage remains an interesting touchstone for comparison, since it is highly probable that Xenophon built up the setting for the dialogue by drawing inspiration from Persian sources. This would not be the first instance of such a technique in the Cyropaedia⁵⁵. Here below is a list of the information on which Xenophon and Movsēs seem to concur and which is not found in either Herodotus or Ctesias: 1) existence of a kingdom of Armenia which, if not fully independent, was at least distinct from that of the Medes; 2) existence of a Tigran of Armenia contemporary with Astyages and hostile to the Medes; 3) rebellion of Armenia against Media during the time of Cyrus; 4) personal alliance between Cyrus and Tigran. Meagre though this information may be, it offers a starting point for reflection. Firstly, the two texts support each other, thereby strengthening the hypothesis put forward above of a tradition that both of them, independently of one another, utilized as a source⁵⁶. This tradition was not integrated into Greek historiography, as the latter acted according to its own conventions and used them as a criterion for judging the reliability of the data to be included in the account of the historical events⁵⁷. Secondly, if this information from Xenophon and Movsēs is felt to

54 Cf. Levine Gera 1991, 78–98 and Mueller-Goldingen 1995, 150–164; see also the contribution by G. Gaggero, in this volume. 55 See Xen. Cyr. 1.2.1: “Even to this day the barbarians tell in story and in song that Cyrus was most handsome …” (φῦναι δὲ ὁ Κῦρος λέγεται καὶ ᾄδεται ἔτι καὶ νῦν ὑπὸ τῶν βαρβάρων εἶδος μὲν κάλλιστος.); cf. also Cyr. 1.4.25. For Xenophon’s use of oral stories heard from Persians, cf. Knauth 1975, 30–31 and the argument put forward in Lenfant 2001. See also the interesting, though at times somewhat excessive, Parpola 2003. 56 If, that is, Movsēs did not draw inspiration from Xenophon, a hypothesis for which we find no solid base. Furthermore, even Gaggero, in this volume, who gives the impression of moving in this direction, then goes on to clarify that – while maintaining the actual name of the Armenian prince – Movsēs’ treatment of Tigran seems to draw on an indigenous source. This is too little to assume that Movsēs Xorenac‘i had direct knowledge of the Cyropaedia, or to rule out the possibility that Xenophon himself had drawn on an Iranian source akin to that exploited by Movsēs. 57 The reference, in particular, is to the idea of the translatio imperii, see Wiesehöfer 2003; Rollinger 2010, 63; Roberto 2011, 114; Gazzano, in this volume.


133

be acceptable, this would reinforce not only the hypotheses that have recently cast doubt on the unitary nature of the Median empire as it is described by Herodotus, but would also give further impetus to the doubts concerning the date at which Urartu was overpowered by the Medes⁵⁸. Moreover, one may speculate that the prolonged period affected by Scythian forays into Median territory⁵⁹ possibly acted as a factor strengthening the bonds among the different princedoms attested by the Assyrian sources⁶⁰. If this conjecture is viable, then the memory of such a phenomenon – on the nature and duration of which one can only hazard the vaguest of guesses – could easily have eventually become blended into the tradition on which Herodotus drew for his Medikos logos, and it would simply have taken its place alongside the elements dating back to the Achaemenid period. As far as concerns the definitive downfall of Urartu (which would later appear under the name of satrapy of Armina in the empire of Darius⁶¹), this is generally considered to have occurred prior to the battle of the Halys fought between the Medes and the Lydians in 585 BC⁶², yet there is no mention of it in the sources. In short, the data appear to be insufficient to set a precise date either on the moment when the kingdom of Urartu was subjugated to the empire (whether Median or Persian or other) or on its reorganization from a tributary country to a satrapy properly speaking. Moreover, at the time of the clash, the very name of Urartu for the kingdom surrounding Lake Van – and consequently the identity of the

58 The first, serious, discussion of the classical tradition on the Median empire is owed to Heleen Sancisi-Weerdenburg: see above all 1988 and 1994; her radical claim, namely that the empire of the Medes did not exist other than in Herodotus’ inkwell, has led to extensive debate, culminating in a conference in 2001 (Lanfranchi/Roaf/Rollinger 2003). The positions taken by the major scholars range from staunch negationism (Liverani 2003) to confident assertionism (Roaf 2003). However, the main problem still concerns the total absence of the Medes from Mesopotamian sources for the period extending from 612 to 550 BC, which raises suspicion but makes any definitive conclusion impossible. Given the circumstances, one cannot but proceed with extreme caution, assuming as the most likely scenario the hypothesis of “light” and decentralised Median power, with the additional assumption that the account of the events was based primarily on Middle Eastern sources (Stronach 2003, 234; Rollinger 2008) or on a comparison between the latter and the Herodotean testimony (Tuplin 2004). 59 Hdt. 1.103–106. Cf. also the allusions (admittedly somewhat unclear in relation to Herodotus’ version, but concerning the same span of years) to the fear aroused by the Scythians in the book of Jeremiah (passim). 60 Radner 2003, 38 f.; for the Babylonian sources on the Medes, cf. Jursa 2003. For an overview, see Rollinger 2010, 68–81. 61 Cf. above all the inscription DB § 1.6 and Hdt. 3.93.1. But see Asheri 1990, 382–383 and Petit 1990. 62 Burney 2007, 94 dates the events to approx. 590 BC.

134

Francesco Mari

peoples who dominated it – are now to be considered as seriously in doubt. In effect, it is likely that the disintegration of the central Urartian power should be pushed back by several decades in comparison to the traditional date, and should be seen in relation with both the above mentioned Scythian incursions as well as the migratory phenomenon that was to result in the arrival of Armenian people properly speaking, who had migrated from a westerly direction. Upon settling in the region, these people would subsequently establish some sort of organization or form of government, albeit rudimentary and imperfect⁶³. Thus if – instead of direct control exerted by the Medes in the Middle East – one were to accept the hypothesis of the existence of a multiplicity of state-like formations that were under the sphere of influence of Ecbatana in one way or another, then the historical existence of a Tigran of Armenia who paid tribute to the Medes or was a relative of the Median sovereign (at least in the same manner as a Cyrus “king of Anšan”⁶⁴) is a possibility which, on the basis of the tradition transmitted by Xenophon and Movsēs Xorenac‘i, should once more be taken into consideration.

63 See again Burney 2007, 95–96 who argues that a general pattern of this kind is extremely plausible. Very interesting, in this line of interpretation, is also the suggestion by Chaumont/Traina 2007, 101–102, who (partly also on the basis of toponomastics) suggests that the Χαλδαῖοι, whom Xenophon describes as adversaries of Tigran at the time of Cyrus’ expedition into Armenia (Cyr. 3.2.1 f.; not to be confused with the Chaldeans of southern Babylonia) are the romanticized reflex of the historical Urartian peoples, gradually driven back towards the mountains and subsequently absorbed by the new inhabitants of the district. But cf. already Parpola 2003, 342. 64 Cf. Zournatzi 2013 and forthcoming. On the titles of Cyrus the Great, see also Lincoln 2007, 6–7; on the specific point of the title of “king of Anšan”, see Schaudig 2001, 26.


135

Appendix¹ Movsēs Xorenac‘i, History of Armenia, 1.24–31 24. But let us now pass on to discuss Tigran and his deeds, for he, of all our kings, was the most powerful and intelligent and the most valiant of these and of all others. He assisted Kiwros in overthrowing the dominion of the Medes, and he brought the Greeks into subjection to himself for no little time. He extended the borders of our territory and established them at their extremes limits in antiquity. He was envied by all who lived in his time, while he and his epoch were admired by posterity. (…) [He was] blond with greyflecked hair, of ruddy complexion and gentleeyed, personable and broad shouldered, strong legged and with noble feet, continent in eating and drinking and orderly at feasts, and – as those among our ancient who sang to the p‘andir used to say – moderate in the pleasure of the flesh (…). He was just and equal in every judgement, and he weighed all the circumstances of each case impartially. He did not envy the noble nor he despise the humble, but over all alike he spread the mantle of his care. At first he was allied with Aždahak who was a Mede, and gave him his sister Tigranuhi in marriage after Aždahak had sought her with insistence (…). For he [Tigran] was an object of suspicion to him since an unexpected prophecy [had revealed] to him his future destiny. (…) 26. In those days, says [Mar Abas Catina], there was no little danger to Aždahak the Mede from the alliance of Kiwros and Tigran. For that reason, and from the severe agitation of his thoughts, a vision appeared to him during his sleep at night, [the like of] which in his waking hours he had never seen with his eyes or heard with his ears. (…) Accordingly, Aždahak plans to have Tigran killed 28. (…) Thus, a messenger arrived and accomplished [the mission] concerning the beautiful maiden, for Tigran agreed and gave his sister Tigranuhi in marriage to Aždahak (…). Aždahak received her, and not only because of the deceit of his heart but also because of her beauty made her the first in rank of his wives, though underneath he was spinning an evil web.

1 Translation by Thomson 1978, revised and corrected by Giusto Traina.

136

Francesco Mari

29. (…) But the prudent and beautiful woman discerned his treachery (…) and swiftly informed her brother of the deceit through friends. Thenceforth [Aždahak] set to work, [suggesting to Tigran] through an embassy that they should meet midway on the borders of their two realms for mutual discussions, as if some important matter or business had arisen that could not be treated by means of a letter or an exchange of messengers but only if the two met face to face. However, Tigran knew the purpose of the message and hid nothing of Aždahak’s plot, but in a letter revealed what was in depths of his heart. And once such baseness had been revealed there was thenceforth no excuse or deceit that could veil such wickedness, but then war broke out openly. The Armenian king gathered [troops] (…) and marched with all his host to the borders of Media. The danger then forced Aždahak to oppose the Armenian with no less a force. (…) When the battle was joined, with his lance Tigran split Aždahak’s iron armor like water; the point of the lance ran right through him, and as he drew it back again he brought out with his weapon half of his lungs. The combat was magnificent, for heroes were facing heroes, and not straightaway did they turn their backs to each other. Therefore the struggle was drawn out over long hours. But the death of Aždahak brought it to an end. And this feat, added to his good fortune, increased Tigran’s glory. (…) 31. (…) There are many called Tigran, but only one descended from Hayk, who killed Aždahak and led into captivity his house and Anoyš the mother of dragons, and with the willing help and the encouragement of Kiwros seized for himself the empire of the Medes and the Persians. Movsēs Xorenac‘i, History of Armenia, 2.13 13. Those events are described by the Greek historians, not by one or two but by many. Being doubtful of these same events we made many researches because we heard that Kiwros had killed Kriwsos and had destroyed the Lydian kingdom. And again the proposition of Kriwsos and Nek‘tanebos is reported. And this Nek‘tanebos is said by Manet‘os to have been the last king of Egypt, while some have called him the father of Ałek‘sandros. But we have found the period of Kriwsos to be two hundred years before that of Artašēs the First, king of Armenia. But because there are many who say that our Artašēs took Kriwsos prisoner and narrate this plausibly, I am so persuaded. Polykratēs speaks as follows: «Artašēs the Parthian is for me superior to Ałek‘sandr the Macedonian, because although he remained in his own country he ruled over T‘ebē and Babelon; and without crossing the River Aliws he destroyed the Lydian army and captured


137

Kriwsos; and before arriving in Asia he was announced in the fortress of Attikē. Alas for his fate! If only he had died in power and not in flight!» In agreement with him speaks Evagaros: «The war of Ałek‘sandr and Dareh is minor compared of that of Artašēs. For “the light of the day was obscured by the dust of the former”, but the latter hid and darkened the sun by his volleys of arrows, turning midday into artificial night. He did not allow the Lydians to flee and bring the news, but even their king Kriwsos he ordered to be placed in a cauldron. Because of him the torrents did not swell the river, for by the drinking [of his soldiers] it shrank to its winter level. By the multitude of his army he rendered the use of numbers inadequate so that there was need for measurement rather than counting. At this he was not boastful, but wept, saying: “Alas for this transitory glory”». Skamadros also writes as follows: «The haughty Kriwsos of Lydia was deceived by the reply of the Pythian oracle: “After passing the Aliws river Kriwsos will break the powers.” He thought that it referred to that of others, but he broke himself. For the Parthian Artašēs captured him and ordered him to be thrown into an iron cauldron. But Kriwsos, remembering the saying of Sołon the Athenian, said in his own tongue: “O Sołon, Sołon, you said well that one should not call a man’s fate happy until his death.” When those standing nearby heard this, they told Artašēs that Kriwsos was invoking some new god. Artašēs had pity and ordered him to be brought. And when he had inquired and learned what it was that he had cried out, he ordered him to be spared torments. P‘łegonis also writes: «The most powerful of all kings was the Parthian Artašēs. Not only did he put the Lydians to flight and capture Kriwsos, but in the Hellespont and in Thrace he changed the nature of the elements. On the land he was borne along as if sailing the sea, over the sea he marched on foot. He threatened the Thessalians, and his repute made the Hellenes wonder. He destroyed the Lacedemonians, he put the Phocians to flight; the Locrians surrendered, the Bithynians were a part of his forces; all Hellas held him in awe. A short time later his disasters surpassed all others. Not so unfortunate was Kiwros warring against the Massagetae; not so many misfortunes did Dareh suffer in flight from the Scythians, or Kambiwsēs the Ethiopians; insignificant was K‘sersēs‘ march against Hellas with an army, when he left them his treasures and tents and alone escaped alive by a hair’s breadth. But he [Artašēs], overweening through his splendid victories, was slaughtered by his own army.» Now I think that these accounts are worthy of belief, and that the Kriwsos whom they say lived in the time of Kiwros and Nek‘tanebos either is fictitious, or else there lived many kings with the one name – as is the custom of many.

138

Francesco Mari

Bibliography Adler, W. / Tuffin, P. (2002), The Chronography of George Synkellos. A Byzantine Chronicle of Universal History from the Creation, Oxford. Asheri, D. (ed.) (1988), Erodoto, Le Storie. I, La Lidia e la Persia, Milano. Asheri, D. (ed.) (1990), Erodoto, Le Storie. III, La Persia, Milano. Assman, J. (1997), La memoria culturale. Struttura, ricordo e identità politica nelle grandi civiltà antiche, Torino (transl. of Das kulturelle Gedächtnis. Schrift, Erinnerung und politische Identität in frühen Hochkulturen, München 1992). Azoulay, V. (2004), “The Medo-Persian Ceremonial: Xenophon, Cyrus and the King’s Body”, in: C. Tuplin (ed.), Xenophon and his World (Historia Einzelschriften 172), Stuttgart, 145–170. Bichler, R. (2010), “Der Hof der Achaimeniden in den Augen Herodots”, in: B. Jacobs / R. Rollinger (eds.), Der Achämenidenhof / The Achaemenid Court. Akten des 2. Internationalen Kolloquiums zum Thema “Vorderasien im Spannungsfeld klassischer und altorientalischer Überlieferungen”, Landgut Castelen bei Basel, 23.-25. Mai 2007, Wiesbaden, 155–188. Borgogno, A. (2005), Romanzi Greci, Torino. Breglia Pulci Doria, L. (1979), “Tigrane il Grande di Armenia in Mosè di Corene. Tradizione classica e tradizione locale”, in: Dialoghi di Archeologia n. s. 1, 95–108. Briant, P. (2002), From Cyrus to Alexander. A History of the Persian Empire, Winona Lake (IN) (transl. of Histoire de l’empire perse de Cyrus à Alexandre, Paris, 1996). Burney, C. (2007), “Avant les Arméniens : les Ourartéens, guerriers et bâtisseurs”, in: Dédéyan 2007, 67–100. Carrière, A. (1894), Nouvelles sources de Moïse de Khoren. Études critiques, Supplément, Wien. Chaumont, M.-L. / Traina, G. (2007), “Les Arméniens entre l’Iran et le monde Gréco-Romain (Ve siècle av. J.-C.-vers 300 ap. J.-C.)”, in: Dédéyan 2007, 101–162. Daryaee, T. (ed.) (2013), Cyrus the Great. An Ancient Iranian King, Santa Monica (CA). Daude, C. (2001), “Le personnage d’Artaxerxès dans le roman de Chariton, Chairéas et Callirhoé: fiction et histoire”, in: B. Pouderon et alii (eds.), Les personnages du roman grec. Actes du colloque de Tours, 18–20 novembre 1999, Lyon, 137–148. Davidson, O. (1994), Poet and Hero in the Persian Book of Kings, Ithaca-London. Dédéyan, G. (ed.) (2007), Histoire du peuple arménien, Toulouse. Due, B. (1989), The Cyropædia, Xenophon’s Aim and Method, Aarhus. Facella, M. (2006), La dinastia degli Orontidi nella Commagene ellenistico-romana (Studi Ellenistici 17), Pisa. Frye, R. N. (1982), “The ‘Aramaic’ Inscription on the Tomb of Darius”, in: Iranica Antiqua, 17, 85–90. Frye, R. N. (2005), The Sassanians, in: The Cambridge Ancient History, XII, Cambridge, 461–480. Geller, M. / Traina, G. (2013), “‘Tigranu, the Crown Prince of Armenia’: Evidence from the Babylonian Astronomical Diaries”, in: Klio, 95/2, 447–454. Grainger, J. D. (1997), A Seleukid Prosopography and Gazetteer, Leiden-New York-Köln. Havelock, E. A. (1963), Preface to Plato, Cambridge (MA)-London. Ivanov, V. I. (2011), “A Probable Structure of a Protoform of the Ancient Armenian Song of Vahagn”, in: Aramazd, 6/1, 7–23. Jones, H. L. (ed.) (1928), The Geography of Strabo, V, Cambridge (MA)-London.


139

Jursa, M. (2003), “Observations on the Problem of the Median ‘Empire’ on the Basis of Babylonian Sources”, in: Lanfranchi/Roaf/Rollinger 2003, 169–179. Kent, R. G. (19532), Old Persian. Grammar, Texts, Lexicon, New Haven. Khatchadourian, L. (2007), “Unforgettable Landscapes. Attachments to the Past in Hellenistic Armenia”, in: N. Yoffee (ed.), Negotiating the Past in the Past. Identity, Memory and Landscape in Archaeological Research, Tucson (AZ), 43–75. Knauth, W. (1975), Das altiranische Fürstenideal von Xenophon bis Ferdousi. Nach den antiken und einheimischen Quellen dargestellt, in Verbindung mit Sejfoddin Nadjmabadi, Wiesbaden. Lanfranchi, G. B. / Roaf, M. / Rollinger, R. (eds.) (2003), Continuity of Empire (?). Assyria, Media, Persia, Padova. Lenfant, D. (2001), “La ‘décadence’ du Grand Roi et les ambitions de Cyrus le Jeune : aux sources perses d’un mythe occidental ?”, in: Revue des Études Grecques 114, 407–438. Lenfant, D. (ed.) (2004), Ctésias de Cnide, La Perse. L’Inde. Autres fragments, Paris. Lenfant, D. (ed.) (2011), Les Perses vus par les Grecs, Paris. Levine Gera, D. (1991), Xenophon’s Cyropaedia. Style, Genre, and Literary Technique, Oxford. Lincoln, B. (2007), Religion, Empire and Torture. The Case of Achaemenian Persia, with a Postscript on Abu Ghraib, Chicago-London. Liverani, M. (2003), “The Rise and Fall of Media”, in: Lanfranchi/Roaf/Rollinger 2003, 1–12. Mahé, J.-P./ Mahé, A. (eds.) (1993), Moïse de Khorène, Histoire de l’Arménie (introduction, traduction de l’arménien et notes), Paris. Mahé, J.-P.e (1994), “Un dieu guerrier à la campagne : l’exemple du Vahagn arménien”, in: Comptes rendus des séances de l’Académie des Inscriptions et Belles-lettres CRAI, 779–804. Miller, W. (ed.) (1925), Xenophon, Cyropaedia, I, London-New York. Mosshammer, A. A. (ed.) (1984), Georgius Syncellus, Ecloga chronographica, Leipzig. Mueller-Goldingen, C. (1995), Untersuchungen zu Xenophons Kyrupädie, Leipzig. Muradyan, G. (1998–2000), “The Retorical Exercises (Progymnasmata) in the Old Armenian Book of Chreia (Girk‘ Pitoyic‘). Translation or Original Composition?”, in: Revue des Études Arméniennes n. s. 27, 399–415. Nadon, C. (2001), Xenophon’s Prince: Republic and Empire in the Cyropaedia, Berkeley. Parpola, S. (2003), “Sakas, India, Gobryas, and the Median Royal Court: Xenophon’s Cyropaedia through the Eyes of an Assyriologist”, in: Lanfranchi/Roaf/Rollinger 2003, 337–350. Perikhanyan, A. G. (1971), “Les inscriptions araméennes du roi Artachès”, in: Revue des Études Arméniennes n. s. 8, 169–174. Petit, T. (1990), Satrapes et satrapies dans l’empire achéménide de Cyrus le Grand à Xerxès Ier, Paris. Radner, K. (2003), “An Assyrian View on the Medes”, in: Lanfranchi/Roaf/Rollinger 2003, 37–64. Roaf, M. (2003), “The Median Dark Age”, in: Lanfranchi/Roaf/Rollinger 2003, 14–22. Roberto, U. (2011), Le Chronographiae di Sesto Giulio Africano. Storiografia, politica e cristianesimo nell’età dei Severi, Soveria Mannelli. Rollinger, R. (2008), “The Median ‘Empire’, the End of Urartu and Cyrus the Great’s Campaign in 547 BC (Nabonidus Chronicle II. 16)”, in: Ancient East and West 7, 51–65. Rollinger, R. (2010), “Das Medische Königtum und die Medische Suprematie im sechsten Jahrhundert v. Chr.”, in: G. B. Lanfranchi / R. Rollinger (eds.), Concepts of Kingship in

140

Francesco Mari

Antiquity. Proceedings of the European Science Foundation Exploratory Workshop. Held in Padova, November 28th-December 1st 2007, Padova, 63–86. Russell, J. R. (1986), “Artaxias I”, in: The Encyclopaedia Iranica, London-Boston, II 6, 659–660. Russell, J. R. (1986–1987), “Some Iranian Images of Kingship in the Armenian Artaxiad Epic”, in: Revue des Études Arméniennes, n. s. 20, 253–270. Russell, J. R. (1989), “Carmina Vahagni”, in: Acta antiqua Academiae Scientiarum Hungaricae, 32, 317–330. Sancisi-Weerdenburg, H. (1988), “Was there ever a Median Empire?”, in: Achaemenid History. Proceedings of the Achaemenid History Workshops and Other Studies on the Persian Empire, 3, Leiden, 197–212. Sancisi-Weerdenburg, H. (1994), “The Orality of Herodotus’ Medikos Logos or: The Median Empire revisited”, in: Achaemenid History. Proceedings of the Achaemenid History Workshops and Other Studies on the Persian Empire, 8, Leiden, 39–55. Sargsyan, A. (2012–2013), “The Iranian Interpretation of the Greco-Persian Wars reflected in the History of Armenia by Movses Khorenatsi”, in: Iran-Nameh. Armenian Journal of Oriental Studies, 81–89. Sargsyan, G. (1956), “Աղբյուրների օգտագործման եղանակը Մոկսես Խորենացու մոտ” [The Way of using Sources in Movsēs Xorenac‘i], in: Banber Matenadarani, 3, 31–42. Schaudig, H. (2001), Die Inschriften Nabonids von Babylon und Kyros’ des Großen samt den in ihrem Umfeld entstanden Tendenzschriften, Münster. Schmeling, G. (ed.) (1996), The Novel in the Ancient World, Leiden. Schmitt, R. (1977), “Thronnamen bei den Achaimeniden”, in: Beiträge zur Namenforshung, n. F. 12, 422–425. Serbat, G. (1973), “La référence comme indice de distance dans l’énoncé de Pline l’Ancien”, in: Revue de Philologie, de Littérature et d’Histoire Anciennes, 47, 38–49. Stramaglia, A. (ed.) (2011), Phlegon Trallianus, Opuscula de rebus mirabilibus; de longaevis, Berlin-New York. Strassler, R. B. (ed.) (2008), The Landmark Herodotus. The Histories, London. Stronach, D. (2003), “Indipendent Media: Archaeological Notes from the Homeland”, in: Lanfranchi/Roaf/Rollinger 2003, 234–248. Tatum, J. (1989), Xenophon’s Imperial Fiction, Princeton. Thomson, R. W. (ed.) (1978), Moses Khorenats‘i, History of the Armenians, Cambridge (MA)-London. Topchyan, A. (2006), The Problem of the Greek Sources of Movsēs Xorenac‘i’s History of Armenia, Leuven-Paris/Dudley (MA). Traina, G. (1991), Il complesso di Trimalcione (Movsēs Xorenac‘i e le origini del pensiero storico armeno), Venezia. Traina, G. (1995), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. I”, in: Le Muséon, 108, 279–233. Traina, G. (1998), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. II”, in: Le Muséon, 111, 95–138. Traina, G. (1999–2000), “Épisodes de la rencontre avec Rome (IIe siècle av. J.-C.-IIIe siècle ap. J.-C.)”, in: Iran & the Caucasus, 3/4, 59–78. Traina, G. (2007), “Moïse de Khorène et l’empire sassanide”, in: Res Orientales, 17, 158–179. Traina, G. (2011), “Note sulla montagna nell’Armenia antica”, in: S. Giorcelli Bersani (ed.), Gli antichi e la montagna. Ecologia, religione, economia e politica del territorio. Atti del convegno di Aosta, 21–23 settembre 1999, Torino, 213–221.


141

Traina, G. (2015), “Tradition et innovation dans la première historiographie arménienne”, in: P. Blaudeau / P. Van Nuffelen (eds.), L’historiographie tardo-antique et la transmission des savoirs, Berlin-New York, 153–164. Traina, G. (forthcoming), “Strabo and the History of Armenia”, in: D. Dueck (ed.), The Routledge Companion to Strabo, London. Tuplin, C. (1996), “Xenophon’s Cyropaedia: Education and Fiction”, in: A. H. Sommerstein / C. Atherton (eds.), Education in Greek Fiction, Bari, 65–162. Tuplin, C. (2004), “Medes in Media, Mesopotamia, and Anatolia”, in: Ancient West & East, 3/2, 223–251. Tuplin, C. (2013), “Xenophon’s Cyropaedia: Fictive History, Political Analysis and Thinking with Iranian Kings”, in: L. Mitchell / C. Melville (eds.), Every Inch a King, Leiden-Boston, 67–90. Von Gutschmid, A. (1876), “Über die Glaubwürdigkeit der armenischen Geschichte des Moses von Khoren”, in: Berichte über die Verhandlungen der königlichen sächsischen Gesellschaft der Wissenschaften zu Leipzig, Phil.-Hist. Classe, 28, 1–43. Wallraff, M. (ed.) (2007), Julius Africanus und die christliche Weltchronik, Berlin-New York. Wallraff, M. / Roberto, U. / Pinggéra, K. (eds.) (2007), Iulius Africanus, Chronographiae. The Extant Fragments, transl. by William Adler (Die griechischen Christlichen Schriftsteller, Neue Folge 15), Berlin-New York. Waters, M. (2014), A Concise History of the Achaemenid Empire, 550–330 BCE, New York. Wiesehöfer, J. (2003), “The Medes and the Idea of the Succession of Empires in Antiquity”, in: Lanfranchi/Roaf/Rollinger 2003, 391–396. Xalatjanc, G. (1896), Apмянский эпос в Истоpии Apмении Mоисея Xopенского, [The Armenian Epos in Movses Xorenac‘i’s History of Armenia], 2 vols., Moscow. Xalatjanc, G. (1903), Apмянские Apшакиды в Истоpии Apмении Mоисея Xopенского, [The Armenian Arsacids in Movsēs Xorenac‘i’s History of Armenia], 2 vols., Moscow. Zournatzi, A. (2013), “The Median logos of Herodotus and the Persians’ legitimate rule of Asia”, in: Iranica Antiqua, 48, 221–252. Zournatzi, A. (forthcoming), “Early Cross-Cultural Political Encounters Along the Paths of the Silk Road: Cyrus the Great as a ‘King of the City of Anshan’”, in: D. Akbarzadeh (ed.), Proceedings of the First International Conference “Iran and the Silk Road”, Teheran, 12th–15th February 2011, Teheran (a pre-publication version is available at www.achemenet. com/document/ZOURNATZI_Cyrus_of_Anshan.pdf).

Anahide Kéfélian

The Roman Army and the Transmission of Latin Loan Words in Old Armenian* Abstract: The aim of this paper is to understand, by means of an interdisciplinary study, how loan words passed from Latin to old Armenian. After considering the languages that were used in the Armenian Kingdom between locals and Roman soldiers, a linguistic analysis is performed. Thus this analysis allows to determine whether the loan words cited by Bolognesi in his seminal 1965 paper are today still considered to have been borrowed directly from Latin or indirectly from Greek. The historical context in which these loan words were borrowed is also discussed, in order to determine if the borrowing resulted from translation of the Bible or from relationships between Romans and Armenians. For this purpose, the Roman policies in the Armenian Kingdom and the origins of the Roman soldiers posted in the Armenian Kingdom are studied.

The Greek language was present in Armenia from as early as the Orontid dynasty and later with Tigranes the Great¹. Some Greek inscriptions testify that the Greek language was also spoken under the Armenian Arsacid dynasty². Thus it seems that two different languages were used at the Armenian court: Aramaic (for both the Semitic and Iranian languages), and Greek³. Greek was also used for diplomatic events as this language was spoken from the East to Egypt as well as at the Parthian court⁴.

* First of all, I would like to thank my supervisor Giusto Traina. I am also indebted to all the specialists who kindly answered my questions: Agnès Ouzounian, Eric Dieu, Patrick Donabédian, Julien Dufour Pierre Larcher, Andrea Scala. I would also like to thank Cécile Kéfélian and Hayden Brian for their valuable help. Finally, my deepest thanks go to the Gulbenkian Foundation which allowed me to carry on my studies. 1 For the Orontid dynasty: De Lamberterie 1991; Canali De Rossi 2004, nos. 9–15. For the period of the reign of Tigranes the Great cf. Plut. Luc. 29.4; Crass. 33.1.7; Traina 2008; cf. infra, n. 15. 2 Canali De Rossi 2004 (although handy and useful, this catalogue needs thorough revision), nos. 2–8, 17–18; Traina 2005. See also the dedication by Pacorus, king of Armenia, found in the city of Rome: CIG III, no. 6559; IG XIV, no. 1472; OGIS, no. 382; IGR I, no. 222; Trever 1953, 581–583; Canali De Rossi 2004, 65, no. 22. 3 Łazar P‘arpec‘i, 10 (1933 ed.). This hypothesis is confirmed by the discovery of bullae from the archives in Artašat, probably burnt in 58 AD during Corbulo’s campaign. Some bullae are inscribed in Greek or in Aramaic. 4 Traina 2007, 30 f.

144

Anahide Kéfélian

We cannot easily estimate how many Armenians spoke Greek. With regard to the fourth century AD, during the evangelization of Armenia, the local population did not understand the liturgy held in Greek or Aramaic (the same situation occurred in St Augustine’s African countryside). Possibly, this was one of the reasons for translating the Bible into Armenian, which implies that a small part of Armenians spoke Greek and Aramaic⁵. Greek may have been spoken more in the urban centers, where important ‘Jewish’ communities resided⁶. Roman military operations brought Armenia into contact with Latin as well. Official inscriptions confirm that Latin was mainly used by the military units. Out of ten titles found in the ancient Kingdom of Armenia, seven are honorific (dedications to Nero, Trajan, Marcus Aurelius and Commodus), two are funerary inscriptions of soldiers, and one is a religious dedicatory. Roman soldiers were stationed in the Armenian capitals Artašat and Vałaršapat. Roman troops left traces of their presence in the kingdom of Armenia. Inscriptions demonstrate the presence of the legions IVth Scythica and VIth Ferrata and a vexillation of the Ist Italica under the reign of Trajan in Artašat⁷, as well as the presence of the legions XVth Apollinaris, XIIth Fulminata under the reign of Marcus Aurelius⁸ and finally of the XVth Apollinaris legion under Commodus in Vałaršapat⁹. This policy enabled the army to protect the Roman citizens and the Armenian king¹⁰. Besides their role of conquest and defense, Roman soldiers protected cities against brigands, spies and local inhabitants in order to maintain order and authority. They were also used for construction projects and collecting taxes¹¹. Moreover, the Roman army could also be moved to strategic points such as roads or fortresses¹². But

5 Donabédian 1994, 7–10. 6 Topchyan 2007. 7 For the dedication of an inscription to Trajan by the IVth Scythica: Aṙakelyan 1967, 114–116 = Id. 1968, 135 = AE 1968, no. 510; for the epitaph of a legionary discovered in Artašat: Aṙakelyan 1967, 116–118 = Aṙakelyan 1968, 136–138 = AE 1968, no. 511. 8 And a cohors eq(uitatae) c(ivium) R(omanorum): BCIASP 33, 1909, 2 = AE 1910, no. 161 = Macler 1910 = AE 1911, no. 83 = Trevner 1953, 262–267, app. 85 = ILS III, 2, 1955, no. 9117 = Mosser 2003, no. 236. 9 Macler 1910, nos. 1–2 = CIL III, no. 6052; no. 13627 = Trevner 1953, 267–270, app. 86 = ILS I, 1955, no. 394 = Saxer 1967, no. 276 = Mosser 2003, no. 237. There are also three inscriptions engraved under the reign of Nero which were found in Castellum Ziata: Curtius 1874 = Wünsch 1884, 241– 242 = CIL III, nos. 6741, 6742, 6742a = ILS 1955, no. 232. 10 Inner revolts were indeed fairly widespread and were particularly numerous under the reign of Augustus: Chaumont 1976, 73–84; Tac. Ann. 2.56.3; cf. Kéfélian 2015, 282–284. 11 Isaac 1986, 383–395; Lee 1986, 455–461; Pollard 2000, 86–110; Butcher 2003, 404. 12 For the garrison in Gaṙni, managed by the prefect Celius Pollio and the centurion Casperius, see Tac. Ann. 12.45. Armenia was often chosen as a platform to attack the Parthian Empire: see e.g. Hdn. 6.12.


145

Fig. 1: Epitaph dedicated to Aelia Maxima and Aelia Valentina by the tribune Publius Aelius Valens (© Erevan History Museum of Armenia, inv. n. 793; photograph by Giusto Traina).

a number of Roman soldiers in Armenia were also Greek-speaking, as shown at least by two inscriptions: the consecration to the goddess Ge Meter Olybris from Areni¹³, and the epitaph from Vałaršapat of the Roman tribune Publius Aelius Valens, dedicated to his wife Aelia Maxima and his daughter Aelia Valentina (Fig. 1)¹⁴. As for the second inscription, Mihail Ivanovič Rostovcev estimated that the cognomen of Publius Aelius Valens suggests a Cilician origin¹⁵. Nevertheless, this is not enough to reach a definitive conclusion on the origin of this knight. In fact, Valens is a typical military cognomen, rather common in the Balkans and in the Near East¹⁶.

13 Ter-Martirosov 1989, 1, 177–189 = BE CIX, no. 308 = Vinogradov 1992, 13–26 = SEG 42, no. 1322 = SEG 47, no. 1163 = SEG 53, no. 2228 = Canali de Rossi 2004, 13–14. See also Traina 1999. 14 Trever 1953, 251–261 = Moretti 1955, 45–46 = BE 1956, no. 345 = AE 1956, no. 31 = SEG XV, no. 839 = Vinogradov 1992, 14–15 = Canali de Rossi 2004, 15–16. 15 Rostovtzeff 1914 (on this study, see Traina 1999); Vinogradov 1992. 16 See e.g. CIL III, nos. 1559, 8009; AE 1960, no. 339; Russu 1973, 110; IDR III, 1, no. 118; AE 2007, no. 1205; 111 no. 36EDH-Nr. HD044376 (www.epigraphische-datenbank-heidelberg.de/ 22/09/2013).

146

Anahide Kéfélian

It is worthwhile analyzing the place of Latin and Greek in the military sphere as well as in the private sphere of soldiers, in order to understand the context facing the soldiers who were stationed in Armenia. The status of the Roman military presence prior to the reign of Nero has so far not been the subject of in-depth study. It is known that up to the reign of Claudius, Roman soldiers were mainly recruited in the Italian peninsula¹⁷. Under Nero, a significant change took place, with 50 % of recruits coming from other areas of the Roman Empire¹⁸. At that time, recruits were usually sent away from their homeland. Oriental recruits were moved to the Rhine and the Danubian fronts and soldiers stationed in the Armenian Kingdom came from three extensive areas: the Danubian, the Oriental and the African areas. The cohors I Syrorum and the numerus Palmyrenorum from Syria were sent to Africa and the cohors I Ituraeorum was sent to Mauretania Tingitana, Germania and Dacia¹⁹. In contrast to these cohorts, the Moors (nomad riders) were sent to the Orient²⁰. Evidence of the presence of Syrian soldiers on the Danubian front has been discovered²¹. Other attestations indicate that Danubian legions and especially the Vth Macedonica and Thracian auxiliaries were sent to the Orient and more especially to the Armenian war front²². The underlying aim of these movements of troops was to transfer new recruits away from their homeland. Moreover, the Romans sought to use auxiliary techniques similar to those of their Armenian, Parthian and Sasanian enemies. A case in point is that of Danubian and African auxiliaries, who were especially good riders, archers and slingers. Soldiers from the XVth Apollinaris were transferred from the Danubian limes to Satala in the first part of the second century AD²³. Until the end of Commodus’ reign, inscriptions attest to the presence of XVth Apollinaris in the Armenian Kingdom²⁴. An important change took place from the second century AD: Roman troops were more and more often stationed in their homeland. This tendency was confirmed frequently over time. The name of the new troops Indigenae or Saraceni,

17 Forni 1953, 159–241; Mann 1983, 49–54; Pollard 2000, 114; Le Bohec 1989a, 71–107. 18 Forni 1953, 65 f.; Pollard 2000, 114. 19 Spaul 2000, 402, 440–444; Le Bohec 2007, 221–295. 20 Le Bohec 1989b, 163–164. Cf. Mann 1983, 147–149, table 25A. 21 For example, the Cohors I Flavia Commagenorum served in Moesia Inferior and Dacia Inferior: Tudor 1971; Spaul 2000, 402–407. 22 CIL XIV, no. 3608; EE IX, 470; InscrIt IV, 1, no. 125; ILS I, no. 986; Butcher 2003, 412. 23 The date of departure of the legion from the Danubian area to Armenia is not well-known: cf. Wheeler 2000, 282–308. 24 Cf. supra, n. 9. Currently, no inscription has been discovered after the reign of Commodus but the absence of inscriptions does not imply the absence of Roman troops.


147

recruited and stationed in the East, seem to attest a local origin²⁵. In addition, the XVth Apollinaris legion established in Satala from 115 AD also recruited locally over time, even though it came from the Balkans. According to K. Butcher, Roman recruits in the East seem to have been mostly Greek-speaking²⁶. In the third century AD, Caracalla’s edict²⁷ indirectly induced a change in recruitment conditions. Before the constitutio Antoniniana, non-citizens frequently enlisted in the Roman army in order to obtain Roman citizenship²⁸ but the Roman army progressively lost its social prestige. To fill this decrease in soldiers wishing to join the army, a new reform was necessary: recruitment criteria became less strict and this led to barbarization of the army. Other modes of recruitment were also set up to overcome this barbarization²⁹. So, unlike in the Early Roman Empire, soldiers recruited into the Roman army at the end of the Roman Empire were posted in their homeland³⁰. Latin subsequently declined noticeably over the centuries in the East. This decline was influenced by several parameters such as the region, the status of the colony and the policy adopted by the Emperor. Furthermore, in order to determine which languages were used in the Roman army, a useful element of comparison may be detected in Judaea, where epigraphy confirms that Latin was used in the official military sphere³¹, and Rabbinic literature mentions the use of four languages, notably Latin³². But even though Latin was the official language in the military sphere, many terms were taken from Greek. Rabbinical texts use many Latin technical terms and their Greek equivalents. For example, the Latin term praefectus could have been replaced by the Greek term ἔπαρχος and the term tribunus militum was translated by the Greek term χιλίαρχος³³. Greek terms transcribed from Latin were also used to characterize soldiers’ duties such as κεντηνάριος which refers to the Latin term for the

25 Pollard 2000, 139, 141. 26 Butcher 2003, 400–401; Pollard, 2000, 116. 27 This edict granted citizenship to all free inhabitants of the Roman Empire. 28 Citizenship was given first to legionaries and finally to auxiliaries. Butcher 2003, 401. 29 Barbero 2011, 85–105; Le Bohec 2006, 55–66. 30 Pollard 2000, 114, 138–141. 31 Suet. Tib. 71; Dio Cass. 57.5. Le Bohec 1989a, 247–249; Pollard 2000, 134–138; Balsdon 1979, 127, 131–135. 32 Talmud, Megillah 1.11.71b; Sota 7. 2.21c; Hadas-Lebel 1990, 216: “quatre langues meritant d’être usitées: le grec pour le chant, le latin pour la guerre, l’araméen pour l’élégie et l’hébreu pour le discours”. See Hadas-Lebel 1990, 216–218. 33 This last term is notably used in one of the Roman inscriptions discovered in Armenia. Cf. infra.

148

Anahide Kéfélian

wage echelon centenarius³⁴. The difference between the Roman and the Greek institutions is reflected in the lack of appropriate terms. Furthermore, even if a Greek equivalent for a Latin term existed, the Latin term transcribed in Greek was often favored, as shown by the use of χώρτη, κοόρτη (Latin cōhors) instead of the Greek term σπεῖρα; of ἄλα (Latin āla), instead of the Greek ἴλη; of λεγιών (Latin legio) instead of τάγμα³⁵. Hugh Mason points out that interactions between two languages are more significant when the traditional culture of the borrower language is weak³⁶. Unfortunately, because of the lack of Armenian texts until the fifth century AD, this analysis cannot be carried out as regards the Armenian language. There were undoubtedly many loan words used in oral discourse which escape this study. It is also difficult to obtain evidence for a diachronic overview of the Armenian language at that time. Nevertheless, Rabbinical texts show that Latin was the official language in the army and that several languages coexisted in the recruits’ private sphere, including Greek³⁷. Despite divergent opinions of researchers, Latin clearly declined over the centuries in the East. Nevertheless, this decline depended on several parameters such as the region, the status of the colony and the policy adopted by the emperor. Linguistics may help to better understand the relationship among Greek, Latin and Armenian, through the study of loan words. The small group of Latin loan words in Armenian is particularly interesting. They were first taken into consideration by the great linguist Heinrich Hübschmann, in the first and only published volume of his grammar³⁸, and then by Alfred Thumb, in his paper on Greek borrowings in Armenian from grabar to ašxarhabar, also concerning phonological features and the vernacular lexicon³⁹. In the first contribution expressly devoted to Latin loan words in Armenian, Antoine Meillet considered as Latin borrowings five Armenian lemmas⁴⁰. In 1961, Günter Reichenkron reviewed these

34 The borrowings can be categorized into two groups: those which were borrowed strictly as standard words and those which became integrated into the Greek linguistic system by taking phonetic or morphological Greek features. As Mason shows, all types of Greek borrowings from Latin are concerned with official contexts, both civil and military: Mason 1974, 4–9; Balsdon 1979, 123. 35 Mason 1974, 7. 36 Mason 1974, 8. 37 Pollard 2000, 136. On significant use of Greek in the East cf. Butcher 2003; Balsdon 1979, 123–136; Sartre 2001. 38 Hübschmann 1895. 39 Thumb 1900. 40 Meillet 1913. The lemmas are arkł, sitł, skutł, kaṙk‘, to which Meillet also added t‘ew. See infra.


149

findings in a brief overview of the Roman-Armenian relationship⁴¹. Finally, in a seminal 1965 paper, where he examined all types of Armenian loan words possibly borrowed from Parthian and Greek⁴², Professor Giancarlo Bolognesi also gave a brief overview of Latin loan words in grabar⁴³. Despite these authoritative studies, many elements have hindered the progress and comprehension of the mechanisms of Latin borrowings. Moreover, apart from some historical considerations by Reichenkron on the Roman presence in Armenia, the study of Latin loan words has been undertaken only from a purely linguistic point of view, with no particular interest in historical, archaeological, iconographical and ethnological issues, concerned with the relationship between Romans and Armenians. As is well known, loan words may be difficult to detect. In particular, the study of anomalies manifested as phonological and morphological distortions, or as syntactic and semantic changes, can provide insight into the origin of loan words. As a matter of fact, a loan word may be rejected or only partially borrowed if it is not favorably received in a suitable context, whereas when a loan word is completely integrated into a new language, it takes on localized features of this language⁴⁴. Accordingly, loan words must be analyzed from the historical, phonetic, morphological and semantic point of view⁴⁵. The historical criterion is of significance for the insight it affords into cultural exchanges, but it is the phonetic criterion that is the most important for linguists: by detecting and tracking phonetic alterations, linguists seek to put forward hypotheses that allow the identification of the putative original language from whence this borrowing derives, and its transitional term. Conjectured phonetic changes may also help to date a borrowing and can sometimes point to its geographic origin. The morphological criterion is more unusual: it implies that its structure reflects the original language. As far as the semantic criterion is concerned, this is less trustworthy than the other criteria because of the difficulty in attributing the origin of the meaning. Loan words may arise on account of historical circumstances, as borrowings are frequently determined by the introduction of material culture. For instance,

41 Reichenkron 1961, 1021–1028, on the evidence of a direct relationship between Armenians and Romans. 42 Greek loan words could in some cases be transmitted through the intermediary of the Parthians: Bolognesi 1966, 583–602. 43 Bolognesi 1966, 569–603. For an overview on the work of G. Bolognesi, see the contribution by M. Morani, in this volume. 44 Deroy 1956, Introduction, §§ 12–16. 45 Deroy 1956, III, § 1–31.

150

Anahide Kéfélian

a borrowing often results from the need to name new items, ideas, institutions, weights and currencies. But other types of contact among different language groups may also play a major role. This is the case of the ancient Kingdom of Armenia, whose political and economic position between the Roman and Parthian empires is well known⁴⁶. No doubt, this strategic location created a favorable situation for intense language contacts⁴⁷. As Armenian did not become a written language until the fifth century AD, loan words were orally transmitted. Nevertheless, in some cases it is possible to determine whether Latin terms documented in later texts were borrowed for purposes of translation or if they were oral borrowings introduced earlier. Let us now consider the first three Latin loan words identified by Meillet: arkł, sitł, skutł. Morphologically speaking, all three borrowings belong to the ł-class of inflexion. No doubt they all are Latin borrowings, although we do not know whether they passed into Armenian directly from Latin or via the intermediation of Greek, for example through the translation of the Bible⁴⁸. Thus, the analysis of lexical, historical and semantic criteria will help to provide a conclusion on their origin. a. Arkł. According to Hübschmann, this loan word comes from the Greek ἅρκλα whereas Meillet concludes that it could have come either from the Greek ἅρκλα or the Latin arcūla⁴⁹. Hübschmann regarded all the borrowings studied by Meillet as borrowed from Greek. Thumb and Meillet were the first to suggest the possibility of Latin loan words. It is difficult to conclude with certainty from phonetics whether they were direct Latin borrowings or whether they were introduced indirectly from the Greek. The syncope of u in arcūla is known to be a possible outcome of this form in classical Armenian. Many languages such as the Celtic, Germanic and Slavonic families have borrowed this term from the Latin arcūla which is considered as a “traveling word”⁵⁰. From a morphological point of view, for this category of inflexion, it is easier to explain the term if arkł was borrowed from a syncopated term⁵¹. However, it is difficult to determine if it comes from the

46 Manandian 1965, 80; Chaumont 1976; Traina 1999–2000; Hewsen 2001, 42–49; Mutafian/Van Lauwe 2001, 28–37. 47 See Bolognesi 1966. 48 For example, Arm. legēōn is a loan word from the Greek λεγεών, transmitted through the translation of the Bible: Thumb 1900, 432. 49 Thumb 1900, 431; Meillet 1913, 349; Meyer-Lübke 1911, no. 611; Reichenkron 1961, 1028; Bolognesi 1966, 603. 50 Meillet 1913, 349. 51 I thank Andrea Scala for this suggestion.


151

Greek, which has an asyncopated shape, or from an informal syncopated shape in Latin. In Latin, arcūla was common and it had a variety of uses⁵²: it simultaneously means wood chest, jewelry box, perfume box, wood or stone coffins, and a frame to reinforce foundation works in wetlands. Thus the primary meaning of arcūla has the sense of containing⁵³. Arcūla was also used in Roman administration to symbolize public funds. This word is also employed five times in the Bible⁵⁴, meaning a bag for packaging, a tax chest, or even a coffin⁵⁵. It is also mentioned in Movsēs Xorenac‘i in reference to a sort of hanging nacelle, carrying men down to the entrance of a cave⁵⁶. In Koriwn, this idea of a container is used to indicate a shrine⁵⁷. In Agat‘angełos, this word translates a coffin or a relic box⁵⁸. Therefore, this term is polysemous and takes on numerous definitions in the transition between the Latin and the Armenian word. In contrast to the term in the Bible, which exclusively refers to a bag or a tax chest, Armenian texts refer to various senses. Thus this word is not only a lexical but also a semantic borrowing, the idea of a container having been carried over into the Armenian language. As the equivalent in the Greek translation of Bible is γλωσσόκομον (see Syr. glwsqwm), we may deduce that arkł came by contact with Latin or its Greek equivalent. b. Sitł. This loan word could come directly from Lat. sĭtŭla or indirectly from Greek σίτλα, borrowed from Latin⁵⁹. Because of the absence of this term in early Armenian texts and in the Bible, Meillet concludes that it was borrowed later. However, the absence of attestation in texts before the fourteenth century does not demonstrate that it was not verbally borrowed during Antiquity. An important part of borrowings may escape our knowledge as they may have originated purely as oral borrowings. Firstly, sĭtŭla is used to designate a container to carry

52 Daremberg/Saglio/Pottier 1873–1919, 1.1, 361–366. 53 Daremberg/Saglio/Pottier 1873–1919, 1.1, 362–366. For the meaning “frame” cf. Vitr. 5.12.3. 54 Leiden Armenian Lexical Textbase: www.sd-editions.com/LALT/home.html, accessed July 10, 2013 and Titus database: http://titus.uni-frankfurt.de/indexe.htm, accessed August 01, 2013. 55 VT 2 Ch. 24.8; NT Jo. 6.6, 12.6, 13.29, 29.29. 56 Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘ 3.45.7. 57 Koriwn, Life of Maštoc‘, 24. 58 Agat‘angełos, Patmowt῾iwn Hayoc‘ 760.3 (394.31), 760.7 (394.31), 764.2 (397.2), 765.2 (397.2), 765.5 (397.12), 766.1 (397.16). 59 Hübschmann 1895, 379, no. 404; Meillet 1913, 349; Meyer-Lübke 1911, no. 7962; Reichenkron 1961, 1029; Bolognesi 1966, 603.

152

Anahide Kéfélian

Fig. 2: The use of the sĭtŭla and the scŭtella in the Roman army: example in Trajan’s Column (Courtesy of Soprintendenza speciale per i Beni archeologici di Roma).


153

dripping water or liquids and can be translated into English by “bucket”⁶⁰. It was also used to receive lustral water during religious ceremonies or to receive ballots for the drawing of lots. However, it was widely used in daily life, especially in the Roman army. Two scenes of Trajan’s column explicitly represent sĭtŭlae as used by Roman soldiers. On Trajan’s column (Fig. 2), Roman troops are represented crossing a river and carrying their packs fastened to a stick of wood on their shoulders. This depiction clearly shows a sĭtŭla and a scŭtella. It is well known that each Roman soldier took with him a bag named sarcĭna, in which he carried his bowl, personal effects, and working tools to build camp and food for seventeen days. In the next representation (Fig. 3), Trajan is visiting a fortress while a soldier is collecting water in a sĭtŭla at the bottom of this scene. From a morphological viewpoint, the case is the same as arkł. It seems to be easier to explain it by a syncopated shape from Greek or Latin. Currently, no text proves that sĭtŭla was borrowed during Antiquity, but it could also have been an oral borrowing. A study of this term and its use during the Byzantine period would be useful to determine the period in which this loan word was borrowed. However, a third hypothesis might be formulated: the Arabic term saṭl for bucket is close to the Latin sĭtŭla. The first Arabic term seems to be sīṭal, later transformed into sayṭal, which is more classical, and then saṭl⁶¹. The Arabic term was probably borrowed from the Greek σίτλα⁶². Armenian could have borrowed sitł from the Arabic term sīṭal, as the syncope of the a explains sitł in Armenian, but this would be unexpected. From a historical point of view, the Arabs dominated Armenia from the middle of the seventh century to the end of the tenth century AD. During this prosperous period, Arabic was the language for the administration of the Armenian kingdom. Furthermore, Arabic became the international language at that time and was spoken in the countries surrounding Armenia. There were also Arabian colonies in the countryside of Armenia⁶³. Nevertheless, words loaned from Arabic are not well-known, and the same holds true for their borrowing dates. Currently, in the Armenian language only fifty Arabic terms are known to have been borrowed before the eleventh century AD. No dated information on the borrowing of sitł is available to help in determining whether this word was borrowed before or after the Arabian domination. Its first witness by Barseł Maškeworc῾i in the fourteenth century AD only conveys a terminus ante

60 Daremberg/Saglio/Pottier 1873–1919, 4.2, 1357–1360. 61 The diphthong ay usually became a long î. 62 Baldi 2008; Rajki 2005. 63 Nichanian 1989, 193.

154

Anahide Kéfélian

Fig. 3: The use of the sĭtŭla in the Roman army: example in Trajan’s Column (Courtesy of Soprintendenza speciale per i Beni archeologici di Roma).


155

quem and sitł could have been borrowed after the eleventh century AD. These three hypotheses are valid but the Greek or Latin origins are favored, because the number of Arabic borrowings is smaller than those from Greek. Sitł could have been borrowed during Antiquity from Latin or Greek, or later from Greek during the Byzantine period. c. Skutł. According to Hübschmann, this lemma could stem from Lat. scŭtella or from the Greek loan words Greek σκουτέλα or σκουτέλλιον⁶⁴. This “traveling word” survived in the Roman, Germanic, and Slavonic families⁶⁵. It was a term used in reference to daily life and implies a bad quality bowl, goblet or tray, in contrast to its use in the Byzantine period, where it appears as a richly decorated plate⁶⁶. Lexically, it could also come from the Latin scŭtŭla, or from its derivative Greek form σκούτλα⁶⁷. From the morphological and lexical viewpoint, it is more likely to have come from scŭtŭla or σκούτλα than from scŭtella or σκουτέλα/ σκουτέλλιον. In effect, it is impossible that scŭtella, and its Greek equivalents, lost their tonic vowel by syncopation. This diminutive form of scuta also means a “bowl, goblet” and a “plate”. It described a bowl that was used by the Roman army and belonged to each legionary’s equipment, as testified by the first scene studied on Trajan’s column where soldiers are carrying their personal effects and more specifically a scŭtella or scŭtŭla (Fig. 2). The lemma also occurs in the Armenian Bible, where it is described as a plate⁶⁸, whereas two Armenian texts mention it as a flower vase⁶⁹. The meaning of these loan words in Armenian texts is different from other quotations in the Bible. The borrowing of skutł is not the result of a translation from the Bible, since πίναξ is used in the Greek Bible and pynk in the Syriac Bible as a translation for “plate”. Therefore, this loan word may have been transmitted through contact with the Roman army, although it is not possible to determine if this term was borrowed directly from the Latin or indirectly from the Greek equivalent.

64 Hübschmann 1895, 380; Meillet 1913, 349; Meyer-Lübke 1911, no. 7756; Reichenkron 1961, 1028–1029; Bolognesi 1966, 603. 65 Meillet 1913, 349. 66 Daremberg/Saglio/Pottier 1873–1919, 4.2, 1156. 67 Sophocles 1870, 998. 68 NT Mt. 14.8, 14.11; Mc. 6.25, 6.28. 69 Movsēs Xorenac‘i, Patmowt῾iwn Hayoc‘ 2.62.15; Movsēs Kalankatowac‘i, Patmowt῾iwn Ałowanic῾ Ašxarhi 2.14.

156

Anahide Kéfélian

d. Kaṙk. In contrast to the first three borrowings, it is not clear whether kaṙk῾ has a Latin origin⁷⁰. Meillet suggested that this term could have come from the Latin carrus and eventually rejected Marr’s hypothesis of a Syriac etymology⁷¹. As Meillet explained, Syriac qarūxa cannot change to kaṙk῾ in old Armenian⁷². The Syriac term qarūxa refers to another Latin borrowing, carrūca (“coach”), a technical term used in reference to the transportation of furniture. The correct Syriac word for “carriage” is rkb (rekeb) and no similarity with kaṙk῾ is conceivable. From a phonetic point of view, kaṙk῾ could have either a Latin (carrus) or a Celtic (Galatian) origin (carros)⁷³. As Meillet explained, the lemmas carrus, carrūca⁷⁴ are considered as “traveling words”. Borrowed from Latin, they passed into the German, Roman and Syriac languages⁷⁵. Thanks to the study of the relationship between Rome and the Armenian Kingdom, it is more likely that these borrowings came from Latin than from Galatian. In the Roman Empire, this Gallic term was used not only for the transportation of food and commercial furniture, but also in troop movements for packs called impedimenta. On Trajan’s column, carri convey weapons, food, barrels, war machines and packs⁷⁶ (Fig. 4). Arm. kaṙk῾ appears in the Old Testament and in Eznik, but in these cases the terms refer to a war chariot. This term does not stem from the translation of the Armenian Bible because the term used in the Greek Bible is ἅρμα whereas rkb (rekeb) is used in the Syriac Bible⁷⁷. Therefore kaṙk῾ seems to have been borrowed before the translation of the Bible. From a semantic point of view, this term was employed for “chariot” and “war chariot”, in contrast to Latin, which has two different terms to describe these two types of chariot. Unlike carrus, -i which refers to a transport chariot, currūs, -us denotes a generic term for chariot. A carrus consists of a horizontal platform with boards to transport furniture and has either two or four wheels. In Armenian, several words also exist for a more specific meaning of “chariot”: vartakaṙk῾ expresses a warlike chariot whereas vač‘, vač‘ē, yopnak, sayl refer to a particular type of chariot. Nevertheless, their semantic differences are poorly known and no definitive conclusion can be reached.

70 Hübschmann 1895, 458; Meillet 1913, 349; Meyer-Lübke 1911, no. 1721; Reichenkron 1961, 1029–1030; Bolognesi 1966, 603. 71 Marr 1912, 545 ff. 72 Meillet 1913, 349–350. 73 Hübschmann 1895, 458, no. 200. A Celtic origin of carrus is supported by Caes. B. Gall. 1.26.3. 74 Daremberg/Saglio/Pottier 1873–1919, 1.2, 928. 75 Meyer-Lübke 1911, no. 1721. 76 Depeyrot 2007; Depeyrot 2008, 69, 83, 162–163; 95–96, 196–197; 103–104; 164–165. 77 In the Latin Bible, the term is currŭs.


Fig. 4: The use of the carrus in the Roman army: examples in Trajan’s Column (Courtesy of Soprintendenza speciale per i Beni archeologici di Roma).

157

158

Anahide Kéfélian

e. T‘ew. This lemma is not a lexical loan word, but a loan translation⁷⁸. T‘ew has two meanings: it is used to name the wing of a bird and the wing of an army. This loan translation could have arisen in Armenian during the translation of the Greek or Syriac Bible to overcome the absence of an equivalent term⁷⁹. In Greek, unlike Armenian, two terms are used in the Bible for these two concepts: κέρας⁸⁰ is used only for the wing of an army, whereas πτερύγιον is only used for the wing of a cherub⁸¹. More generally in Greek, neither of these two terms alludes to both the wing of a bird and the wing of an army. In the Syriac Bible, two words, qrn and knp, are equally used for both senses. Therefore, the loan translation does not come from the Greek Bible, where, as mentioned, two different terms are used to name the wing of an army and the wing of a cherub. It could have come from Syriac but additional studies should be performed to check this hypothesis. As explained by Sandfeld Jensen, it is sometimes difficult to differentiate between a loan translation and a coincidence⁸². Besides the Syriac origin, this loan translation could also have an Indo-European or an Armenian origin. Additionally, it exists in Greek and the Germanic languages. According to Sandfeld Jensen, there is no doubt that Eastern and Northern languages borrowed this loan translation from the Latin technical term āla, but the cultural and historical differences make it difficult to compare the Armenian Realien with the other contexts. There were fewer contacts between Armenia and Romans than with Germany, which was partly reduced to two Roman provinces. Sandfeld Jensen cites many languages that were spoken along the Rhine, and from the Danube to Thrace. It should be borne in mind that these were among the most important recruitment areas for auxiliary units divided into cōhors and āla. We do not know whether these loan translations are coincidences and whether they have a link with auxiliary units⁸³. From a linguistic point of view, all these loan words became assimilated to the Armenian phonetic system. This indicates that they had long existed as loan

78 This kind of borrowing is translated literally into the new language thanks to an equivalent. A loan translation can also be added to an existing term. Cf. Dubois et al. 1973, 72–73. 79 During the first phase of Christianization, the Greek and Syriac versions of the Bible circulated in Armenia. 80 Κέρας is notably used to name the horn of animals as well as the wing of an army. It corresponds to the Latin term cornū, -ūs, different from āla which is used for a wing of a bird, whereas cornū is used for the horn of an animal. 81 Occurrences for t‘ew: VT. 3 (1) Reg. 6.24. 6.27; 2 Ch. 3.11; Judith 14.2; 1 Macc. 3.23, 9.1, 9.12, 9.14; 9.16. 82 Sandfeld Jensen 1912, 167–168; Deroy 1956, chapter IX, § 3–7, 22. 83 See the case study in Traina 2013. The question of auxiliary units of Armenians will be treated in my PhD dissertation.


159

words in Armenian. In order to determine the context in which the loan words were borrowed, several fields such as archaeology, history, and ethnology are useful. Each source has its own limitations and the gaps may be filled by combining several different fields. As far as the Roman-Armenian relationship is concerned, literary sources are the first field that was studied. But these sources are biased in favor of the point of view of Romans and literature, as numismatics and epigraphy served to express the power and ideology of the Roman Empire. These textual sources must be complemented with archaeological research which supplies geographical data and artifacts such as militaria. Such data confirm the real evidence of the Roman presence and interaction in the Kingdom of Armenia, whereas literature brings us mostly diplomatic and factual information in the context of war. The Latin loan words into Armenian are polysemous lemmas. These “traveling words” are military borrowings which were used for technical terms or for the artifacts of Roman soldiers. Armenian soldiers in the Roman army may also have played a role in favor of the introduction of some loan words. However, no information about the context of their mission in the Roman army has come down to us. Therefore, it is difficult to draw a conclusion from such meagre information. Any inference about these five loan words must be made case by case. For the first three lexical loan words arkł, sitł, and *skutł it is still difficult to draw a conclusion. Further to this analysis, an uncertainty remains between the Latin and Greek origin for some lexical borrowings. The inscriptions discovered in the Armenian Kingdom demonstrate that Latin was used as the official language. However, as the recruitment of soldiers in the Roman army became more local, the Greek language was reinforced. The diplomatic exchanges between Rome and the Armenian Kingdom, as well as soldiers’ communication with local inhabitants, may well have been carried out in Greek, although the proportion of Armenians who were Greek-speakers is unknown. As far as the lexicological, archaeological and historical perspective is concerned, both hypotheses are valid. With regard to the borrowing sitł, this term could also have been borrowed during the Byzantine period, while as regards the origin of kaṙk῾, the study of the linguistic context performed in this article shows that the Latin and Galatian hypotheses remain. However, historical investigations indicate that an origin from Latin is more probable than from Galatian. Greek and Syriac hypotheses can be dismissed thanks to this interdisciplinary approach. As for the loan translation t‘ew, there can be no certainty on its origins. Perhaps it comes from the translation of the Syriac Bible which used qrn and knp with both different meanings of āla. The Greek hypothesis can be dismissed due to the use of different terms in the Bible. This loan translation could also

160

Anahide Kéfélian

have arisen from contacts with Roman soldiers and Armenians in the kingdom of Armenia, or from the presence of Armenian soldiers in the Roman army. It is also possible that this term has an Armenian origin. The second important issue is to determine whether the borrowings were due to contacts between Roman soldiers and Armenians, or were the consequence of the translation of the Bible, as in the case of the term legēōn. The latter is likely to have been a more recent borrowing, in contrast to the five other loan words, which, as suggested by their Armenian lexical and morphological features, were borrowed much earlier. Therefore, it would appear that arkł, sitł, *skutł, and kaṙk῾ were borrowed thanks to relationships among different ethnic groups. No doubt, further investigation will allow greater insight into the Roman relationship with local inhabitants and auxiliaries.

Bibliography Aṙakelyan, B. (1967), “Латинские надписи из древней столицы Армении-Арташат [Latin inscriptions from the ancient capital of Armenia, Artaxata]”, in: Patma-banasirakan Handes 4, 302–311 = Italian translation: “Epigrafi latine nella capitale Artašat”, in: Bazmavēp 126 (1968), 135–128. Baldi, S. (2008), Dictionnaire des emprunts arabes dans les langues d’Afriques de l’Ouest et en swahili, Paris. Balsdon, J. P. V. D. (1979), Romans and Aliens, London. Barbero, A. (2011), Barbares: immigrés, refugiés et déportés dans l’Empire romain, Paris. Bolognesi, G. (1966), “La tradizione culturale armena nelle sue relazioni col mondo persiano e col mondo greco-romano”, in: Atti del Convegno sul tema: La Persia e il mondo Greco-Romano, Roma 11–14 aprile 1965, Roma, 569–603. Butcher, K. (2003), Roman Syria and the Near East, Los Angeles. Canali De Rossi, F. (2004), Iscrizioni dello estremo Oriente Greco : un repertorio, Bonn. Curtius, E. (1874), Ephesos: Ein Vortrag gehalten im wissenschaftlichen Verein Berlin am 7. Februar 1874, Berlin. De Lamberterie, C. (1991), “Un poète hellénistique en Arménie”, in: A. Blanc / A. Christol (eds.), Langues en contact dans l’Antiquité: aspects lexicaux, Nancy-Paris, 151–167. De Rossi, G. B. (1880), “Adunanze dell’Instituto”, in: Bullettino dell’Instituto di corrispondenza archeologica 6, 139. Chaumont, M.-L. (1976), “L’Arménie entre Rome et l’Iran: de l’avènement d’Auguste à l’avènement de Dioclétien”, in: ANRW II, 9/1, 71–194. Daremberg, C. / Saglio, E. / Pottier, E. (eds.) (1873–1919), Dictionnaire des antiquités grecques et romaines d’après les textes et les monuments, Paris. Depeyrot, G. (2007), Optimo principi: iconographie, monnaie et propagande sous Trajan. I, La colonne Trajane, Wetteren. Depeyrot, G. (2008), Les légions romaines en campagne: la colonne trajane, Paris. Deroy, L. (1956), L’emprunt linguistique, Paris.


161

Donabédian, A. (1994), “Quelques remarques sur l’alphabet arménien”, in: Slovo 14, 7–21. Dubois, J. et al. (1973), Dictionnaire de linguistique, Paris, 1973. Forni, G. (1953), Il reclutamento delle legioni da Augusto a Diocleziano, Roma. Hadas-Lebel, M. (1990), Jérusalem contre Rome, Paris. Hewsen, R. H. (2001), Armenia: A Historical Atlas, Chicago. Hübschmann, J. H. (1895), Armenische Grammatik. I, Armenische Etymologie, Leipzig. Isaac, B. (1986), “Reflections on the Roman Army in the East”, in: P. Freeman / D. L. Kennedy (eds.), The Defense of the Roman and Byzantine East, Oxford, 383–395. Kéfélian, A. (2015), “L’autorité de Rome sur le Royaume d’Arménie: aperçu numismatique”, in: Laboratoire ERAMA, Les mises en scènes de l’autorité dans l’Antiquité, Paris-Nancy, 275–291. Khatchatrian, J. (1996), “The Archives of Sealings found at Artashat”, in: M.-F. Boussac / A. Invernizzi (eds.), Archives et sceaux du monde hellénistique en Arménie, Athènes-Paris, 365–370. Le Bohec, Y. (1989a), L’armée romaine sous le Haut-Empire, Paris. Le Bohec, Y. (1989b), Les unités auxiliaires de l’armée romaine en Afrique Proconsulaire et Numidie sous le Haut-Empire, Paris. Le Bohec, Y. (2006), L’armée romaine sous le Bas-Empire, Paris. Le Bohec, Y. (2007), L’armée romaine en Afrique et en Gaule, Stuttgart. Lee, A. D. (1986), “Embassies as Evidence for the Movement of Military Intelligence between the Roman and Sassanian Empires”, in: P. Freeman / D. L. Kennedy (eds.), The Defense of the Roman and Byzantine East, Oxford, 411–429. Macler, F. (1910), “Rapport sur une mission scientifique en Arménie russe”, in: Nouvelles Archives des Missions Scientifiques et Littéraires n. s. 2. Manandian, H. (1965), Trade and the Cities of Armenia in Relation to Ancient World Trade, Lisbon (transl. by Nina Garsoïan of O torgovle i gorodakh Armenii v svyazi s mirovoĭ torgovleĭ drevnikh vremyon, Erevan 1945). Mann, J. C. (1983), Legionary Recruitment and Veteran Settlement during the Principate, London. Marr, N. (1912), Известия Императорской Академии наук [Izvestija Imperatorskoj Akademii nauk], Saint Petersbourg. Mason, H. (1974), Greek Terms for Roman Institutions. A Lexicon and Analysis, Toronto. Meillet, A. (1913), “Hypothèses sur quelques emprunts de l’ancien arménien au latin”, in: Mémoires de la Société Linguistique de Paris 18, 248–350. Meyer-Lübke, W. (1911), Romanisches etymologisches Wörterbuch, Heidelberg. Moretti, L. (1955), “Due note epigrafiche”, in: Athenaeum 33, 32–46. Mosser, M. (2003), Die Steindenkmäler der Legio XV Apollinaris, Wien. Mutafian, C. / Van Lauwe, É. (2001), Atlas historique de l’Arménie: Proche-Orient du VIIIe siècle av. J.-C. au XXIe siècle, Paris. Nichanian, M. (1989), Ages et usages de la langue arménienne, Paris. Pollard, N. (2000), Soldiers, Cities, and Civilians in Roman Syria, Ann Arbor. Rajki, A. (2005), Arabic Etymological Dictionnary, [s. l.] (available online: https://archive.org/ details/ArabicEtymologicalDictionary). Reichenkron, G. (1961), “Armeniaca I. Teil a, Betrachtungen zu den latein-armenischen sprachlichen Beziehungen zur Zeit des Imperium Romanum”, in: Handes Amsorya 75, 1021–1043.

162

Anahide Kéfélian

Rostovtzeff, M. I. (1914), “A New Greek Inscription from Echmiadzin”, in: Christian Orient 3/3, 246–248. Russu, I. I. (1973), “Inscripţii romane dîn Praetorium (Mehadia)”, in: Banatica 2, 103–115. Sandfeld Jensen, K. (1912), “Notes sur les calques linguistiques”, in: Festschrift für Wilhelm Thomsen zur Vollendlung des siebzigsten Lebensjahres am 25 Januar 1912, Leipzig, 167–168. Sartre, M. (2001), “Les colonies romaines dans l’Orient grec: rapport de synthèse”, in: Electrum 5, 111–152. Saxer, R. (1967), Epigraphische Studien. I, Untersuchungen zu den Vexillationen des römischen Kaiserheeres von Augustus bis Diokletian, Köln. Sophocles, E. A. (1870), Greek Lexicon of the Roman and Byzantine Periods (from BC 146 to AD 1100), Boston. Spaul, J. (2000), Cohors²: the Evidence for and a Short History of the Auxiliary Infantry Units of the Imperial Roman Army, Oxford. Thumb, A. (1900), “Die griechischen Lehnwörter in Armenischen”, in: Byzantinische Zeitschrift 9/2, 388–452. Topchyan, A. (2007), “Jews in Ancient Armenia (1st Century BC-5th Century AD)”, in: Le Muséon 120, 435–476. Traina, G. (1999), “Rostovcev e l’epigrafia greco-latina dell’Armenia antica”, in: Arnaldo Marcone (ed.), Rostovcev e l’Italia, Napoli, 441–448. Traina, G. (1999–2000), “Épisodes de la rencontre avec Rome (IIe siècle av. J.-Chr.-IIIe siècle ap. J.-Chr.)”, in: C. Mutafian (ed.), Roma/Armenia, Catalogue de l’exposition (Vatican 1999), Roma, 32–41. Traina, G. (2005), “Notes on Hellenism in the Iranian East (Classico-Oriental Notes, 6–8)”, in: Iran and Caucasus 9/1, 1–14. Traina, G. (2007), “Les écritures ‘classiques’. L’utilisation du grec et du latin dans l’Arménie ancienne”, in: Claude Mutafian (ed.), Arménie: la magie de l’écrit, Paris-Marseille, 28–33. Traina, G. (2008), “Plutarque et le théâtre grec dans l’Arménie ancienne”, in: B. Der Mugrdechian (ed.), Between Paris and Fresno. Armenian Studies in Honor of Dickran Kouymjian, Costa Mesa, 311–319. Traina, G. (2010), “L’Armenia tra Roma e l’impero sassanide: aspetti diplomatici”, in: Bizantinistica. Rivista di Studi Bizantini e Slavi 11, 327–335. Traina, G. (2013), “Beobachtungen zur Inschrift von Maris, Casiti filius”, in: Zeitschrift für Papyrologie und Epigraphik 185, 279–285. Ter-Martirosov, F. (1989), “Греческая надпись из Арени” [Grečeskaja nadpis’ iz Areni], in: Istorico-filologiceskij zurnal, 1, 177–189. Trever, K. V. (1953), Očerki po istorii kul’tury drevnej Armeni, Moskva-Leningrad. Tudor, D. (1971), “Sirienii in Dacia Inferiora”, in: Apulum 9, 659–664. Vinogradov, J. (1992), “The Goddess Ge Meter Olybris. A New Epigraphic Evidence from Armenia”, in: East and West 42/1, 13–26. Wheeler, E. L. (2000), “Legio XV Apollinaris. From Carnutum to Satala – and Beyond”, in: Y. Le Bohec / C. Wolff, Les Légions de Rome sous le Haut-Empire. Actes du congrès de Lyon, 17–19 Septembre 1998, 261–308. Wünsch, J. (1884) “Inschriften aus Armenien”, in: Archäologisch-epigraphische Mitteilungen aus Österreich-Ungarn 8, 241–242.

Federico Frasson

Armenia and Armenians in Asinius Quadratus’ Παρθικά* Abstract: This paper examines the fragments of Asinius Quadratus’ Parthica concerning Armenia and suggests some hypotheses about the narrative context from which they were drawn. In particular, Quadratus’ words are examined in relation to specific historical events that took place near the eastern frontier of the Roman Empire.

On account of its distinctive geographic position, for centuries Armenia held an important strategic position in the political and military events that saw the Roman empire pitted first against the Parthians and subsequently against the Sassanids. Armenia is thus seen as having played a pivotal role in the complex political and diplomatic relations and conflicts among the powers contending for ascendancy in the region, although the classification – traditionally adopted in the critical literature – that views Armenia as a mere buffer state¹ may be an oversimplification. However, it is undeniable that both Rome and the Iranian empires expended considerable efforts on attempting to draw the kingdom of Armenia into their own sphere of influence. Eager to take advantage of Armenian territory as a bridgehead from which to launch their attacks, they sought, in particular, to assure the rise to the throne of candidates who would be favourable to their cause². It is therefore hardly surprising to find some names of sovereigns, peoples, cities and regions of Armenia – a land repeatedly traversed or invaded by the Roman legions and the Iranian armies – appearing in a work such as the Παρθικά by Asinius Quadratus, as it is specifically devoted to the wars fought by

* This paper was composed in the framework of the University research project “Territorio e paesaggio: percorsi storico antichi e storico artistici”, coordinated by Prof. Giovanni Mennella and carried out at the University of Genoa, Italy. I wish to thank Prof. Giusto Traina and Prof. Francesca Gazzano for their valuable suggestions and for allowing me to publish the results of my research in the present collection. 1 Cf. Traina 1998, 112. 2 The bibliography concerning the relations and conflicts between the Romans and the Iranian empires is boundless; here we give a generic reference to several texts that provide a broad overview, together with the further bibliographical references contained therein: Von Gutschmid 1888; Debevoise 1938; Angeli Bertinelli 1976; Angeli Bertinelli 1979; Wolski 1993; Verstandig 2001; Winter/Dignas 2001; Zecchini 2005.

164

Federico Frasson

the Romans against the Parthians during the second half of the second century AD, and perhaps also to the conflicts that occurred in the first half of the following century against the Parthians themselves and the first Sassanids. The Παρθικά are thought to have perhaps been written in order to celebrate the military campaign conducted by Alexander Severus in 232–233 against the first Sassanid sovereign Ardashir, or alternatively after the war which, ten years later, featured Gordian III, the Praetorian Prefect Timesitheus and the future emperor Philip (the Arab)³. This work is known above all from fragments cited, in the overwhelming majority of cases, in entries forming part of the Ἐθνικά by Stephanus of Byzantium. The number of fragments ranges from a minimum of 13 to a maximum total of 22⁴, the latter number being reached on the basis of a twofold assumption: firstly, that the probable double citations of a certain passage from Quadratus (frr. 8a–b, 29a–b = FRH, 2, 1136, 1142, frr. 15a–b, 29–30) are not conflated, and secondly, that the count includes not only certain fragments scholars usually attribute to the Παρθικά, although an indication of the name of the work from which they were taken is lacking, but also the controversial fr. 20⁵, whose inclusion in the Παρθικά is open to doubt as it cannot be excluded that it stems from the Χιλιετηρίς⁶. The uncertainty concerning the period in which the work was composed, which is an aspect of the broader chronological problem encountered when detailing the historiographic activity of Asinius Quadratus⁷, also affects judgments concerning the extension of the Παρθικά and the topics addressed therein. It would appear that the only points on which clear-cut statements can be made concern the information that can be derived from fr. 9 and from frr. 16–18. In par-

3 The first hypothesis is indirectly suggested in HRR, CLXXXXVII (cf. recently FRH, 1, 613), while for the second see Manni 1971, 195 and, in particular, Zecchini 1998, 3004–3006, 3009. 4 In the present paper, as for the reference to the individual fragments, unless otherwise indicated, the same numbering and text as adopted by Jacoby (Asin. FGrHist 97 frr. 5–20, 27–30) will be used. 5 Asin. FGrHist 97 frr. 5–18, 27–30 = FRH, 2, 1134, 1136, 1138, 1140, 1142, frr. 12–24, 26–31 (cf. Steph. Byz. s. vv. variis); cf. formerly FHG, 659–661, frr. 1–19; HRR, 144–146, frr. 11–29. Asin. FGrHist 97 fr. 19 = FRH, 2, 1140, fr. 25 (cf. SHA Verus 8.1–4); cf. formerly FHG, 661, fr. 20; HRR, 146, fr. 30. Asin. FGrHist 97 fr. 20 = FRH, 2, 1132, fr. 6 (cf. SHA Avid. Cass. 1.1–3); cf. formerly FHG, 661, fr. 21; HRR, 146–147, fr. 31*. 6 On the problem of the attribution of fr. 20, as well as of its authenticity as a citation – direct or indirect – of Asinius Quadratus, cf. recently Frasson 2013, 318–319, with the bibliography cited therein, to which should be added FRH, 3, 649, fr. 6. 7 Although other proposals have been put forward, the main doubt concerning the period in which he wrote focuses mainly on whether he wrote during the reign of Alexander Severus or that of Philip the Arab; according to a recent hypothesis, it is conceivable that Quadratus may have been the court historian of the latter (for an overview of the chronological question, see, most recently, Frasson 2013, 310–312).

Armenia and Armenians in Asinius Quadratus’ Παρθικά

165

ticular, fr. 9, as will be described in greater detail below, almost certainly involves a reference to events dating back to 163, while the other fragments are taken from book 9 of the Παρθικά, i.e. from the last one explicitly mentioned by Stephanus. It follows that the Παρθικά most certainly dealt with the Parthian war conducted by L. Verus (or rather, by his generals) between 161/2 and 166, and that the main body of the work must have been composed of at least 9 books, but the impossibility of identifying with absolute certainty the background context in the fragments belonging to book 9 leads to considerable complications in estimating the chronological extension of the period covered by the Παρθικά. However, it seems fairly likely that Quadratus did at least describe the expeditions of Septimius Severus pertaining to the years 195 and 197, and that he probably also dealt with the expedition headed by Caracalla (216–217) and that of Alexander Severus (232–233). He may conceivably even have proceeded as far as the campaign led by Gordian III (241–244), if one allows the conjecture that the author of the Παρθικά held the position of historian at the court of Philip the Arab⁸. Given the probable extension of the work, as has already been hypothesized⁹, the number of books is likely to have been between 10 and 12. A total of 9 books would hardly seem sufficient, although a subdivision into 9 books could possibly have been chosen as a significant echo of Herodotus’ Histories, which certainly acted as a model for Quadratus. For instance, it is no coincidence that Quadratus wrote his history of Rome, the Χιλιετηρίς, in Ionian dialect, thus imitating the historian from Halicarnassus¹⁰. Moreover, precisely because Quadratus shared with Flavius Arrianus (henceforth: Arrian) a sense of following in the footsteps of the Herodotean historiographic approach, it has been suggested that he may have wished to portray himself as a continuator of Arrian¹¹, whose Παρθικά, according to the testimony of Photius¹², dealt with the question of the emperor Trajan’s Parthian war (115–117) in 17 books.

8 Thus in Zecchini 1998, esp. 3010; cf. supra, n. 7. For an extension of the work up to the military campaign of Alexander Severus, see, most recently, FRH, 1, 615. 9 Zecchini 1998, 3010. 10 Suda κ 1905, s.v. Κοδράτος = Asin. FGrHist 97 T 1 = FRH, 2, 1128, T 5; cf. formerly FHG, 659; HRR, CLXXXXV. On the Χιλιετηρίς, see recently Zecchini 1998, 3014–3018; Frasson 2013; FRH, 1, 613–615. 11 FGrHist, IIC, 97, 301–302; Zecchini 1998, 3009–3010; FRH, 1, 615. 12 Phot. Bibl. 58, 17a Bekker = Arr. FGrHist 156 T 2; fr. 30 = fr. 1 (Parth.) Roos. It may not be a coincidence that Arrian and Quadratus are mentioned in close succession, as sources for the imperial era, on a list of historiographers drawn up by Evagrius Scholasticus (sixth century AD), who was probably referring to the Παρθικά by these authors (Evagr. HE 5.24 = Asin. FGrHist 97 T 3 = FRH, 2, 1128, T 3; cf. previously FHG, 659; HRR, CLXXXXV. See HRR, CLXXXXVI; FGrHist, IIC, 97, 301; Zecchini 1998, 3005, 3009).

166

Federico Frasson

Study of the fragments on Armenia from Asinius Quadratus’ Παρθικά, preserved by Stephanus of Byzantium, allows a further hypothesis on the structure and content of Quadratus’ work. The fragments are listed here according to the numbering and the text of Jacoby: Asin. FGrHist 97 fr. 8a = FRH, 2, 1136, fr. 15a (cf. Steph. Byz. 713.5–7 Meineke, s.v. Ὠτηνή); cf. previously FHG, 660, fr. 6: Ὠτηνή· … τὸ ἐθνικὸν ὁ αὐτὸς ἐν τῇ βʹ· “περὶ δὲ Κῦρον ποταμὸν Ὠβαρηνοί τε καὶ Ὠτηνοὶ νέμονται, μέγα μέρος Ἀρμενίας ὄντες”. Asin. FGrHist 97 fr. 8b = FRH, 2, 1136, fr. 15b (cf. Steph. Byz. 705.10–13 Meineke, s.v. Ὠβαρηνοί); cf. previously FHG, 660, fr. 8; HRR, 145, fr. 18: Ὠβαρηνοί· μέρος Ἀρμενίας παρακείμενον Κύρῳ ποταμῷ, ὅστις Εὐφράτην παραλλήλως ἔχει. Μέμνηται δ᾽ αὐτῶν καὶ Κουάδρατος ἐν ϛʹ Παρθικῶν· “παρὰ δὲ Κύρῳ ποταμῷ Ὠβαρηνοί τε καὶ Ὠτηνοὶ νέμονται, μέγα μέρος Ἀρμενίας ὄντες”. Asin. FGrHist 97 fr. 9 = FRH, 2, 1136, fr. 16 (cf. Steph. Byz. 713.3–5 Meineke, s.v. Ὠτηνή); cf. previously FHG, 660, fr. 6; HRR, 144, fr. 16: Ὠτηνή· μοῖρα Ἀρμενίας. Κουάδρατος ἐν Παρθικῶν γʹ· “ὁ δὲ τῆς Ἀρμενίας βασιλεὺς Πάκορος ἐν τούτῳ περὶ Ἀρτάξατα καὶ τὴν Ὠτηνὴν τῆς Ἀρμενίας διάγων”. Asin. FGrHist 97 fr. 18 = FRH, 2, 1140, fr. 26 (cf. Steph. Byz. 623.9–12 Meineke, s.v. Τιγρανόκερτα); cf. previously FHG, 660, fr. 14; HRR, 145, fr. 24: Τιγρανόκερτα· πόλις, οὐδετέρως, πρὸς Ἀρμενίαν, ἀπὸ Τιγράνου βασιλέως Ἀρμενίας. Κουάδρατος θʹ· “καὶ ᾤκισε τὰ Τιγρανόκερτα. Τὸ δ᾽ ἐστὶ τῇ Παρθυαίων φωνῇ Τιγρανούπολις”.

As a first level of analysis, it is helpful to take into consideration the geographic names present in each fragment, in order to formulate hypotheses concerning the narrative context in which they were set. Frr. 8a and 9 are taken from the entry Ὠτηνή of the Ἐθνικά, in which it is explicitly stated that Otene was a region of Armenia. Its geographic identification is aided by the citation of Quadratus (fr. 8a), which Stephanus quotes in order to provide an example of the ethnonym of the population of the region in question, namely the Ὠτηνοί. According to the historiographer, they lived in the area through which there flowed the river Κῦρος, the present-day Kura¹³, which rises near the north-eastern border of Turkey and then continues through Georgia and Azerbaijan, until it eventually reaches the Caspian Sea. Quadratus further states

13 On the river Κῦρος and on the Greek and Latin variants of the hydronym, cf. Weissbach 1924; the same river also appears in another fragment of Asinius Quadratus: FGrHist 97 fr. 27 = FRH, 2, 1140, fr. 27 (cf. Steph. Byz. β 112 Billerbeck, s.v. Βόγχαι); cf. formerly FHG, 660, fr. 15; HRR, 145, fr. 25.


167

that the banks of this river were also home to the Ὠβαρηνοί, to whom Stephanus of Byzantium devotes a specific entry, citing once again Asinius Quadratus. However, since the wording of the statement by the historian as reported in the entry for Ὠβαρηνοί (fr. 8b) is virtually identical to the words given in the entry for Ὠτηνή (fr. 8a), it is likely that the lexicographer cited the same passage of the Παρθικά twice, wrongly attributing its origin to two different books (respectively 6 and 2)¹⁴, although it cannot be ruled out that the discrepancy should instead be ascribed to the manuscript tradition of Stephanus. The length of the river Kura, which extends for over 1500 km with a course roughly parallel to that of the Euphrates¹⁵, or at least to the Mesopotamian stretch of the latter, means that the watercourse is in itself not a reliable indicator that could demarcate specific areas where the Ὠτηνοί and the Ὠβαρηνοί may have dwelt. Thus the information provided by fragments of Asinius Quadratus can do no more than place them generically near the north-eastern borders of ancient Armenia¹⁶, but further insight can be gleaned from other literary sources. In particular, as far as Otene is concerned, a significant statement can be found in Pliny, who asserts that Atrapatene ab Armeniae Otene regione discreta Araxe¹⁷, in other words by the present-day river Aras¹⁸, which marks the boundary between several different States (Turkey, Armenia, Iran and Azerbaijan). Accordingly, if credence is given to Pliny, the region of Otene can be assumed to have been located in the north-eastern corner of ancient Armenia, between the Kura and the Aras¹⁹, taking into account that before the Aras became an affluent of the Kura, it was the Aras itself that marked the boundary between Armenia (Otene) and Media Atropatene, exactly as it divides Azerbaijan from Iran over a certain stretch of its course, in the present-day situation.

14 See HRR, 145, fr. 18, where the words of Quadratus are attributed to book 6, and FGrHist 97 frr. 8a-b, in which preference seems to be awarded to book 2; cf. FRH, 3, 652, fr. 15. In contrast, the argument that these are two distinct fragments is put forward by Zecchini 1998, 3011, 3013, who associates the former (fr. 8a) with the military operations that culminated in the death of M. Sedatius Severianus in 161 (cf. infra) and the latter (fr. 8b) with the description of the new set-up imposed on Armenia and Mesopotamia at the end of the war waged by L. Verus’ generals between 162 and 166. 15 Cf. fr. 8b: ὅστις Εὐφράτην παραλλήλως ἔχει. 16 The river Κῦρος separated Armenia from Iberia and from Albania: cf. Strab. 1.3.21; 11.3.2; Plin. HN 6.10.29; 6.13.39; Ptol. Geog. 5.12.1; 5.13.1 Nobbe = Stückelberger/Graßhoff; see e.g. Weissbach 1924, 186. On this river as a boundary between Albania and Armenia, cf. recently Bais 2001, 156–169. 17 Plin. HN 6.13.42; the same author mentions Otene in 12.13.49 as well. 18 Cf. e.g. Tomaschek 1895. 19 This location also seems to be preferred in Barrington Atlas 88 F4; Talbert 2000, 1296; Mutafian/Van Lauwe 2001, 31, 33.

168

Federico Frasson

Otene, which can plausibly be regarded as corresponding to the region which Armenian historians and geographers designate by the name Utikʽ²⁰, is also mentioned by Ptolemy and Eusebius of Caesarea, as well as in Herodian’s De prosodia catholica²¹. In particular, Ptolemy’s list of the regions of Greater Armenia makes a reference to Ὠτηνή²² together with Τωσαρηνή, as these are both situated παρὰ δὲ τὸν Κύρον ποταμόν. Precisely the term Τωσαρηνή, which in the manuscripts is transmitted in a number of variant forms, including Ὀσαρηνή and Ὀσσαρηνή, has sometimes been set in relation with the above cited ethnonym Ὠβαρηνοί and has been dubitatively corrected to Ὠβαρηνή²³. But according to another hypothesis, the term should be considered in association with the choronym Γωγαρηνή²⁴, although the eventuality that reference is being made to the same region in all these cases can by no means be ruled out. Furthermore, if this region were Gogarene²⁵, then it would effectively not be far from the river Kura²⁶, like the historical province of Gugarkʽ, to which it would most likely correspond²⁷. If one accepts the latter solution, and taking into account not only the geographic position of Gugarkʽ but also the words of Strabo and Stephanus of Byzantium concerning the Γωγαρηνή²⁸, the hypothesis can be put forward that the Ὠβαρηνοί, mentioned together with the Ὠτηνοί by Quadratus (frr. 8a-b), lived in an area that may have

20 Cf. e.g. Polaschek 1961; Hewsen 1982, 114, 148; see formerly Saint-Martin 1818, 86–87; James 1873c; Müller 1901, 938; Hübschmann 1904, 270–275. See also the maps in Mutafian/Van Lauwe 2001, 33, 37, 39, 41, 129. 21 Cf. respectively Ptol. Geog. 5.13.9 Nobbe (cf. 2, 550 Stückelberger/Graßhoff; 5.12.4 Müller); Eus. Praep. Evang. 6.10.31; Hdn. Pros. cath. 332.10 Lentz (but in the latter, Otene is defined as μοῖρα Ἀραβίας); see also Rav. Cosm. 2.12, p. 22 Schnetz. It has sometimes been suggested (Sturm 1935) that the region may have corresponded to the Madena of Eutr. 8.3.2 and the Madaena of Ruf. Fest. 15. 22 It should be noted, however, that the form Ὠτηνή, which is followed in the editions of Ptolemy’s Geographia edited by Nobbe and by Müller, is a correction; the variants actually present in the manuscripts are Μοτηνή, Μωτηνή, Ὠπηνή, Τωμηνή, Τοτηνή and Τωτηνή (see Müller 1901, 938; cf. Polaschek 1961, 1179; see also Sturm 1935), the latter being adopted in the recent edition by Stückelberger and Graßhoff (2, 550), although in the text Ὠτηνή is nevertheless indicated, in brackets, as an alternative. 23 Müller 1901, 937; Sturm 1937a; Sturm 1937b. 24 Cf. Sturm 1937a; Sturm 1937b. 25 For this hypothesis, see formerly De Sainte-Croix 1798, 115; Saint-Martin 1818, 81; James 1873b; cf. Müller 1901, 937–938 and more recently Hewsen 1982, 148. 26 Strab. 11.14.4–5. 27 See e.g. Hübschmann 1904, 275–276; Kiessling 1912; Hewsen 1982, 114, 148; Bøe 2001, 209; cf. also the maps in Mutafian/Van Lauwe 2001, 29, 33, 35, 37, 39, 41, 45. 28 Strab. 11.14.4–5; cf. Steph. Byz. γ 122 Billerbeck, s.v. Γωγαρηνή (Γωγαρηνή· χωρίον μεταξὺ Κόλχων καὶ Ἰβήρων τῶν ἀνατολικῶν. Tὸ ἐθνικὸν Γωγαρηνός); cf. also Hdn. Pros. cath. 332.16–17 Lentz.


169

bordered on the land of the Ὠτηνοί, but was probably located somewhat further north-west, in the area of the present-day Tashir and Ashotsʽ. It may even have extended as far as the regions of Javakhetʽi, Artani and Kola, as well as to the valleys of the Debeda, the Aghstafa and the Inja²⁹. Fr. 9 mentions not only Otene, but also the city of Ἀρτάξατα, in other words the Armenian Artašat (Lat. Artaxata), one of the ancient “capitals”³⁰ of Armenia. Built between 180 and 170 BC during the reign of Artaxias I, it stood, according to Strabo, on a small peninsula surrounded on three sides by the waters of the river Aras³¹. The probable remains of the city have been discovered in the area of the monastery of Khor Virap³², in the Armenian province of Ararat, roughly 30 km south of Erevan and not far from the Turkish city of Aralık, i.e. about 5 km south of the city which today bears the name of Artašat³³. Like fr. 9, fr. 18 also contains a reference to an Armenian “capital”, bearing a name, Τιγρανόκερτα, which also appears elsewhere as a recurrent designation of ancient Armenian cities³⁴. But it seems extremely likely that the city cited by Asinius Quadratus should be identified with the more famous Tigranocerta (Arm. Tigranakert) founded by Tigran II and known above all for the great battle that was fought below its walls in 69 BC³⁵. As far concerns the site where this

29 Toumanoff 1963, 185–186; cf. Barrington Atlas 88 B3, C3. 30 As has sometimes been underlined (Traina 2010, 45), it is incorrect and anachronistic, as far as ancient Armenia is concerned, to speak of capital cities, since the royal power was not linked to a specific centre but moved with the presence of the king as the latter travelled through his lands. 31 Strab. 11.14.6 (in actual fact, the peninsula was perhaps created by the confluence of the Metsamor with the Aras: cf. MX 2.49); on the city – and for further references to the latter in the ancient sources, omitted here for the sake of brevity – cf. among others James 1873a; Baumgartner 1895 and more recently Aṙakʽelyan 1984; Grousset 1984, 82; Pasdermadjian 1986, 32–33; Chahin 1987, 222–223; Hewsen 1987; Mahé/Mahé 1993, 363 n. 3 ad 2.49; Traina 1999–2000, 63–64; Traina 2001, 146, 149; Traina 2010, 45–46; Khatchatrian 2010. 32 On the probable ruins of Artaxata and the results of the archaeological investigations, cf. e.g. Aṙakʽelyan 1984; Kanetsian 1998, 6–7, 15–26, 33–35, 43–49, 51, 55–63, 65–73, 75–78, 80–83, 86–94; Khachatrian 1998; Khatchatrian 2010; on the probable site of the ancient city, cf. also, among others, Hewsen 1987, 653; Barrington Atlas 89 G1; Talbert 2000, 1271; see earlier Baumgartner 1895. 33 On the contrary, Ἀρτάξατα was identified with Artašat by Lehmann-Haupt 1910, 173–176. 34 One may call to mind, for example, the city of Tigranakert in Artsakh, where in recent years archaeological excavations have been carried out (cf. e.g. Petrosyan 2010a; Petrosyan 2010b; Petrosyan et alii 2012, with the earlier bibliography indicated therein). 35 On Tigranocerta and on the ancient sources concerning the city, cf. e.g. Dyer 1873; Henderson 1903; Lehmann-Haupt 1910, 381–429, 501–523; Markwart 1930, 86–92, 112–116, 119–121; Lehmann-Haupt 1936; Chaumont 1982; Grousset 1984, 88–89; Pasdermadjian 1986, 38–39; Chahin 1987, 229, 241; Sinclair 1994–1995; Sinclair 1996–1997; Kanetsian 1998, 8–9; Traina 1999–2000,

170

Federico Frasson

ancient city stood, there is no unanimous agreement among scholars. Among the various solutions proposed, preference has often been awarded to an identification with the present-day city of Silvan (Byzantine Μαρτυρόπολις; Ott. Turk. Maiyāfāriḳīn)³⁶, which lies about 80 km north-east of Diyarbakır, in south-eastern Turkey. Recently, however, Sinclair has revived the proposed identification with a site slightly further east, namely the ruins of the walled city of Arzan (district of Kurtalan, in the province of Siirt), on the eastern banks of the Garzan Su³⁷. In the light of what it has been possible to ascertain regarding the localization of each of the toponyms cited in the fragments examined, some hypotheses can now be put forward on the background context to which the fragments belong. It is helpful to start with fr. 9, which seems to present fewer difficulties, since it includes not only a mention of the poleonym Ἀρτάξατα, but also a fairly significant personal name, Πάκορος. The latter can almost certainly be identified as the Pacorus who, after becoming the sovereign of Greater Armenia upon the initiative of the king of the Parthians Vologaeses IV (III) around 161³⁸, is said to

69–71; Traina 2001, 141–143, 150–154; Chaumont 2002, 84–85; Garsoïan 2005; Avdoyan 2006; Traina 2010, 45–47; Hakobyan 2010. 36 See e.g. Lehmann-Haupt 1910, passim; Lehmann-Haupt 1936, 985–989, 1007; Manandian 1963, 23, 54–58, 88, 91 (map), 93, 135 (map); Hewsen 1982, 135, 137–138, 140, 143, 150; Pasdermadjian 1986, 39; Hakobyan 2010, 99–100; for an overview of other possible locations of the ancient Tigranocerta and on its identification with Tel Ermen, south of the Turkish city of Mardin, cf. Henderson 1903; Holmes 1917; this identification has been favoured more recently by Chaumont 1982; Chaumont 2002, 84. 37 Sinclair 1994–1995; Sinclair 1996–1997; see also Barrington Atlas 89 D3; Talbert 2000, 1280; Avdoyan 2006, 95; Traina 2010, 45–47. Described by Taylor 1865, 26–28, Arzan was identified with Tigranocerta as early as 1873 (cf. Kiepert 1873) and was taken into consideration as a possible site of the ancient city by, among others, Lehmann-Haupt 1910, 385; Lehmann-Haupt 1936, 987, who however eventually opted in favour of the present-day Silvan, which he referred to with its Kurdish name (Farkin). A proposal to reconsider the site of Arzan is already contained in Nogaret 1984 (esp. 424–425), where the limits of the Lehmann-Haupt hypothesis are pointed out. 38 Pacorus is thought to have come to power after the Parthian invasion of Armenia and the victory that the army led by General Osroes obtained over the legions of M. Sedatius Severianus at Elegeia (cf. infra, n. 41). Nevertheless, it cannot be completely ruled out that he already reigned over Armenia before the arrival of the Parthians; if this were the case, then one could assume that the Parthians had simply decided to leave him on the throne. Such a decision would be even more understandable if the sovereign was indeed of Arsacid descent, as has sometimes been suggested (cf. e.g. OGIS, 590 ad 382; Chaumont 1976, 147–149; contra Juntunen 2013, 169–170). Pacorus is generally identified – albeit with a degree of uncertainty as regards the correspondence – with Αὐρήλιος Πάκορος, βασιλεὺς Μεγάλης Ἀρμενίας who, as can be read in a funerary inscription deriving from Rome (CIG 3, 6559 = IG 14, 1472 = OGIS 382 = IGRR 1, 222 = IGUR 2, 415), dealt with purchasing the sarcophagus for his deceased brother; Pacorus has also been identified with the character whose name is engraved on a silver bowl found in 1934 or 1935 in


171

have been deposed and replaced by Sohaemus. The source of this information is found in the writings of Fronto³⁹, who refers to the events that took place after the Roman victory of 163 AD, when the legate of Cappadocia M. Statius Priscus penetrated into Armenian land and took possession of Artaxata. If, as is very likely, the identification of the Πάκορος in fr. 9 of Asinius Quadratus with the sovereign deposed in 163 were to prove to be correct⁴⁰, then book 3 of the Παρθικά, from which Stephanus of Byzantium claims to have taken the historian’s words, presumably dealt with the so-called Armenian phase of L. Verus’ Parthian campaign⁴¹. In any case, the fragment appears to allude to a time when Pacorus was still on the throne, given that Quadratus cites the places where the sovereign habitually spent time, namely the “capital” Artaxata, where he certainly had his palace, and the region of Otene, which has sometimes been suggested as the location of his winter residence⁴². While the context can thus be identified with a fair degree of certainty for fr. 9, the same cannot be said for the other fragments concerning Armenia, starting from fr. 8a, which provides information of an exclusively geographic nature. As pointed out earlier, it speaks of the Obareni and the Oteni, whose territory must have extended over a large part of north-eastern Armenia, so much so that for Quadratus Ὠβαρηνοί τε καὶ Ὠτηνοὶ were μέγα μέρος Ἀρμενίας. Assuming that the historiographer’s words were genuinely taken from book 2 of his Παρθικά and not from book 6, as indicated elsewhere by Stephanus of Byzantium (fr. 8b; cf. supra),

the northern Caucasus (SEG XV 838; cf. earlier e.g. Moretti 1955, 45). On the figure of Pacorus, cf. among others Marquart 1905, 225–227; PIR2 1, 320, A 1566; Debevoise 1938, 249; Chaumont 1969, 16–17; IGUR 2.1, 85 ad 415; Chaumont 1976, 147–149; Braund 1984, 43, 45; Van den Hout 1999, 302; Chaumont 2002, 98; FRH, 3, 652, fr. 16; Juntunen 2013, 169–170. 39 Fronto Ep. ad Verum 2.18 Van den Hout; cf. Van den Hout 1999, 301–302. On Sohaemus, see most recently Juntunen 2013, 158–160, 163. 40 Among those who have published the fragments of Asinius Quadratus, only Karl Müller has put forward the suggestion that Pacorus should be seen as the king of the Parthians who fought against Trajan (FHG, 659), whereas the others, like almost all the scholars (see e.g. supra n. 38), have opted for the sovereign from the era of Marcus Aurelius (cf. HRR, 144 ad fr. 16; FGrHist, IIC, 97, 302; FRH, 3, 652, fr. 16; cf. also Zecchini 1998, 3011). 41 On the issue of the early signs of the impending war that would be waged by L. Verus, on the defeat of Severianus, on the so-called Armenian phase of the conflict and on the relevant sources, cf. e.g. Von Gutschmid 1888, 147–148; Marquart 1905, 224–227; Debevoise 1938, 246–249; Pareti 1960, 309–312; Chaumont 1969, 16–17; Angeli Bertinelli 1976, 25–27; Chaumont 1976, 147– 149; Angeli Bertinelli 1979, 76–77; Mitford 1980, 1203–1205; Bivar 1983, 93; Grousset 1984, 111–112; Pasdermadjian 1986, 82; Birley 1987, 121–126, 128–129; Wolski 1993, 185; Verstandig 2001, 324, 326; Chaumont 2002, 98; Fraschetti 2008, 64–68. 42 Marquart 1905, 225 n. 2, who hypothesizes that it was located at Chałchał, as in the Sassanid era (cf. Hübschmann 1904, 272–273).

172

Federico Frasson

one may conjecture that Quadratus made a reference to the aforementioned populations in the initial part of his work, certainly before narrating the events of 163, which, as stated above, are outlined in book 3. However, the interpretation of fr. 8a cannot be put forward independently from that of the other fragments pertaining to the first two books of the Παρθικά, namely frr. 5–6, deriving from book 1, and fr. 7, taken from book 2, because, although these fragments do not refer to Armenia, they are probably set in a chronological background not unlike that of fr. 8a. Jacoby, in his edition of the fragments of Quadratus, suggested that book 1 may have contained “eine kurze geographische Einleitung und Zusammenfassung der älteren Geschichte”, and that this introductory part may have included at least fr. 5, concerning the Gelae, a population of Media Atropatene (see infra)⁴³, and fr. 28, in which the German scholar saw a reference to the battle of Mount Gindarus of 38 BC⁴⁴. Jacoby himself justified the presence of the Moors of fr. 6 as a reference to a unit of the army, though he gave no explicit indications concerning fr. 7 (on the Zobidae) and frr. 8a-b⁴⁵. On the other hand, in a recent re-examination of the fragments of Asinius Quadratus, Zecchini has hypothesized that all the fragments taken from books 1 and 2 referred to events immediately preceding 163. In particular, Zecchini argues that the names of populations present in frr. 5–7 were quoted by Quadratus in a

43 FGrHist, IIC, 97, 302; cf. FRH, 1, 616: “it perhaps began with an ethno-geographical introduction”. For the fragment on the Gelae, see Asin. FGrHist 97 fr. 5 = FRH, 2, 1134, fr. 12 (cf. Steph. Byz. γ 72 Billerbeck, s.v. Γηλύς); cf. earlier FHG, 659, fr. 1; HRR, 144, fr. 11. According to FRH, 3, 651, fr. 12, the fragment in question, seen as belonging to an ethno-geographic introduction, very likely formed part of a passage concerning the origins of the Parthians. 44 FGrHist, IIC, 97, 302–303; FGrHist 97, 448, 451 ad fr. 28. Cf. Asin. FGrHist 97 fr. 28 = FRH, 2, 1142, fr. 28 (cf. Steph. Byz. γ 79 Billerbeck, s.v. Γίνδαρα); cf. earlier FHG, 661, fr. 16; HRR, 146, fr. 26. Fr. 28 could perhaps be regarded as forming part of the Παρθικά, but this attribution is a mere conjecture, since the fragment is transmitted by a figure (Stephanus of Byzantium) who restricts himself to naming Quadratus as his source, without any mention of the name of the work from which he took it. Although the existence of an introductory part would accord well with the characteristics of a broad-based work such as that of Quadratus, which was probably made up of over 10 books, the problems concerning fr. 28 suggest the fragment in question should not be taken into consideration in the framework of this conjectural hypothesis. The cautionary approach is due not only to the aforementioned problems of attribution, but also to doubts concerning the background setting to which the fragment could have belonged: for instance, it could have been a reference to Avidius Cassius’ march from Antioch to the Euphrates (cf. Zecchini 1998, 3012) rather than to Ventidius Bassus’ victory over Pacorus in 38 BC. 45 FGrHist, IIC, 97, 302; cf. Asin. FGrHist 97 fr. 6 = FRH, 2, 1134, fr. 13 (cf. Steph. Byz. μ 101 Billerbeck, s.v. Μαυρούσιοι καὶ Μαῦροι); cf. earlier FHG, 659, fr. 2; HRR, 144, fr. 12; Asin. FGrHist 97 fr. 7 = FRH, 2, 1136, fr. 14 (cf. Steph. Byz. ζ 32 Billerbeck-Zubler, s.v. Ζωβίδαι); cf. earlier FHG, 659, fr. 3; HRR, 144, fr. 13.


173

list of the contingents that composed the Parthian and the Roman armies. This solution has been deemed preferable as compared to the hypothesis of an introduction of a geographic nature on the area involved in the military operations, since the lands of the Gelae (who dwelt in the northern part of Media Atropatene, not far from the Caspian sea) and of the Zobidae (who lived near Carmania) did not form part of the area involved in the combats of the Armenian phase of the conflict⁴⁶. Furthermore, the reference to the Moors, as in some sense already hinted by Jacoby, can be explained only in terms of a reference to the auxiliary cavalry that this North African population supplied to the Romans⁴⁷, who exploited its dynamism as a means to fight against the cavalry of the Parthian armies⁴⁸ and probably also made use of it during the expedition headed by L. Verus⁴⁹. However, since fr. 6 concern the description of the Roman army, it would surely have preceded fr. 5 which, like fr. 7 that is taken from book 2, concerns the Parthian army. In turn, fr. 8a should be seen in connection with the events of 161, under the hypothesis that book 2 described the successes achieved by the Parthians in 161–162⁵⁰. This is undoubtedly an attractive hypothesis, although for fr. 5 and fr. 7 it requires the further hypothesis, as already pointed out by the original proponent, that the description of the Parthian army was contained partly in the first book

46 Cf. Zecchini 1998, 3010, where it is argued that in addition to the Gelae and the Zobidae, the Tapurri whose name is mentioned in fr. 30 = FRH, 2, 1142, fr. 31 (Stephanus does not indicate the number of the book of the Παρθικά from which he drew the fragment) were likewise stated to have formed part of the contingents that made up the Parthian army (cf. FRH, 3, 655, fr. 31). On the subject of the Gelae and the lands where they dwelt, cf. e.g. Weissbach 1910; see recently Zecchini 1998, 3010; Barrington Atlas 90 D3, E3; Talbert 2000, 1294; FRH, 3, 651, fr. 12; on the Zobidae, see Kretschmer 1927; Weissbach 1932; cf. recently Zecchini 1998, 3010; FRH, 3, 652, fr. 14. 47 Zecchini 1998, 3010; cf. also FRH, 3, 651–652, fr. 13. 48 Gabba 1966, 72–73 = Gabba 1974, 40–41; Zecchini 2005, 77–78; on the utilization of Moorish auxiliaries during the wars the Romans waged in the East, cf. e.g. also FRH, 3, 651–652, fr. 13. 49 Evidence for the presence of Moorish cavalry in the armies that took part in L. Verus’ Parthian war is usually based on a passage from Lucian, where the author ridicules a historian who, it was said, included in his story a whole host of events that had not yet happened, such as a far-fetched account of an expedition by Avidius Cassius all the way to India: among the various other contingents he took with him, Cassius is portrayed as having also enlisted a small detachment of Moors (Lucian Hist. conscr. 31). A further anecdote that could perhaps be adduced is found in a passage where Lucian (Hist. conscr. 28) criticizes a historian who, instead of describing the salient aspects of the battle of Dura-Europos, dwelt at length on the futile endeavours of the Moorish cavalryman Mausacas – if, that is, the passage in question does indeed refer to L. Verus’ campaign (cf. e.g. Stein 1930; Canfora 1989, 260–262). 50 Zecchini 1998, 3011.

174

Federico Frasson

and partly in the second, possibly between the end of the one and the beginning of the other⁵¹. Additionally, for fr. 8a, it calls for the assumption that some event occurred which affected the north-eastern sector of the kingdom of Armenia, although the available sources concerning that particular period⁵² do not seem to offer support for such a view. Rather, they focus only on the site of the battle between Osroes and M. Sedatius Severianus, namely the city of Elegeia (cf. supra n. 38), which, however, was situated in the western part of Armenia, between the present-day Turkish cities of Erzincan and Erzurum⁵³. But since the surviving literary sources provide only brief flashes of information, often not interpretable as an organic whole, and as there is a total lack, in the ancient texts, of any systematic and in-depth treatment of the way the circumstances unfolded⁵⁴, it cannot be excluded that the regions of the north-eastern tip of Armenia were to some extent involved, given that they were relatively close to the “capital” Artaxata. This view may perhaps be further substantiated by the fact that king Pacorus, who was supported by the Parthians, would periodically spend time in Otene (fr. 9; cf. supra). Accordingly, Quadratus’ words may genuinely be describing events that occurred in 161–162. Such an interpretation has the benefit of placing frr. 5–8a in a recognizable descriptive context, but an alternative solution can also be put forward, albeit purely as a working hypothesis. Thus it could be suggested that the initial part of Quadratus’ work consisted of an introduction having the function of a linking element, containing an overview of the main events that had affected the eastern lands of the Roman empire during the period extending from the conclusion of Trajan’s Parthian campaign (117) – the point reached by the description given in Arrian – up to the invasion of Armenia and Syria by the armies of Vologaeses (161), which provoked the reaction of Rome and resulted in the campaign coordinated by L. Verus. On the basis of this hypothesis, and assuming that the description of the facts continued at least in part into book 2, an attempt can be made to reconsider the contents of frr. 5–8a, which could thus be seen as referring to events that took place during the time period in question. For instance, fr. 5 could have contained a description of the recently founded province of Armenia (administratively combined, it would seem, with Lesser Armenia and Cappadocia) which had

51 Zecchini 1998, 3010; cf. also FRH, 3, 651, fr. 12. 52 Dio Cass. 71.2.1 Boiss.; Fronto Princ. hist. 16, 19; Lucian Hist. conscr. 21, 25–26; Alex. 27; Eutr. 8.10.2; SHA Marc. 8.6; Verus 6.9; Oros. 7.15.2; Zonar. 12.2. 53 See e.g. Baumgartner 1905; Manandian 1965, 100; Hewsen 1982, 125; Bosworth 1983, 271; Barrington Atlas 89 C1; Talbert 2000, 1273; cf. Mutafian/Van Lauwe 2001, 33. 54 Cassius Dio’s account of the events survives only in the version epitomized by Xiphilinus.


175

been set up by Trajan and was then immediately abandoned by Hadrian⁵⁵; in this perspective, the Gelae could have been mentioned together with other populations who lived beyond the south-eastern border of Armenia, i.e. in the northern part of Media Atropatene. As far as fr. 6 is concerned, it may well be that the mention of the Moors genuinely referred to the auxiliary cavalry contingents, although it may not have been merely an offhand comment, thrown in simply as part of a list of army units; rather, it may instead have been developed in the context of a specific episode. For instance, what might come to mind is Hadrian’s dramatic decision, in 117, to strip Lusius Quietus – who at that time held a post in Judea – of his command over the Moorish contingents and to dismiss him, on suspicion that he nurtured an ambition to seize power and take over the reins of government⁵⁶. Not long afterwards (late summer 117 – early summer of 118) Quietus, who had been accused of hatching a plot against the emperor together with three accomplices, was murdered while he was on a journey⁵⁷. An event of this kind would certainly have fitted into Quadratus’ story, given that Quietus, originally from independent Mauritania (where he held a prominent role in his tribe⁵⁸) and an official at the head of the Moorish cavalrymen in the Dacian wars, had become a member of Trajan’s general staff and followed the latter to the Orient, where he acquired fame and glory through his deeds and exploits in the victorious campaign against the Parthians. The fall into disgrace of such an important figure, who had even been raised to the status of consul in 117⁵⁹ and was very closely linked to the question of the Parthian wars⁶⁰ and of Near Eastern affairs (where he had also

55 Dio Cass. 68.20.3 Boiss.; Eutr. 8.3.2; 8.6.2; Ruf. Fest. 14, 20; cf. also SHA Hadr. 21.11; Hieron. Chron. a. 113, 117 p. Chr., pp. 196–197 Helm; see e.g. Debevoise 1938, 223, 225, 229, 237, 240; Lepper 1948, 136–141; Chaumont 1969, 10–12, 14; Angeli Bertinelli 1976, 13, 17–18; Chaumont 1976, 136, 138–139, 143–144; Angeli Bertinelli 1979, 74–76; Mitford 1980, 1199–1200; Bivar 1983, 88; Grousset 1984, 110–111; Pasdermadjian 1986, 81–82; Chahin 1987, 250; Lightfoot 1990, 121–123; Wolski 1993, 179, 183; Traina 1999–2000, 74–75; Chaumont 2002, 97–98; Patterson 2013, 173, 187, 189–190, 193. 56 SHA Hadr. 5.8; it is probable that the procedure involved disarming and then discharging the Moorish units which were under the command of Quietus in Palestine (cf. Bussi 2006, 727; Roberto 2008, 83). 57 SHA Hadr. 7.1–2; on the dating, cf. Roberto 2008, 84. 58 Cf. Dio Cass. 68.32.4 Boiss.: τῶν Μαύρων ἄρχων. 59 On Lusius Quietus and the sources pertaining to this figure, cf. e.g. Groag 1927; Leschi 1945, 148–149; Smallwood 1962; Gabba 1966, 72–73 = Gabba 1974, 40–41; PIR2 5.1, 113–114, L 439; Devijver 1977, 542–543 L 41 (cf. Devijver 1987, 1632 L 41; Devijver 1993, 2162 L 41); Southern 1989, 92–93; Bussi 2006; Roberto 2008; the reader is referred to these contributions also for additional bibliographical references. 60 It is very likely that the figure of Lusius Quietus was one of the topics extensively addressed by in Arrian’s Παρθικά: see Arr. FGrHist 156 fr. 140 = fr. 53 (Parth.) Roos (cf. Suda ε 147, s.v.

176

Federico Frasson

violently put down a series of insurrections of the Jews) could, in the eyes of the historiographer, have represented a significant example underlining the contrast between the policy of Trajan and that of Hadrian. In particular, the latter is thought, among other things, to have been wary of men like Quietus who had achieved enormous power under Trajan⁶¹. The unseating of Lusius Quietus is sometimes thought to have been the factor underlying the revolt that broke out in Mauritania in 117, which was put down by Q. Marcius Turbo. According to some scholars, in fact, the Moorish units, shorn of the command of Quietus and discharged from service, could have fomented an insurrection once they returned home⁶². Moreover, while it is more likely that fr. 6 concerns the unseating of Quietus, the possibility cannot be excluded that it may have referred precisely to the Moorish rebellion of 117, because although the unrest did not affect the eastern part of the empire, such a serious event might have led to its being mentioned by Quadratus. Turning now to fr. 7, it is not easy to find a context suited to the citation of a population such as that of the Zobidae⁶³, who lived in Parthia⁶⁴, near the present-day region of Kermān, the ancient region of Carmania⁶⁵. Among the many possible solutions, one might suggest that Quadratus’ reference to the Zobidae was made in connection with the struggles which, during the era of Hadrian, saw the various candidates to the throne of the Parthian empire pitted against one another. The name of the Zobidae may have been brought up in connection with the military contingents, or perhaps in relation to the territories that were to be kept under the control of a specific contender⁶⁶.

ἐγνωσμένον); Arr. fr. *50 (Parth.) Roos (cf. Suda ε 2773, s.v. ἐπιχειρήσειν; FGrHist, IID, 156, 576– 577); Arr. fr. *52 (Parth.) Roos (cf. Suda μ 298, s.v. Μαυρούσιοι; FGrHist, IID, 156, 577). See Groag 1927, 1878; Roberto 2008, 86. As mentioned earlier, it seems highly plausible that Quadratus saw himself as a continuator of Arrian’s work. 61 On the reasons underlying the elimination of Quietus, see recently Roberto 2008, 84. 62 Cf. e.g. Southern 1989, 93; Bussi 2006, 727; Roberto 2008, 83–84. 63 As well as the mention of Stephanus of Byzantium (vd. supra n. 45), cf. Ptol. Geog. 6.5.1 Nobbe = Stückelberger/Graßhoff (τοὺς Σωβίδας). 64 Cf. e.g. Kretschmer 1927 and most recently FRH, 3, 652, fr. 14. 65 See e.g. Schmitt 1990. 66 We know very little, and above all mainly only on the basis of numismatics, about the turbulent situation within the Parthian empire. This empire, laid waste by Trajan’s expeditions, was the scene of the fierce conflict between Osroes I and Vologaeses III (II). The latter, also known as Balāš III, gained the upper hand in clashes against Osroes I that took place roughly during the second decade of the second century, but he had to expend considerable effort in order to hold his position against at least two pretenders: Mithridates IV, who apparently controlled the regions of the Iranian plateau, and another prince who did not indicate his name on the coins


177

Fr. 8a (= 8b) may likewise have been referring to events that took place in the era of Hadrian, and in this case to something that directly involved the Caucasian area. Of interest in this context is a significant comment by Cassius Dio⁶⁷, who reports that after the end of the revolt of the Jews, a new war broke out, due to the fact that the Alans, at the instigation of Pharasmanes, swept into Albania and Media (Atropatene), wreaking enormous devastation. The incursions, which partly also affected Armenia and the Roman province of Cappadocia, ceased only when Vologaeses offered gifts to the raiders, and Arrian, the governor of Cappadocia, succeeded in instilling fear in them. Through comparison with other ancient texts, modern scholars have determined that the Alans surged through the Caucasus around 135, and have identified Pharasmanes II, the king of Iberia, as the character who backed their invasion and who probably allowed the horde to transit through his lands in order to cause damage to the Albanians and the Parthians, more than to the Romans⁶⁸. There is considerable uncertainty, on the other hand, concerning the identity of the character referred to as Vologaeses, whose gifts are supposed to have played a role in persuading the Alans to put an end to their raids. Suggested identifications include the king of the Parthians Vologaeses III (II), or the king of Armenia⁶⁹. In the passage from Cassius Dio cited

(cf. e.g. Debevoise 1938, 242, 244; Sellwood 1980, 263–267; Schippmann 1989; Wolski 1993, 184; Verstandig 2001, 317–319). 67 Dio Cass. 69.15.1 Boiss. = Arr. FGrHist 156 T 5 = T 12 Roos; the text used here is the one reconstructed by Boissevain on the basis of the epitome of John Xiphilinus and on that of Zonaras (11.23–24), as well as the Excerpta de legationibus by Constantinus Porphyrogenitus, given that the original of Cassius Dio is lost. 68 Pharasmanes II’s intention was, it would seem, to damage the Albanians and Media Atropatene, a vassal of the Parthians. This is the opinion of Belfiore 2012, 16–17, whereas Bais 2001, 94–96 sees the scheme excogitated by the Iberian king as arising within the context of a crisis in Ibero-Roman relations (cf. earlier Carrata Thomes 1958, 21–22; Giardina 1996, 116–119, 134–135); on the invasion by the Alans during the era of Hadrian, cf. e.g. Von Gutschmid 1888, 146–147; Pelham 1896, 630, 635–639 = Pelham 1911, 218, 226–232; Mommsen 1921, 405; Debevoise 1938, 242–244; Magie 1950, 1, 621, 659; 2, 1483 n. 43; Carrata Thomes 1958, 21–22; Altheim 1959, 12, 249– 250; Chaumont 1969, 14–15; Bachrach 1973, 8–10, 126–132; Chaumont 1976, 145–146; Bosworth 1977; Mitford 1980, 1202; Grousset 1984, 111; Bosworth 1993, 228, 231; Wolski 1993, 184; Braund 1994, 233; Giardina 1996, 134–135; Schottky 1998, 452–453; Alemany 2000, 84–85; Bais 2001, 94–95; Chaumont 2002, 98; Belfiore 2012, 11–19. The Alans were by no means new to large-scale incursions of this kind, as shown by the great invasion of 72, which humiliated the sovereigns of Media and Armenia (Joseph BJ 7.7.4; cf. e.g. Täubler 1909, 18–21; Carrata Thomes 1958, 13; Altheim 1959, 11–12; Bosworth 1977, 223–224; Alemany 2000, 91–93; Bais 2001, 88–89; Belfiore 2012, 19–20). 69 The first solution is preferred by such authors as Pelham 1896, 635–636 = Pelham 1911, 227; Mommsen 1921, 405; Debevoise 1938, 242–244; Magie 1950, 2, 1528 n. 2; Carrata Thomes 1958, 22; Bosworth 1977, 219–220; Mitford 1980, 1202 n. 96; Grousset 1984, 111; Schippmann 1989;

178

Federico Frasson

above, it is particularly interesting to note the mention of Arrian, the well-known historian who, in his capacity as legate of Cappadocia, apparently found himself contending with the terrible threat of the Alans. But it is not clear whether he restricted himself to organising defensive operations and preparing the army, in accordance with the arrangements stated in his Ἔκταξις κατὰ Ἀλανῶν⁷⁰, planning a battle that never took place, or whether real fighting actually broke out, as is perhaps suggested by a later source⁷¹. Whatever the truth of the situation, it is likely that the Alans, during their outbound trip and/or on the return journey, passed through and very probably wrought destruction on the lands of the Obareni and the Oteni, which were situated in the broad valley of the river Kura that divided Iberia and Albania from Armenia (cf. supra n. 16)⁷².

Bosworth 1993, 231, 270 and recently Belfiore 2012, 13, whereas the second solution is favoured by, for instance, Von Gutschmid 1888, 147; Boissevain 1890, 336–337; PIR 3, 477, V 633; Marquart 1905, 221; Chaumont 1969, 14–15; Pflaum 1970, 196; Bachrach 1973, 10; Chaumont 1976, 145–146. 70 On this work by Arrian, see, recently, Bosworth 1993, 264–272; Alemany 2000, 79–82; Saxtorph 2002; Belfiore 2012. 71 According to Themistius (Them. Or. 34.8 = Arr. T 13 Roos; cf. Alemany 2000, 108–109), who is worthy of credibility in the view of Bosworth 1977, 229–232, 254–255; Braund 1994, 233, Arrian chased the Alans out of Armenia and even crossed the Caspian Gates (see infra n. 72), additionally turning his attention to the question of drawing up the boundaries between the Iberians and the Albanians (cf. e.g. Bais 2001, 95–96). In this regard, note also the words of John the Lydian (Lydus Mag. 3.53.4 = Arr. FGrHist 156 fr. 37 = T 14 Roos; cf. Alemany 2000, 81, 90), who argued that Arrian gave an accurate description of the Caspian Gates, both in his Ἀλανική ἱστορία and also in the Παρθικά. Arrian’s knowledge, John the Lydian maintained, was due to the fact that Arrian was stationed near that area when he was in command of the region in question, in the era of Trajan. However, even if one hypothesizes, as has already been suggested, that the command in question was that of Cappadocia and that Trajan was mistakenly written instead of Hadrian (cf. e.g. Bosworth 1983, 265–270), there is insufficient evidence to support the view that the passage in question represents a confirmation of the statement made by Themistius. Therefore the suggestion of a significant military intervention by Arrian, which would have included a veritable chase and pursuit of the Alans, must remain no more than an intriguing hypothesis. On the possibility that there may have been a minor skirmish between the Alans and the Romans, cf. Carrata Thomes 1958, 22; the question of a battle between Arrian and the Alans is also addressed by Bachrach 1973, 8. On this whole question and on Arrian’s Ἀλανική ἱστορία, of which the Ἔκταξις κατὰ Ἀλανῶν could be an abridgment or a complement, cf. most recently Belfiore 2012, 13–19, 37–40. 72 Had Pharasmanes allowed the Alans to pass through the so-called Caucasian Gates or Caspian Gates, i.e. the Dar’jal Pass, near the present-day boundary between Georgia and Russia (cf. Bosworth 1977, 219, 228–229; Bosworth 1983, 266–267; Giardina 1996, 100, 134; Gnoli 1996, 836–837, 849; Belfiore 2012, 14, 16; see Ognibene 2009, 262–263 for the close relation between the pass and the Alans), the population from the North would have had free access to the valley of the river Kura and, advancing in the direction of the mouth of the Kura, would have reached Albania. Situated just beyond the Kura, the lands of the Obareni and the Oteni (cf. supra) would


179

Given the extent of the Alan invasion, which, indirectly at least, involved even the Romans themselves, and which may have been alluded to by other sources as well, among which the History of Armenia by Movsēs Xorenacʽi⁷³, it seems quite probable that Quadratus himself also addressed the subject, since he may have been interested in the role played in this circumstance by Arrian, his literary model. Thus if Quadratus effectively devoted the first part of his work to events that took place in the East during the reigns of Hadrian and Antoninus Pius, then he may well have dealt with the Alan invasion. Naturally, even if these suppositions were to be awarded some degree of credibility, the hypothesis that

have been an attractive target and, in this phase, marauding Alans might have set their sights on extending their raids into this area (on the rich resources of Gogarene, which can perhaps, as mentioned earlier, be identified with the region inhabited by the Obareni, cf. Strab. 11.14.4). In effect, if at least a part of the invaders who were aiming to make their way into Media had decided to cross the Kura upstream of the point where the Aras flows into the Kura, then the region they would have crossed through would have been none other than Otene. But the latter could in any case have been ravaged by the plundering and pillaging wreaked by the Alans even if the latter were to have taken a different route, travelling not over the Dar’jal pass but rather, as has been put forward by some authors as an alternative hypothesis (see e.g. Belfiore 2012, 18), through the narrow Derbent passage (sometimes likewise called the Caspian Gates; cf. also infra, Appendix) between the Caucasus and the Caspian Sea, in present-day Dagestan. However, the horde of Alans could also have crossed through the lands of Armenia as they marched back towards the North: in such a case, they might have seized the opportunity to attempt an attack on the province of Cappadocia before resuming their march towards the mountains, laden with booty; thus they may have crossed the Caucasian Gates in the opposite direction (on the possibility that there may have been a skirmish between the army commanded by Arrian and the Alans in the vicinity of a mountain pass situated to the east of present-day Erzurum, cf. Bosworth 1977, 234, 246–247; see, however, Bosworth 1993, 268, 270). 73 On this hypothesis, cf. Chaumont 1976, 146; see earlier Chaumont 1969, 15; cf. also infra, Appendix. Debevoise (1938, 243; see also Altheim 1959, 12 and recently Alemany 2000, 389–390), on the other hand, hypothesized that a reference to the surge of the invading Alans in 135 can be found in a passage of the so-called Chronicle of Arbela (chap. 3, 8–10 Mingana = 7–10 Kawerau = 30–31 Ital. transl. Ramelli; on the authenticity of this work, long considered to be a modern forgery, cf. recently Ramelli 2002, 7–8; Ramelli 2006), which contains a description of a war fought by the Parthians against “popoli ribelli che erano irrotti nelle terre dei monti di Qardū e avevano depredato e devastato molte città” (30 Ramelli); more recently, however, the hypothesis has been put forward that the passage in question may allude to the rebellion by the head of the Gordyeni at the time of Trajan’s invasion (cf. Ramelli 2006, 153–154). Finally, there is a literary mention, in the Babylonian Story, a novel written by Iamblichus in the second century AD (Iambl. Bab. 66 Habrich; cf. Phot. Bibl. 94, 78a Bekker; see Altheim 1959, 15; Alemany 2000, 89), of a certain Ἀλανῶν στρατός on the payroll of the Babylonian king Garmus; this episode has sometimes been dubitatively linked to the invasion that took place in 135 (cf. Belfiore 2012, 38 n. 93), though it does not appear that any particular significance can be attributed to the literary mention in itself.

180

Federico Frasson

the Obareni and Oteni (fr. 8a = 8b) were effectively mentioned by name in connection with the events of 135 remains merely one of the many possibilities. Overall, however, while re-examination of the fragments of Asinius Quadratus taken from books 1 and 2 of the Παρθικά provides no definite confirmation of the various hypotheses examined here, it does seem to give a certain plausibility to the suggestion that the opening part of his work was devoted to an overview of the events that occurred between 117 and 161 and which fairly directly involved the eastern border of the Roman empire. But even if this interpretation is accepted, it remains unclear whether the description of the war that broke out in 161 with the Parthian invasion of Armenia began in book 3 or whether – although there is no clear evidence of this further suggestion – it already began in book 2. What remains to be identified, at this point, is a possible background setting for fr. 18, which is taken from book 9⁷⁴ and concerns Tigranocerta, very likely in the context of a flashback⁷⁵. The few words that have come down to us, which refer to the king of Armenia Tigran II and, in particular, to the fact that he founded the new “capital” Tigranocerta (cf. supra)⁷⁶, suffice to highlight two themes that were of special interest to the historiographer, namely κτίσεις and etymology⁷⁷.

74 Although the work from which the fragment is taken is not explicitly indicated by Stephanus of Byzantium, there appears to be no doubt at all that it was the Παρθικά, since the passage deals with Armenia, which on a number of occasions was involved in the wars between the Romans and the Parthians; moreover, in the fragment Quadratus refers directly to the Parthians, providing the etymology of the poleonym Tigranocerta. Cf. FRH, 3, 654, fr. 26. 75 A digression is the simplest explanation that would justify the mention of a rather ancient event such as the founding of Tigranocerta (first half of the first century BC) in book 9, which was presumably one of the last books of the Παρθικά, since the hypothesis that book 9 dealt with the events of the reign of Tigran II can definitely be rejected. By the same token, it seems realistic to rule out the suggestion that the subject-matter, in the Παρθικά, was set out according to a geographic order or some other criterion. Finally, one can likewise dismiss the possibility that the fragment might have been taken from a work other than the Παρθικά (see supra n. 74). On the entire question, cf. FRH, 1, 616; 3, 654, fr. 26. 76 Here we accept the reading ᾤκισε, which was already adopted in Meineke’s edition of Stephanus’ Ἐθνικά (623) and in FGrHist 97 fr. 18 and FRH, 2, 1140, fr. 26 (see also FRH, 3, 654, fr. 26), whereas FHG, 660, fr. 14 and HRR, 145, fr. 24 have the form ᾤκησε, which considerably changes the meaning of the sentence. 77 Quadratus’ interest in the founding of the city (cf. e.g. FGrHist, IIC, 97, 302; FRH, 1, 616) is also well documented by another fragment of book 9 of the Παρθικά (fr. 16 = FRH, 2, 1138, fr. 23) and by a fragment concerning Ravenna, very likely taken from the Χιλιετηρίς (fr. 26 = FRH, 2, 1132, fr. 7). Taken together with a fragment pertaining to the Alamanni that belongs either to the Χιλιετηρίς or to a monograph on Germany (fr. 21 = FRH, 2, 1132, fr. 8), the latter piece of evidence also highlights the historiographer’s curiosity in questions of an etymological nature. On the cited fragments taken from the Χιλιετηρίς, cf. most recently Frasson 2013, 319–320, 324–325, with


181

In an etymological perspective, Quadratus strongly maintained that the poleonym Tigranocerta was the term corresponding to (τῇ Παρθυαίων φωνῇ) the Greek Τιγρανούπολις. On the basis of this statement, many modern scholars have set up the equivalence κέρτα = πόλις⁷⁸, which would seem to be supported by Appian and Hesychius as well⁷⁹. In actual fact, the true meaning of -kerta/-karta in place names is the subject of intense debate, and numerous different solutions have been put forward in addition to the acceptation of “city”, including a putative loan from Aramaic. Other perhaps more plausible proposals include “made, built” or “residence, house”, which derive from ancient Iranian, a language from which Armenian did indeed draw a large number of loanwords⁸⁰. Moreover, precisely for the latter reason, the allusions by Quadratus and Hesychius to the Parthian and Armenian language respectively (see n. 79), should not be considered as in contradiction with one another. In any case, whatever the correct etymology of the poleonym, both Asinius Quadratus and, earlier, Appian, can be considered sufficiently precise, because their words are not so much a literal translation as, rather, an apparent suggestion of the name the city would have been given had it been a city originally founded in a Greek context. In effect, the eponymous cities

the bibliography indicated therein; on Asinius Quadratus’ etymological interests, see e.g. Norden 1920, 495–496; Zecchini 1998, 3013, 3017–3018. 78 This view, which was already prevalent in erudite contexts, as shown for instance by the Genevan edition, published in 1553, of the historic Calepino dictionary and by Gesner’s Novus linguae et eruditionis Romanae thesaurus (cf. respectively Calepinus 1553, 261r, s.v. Tigranocerta; Gesnerus 1749, 798), received support even in later centuries (cf. e.g. Dyer 1873). 79 Cf. respectively App. Mith. 67.285 (καὶ Τιγρανόκερτα ἀφ’ ἑαυτοῦ προσεῖπε· δύναται δ’ εἶναι Τιγρανόπολις) and Hsch. κ 2361, s.v. Κέρτα (Κέρτα· πόλις, ὑπὸ Ἀρμενίων). 80 On the etymological debate, and for a long list of Armenian (and other) poleonyms ending in -kerta/-karta or in forms that can be traced back to similar endings, cf. earlier Blau 1877 (see more recently e.g. Chaumont 1974, passim); on the equivalence -kerta = city, with specific reference to Tigranocerta, cf. supra n. 78. On the acceptation “made/built by Tigran” or “foundation by Tigran”, presupposing a derivation of the characteristic ending from Old Iranian karta-, the perfect passive participle of the verb *kar = “to do, make” (Bartholomae 1904, 444; Cheung 2007, 236), cf. e.g. Hübschmann 1904, 474 and more recently Chaumont 1982, 89; Schmitt 1996, 701– 702; Garsoïan 2005; Avdoyan 2006, 83; Hakobyan 2010, 98. However, it has been suggested that a different interpretation should be given to the ending -kerta/-karta in the case of place names, for which the meaning of “residence, house” is proposed by Blau 1877, 504–505 (cf., however, Nöldeke 1879); Bartholomae 1904, 448 n. 22 (the latter puts forward a connection with another Iranian verb form: kar, no. 5, 449–451).

182

Federico Frasson

founded in Armenia sometimes did have the ending -kert⁸¹, just as those in the Greek-speaking part of the Roman empire often ended in -πολις⁸². Let us now return to the narrative setting of fr. 18. If one bears in mind that the Παρθικά, as already mentioned, were probably composed of a total of 10–12 books, it is likely that the name of Tigranocerta mentioned in book 9 was brought up in relation to the expeditions of Septimius Severus, Caracalla or Alexander Severus⁸³. In the sources pertaining to the expeditions of Septimius Severus the only reference to Armenia is contained in a passage from Herodian (on the second Parthian campaign of 197–199), in which it is said that the emperor, determined to invade the Armenian kingdom, was dissuaded from this undertaking precisely by the Armenian sovereign, who, the passage claims, gave him money, gifts and hostages⁸⁴. Although Herodian’s words have at times been regarded as somewhat

81 The ending -kert characterizes many Armenian cities over the centuries: in addition to the numerous cases of Tigranakert, mention should be made, among others, of Tʽsolakert, Bakourakert, Hadamakert (cf. the map in Mahé/Mahé 1993, 449), Arcatiokert (cf. e.g. Mutafian/Van Lauwe 2001, 29, 31), Vardanakert (cf. e.g. Mutafian/Van Lauwe 2001, 31, 33, 35, 37), Manazkert (cf. e.g. Mutafian/Van Lauwe 2001, 39, 41, 43); see also supra n. 80. Among the many modern centres, and restricting attention to the current territory of Armenia, one finds cases such as the city (now extensively depopulated) of Dastakert and numerous villages, among which Darakert, Geghakert, Karakert, Norakert and Noyakert. 82 The names of some cities founded in the first half of the second century come to mind, such as Marcianopolis, Traianopolis (various instances), Antinoopolis and Hadrianopolis or, with reference to a period closer to Asinius Quadratus’ lifetime, Philippopolis of Arabia, not to mention an obvious fourth century case, the capital Constantinopolis. 83 Lehmann-Haupt 1936, 1000–1002 (see earlier Lehmann-Haupt 1910, 506) sought a reason that would explain why, in fr. 18, Tigranocerta was not located in Armenia, but rather, πρὸς Ἀρμενίαν; he thus speculated, following in the footsteps of Mommsen, that these words were already present in Quadratus’ original text, and that Quadratus was writing at a time when, from the administrative point of view, the city could well have come to be situated within the borders of the province of Mesopotamia set up by Septimius Severus, perhaps in 198–199. However, this argument rests on the assumption – itself based on the authority of Mommsen’s account – that the border between Mesopotamia and Armenia ran along the natural barrier of the Eastern Taurus Mountains and in particular along the Hazro dağları. Should the Lehmann-Haupt hypothesis prove to be well-founded, one might also conjecture that Tigranocerta was within the boundaries of the province of Mesopotamia not only at the time when Quadratus was writing, but even as early as the actual time of the episode the historiographer was relating in his mention of the city. If so, this would imply that fr. 18 referred to an event that was subsequent to the time when Mesopotamia was reduced to the status of a Roman province. On the transformation of Mesopotamia into a province of the Empire, cf. e.g. Platnauer 1918, 194–195; Magie 1950, 1, 675; 2, 1544–1545 n. 26; Pareti 1960, 400; Angeli Bertinelli 1976, 40–41; Angeli Bertinelli 1979, 79; Mitford 1980, 1206; Birley 1988, 132; Patterson 2013, 190–191. 84 Hdn. 3.9.2, in which the event is described as occurring earlier than Septimius Severus’ attempted attack on the stronghold of Hatra (and not subsequent to this attempt, as can be read in


183

unreliable⁸⁵, the fragment of Quadratus concerning Tigranocerta has recently been seen as a confirmation of the reliability of Herodian’s statement on Septimius Severus’ projects with regard to Armenia⁸⁶. Armenia played a more substantial role in the eastern policy of Caracalla, who, after capturing the king of Osroene by trickery, did the same with the Armenian sovereign, thereby triggering an anti-Roman revolt. The rebellion is thought to have been headed by the future Tiridates II, who was perhaps a son of the imprisoned king. According to Cassius Dio (78.21.1 Boiss.), Caracalla aimed to put down the revolt by sending an army to Armenia, under the command of Theocritus, a freedman who in the past had been a dancer, but the Roman army suffered a heavy defeat⁸⁷. Although we have no knowledge of where the battle took place, Tigranocerta may have been mentioned in connection with this wartime episode or in relation to the expedition conducted against Adiabene in 216 by the emperor in person⁸⁸, who could have reached his destination by travelling through Gordiene and Calachene⁸⁹, after crossing and laying sack to part of the territory of Armenia⁹⁰. In this case, the affected area would presumably have been none

Chaumont 1976, 153; Zecchini 1998, 3014). Furthermore, in contrast to the versions handed down by the other ancient sources, the historian places Septimius Severus’ attack prior to the victorious expedition that culminated in the capture of Ctesiphon. On Septimius Severus’ second campaign in the East, cf. e.g. Platnauer 1918, 114–121; Debevoise 1938, 259–262; Magie 1950, 1, 674–675; 2, 1542–1545 nn. 25–26; Pareti 1960, 398–400; Angeli Bertinelli 1976, 37–39; Angeli Bertinelli 1979, 79; Bivar 1983, 94; Birley 1988, 129–133; Wolski 1993, 190; Verstandig 2001, 334–336; on the passage from Herodian that relates to Armenia, cf. most recently Patterson 2013, 176. 85 Cf. e.g. Platnauer 1918, 118 n. 1; Chaumont 1976, 153. 86 Zecchini 1998, 3014. 87 On the policies adopted by Caracalla, with reference to Armenia, cf. e.g. Asdourian 1911, 118; Debevoise 1938, 263–264; Magie 1950, 1, 685; 2, 1553–1554 n. 43; Chaumont 1969, 20–21; Chaumont 1976, 154–155, 157; Grousset 1984, 112; Pasdermadjian 1986, 83; Wolski 1993, 192; Verstandig 2001, 339–340; Chaumont 2002, 99; Patterson 2013; on the figure of Theocritus, cf. Dio Cass. 78.21.2–4 Boiss.; see e.g. Von Domaszewski 1903, 224–225; Stein 1934. 88 On this act of warfare, cf. e.g. Debevoise 1938, 265; Magie 1950, 1, 686; 2, 1554 n. 45; Pareti 1960, 430–431; Chaumont 1976, 155–156; Bivar 1983, 95; Wolski 1993, 192; Verstandig 2001, 342; Patterson 2013, 178. Although Zecchini (1998, 3014) does not rule out the eventuality that fr. 18 of Quadratus may be alluding to events pertaining to the era of Caracalla, he believes it is more likely that in book 9 the historiographer dealt only with the wars of Septimius Severus, in which case the wars waged by Caracalla would have begun in book 10. 89 On this hypothesis, see Von Gutschmid 1888, 155; cf. also Chaumont 1976, 156 n. 474. 90 The possibility that on this occasion Caracalla’s troops may have marched through part of the Armenian territory, ravaging and plundering the land as they went, is not excluded by Asdourian 1911, 118, and is also endorsed by Chaumont 1976, 156, given that when the hostilities came to an end, in 217, the new king of Armenia, Tiridates II, demanded that Macrinus return the booty the Romans had looted from Armenian land (cf. Dio Cass. 79.27.4 Boiss.).

184

Federico Frasson

other than Arzanene, where Tigranocerta probably stood, given that in the south this region was separated from Gordiene only by the course of the river Tigris⁹¹. Finally, it appears less probable that book 9 of the Παρθικά already recounted the expedition of Alexander Severus⁹², with an account, in particular, of troops acting under the orders of Iunius Palmatus and crossing through Armenia with instructions to attack Media Atropatene. By the same token, it appears unlikely that there was a mention of Tigranocerta precisely in this connection⁹³. Irrespective of the reliability or otherwise of the hypotheses put forward here concerning the specific episodes to which the various fragments refer, the analysis of the individual fragments of the Παρθικά concerning Armenia highlights some of the main characteristics of Asinius Quadratus’ manner of writing history. His manner of presentation is comparable to the approach emerging in his grandiose work celebrating the history of Rome, the Χιλιετηρίς which, like the Παρθικά, survived until the Byzantine era⁹⁴. The fragments of both these works delineate the portrait of a highly educated historiographer with an extensive cultural background, as testified by his attention to the κτίσεις and etymological questions, and also having considerable geographic knowledge on the regions that were the scene of the events he wished to narrate. His writings, certainly dotted with a great number of names of places and people, did not escape the attention of a lexicographer such as Stephanus of Byzantium⁹⁵, whose Ἐθνικά, despite having come down to us in an epitomated form, represents the main source of knowledge on the literary production of Asinius Quadratus. Indeed, it is precisely through our fractional glimpse of the work of Stephanus of Byzantium that it becomes possible, when examining the fragments described here, to perceive the central role Armenia and the Armenians must have played in the pages of the Παρθικά, a role that clearly reflected the importance – confirmed by many other sources – of Armenia in the context of relations between two among the greatest empires of antiquity.

91 Vd. Barrington Atlas 89; Mutafian/Van Lauwe 2001, 29, 33, 35, 37. 92 Some fragments of the Παρθικά seem to warrant the assumption that books 7 and 8 (and conceivably book 9 itself as well) still dealt with Septimius Severus’ campaigns (cf. Zecchini 1998, 3013–3014). 93 On the expedition, cf. esp. Chaumont 1969, 36–37; Chaumont 1976, 163–165; see also, among others, Magie 1950, 1, 695–696; Pareti 1960, 461–463; Angeli Bertinelli 1979, 109; Frye 1983, 124. 94 On the survival of the works of Asinius Quadratus in the Byzantine era, cf. recently Zecchini 1998, 3020–3021; Frasson 2013, esp. 303–308, 326. 95 Cf. most recently Frasson 2013, 325–326; the topic has also been discussed by P. De Cicco (“How to reduce a historical production into a compendium of geographical knowledge: Asinius Quadratus and his Late-antique readers”) at the workshop entitled Historiography and Space in Late Antiquity (Ghent University, 15–17 January 2015).


185

Appendix. Movsēs Xorenacʽi: a possible source for the Alan invasion of 135? According to Chaumont⁹⁶, a passage from the History of Armenia by Movsēs Xorenacʽi (2.65)⁹⁷ could, despite its anachronisms and exaggerations, allude to the Alan invasion in the era of Hadrian. According to the story related by the Armenian historian, a coalition of peoples from the North, namely the Khazarkʽ and the Basilkʽ, led by their king, Vnasēp Surhap, crossed the Čor Gate (i.e. the Derbent pass)⁹⁸ and the river Kura, at which point it was challenged and defeated by the King of Armenia, Vałarš, who filled the plain with corpses. And then, not content with victory, Vałarš chased the retreating enemy back beyond the Čor pass and succeeded in inflicting a further defeat on the foe, but he was killed in the battle. The valiant Armenian sovereign would later be avenged by his son Khosrov who, after crossing over the mountains, would put his enemies to flight and order that one man out of every hundred be taken hostage, and would subsequently erect a stele with an inscription in Greek as a symbol of his authority and his loyalty to Rome⁹⁹. It is evident that the mention of Khazarkʽ and Basilkʽ (Khazars and Barsils) in relation to events Movsēs sets in the second century AD should certainly be considered as an anachronism¹⁰⁰. However, this may not have been present origi-

96 Chaumont 1976, 146; see earlier Chaumont 1969, 15. 97 On this passage, cf. most recently Rapp 2014, 126–127. 98 On this pass, cf. e.g. Bosworth 1977, 224–225, 226–227; Thomson 1978, 211 n. 13; Mahé/Mahé 1993, 371, II, 65, n. 12; Kettenhofen 1994; Alemany 2000, 279. 99 The source for these episodes, according to MX 2.66, is Bardaisan of Edessa, who, having travelled to the fortress of Ani, supposedly read and translated into Syriac the Olympius’ “temple history” (MX 2.48), updating it with the events of his own era; however, this piece of information is generally held to be a figment of Movsēs’ imagination (cf. Thomson 1978, 16, 35). 100 The most ancient episode with regard to which the Khazars are expressly mentioned in a source written in Greek (Thphn. Chron. 315 De Boor = 446 Engl. transl. Mango/Scott) dates back to the first half of the seventh century AD (cf. e.g. Golden 1980, 51; Alemany 2000, 330; however, on the possibility that the Khazars could correspond to the Ἀκάτιροι or Ἀκάτζιροι of Priscus Panites, i.e. the Acatziri of Jordanes, cf. Henning 1952, 502–507). On the close link between the Khazars and Barsil/Berzilia, a toponym and tribal name associated with the northern part of the Caucasus, but also with the Volga valley, cf. Golden 1980, 143–147. On the anachronism in MX 2.65, cf. e.g. Dunlop 1954, 8–10; Toumanoff 1963, 331; see recently Shapira 2007, 311–316. The mention is also considered anachronistic by Zuckerman 2007, 401, 430, who believes that the passage may have been inspired by the Khazar invasion of Transcaucasia, which took place in 685. The story of an invasion of Transcaucasia, anachronistically attributed to the Khazars, can also be found in a passage from the Kʽartʽlis Cʽxovreba (a collection of Georgian chronicles drawn up in different time periods from the eighth century onwards), which shows some points

186

Federico Frasson

nally in the Armenian historian’s text: it could instead have been the result of a later interpolation – designed to explain what might, at the outset, have been a generic mention of “peoples from the North” –, which later became an integral part of the transmitted text¹⁰¹. Be that as it may, the passage could in any case conceal an indirect reference to the Alans¹⁰². If it was indeed present in the original, the peoples from the North can be assumed to have been cited with the name of peoples that occupied the lands lying north of the Caucasus at the time when the History of Armenia was written, while if it was an interpolation it may have been an expansion of what was simply a generic phrase (“the hosts of the northern peoples united”¹⁰³), to which an unknown interpolator presumably added the name of the peoples who lived beyond the above cited mountain range in the period when the interpolator himself was writing. There may be a more serious problem arising in connection with the dating of the reign of the Vałarš who is the protagonist of the episode. According to Movsēs, he came to power in the thirty-second year of the reign of the king of the Parthians who bore the same name, whom scholars generally identify as Vologaeses IV (III). If this identification is correct, then the reign of Vałarš of Armenia would have begun in 180¹⁰⁴, i.e. over forty years after the Alan invasion. This incongruity would be difficult to brush aside, and would severely weaken the proposal put forward by Chaumont, if the source utilized belonged to the range of texts forming the body of Graeco-Roman historiographic works, or at least of those among them generally considered most reliable. But the History of Armenia presents a number of unique features that make it a special case, and it must be assessed as such. Setting aside the delicate question of the period in which the author lived – which has long been the object of disagreement among scholars¹⁰⁵ – it is important to note that the History of Armenia is the singular outcome of data taken from classical sources, at times with direct or indirect use of works or passages which today are lost and are not known elsewhere¹⁰⁶, and of information derived from,

of contact with the passage from the History of Armenia (cf. Shapira 2007, 320–322; see also e.g. Alemany 2000, 311). 101 Thus Traina 1991, 31–32, who, arguing that the work of Movsēs Xorenacʽi should be seen as dating from the fifth century, believes that the citation of Khazarkʽ and Basilkʽ is the result of an interpolation. 102 On this possibility, as well as on the above cited works by Chaumont, cf. the observations in Alemany 2000, 285. 103 Cf. Thomson 1978, 211. 104 Cf. Thomson 1978, 210 n. 1; Mahé/Mahé 1993, 370, II, 65 n. 1. 105 Cf. e.g. Thomson 1978, 1–8, 58–61; Traina 1991, 19–42, 93–99; Mahé/Mahé 1993, 18–20; for an overview of the problem, see recently Patterson 2013, 179 n. 11. 106 Cf. the contributions by F. Gazzano and F. Mari, in this volume.


187

as it were, “local” sources, both written (including, perhaps, archival documents) and oral, such as songs, tales and fables of the Armenian tradition¹⁰⁷. It is the very fact of the heterogeneous nature of the sources used by Movsēs Xorenacʽi that underlies the difficulty of ascertaining the reliability of his description, the limitations of which are evident above all in the chronological data. Seeking to reconcile two different types of sources, namely the chronology derivable on the basis of the works of the classical world and the native traditions¹⁰⁸, the historian built up a work characterised by numerous inaccuracies, where at times links are forged between figures and events dating from different chronological contexts: for instance, a given sovereign was at times credited with deeds that should more properly have been attributed to a different figure. These incongruities, perhaps favoured by the frequent presence of sovereigns bearing the same name on the Armenian throne¹⁰⁹, should be understood in the framework of the more general “tendency to telescope events”¹¹⁰ that is a typical feature of the work of Movsēs Xorenacʽi, who reduced the number of those who did effectively reign, in order to simplify his narrative. In the light of these considerations, Chaumont’s hypothesis on the Alan invasion of 135 perhaps gains greater credence, because the Armenian historian could have wrongly attributed to a Vałarš – whom he or his sources placed in the closing decades of the second century AD – the deeds accomplished by a previous character likewise named Vałarš, who lived in the first half of the same century. The latter figure could have been precisely the Vologaeses mentioned by Cassius Dio on the subject of the Alan invasion¹¹¹. In sum, it cannot be fully ruled out that the version proposed by Movsēs Xorenacʽi, which was possibly the outcome of tales that had gradually been built up in the Armenian oral tradition, may reflect an Alan attack not attested elsewhere and subsequent to that of 135. However, it appears more likely that the reference concerns the events dating from the age of Hadrian, which are also known

107 On the sources used by Movsēs Xorenacʽi, cf. e.g. Thomson 1978, 8–56; Traina 1991, 41–68, 73–82; Mahé/Mahé 1993, 18–21; Traina 1995; Thomson 1996, 508–510; Traina 1998, 95–112; Topchyan 2006; see the recent mention in Juntunen 2013, 164. 108 Cf. e.g. Thomson 1978, 56–58; Mahé/Mahé 1993, 76; Thomson 1996, 508–509. 109 Cf. most recently Juntunen 2013, 161. 110 Patterson 2013, 179. 111 Dio Cass. 69.15.1 Boiss. = Arr. FGrHist 156 T 5 = T 12 Roos; on the much debated issue of the identity of this Vologaeses, cf. supra. Additionally, another piece of information contained in the same passage taken from Movsēs Xorenacʽi (2.65), the foundation of Vałaršapat, should perhaps be attributed to an earlier sovereign (cf. Mahé/Mahé 1993, 370, II, 65 n. 1, who mention a Vałarš I, said to have reigned between 117 and 140, whereas they name Vałarš II the figure mentioned by the Armenian historian; cf. also Rapp 2014, 127).

188

Federico Frasson

from the writings of Arrian and Cassius Dio. Thus in this as in other cases, the words of Movsēs Xorenacʽi are of great value, as they could be the fruit of an imprecise assemblage of data that are genuine at least in part (for example, the information on the pitched battles between the Armenians and the Alans) and are not available elsewhere because they draw on sources which are now completely inaccessible.

Bibliography Barrington Atlas = R. J. A. Talbert (ed.), Barrington Atlas of the Greek and Roman World, Princeton 2000. FHG = C. Müllerus (ed.), Fragmenta Historicorum Græcorum, 3, Parisiis 1849. FRH = T. J. Cornell (ed.), The Fragments of the Roman Historians, 3 vols., Oxford 2013. HRR = H. Peter (ed.), Historicorum Romanorum Reliquiae, 2, Lipsiae 1906. Alemany, A. (2000), Sources on the Alans. A Critical Compilation, Leiden-Boston-Köln. Altheim, F. (1959), Geschichte der Hunnen. 1, Von den Anfängen bis zum Einbruch in Europa, Berlin. Angeli Bertinelli, M. G. (1976), “I Romani oltre l’Eufrate nel II secolo d.C. (le province di Assiria, di Mesopotamia e di Osroene)”, in: ANRW, 2.9.1, Berlin-New York, 3–45. Angeli Bertinelli, M. G. (1979), Roma e l’oriente. Strategia, economia, società e cultura nelle relazioni politiche fra Roma, la Giudea e l’Iran, Roma. Aṙakʽelyan, B. N. (1984), “Les fouilles d’Artaxata: bilan provisoire”, in: Revue des Études Arméniennes n.s. 18, 367–395. Asdourian, P. P. (1911), Die politischen Beziehungen zwischen Armenien und Rom von 190 v. Chr. bis 428 n. Chr. Ein Abriss der armenischen Geschichte in dieser Periode, diss. Venedig. Avdoyan, L. (2006), “Tigranocerta: The City ‘Built by Tigranes’”, in: R. G. Hovannisian (ed.), Armenian Tigranakert/Diarbekir and Edessa/Urfa, Costa Mesa, 81–95. Bachrach, B. S. (1973), A History of the Alans in the West. From Their First Appearance in the Sources of Classical Antiquity through the Early Middle Ages, Minneapolis. Bais, M. (2001), Albania caucasica. Ethnos, storia, territorio attraverso le fonti greche, latine e armene, Milano. Bartholomae, C. (1904), Altiranisches Wörterbuch, Strassburg. Baumgartner, A. (1895), “Artaxata”, in: PWRE, 2.1, Stuttgart, 1311. Baumgartner, A. (1905), “Elegeia (no. 1)”, in: PWRE, 5.2, Stuttgart, 2258. Belfiore, S. (ed.) (2012), Lucio Flavio Arriano, Schieramento contro gli Alani, Roma. Birley, A. R. (1987), Marcus Aurelius. A Biography, London-New York, 2nd ed. Birley, A. R. (1988), Septimius Severus. The African Emperor, London-New York, 2nd ed. Bivar, A. D. H. (1983), “The Political History of Iran under the Arsacids”, in: E. Yarshater (ed.), The Cambridge History of Iran, 3.1, Cambridge, 21–99. Blau, O. (1877), “Ueber -karta, -kerta in Ortsnamen”, in: Zeitschrift der Deutschen Morgenländischen Gesellschaft 31, 495–505.


189

Bøe, S. (2001), Gog and Magog. Ezekiel 38–39 as pre-text for Revelation 19,17 and 20,7–10, Tübingen. Boissevain, U. P. (1890), “Ein verschobenes Fragment des Cassius Dio (75, 9, 6)”, in: Hermes 25, 329–339. Bosworth, A. B. (1977), “Arrian and the Alani”, in: Harvard Studies in Classical Philology 81, 217–255. Bosworth, A. B. (1983), “Arrian at the Caspian Gates: A Study in Methodology”, in: The Classical Quarterly n.s. 33.1, 265–276. Bosworth, A. B. (1993), “Arrian and Rome: the Minor Works”, in: ANRW, 2.34.1, Berlin-New York, 226–275. Braund, D. (1984), Rome and the Friendly King. The Character of the Client Kingship, London-Canberra-New York. Braund, D. (1994), Georgia in Antiquity. A History of Colchis and Transcaucasian Iberia. 550 BC-AD 562, Oxford. Bussi, S. (2006), “Lusio Quieto: un ‘maghrebino’ ai vertici dell’Impero”, in: A. Akerraz / P. Ruggeri / A. Siraj / C. Vismara (eds.), L’Africa Romana. Mobilità delle persone e dei popoli, dinamiche migratorie, emigrazioni ed immigrazioni nelle province occidentali dell’Impero romano. Atti del XVI convegno di studio, Rabat, 15–19 dicembre 2004, 2, Roma, 721–728. Calepinus, A. (1553), Dictionarium , quarto & postremo ex R. Stephani Latinæ linguæ Thesauro auctum, [Geneva]. Canfora, L. (1989), Una società premoderna. Lavoro, morale, scrittura in Grecia, Bari. Carrata Thomes, F. (1958), Gli Alani nella politica orientale di Antonino Pio, Torino. Chahin, M. (1987), The Kingdom of Armenia, London-New York-Sydney. Chaumont, M.-L. (1969), Recherches sur l’histoire d’Arménie de l’avènement des Sassanides à la conversion du royaume, Paris. Chaumont, M.-L. (1974), “Études d’histoire parthe. III, Les villes fondées par les Vologèse”, in: Syria 51.1–2, 75–89. Chaumont, M.-L. (1976), “L’Arménie entre Rome et l’Iran. I, De l’avènement d’Auguste à l’avènement de Dioclétien”, in: ANRW, 2.9.1, Berlin-New York, 71–194. Chaumont, M.-L. (1982), “Tigranocerte: données du problème et état des recherches”, in: Revue des Études Arméniennes n.s. 16, 89–110. Chaumont, M.-L. (2002), “Tentazioni dell’Iran e del mondo greco-romano (V secolo a.C.-300 d.C. circa)”, in: G. Dédéyan (ed.), Storia degli Armeni, ed. italiana a cura di A. Arslan e B. L. Zekiyan, Milano, 73–114. Cheung, J. (2007), Etymological Dictionary of the Iranian Verb, Leiden-Boston. Debevoise, N. C. (1938), A Political History of Parthia, Chicago. De Sainte-Croix, G.-E.-J. de Guilhem de Clermont-Lodève (1798), “Mémoire sur le cours de l’Araxe et du Cyrus”, in: Voyages historiques et géographiques dans les pays situés entre la Mer Noire et la Mer Caspienne. Seconde partie, Paris, 87–128. Devijver, H. (1977), Prosopographia militiarum equestrium quae fuerunt ab Augusto ad Gallienum, 2, Leuven. Devijver, H. (1987), Prosopographia militiarum equestrium quae fuerunt ab Augusto ad Gallienum, 4, Suppl. 1, Leuven. Devijver, H. (1993), Prosopographia militiarum equestrium quae fuerunt ab Augusto ad Gallienum, 5, Suppl. 2, Leuven. Dunlop, D. M. (1954), The History of the Jewish Khazars, Princeton.

190

Federico Frasson

Dyer, T. H. (1873), “Tigranocerta”, in: W. Smith (ed.), A Dictionary of Greek and Roman Geography, 2, London, 1208. Fraschetti, A. (2008), Marco Aurelio. La miseria della filosofia, Roma-Bari. Frasson, F. (2013), “Una storia di Roma letta a Bisanzio. La Χιλιετηρίς di Asinio Quadrato”, in: F. Gazzano / G. Ottone (eds.), Le età della trasmissione. Alessandria, Roma, Bisanzio. Atti delle Giornate di Studio sulla storiografia greca frammentaria, Genova, 29–30 maggio 2012, Tivoli, 303–326. Frye, R. N. (1983), “The Political History of Iran under the Sasanians”, in: E. Yarshater (ed.), The Cambridge History of Iran, 3.1, Cambridge, 116–180. Gabba, E. (1966), “Sulle influenze reciproche degli ordinamenti militari dei Parti e dei Romani”, in: Atti del Convegno sul tema: La Persia e il mondo greco-romano (Roma 11–14 aprile 1965), Roma, 51–73. Gabba, E. (1974), Per la storia dell’esercito romano in età imperiale, Bologna. Garsoïan N. (2005), “Tigran II”, in: Encyclopædia Iranica, online edition, 2005 (www.iranicaonline.org/articles/tigran-ii). Gesnerus, I. M. (1749), Novus linguae et eruditionis Romanae Thesaurus, 4, Lipsiae. Giardina, A. (1996), “Roma e il Caucaso”, in: Il Caucaso: cerniera fra culture dal Mediterraneo alla Persia (secoli IV-XI). Settimana di studio del Centro italiano di studi sull’alto Medioevo, Spoleto, 20–26 aprile 1995, 1, Spoleto, 85–141. Gnoli, G. (1996), “Il nome degli Alani nelle iscrizioni sassanidi: considerazioni linguistiche e storiche sul tema dell’opposizione tra Iran esterno e Iran interno”, in: Il Caucaso: cerniera fra culture dal Mediterraneo alla Persia (secoli IV-XI). Settimana di studio del Centro italiano di studi sull’alto Medioevo, Spoleto, 20–26 aprile 1995, 2, Spoleto, 831–861. Golden, P. B. (1980), Khazar Studies. An Historico-Philological Inquiry into the Origins of the Khazars, 1, Budapest. Groag, E. (1927), “Lusius (no. 9)”, in: PWRE, 13.2, Stuttgart, 1874–1890. Grousset, R. (1984), Histoire de l’Arménie des origines à 1071, Paris. Hakobyan, H. (2010), “L’éphémère capitale de Tigrane le Grand”, in: P. Donabédian / C. Mutafian (eds.), Les douze capitales d’Arménie, Paris-Marseille, 96–104. Henderson, B. W. (1903), “Controversies in Armenian Topography. I, The Site of Tigranocerta”, in: The Journal of Philology 28, 99–121. Henning, W. B. (1952), “A Farewell to the Khagan of the Aq-Aqatärān”, in: Studies Presented to Vladimir Minorsky by His Colleagues and Friends (= Bulletin of the School of Oriental and African Studies. University of London 14.3), Cambridge, 501–522. Hewsen, R. H. (1982), “Ptolemy’s Chapter on Armenia: an Investigation of his Toponyms”, in: Revue des Études Arméniennes n.s. 16, 111–150. Hewsen, R. H. (1987), “Artaxata”, in: E. Yarshater (ed.), Encyclopædia Iranica, 2, London-New York, 653–654. Holmes, T. R. (1917), “Tigranocerta”, in: The Journal of Roman Studies 7, 120–138. Hübschmann, H. (1904), “Die altarmenischen Ortsnamen. Mit Beiträgen zur historischen Topographie Armeniens und einer Karte”, in: Indogermanische Forschungen 16, 197–490. James, E. B. (1873a), “Artaxata”, in: W. Smith (ed.), A Dictionary of Greek and Roman Geography, 1, London, 227. James, E. B. (1873b), “Gogarene”, in: W. Smith (ed.), A Dictionary of Greek and Roman Geography, 1, London, 1004.


191

James, E. B. (1873c), “Otene”, in: W. Smith (ed.), A Dictionary of Greek and Roman Geography, 2, London, 504. Juntunen, K. (2013), “The Arrogant Armenian. Tiridates (Bagratuni) in Cassius Dio and Movses Khorenatsʽi”, in: Arctos 47, 153–172. Kanetsian, A. G. (1998), “Città e insediamenti nell’Armenia di età classica”, in: A. Invernizzi (ed.), Ai piedi dell’Ararat. Artaxata e l’Armenia ellenistico-romana, Firenze, 1–94. Kettenhofen, E. (1994), “Darband (1)”, in: E. Yarshater (ed.), Encyclopædia Iranica, 7.1, London-New York, 13–19. Khachatrian, Ž. D. (1998), “Artaxata capitale dell’Armenia antica (II sec. a.C.-IV sec. d.C.)”, in: A. Invernizzi (ed.), Ai piedi dell’Ararat. Artaxata e l’Armenia ellenistico-romana, Firenze, 95–158. Khatchatrian, J. (2010), “Artachat. Capitale d’Arménie durant cinq siècles”, in: P. Donabédian / C. Mutafian (eds.), Les douze capitales d’Arménie, Paris-Marseille, 88–95. Kiepert, H. (1873), “Über die Lage der armenischen Hauptstadt Tigranokerta”, in: Monatsberichte der königlich preussischen Akademie der Wissenschaften zu Berlin, 164– 211. Kiessling, E. (1912), “Gogarene”, in: PWRE, 7.2, Stuttgart, 1553–1555. Kretschmer, K. (1927), “Σωβίδαι”, in: PWRE, 3 A.1, Stuttgart, 770. Lehmann-Haupt, C. F. (1910), Armenien Einst und Jetzt. Reisen und Forschungen, 1, Berlin. Lehmann-Haupt, C. F. (1936), “Tigranokerta”, in: PWRE, 6 A.1, Stuttgart, 981–1007. Lepper, F. A. (1948), Trajan’s Parthian War, Oxford-London. Leschi, L. (1945), “La carrière de Q. Marcius Turbo, préfet du prétoire d’Hadrien”, in: Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres 89.1, 144–162. Lightfoot, C. S. (1990), “Trajan’s Parthian War and the Fourth-Century Perspective”, in: The Journal of Roman Studies 80, 115–126. Magie, D. (1950), Roman Rule in Asia Minor to the End of the Third Century After Christ, 2 vols., Princeton. Mahé, J.-P. / Mahé, A. (eds.) (1993), Moïse de Khorène, Histoire de l’Arménie (introduction, traduction de l’arménien et notes), Paris. Manandian, H. (1963), Tigrane II & Rome. Nouveaux éclaircissements à la lumière des sources originales (trad. H. Thorossian), Lisbonne. Manandian, H. A. (1965), The Trade and Cities of Armenia in Relation to Ancient World Trade (trad. N. G. Garsoïan), Lisbon. Manni, E. (1971), “Asinio Quadrato e l’arcaismo erodoteo nel III secolo d.C.”, in: Studi di storiografia antica in memoria di Leonardo Ferrero, Torino, 191–201. Markwart, J. (1930), Südarmenien und die Tigrisquellen nach griechischen und arabischen Geographen, Wien. Marquart, J. (1905), Untersuchungen zur Geschichte von Eran, 2 (Philologus, Supplb. 10.1), Leipzig. Mitford, T. B. (1980), “Cappadocia and Armenia Minor: Historical Setting of the Limes”, in: ANRW, 2.7.2, Berlin-New York, 1169–1228. Mommsen, T. (1921), Römische Geschichte, 5, Berlin, 9th ed. Moretti, L. (1955), “Due note epigrafiche”, in: Athenaeum n.s. 33, 32–46. Müller, K. (ed.) (1901), Κλαυδίου Πτολεμαίου Γεωγραφική ὑφήγησις. Claudii Ptolemæi Geographia, 1.2, Parisiis. Mutafian, C. / Van Lauwe, É. (2001), Atlas historique de l’Arménie. Proche-Orient et Sud-Caucase du VIIIe siècle av. J.-C. au XXIe siècle, Paris.

192

Federico Frasson

Nogaret, M. (1984), “Quelques problèmes archéologiques et topographiques dans la région de Maiyāfāriḳīn”, in: Revue des Études Arméniennes n.s. 18, 411–433. Nöldeke, T. (1879), “Ueber îrânische Ortsnamen auf kert und andre Endungen”, in: Zeitschrift der Deutschen Morgenländischen Gesellschaft 33, 143–156. Norden, E. (1920), Die germanische Urgeschichte in Tacitus Germania, Leipzig-Berlin. Ognibene, P. (2009), “Gli Alani tra la Persia e Bisanzio”, in: Bizantinistica s. II, 11, 261– 272. Pareti, L. (1960), Storia di Roma e del mondo romano, 5, Torino. Pasdermadjian, H. (1986), Histoire de l’Arménie depuis les origines jusqu’au traité de Lausanne, Paris, 4th ed. Patterson, L. E. (2013), “Caracalla’s Armenia”, in: Syllecta Classica 24, 173–199. Pelham, H. F. (1896), “Arrian as Legate of Cappadocia”, in: The English Historical Review 11, 625–640. Pelham, H. F. (1911), Essays, ed. by F. Haverfield, Oxford. Petrosyan, H. (2010a), “La Tigranakert d’Artsakh”, in: P. Donabédian / C. Mutafian (eds.), Les douze capitales d’Arménie, Paris-Marseille, 105–109. Petrosyan, H. (2010b), “Tigranakert in Artsakh”, in: Tigranes the Great, Yerevan, 380–387. Petrosyan, H. / Kirakosyan, L. / Safaryan, V. / Zhamkochyan, A. / Vardanyan, R. / Karapetyan, I. / Vardanesova, T. (2012), “The Discovery and First Results of Archaeological Investigation of Tigranakert in Artsakh, 2005–2009”, in: P. Avetisyan / A. Bobokhyan (eds.), Archaeology of Armenia in Regional Context. Proceedings of the International Conference dedicated to the 50th Anniversary of the Institute of Archaeology and Ethnography Held on September 15–17, 2009 in Yerevan, Yerevan, 223–235. Pflaum, H.-G. (1970), “La valeur de la source inspiratrice de la Vita Hadriani et de la Vita Marci Antonini à la lumière des personnalités contemporaines nommément citées”, in: J. Straub (ed.), Bonner Historia-Augusta-Colloquium 1968/1969, Bonn, 173–232. Platnauer, M. (1918), The Life and Reign of the Emperor Lucius Septimius Severus, Oxford-London. Polaschek, E. (1961), “Uti (no. 2)”, in: PWRE, 9 A.1, Stuttgart, 1178–1179. Ramelli, I. (ed.) (2002), Il Chronicon di Arbela: presentazione, traduzione e note essenziali, Madrid. Ramelli, I. (2006), “Il Chronicon di Arbela: una messa a punto storiografica”, in: Aevum 80.1, 145–164. Rapp, S. H. (2014), The Sasanian World through Georgian Eyes. Caucasia and the Iranian Commonwealth in Late Antique Georgian Literature, Farnham-Burlington. Roberto, U. (2008), “L’eredità di Traiano: la vicenda di Lusius Quietus nel pensiero di Temistio (Or. XVI 204D-205A)”, in: A. Castaldini (ed.), L’eredità di Traiano. La tradizione istituzionale romano-imperiale nella storia dello spazio romeno. Atti del convegno internazionale di studi, Bucarest, 6–7 giugno 2007, Bucarest, 72–90. Saint-Martin, J. (1818), Mémoires historiques et géographiques sur l’Arménie, 1, Paris. Saxtorph, N. M. (2002), “Acies contra Alanos: Arrian on Military Tactics” (translation of Arrian’s text by C. G. Tortzen), in: K. Ascani / V. Gabrielsen / K. Kvist / A. H. Rasmussen (eds.), Ancient History Matters. Studies Presented to Jens Erik Skydsgaard on His Seventieth Birthday, Roma, 221–226. Schippmann, K. (1989), “Balāš III”, in: E. Yarshater (ed.), Encyclopædia Iranica, 3, London-New York, 576.


193

Schmitt, R. (1990), “Carmania”, in: E. Yarshater (ed.), Encyclopædia Iranica, 4, London-New York, 822–823. Schmitt, R. (1996), “Armenische Personennamen und geographische Namen: eine Übersicht”, in: Il Caucaso: cerniera fra culture dal Mediterraneo alla Persia (secoli IV-XI). Settimana di studio del Centro italiano di studi sull’alto Medioevo, Spoleto, 20–26 aprile 1995, 2, Spoleto, 685–708. Schottky, M. (1998), “Quellen zur Geschichte von Media Atropatene und Hyrkanien in Parthischer Zeit”, in: J. Wiesehöfer (ed.), Das Partherreich und seine Zeugnisse. The Arsacid Empire: Sources and Documentation. Beiträge des Internationalen Colloquiums, Eutin (27.-30. Juni 1996), Stuttgart, 435–472. Sellwood, D. (1980), An Introduction to the Coinage of Parthia, London, 2nd ed. Shapira, D. D. Y. (2007), “Armenian and Georgian Sources on the Khazars: a Re-Evaluation”, in: P. B. Golden / H. Ben-Shammai / A. Róna-Tas (eds.), The World of the Khazars. New Perspectives. Selected Papers from the Jerusalem 1999 International Khazar Colloquium hosted by the Ben Zvi Institute, Leiden-Boston, 307–352. Sinclair, T. (1994–1995), “The Site of Tigranocerta. I”, in: Revue des Études Arméniennes n.s. 25, 183–254. Sinclair, T. (1996–1997), “The Site of Tigranocerta. II”, in: Revue des Études Arméniennes n.s. 26, 51–118. Smallwood, E. M. (1962), “Palestine c. A. D. 115–118”, in: Historia 11.4, 500–510. Southern, P. (1989), “The Numeri of the Roman Imperial Army”, in: Britannia 20, 81– 140. Stein, A. (1930), “Mausakas”, in: PWRE, 14.2, Stuttgart, 2408. Stein, A. (1934), “Theokritos (no. 4)”, in: PWRE, 5 A.2, Stuttgart, 2027–2028. Sturm, J. (1935), “Motene (no. 2)”, in: PWRE, Suppl. 6, Stuttgart, 541. Sturm, J. (1937a), “Τωσαρηνή”, in: PWRE, 6 A.2, Stuttgart, 1810. Sturm, J. (1937b), “Ὠβαρηνοί”, in: PWRE, 17.2, Stuttgart, 1705. Talbert, R. J. A. (ed.) (2000), Barrington Atlas of the Greek and Roman World. Map-by-Map Directory, 2, Princeton-Oxford. Täubler, E. (1909), “Zur Geschichte der Alanen”, in: Klio 9, 14–28. Taylor, J. G. (1865), “Travels in Kurdistan, with Notices of the Sources of the Eastern and Western Tigris, and Ancient Ruins in Their Neighbourhood”, in: Journal of the Royal Geographical Society of London 35, 21–58. Thomson, R. W. (ed.) (1978), Moses Khorenats‘i, History of the Armenians, Cambridge (MA)-London. Thomson, R. W. (1996), “The Writing of History: the Development of the Armenian and Georgian Traditions”, in: Il Caucaso: cerniera fra culture dal Mediterraneo alla Persia (secoli IV–XI). Settimana di studio del Centro italiano di studi sull’alto Medioevo, Spoleto, 20–26 aprile 1995, 1, Spoleto, 493–514. Tomaschek, W. (1895), “Araxes (no. 2)”, in: PWRE, 2.1, Stuttgart, 403–404. Topchyan, A. (2006), The Problem of the Greek Sources of Movsēs Xorenacʽi’s History of Armenia, Leuven-Paris-Dudley (MA). Toumanoff, C. (1963), Studies in Christian Caucasian History, Georgetown. Traina, G. (1991), Il complesso di Trimalcione (Movsēs Xorenac‘i e le origini del pensiero storico armeno), Venezia. Traina, G. (1995), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. I”, in: Le Muséon, 108, 279–233.

194

Federico Frasson

Traina, G. (1998), “Materiali per un commento a Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘. II”, in: Le Muséon, 111, 95–138. Traina, G. (1999–2000), “Épisodes de la rencontre avec Rome (IIe siècle av. J.-C.-IIIe siècle ap. J.-C.)”, in: Iran & the Caucasus, 3/4, 59–78. Traina, G. (2001), “Strabone e le città dell’Armenia”, in: Id. (ed.), Studi sull’XI libro dei Geographika di Strabone, Galatina, 141–154. Traina, G. (2010), “Capitales ou résidences royales. La ville dans l’Arménie ancienne”, in: P. Donabédian / C. Mutafian (eds.), Les douze capitales d’Arménie, Paris-Marseille, 45–49. Van den Hout, M. P. J. (1999), A Commentary on the Letters of M. Cornelius Fronto, Leiden-Boston-Köln. Verstandig, A. (2001), Histoire de l’empire parthe (-250 -227), Bruxelles. Von Domaszewski, A. (1903), “Untersuchungen zur roemischen Kaisergeschichte”, in: Rheinisches Museum für Philologie 58, 218–230. Von Gutschmid, A. (1888), Geschichte Irans und seiner Nachbarländer von Alexander dem grossem bis zum Untergang der Arsaciden, Tübingen. Weissbach, F. H. (1910), “Geli”, in: PWRE, 7.1, Stuttgart, 986–987. Weissbach, F. H. (1924), “Kyros (no. 2)”, in: PWRE, 12.1, Stuttgart, 184–187. Weissbach, F. H. (1932), “Ταβιηνή”, in: PWRE, 4 A.2, Stuttgart, 1878. Winter, E. / Dignas, B. (2001), Rom und das Perserreich. Zwei Weltmächte zwischen Konfrontation und Koexistenz, Berlin. Wolski, J. (1993), L’empire des Arsacides (Acta Iranica, 32), Lovanii. Zecchini, G. (1998), “Asinio Quadrato storico di Filippo l’Arabo”, in: ANRW, 2.34.4, Berlin-New York, 2999–3021. Zecchini, G. (2005), “Il bipolarismo romano-iranico”, in: C. Bearzot / F. Landucci / G. Zecchini (eds.), L’equilibrio internazionale dagli antichi ai moderni, Milano, 59–82. Zuckerman, C. (2007), “The Khazars and Byzantium. The First Encounter”, in: P. B. Golden / H. Ben-Shammai / A. Róna-Tas (eds.), The World of the Khazars. New Perspectives. Selected Papers from the Jerusalem 1999 International Khazar Colloquium hosted by the Ben Zvi Institute, Leiden-Boston, 399–432.

III. Biblical, Christian and Byzantine Literature

Theo Maarten van Lint

Among Others: Greek in Context in the Letters of Grigor Magistros Pahlawuni (eleventh century) Abstract: The erudite layman Grigor Magistros Pahlawuni (c. 990–1058) is an example of the appropriation and transformation of Greek learning in Armenian culture. Cultivating an Irano-Armenian heritage he fuses this with his Hellenistic education and Christian faith as professed by the Armenian Apostolic Church. The article concentrates on one of two letters written to Amir Abrehim, devoted to philosophy.

The life and work of Grigor Magistros Pahlawuni, who was born around 990 in Bjni, close to Ani, the capital of the Armenian Bagratid Kingdom located just west of the present border between Turkey and the Republic of Armenia, and died in 1058 in Taron, west of lake Van, can be considered both a late and a prime example of the Armenian appropriation and creative transformation of Greek learning, fusing Hellenistic erudition with the Irano-Armenian matrix of Grigor’s cultural world¹. In this contribution first a brief introduction will be given of Grigor’s views and activities in scholarly and educational matters in the context of the particular circumstances in which this Armenian naxarar or nobleman of most exalted station, later an exile not quite brought to his knees, and ultimately, Byzantine governor, furthered Armenian Christianity and culture². Subsequently, Grigor’s exchanges with a certain Amir Ibrahim or Abrehim will be presented, a

1 A survey article of Grigor Magistros’s life and work is Sanjian 1993, 131–158; see also van Lint 2014, with further bibliography. A recent article directly relevant to the matter in hand is Muradyan 2013. References to literature in this contribution aim to present guides for further reading and do not pretend to be exhaustive. A bibliography of Armenian textual production is Thomson 1995, followed by the supplement Thomson 2007. 2 A succinct introduction to Armenian history, religion and culture is Uluhogian 2009. On the historical background of the period, see Garsoïan 1997a; Ead. 1997b; Ead. 1998; Redgate 1998; Settipani 2006; Garsoïan/Thierry 2007; Cheynet/Dédéyan 2007; Greenwood 2008. The position of Armenia within the Islamic world and its consequences for Armenian politics, culture and outlook is the focus of Dadoyan 2011–2014, of which the first two volumes are most relevant for the period covered in this contribution; finally, Dédéyan 2003 also takes the immediately preceding period into consideration. On the historical geography of Armenia, see Hewsen 2000 and Mutafian/van Lauwe 2001.

198


Muslim with an Armenian mother, who had requested letters on two subjects: faith and philosophy³. In the final part of this contribution the second of these letters, regarding philosophy, will be considered.

1 Eleventh century Armenia and the reception of Greek culture The reception of Greek texts, culture, philosophy, and mythology is usually associated with the period up to the first centuries after the adoption of Christianity in Armenia. A programme of translation and of intellectual and cultural appropriation of non-Armenian texts began, while the absorption of the outlook of the Eastern Mediterranean world had started long before the Armenians officially became Christians. We can follow this process up to the thirties of the eighth century, through the works translated and the language developed in what is called the Hellenising School, but is perhaps better termed more neutrally as the Grecising style⁴. After 735 the transmission of Greek learning to Armenia is less evident, just as it is unclear to what extent it was innovatively absorbed. The Arab conquest of the 650s, and in particular its character after the Abbasid ascendancy in the 750s is usually adduced as a factor that cut off the development, or at least interrupted it. Yet, there was still contact enough with important aspects of Byzantine Greek culture. The ninth century theological exchanges with Photius may serve as a reminder of this⁵. Apart from evidence of contact on ecclesiastical and Christological matters, indigenised Greek learning continued to be transmitted and commented upon. Perhaps the clearest indication of this is the string of commentaries that were made on Dionysius Thrax’s Techne grammatike, which had been available in

3 On Grigor Magistros’s Christian apologetics in his interaction with Muslim interlocutors see Terian 2012 and van Lint 2010. 4 Writing in Armenian started with the invention of the Armenian alphabet c. 405, leading to a veritable flowering of a variety of genres, both in translations from Greek and Syriac and in original work. A succinct overview is Thomson 1997. The translation techniques applied are discussed in Mahé 1996a; see also Id. 1996b. A survey of the scholarship on the Hellenising School, the diverse interpretations of the phases it went through and their characteristics, of the texts concerned and of the grammatical features that are subsumed under this title, followed by a detailed study of the features themselves is presented in Muradyan 2012a. See also Calzolari/Barnes 2009. 5 See Dorfmann-Lazareff 2004; Greenwood 2006.

Among Others: Greek in Context in the Letters of Grigor Magistros Pahlawuni

199

Armenian, with a first commentary, from the sixth century onwards. In a splendid study published first in 1915, Nicholas Adontz has identified various layers of commentary, of which Hamam Arewelc‘i’s (825?-890?) can be ascribed to the ninth century⁶. From that same century onwards, newfound Byzantine confidence and its seemingly unstoppable eastward expansion had consequences for the situation in Armenia. If we do not have much evidence of direct knowledge of Greek texts or of translation activity from Greek in that period, what we do see is an increase in the bestowing of sometimes very high Byzantine titles on members of the Armenian nobility. Armenians were co-opted into the Empire’s system, some while remaining loyal subjects of the Armenian Church and of the Bagratid or Arcruni kingdom, others after being persuaded to change religious and secular allegiance, with yet others seeking it. This process, unfolding over the ninth to the eleventh centuries has been studied to a certain extent, but much remains to be done⁷. What has not been traced is the impact of Greek learning on those nobles and their families who were co-opted into the Byzantine reward system. Did this lead to an increase in familiarity with Greek philosophical thought, Greek poetry and historiography, and with Greek epistolography in Armenia? Whether as a consequence of the absorption of recent Byzantine imperial ideology or for other reasons, during the tenth century some of the most lastingly influential Armenian clerical intellectuals were disposed well to Greek learning as well as to the role of the Byzantine Emperor in the Christian world. They were connected with the monastery of Narek, and three members of one family can be mentioned as good examples: Xosrow Anjewac‘i (900?–963?), Anania Narekac‘i (910?–985?) and Xosrow’s son Grigor Narekac‘i, (950?–1003) whose Book of Lamentation is a profound work of mystical literature and the most revered text in Armenian after Scripture. However, even with them direct, contemporary absorption of Greek texts seems less likely than a continued, and by this time possibly more intense, transmission and elaboration on material translated earlier, although we should not exclude a situation in which both occurred simultaneously⁸.

6 Adontz 1970. On Hamam Arewelc‘i and his work see further Thomson 2005. Studies on the Armenian commentaries on Dionysius Thrax’s Grammar are Ervine 1988; Ead. 1995; Clackson 1995. 7 See Greenwood 2010. 8 See Mahé/Mahé 2000; the introductory chapters (pp. 1–195) give a survey of the political, cultural, theological, and literary background of the school of Narek and its three central representatives. Hrač‘ya T‘amrazyan has shed light on the school of Narek: T‘amrazyan 1986, 1999, 2004, 2013. See further Cowe 1991. Reflection on Grigor Narekac‘i in particular is well represented, in

200


We do know that the court at Ani, since 961 the capital of the Bagratid Kingdom and since 992 also the see of the Catholicos, the Head of the Armenian Apostolic Church, was renowned for its interest in learning, in particular at the end of the tenth and in the early decades of the eleventh centuries. Research in this area is in its initial stages, and thus the following description may have to be adapted as our insight develops⁹. In one of Grigor Magistros Pahlawuni’s nine letters addressed to Petros Getadarj, who was Catholicos of the Armenians between 1019 and 1058, he asks his friend to release from oblivion a tome containing the works of the seventh century polymath Anania of Širak, known as the K‘nnikon, which contains all knowledge applied in the quadrivium¹⁰. Significantly, it is held in the catholicossal library – not, for instance, at the court – and, just as significantly, Grigor humbly adds that, if his Holiness deems it unfitting that the work be brought into circulation, he, Grigor, will obediently abide by his will. Transmission of secular knowledge, albeit interpreted as a cornerstone for the knowledge of God, took place in monastic circles, and the dissemination was controlled by clerical authorities. It is important to stress that while we know of a circle of nobles, including the king, who were interested in learning, from this period we only have the works of Grigor Magistros that attest to the active pursuit of knowledge by a layperson. Leaving aside now the question of the transmission of Greek learning between the eighth and late tenth centuries, it is worthwhile to briefly sketch the life and work of Grigor Magistros Pahlawuni, before we turn to his correspondence.

2 Grigor Pahlawuni Magistros’ life and work Grigor Pahlawuni was born in a family that had long played an important role in the Bagratid Armenian Kingdom, and had been the recipients of Byzantine favour for several generations. It also laid emphasis on its high lineage: that of

monographs as well as in collective volumes, and individual articles. Pĕltean (Beledian) 2008; Zekiyan 1999; Mahé/Zekiyan 2006; Mahé/Rouhana/Zekiyan 2010; the proceedings of the Millennium Conference on Grigor Narekac‘i Prelacy of the Eastern Diocese (held in New York in 2003), published in HASK, 10, 2002–2006. To the (partial) translations of the Narek mentioned above may be added Samuelian 2002 and those, together with studies on the saint, contained in Russell 2004. 9 See Mat‘evosyan 1986–1987 and 1997; the section “L’école d’Ani et la vie intellectuelle” by various authors in Kévorkian 2001, 235–273. 10 On Anania Širakac‘i, see Greenwood 2011; on the identity and importance of his educational work, Mahé 1987.


201

the Pahlawuni, or Part‘ews, in other words, the Parthians. In particular Grigor insisted on the Pahlawuni genealogy, connecting them with Grigor the Illuminator, and with the Arsacid kings. The different strands of his Pahlawuni ancestry made him a scion of two of the most outstanding Parthian families, the Suren and the Karen¹¹. Insistence on their noble lineage was essential for the standing of naxarar families: it anchored and legitimised them in this fleeting world. His was an ancestry therefore that defined Armenianness through two of its main institutions: through royalty, soon to prove elusive, and through the permanence, albeit under almost constant threat, of the Armenian Church. Not only Grigor Magistros, this layman and exception in Armenian literature did so, but also his great-grandson, one of the most illustrious Armenians of all times, Catholicos Nersēs Šnorhali, in his Vipasanut‘iwn, or ‘Epic History’, written in early manhood just after 1120¹². This anchorage through genealogy points us to a triple cultural background – Armenian, Iranian and Greek – present in the Pahlawuni family, in Grigor’s own work, and indeed in Armenian culture. First and foremost, Armenian by prime definition, through religion, language and genealogy and no less consciously, Parthian Iranian, again by genealogy, and by various other aspects such as the persisting complex of pre-Christian beliefs mostly relegated to folklore, but partially co-opted by the Church, and finally Greek, through pre-Chalcedonian or Nicaene Christianity, through Byzantine preferment, and, perhaps more importantly, through the absorption of Classical education in a Christian matrix. In Grigor’s case we can point to his profound interest in the common Armeno-Iranian past, which at times gives the impression of a fusion of the two in his mind, were it not for his very pronounced Apostolic Armenian Christian faith. For this Grigor develops a distinct literary ethos, combining pre-Christian Armenian heroic poetry with Christian precepts. We witness this in his extraordinary panegyric for Vahram Pahlawuni, his uncle, sparapet or commander in chief of the Armenian army, who died in the saddle at eighty years of age¹³. To this triple background we may add a fourth, Arabic and more generally speaking an Islamic one, which entered Armenian culture in the seventh century,

11 Grigor lays emphasis on his lineage, the authenticity of which has not been proven, in two of his letters, nos. 18 and 38 in the recent edition of his works by Gohar Muradyan: Muradyan 2012b, 256–262 and 312. See also Terian 2012, verses 952–953. 12 Mkrtč‘yan (1981), 107–110, verses 1091–1134. See also Mutafian 2012, 248–257 and Genealogical Table 40. 13 Letter no. 18 in Muradyan’s edition (Muradyan 2012b). See van Lint forthcoming and Č‘ugaszyan 1987.

202


and which transpires both in the themes of his work and in his poetics, about which below. Having been educated in the best possible way in the Kingdom of Ani, Grigor became an erudite by volition, mastering theology, natural sciences, and medicine, and further the founder of schools, in which the seven artes liberales were taught. Like other family members, who founded Marmašēn monastery, he was an important patron of architecture, founding among others Kečaṙis monastery. His gift to Havuc‘ Taṙ monastery of a wooden panel depicting Christ victorious on the cross introduced this type of image in Armenia where it was hewn on crossstones or xač‘k‘ars, of the amenap‘rkič‘ or All-Saviour type¹⁴. From 1021, the year his father died, he also assumed political and military roles, supporting, in 1042, the succession to the Bagratid throne of Gagik, with whom he subsequently fell out, which led to his persecution and bitterness on Grigor’s side. However, in 1044, when raison d’êtat called for decisive action Grigor defended the king aga ins t the pro-Byzantine faction’s advice to heed a call to Constantinople, where the Emperor had invited the young and un-experienced monarch. It was a trap. The background to Byzantium’s policy was a promise apparently made by Yovhannēs Smbat, the previous king, to bequeath his kingdom to the Empire upon his death in exchange for protection against rivals. Grigor’s interventions proved in vain and just as king Gagik had had to surrender his kingdom in exchange for lands in the Empire, so did Grigor: between 1045 and 1049 he was made magistros and dux of Mesopotamia and Taron, west of Lake Van, provinces with a large Armenian population, where he undertook a rigorous campaign against the T‘ondrakian heresy¹⁵. After 1045 he continued to teach as well, and on several occasions he pronounced the desire to withdraw to a monastery. It is uncertain whether this was granted by his Byzantine sovereign. He would then have died in a monastic environment in 1058¹⁶. Among Grigor’s works his commentary on Dionysius Thrax’s Techne grammatike must be mentioned, which was the staple text – incorporating previous Armenian commentaries – until it was superseded by Yovhannēs Erznkac‘i’s work, two centuries later¹⁷. It ensured the cohesion of a religious education in

14 Der Nersessian 1978, 111 (image) and 114 (description); Donabédian/Thierry 1987, 128 (description), 389 (image). 15 On Grigor as administrator dealing with heretics, see Garsoïan 1967 and the master’s thesis of Federico Alpi (Alpi 2010). 16 On speculation about this see Sanjian/Terian 1985–1986. 17 See the works by Adontz 1970; Ervine 1988; Ead. 1995; Clackson 1995 and Xač‘erean 1987.


203

which everything, from language to the cosmos hung together in an ordered, rational way, installed by God. Remarkably, we find in it a paragraph where Grigor comments on poetry, and states that he has studied Arabic with much effort, first script, then language, then its poets, with their rhyme, or kafiye¹⁸. This propensity towards rhyme comes to the fore in his retelling of the Bible in verse, where in imitation of the Quran the syllable -in is used for the monorhyme Grigor here employs. It is the result of an encounter in 1045 in Constantinople with the Muslim poet and erudite, al Manazi, an avid collector of Greek manuscripts. It launches the genre of the religious epic in Armenian¹⁹. Grigor’s third contribution to Armenian literature is his epistolary of some eighty-eight letters. Next to the edition by Kostaneanc‘ of 1910 we now have Muradyan’s edition, published in 2012 in volume 16 of the Matenagirk‘ Hayoc‘, using more and better manuscripts than Kostaneanc‘ had done²⁰. The thorny question of the translation of Plato into Armenian and the possible involvement in it of Grigor Magistros has been studied by various scholars and has recently been critically assessed in work by Aimi and Tinti²¹. The Armenian erudite may also have translated Euclid’s work into Armenian²². Recent research has brought to light that Grigor Magistros is the author of several more šarakans or hymns than previously thought²³. He also wrote a panegyric of the Holy Cross.

3 Byzantine epistolary models Grigor’s letters are modelled on Byzantine epistolary practice²⁴. Gregory of Nazianzus’ description reflects what an ideal letter would have to look like. He states that “letters should be brief, clear, and phrased like a conversation with an absent friend and should treat serious topics with elegant expressions”²⁵. Letters were often read out aloud, not only to the immediate addressee, but to a wider audience as well. Letters were entrusted to grammatophoroi or letter bearers. These were supposed to be “living letters” (empsychos epistole), and to

18 Adontz 1970, 232; cf. Cowe 2005, 384–385. 19 Edition, translation and commentary in Terian 2012. 20 Kostaneanc‘ 1910; Muradyan 2012b. 21 See the contribution by I. Tinti and the appendix by C. Aimi in this volume; cf. Tinti 2012. 22 Akinean 1960; Leroy 1936. 23 Arevšatyan 2014, 61–107, 154–161. 24 For this section, see Jeffreys/Kazhdan 1991. 25 Jeffreys/Kazhdan 1991, 719, quoting from Gallay 1964–1967, letter no. 51.

204


convey factual information, while the letters functioned as literary decoration to the message. They could be rather general and even be borrowed wholesale by others. Letters could be accompanied by gifts, such as a book, but edibles like fruit or fish occurred as well. Three types of letters may be distinguished according to purpose: official, private and literary. A division in types of content may be made distinguishing diplomatic, theological and scholarly letters; further letters of recommendation, indoctrination and censure; and letters of consolation. There is no clear distinction in genre between letters, conversation and homily; there exist sermons in the form of a letter; others could take the form of a historical work, the preamble to a major work, or a theological tract – an example of which is Photius letter to Ašot I²⁶, while another is Grigor Magistros’ letter on faith to Amir Ibrahim; both are quite long. Letters could form part of larger works. Formulaic expressions characterise the opening and closure of Byzantine letters. The letter was considered a piece of rhetoric, requiring skill. In the opening part or prooimia the author was expected to give a display of erudition through allusions to and quotations of classical and biblical literature. These varieties are found in Grigor’s letters as well.

4 Armenian praxis: Grigor Magistros’ Epistolary How does this description relate to Grigor’s letters? One gets the impression that in most of them Grigor seeks to convey real information. Matters discussed range from the request to borrow a book, to the accompaniment of a gift – rather than the letter being accompanied by a gift – to stern reprimands to lazy pupils, letters in which Grigor bewails his hard fate, in which he consoles others for what has befallen them: the overall impression of the epistolary is of a greater engagement than perhaps seen in some of the Byzantine counterparts. The opening and closing formulae of the letters are brief in the extreme, the language is usually complex, sliding into the enigmatic in some cases, but very clear in others. Lofty themes alternate with playful images; a storm with whipping rain is the occasion for a brief letter in which Grigor transmits a rare gem of pre-Christian epic, heard from a peasant. The erudition the letters betray can easily overwhelm. There hardly seems an area of knowledge for which Grigor did not have a deep interest and which he did not pursue with a voracious appetite. Human emotions also

26 Laourdas/Westerink 1985, 4–97.


205

abound: deep grief over injustice suffered from the king he helped crown, and nothing less than rage against heretics. He bewails in moving words the death through illness of one of his young sons. He writes with undisguised affection, even admiration to his son Vahram, the later Catholicos Grigor II Vkayasēr (Martyrophilos), encouraging him in his studies. Unexpectedly moving is the description of his tender memories of his doting uncle Vahram, in answer to a letter of condolence by a family friend, Yovhannēs Ark‘episkopos of Siwnik‘, on the occasion of Vahram’s death²⁷.

5 Two letters to Amir Ibrahim, on Faith and on Philosophy²⁸ Two letters in the epistolary are written in answer to a request by Amir Ibrahim (Abrehim): one on Faith, and one on Philosophy. Their context is triple. Firstly, that of Christian-Islamic polemics: Amir Abrehim asks questions that reflect Muslim sensitivities about Christian theology. Grigor develops his answer along the lines of Christian apologetics. It is of particular interest that we have with Grigor’s long letter on faith, together with his Biblical poem aptly entitled by Terian Magnalia Dei, both a theological and an aesthetic apology of Christianity and a multifaceted polemic with contemporary Islam²⁹. A second aspect is the reception of Philosophy in Islam and in Christianity and the relationship of these two to the previous point. The two are indissolubly linked for the rational Christian in Grigor’s view. We do not know why Amir Abrehim asked for one letter on faith and one on philosophy, nor what Grigor Magistros knew about this Armenian Muslim’s predicament or interest, although he was clearly acquainted with him prior to the correspondence. Finally there is Grigor Magistros’ educational programme, encompassing trivium and quadrivium, the Old and New Testaments and their interpretations. Grigor embraces the motto of Plato’s Academy: “Let no one enter who has not mastered Geometry”. Mathematics is the underlying principle and expression of God’s creative order. Grigor’s educational views are thus fully consonant with

27 Cf. supra, n. 13. A complete translation of the letters with commentary is being undertaken by the author of these lines. 28 Letters nos. 9 and 26 in Muradyan’s edition (Muradyan 2012b, 213–237 and 288–294); nos. 70 and 71 in Kostaneanc‘ edition (Kostaneanc‘ 1910, 170–201 and 201–208). 29 Succinct information on both in van Lint 2010; on the latter, see further Terian 2012.

206


the previous two points: the relationship of faith with philosophy in it is fundamental, marrying Greek philosophy as received among others through Philo of Alexandria and Clement of Alexandria’s Protreptics with general and Armenian patristics.

6 Structure and content of Grigor’s Letter to Amir Abrehim on Philosophy The easiest way to approach the contents of the second letter addressed to Amir Ibrahim, On Philosophy, which is much shorter than the one On Faith – by some margin the longest in the collection – but more complex in language and images, is to present a summary revealing its structure. It abundantly employs Grecisms and complex mythological imagery. A brief summary of each paragraph is followed by a word of comment³⁰. Answer to Amir Ibrahim, who requested one on philosophy and one on faith § 1.1–3 Introduction with praise for the wisdom of the interlocutor’s request In earthly life faith and philosophy are honourable; they contain all good things in the present and are a pledge for the future. Comment. Amir Ibrahim is discerning what is essential in life – a good start. § 2.4–7 Faith stems from the Holy Spirit: drink from the cup of faith like the prophets and kings did Grigor refers to the previous letter, on faith: he gave of his wisdom, but its source is the Holy Spirit. Amir Abrehim’s is not a shameful request, and Grigor now fulfils it in humility. Comment. Grigor takes a humble stance, explaining that his capability to teach on faith and philosophy stems from the Holy Spirit. He does it in his own particular way – “his cup”. The fact that Ibrahim asks for insight shows his humility, which is a virtue and not shameful at all. Humility thus unites the one who asks with the one who answers. Personal pride, stressing division, is out of the question.

30 A translation of the letter, with more extensive comment is under preparation for publication in the complete translation of Grigor Magistros’ epistolary.


207

§ 3.8–9 Grigor provides assistance, not with the ready product, like a pearl fished up from the bottom of the sea, but by pointing to the need for preparation and inner growth Wisdom and judiciousness grow and develop gradually. Comment. In asking for these letters, Amir Abrehim should not be expecting a quick fix and a ready answer, like a pearl fished up from the bottom of the sea, ready for the taking. Insight is the result of a learning process that takes time. § 4.10–12 Philosophy and royal lineage: Salomon, Plato, Nektanebos; to be connected with Amir Abrehim’s house The most exalted among human lineage strove after wisdom, it behoves Amir Abrehim to do the same. Comment. Grigor combines the desire for wisdom with the noblest lineage in royalty and philosophy, both of Biblical and of Greek background. Nobility of lineage and nobility of spirit coincide in Amir Abrehim. Grigor sounds a note of affinity with his interlocutor, praising him before he takes the plunge and sets out on a difficult journey leading to insight in reality through philosophy. § 5.13–15 Philosophy is a heady drink that will make one feel as if being tossed about at sea, longing for the captain of the ship Grigor will now “philosophise”: the readers are welcomed to the feast to eat a bit, then they will become drunk with heavy stuff and totter about as in their cups, like the Greek god Dionysius, and will feel on fire like Hephaistos; they will feel as if at sea, called Pisidon, that is Poseidon, by Grigor Magistros, and they will realise their need of a nawapet, a captain and guide. Comment. In their complexity and winding character the sentences are an image of someone gradually getting drunk and looking for help: in order to go and see a doctor, you have to realise you are ill. Grigor mingles images of Greek mythology with early Christian ones, ending with Christ, who is implied by the use of the word nawapet, behind which stands the Greek kubernētēs. § 6.16–19 Philosophy is hard as a stone, like a ruby: it stems from the Logos, born from a Virgin Comment. The development one undergoes in obtaining wisdom is a painful process that requires full engagement. Grigor seems to use images from alchemy, involving Hephaistos, and invokes the process of the formation of a ruby. A wide array of names, geographic and otherwise is used: Basra, Yemen, Arabic and Laconian rock, and Persian ways. This wide ranging list of places and images does not obfuscate the message at the heart of the letter: philosophy is an applied

208


kind of knowledge, it leads to insight that must transform, and transform in a particular way, namely in accordance with the wisdom and love of Christ. Grigor here brings philosophy and theology together, which turns this letter into a direct continuation of the one on faith, with a subject that is not an alternative to it, but inextricably bound to it, as it has its origin in the logos, Christ, the source of rational insight in reality and truth. Logos as the source of logic (logikē) is not just delightful wordplay, but delectable reality. § 7.20–24 Not all philosophers make quality goods, just as not all poetry is excellent Comment. Grigor seems to warn against so-called philosophers who peddle coarse ware and are in no way comparable to real philosophers, who offer quality goods. The opposition Sophist – Socratic philosopher springs to mind. § 8.25–29 Good philosophising is transparent in its argumentation and regards the general principles, explaining our existence from the essence Comment. Grigor makes a new start here, going into what he has said before. He brings God and philosophy closer together in his exposition. § 9.30–34 God’s creation is understandable through the ratio, which is given to us: use cosmology and numerology. We need to use both our senses and our ratio to seek wisdom Comment. Grigor brings faith and reason together, and introduces divisions: “God the immovable world, heaven the moving world, human beings the rational world; earth and the units [that is, 1 to 9] (are) God”. He quotes here Hermes Trismegistos, possibly, as he does in the other letter to Amir Abrehim, through Clement of Alexandria’s Protrepticus³¹. § 10.35–38 Our physical presence on earth has meaning: genealogy counts. Amir Abrehim and Grigor are blood relatives through Zrvan and Hayk, but this should not lead to pride in a worldly way Comment. Grigor seeks common ground with Amir Ibrahim, using the genealogy of Movsēs Xorenac‘i, whom Grigor often quotes precisely due to the historian’s emphasis on Armenian genealogy, tying it in with Biblical history.

31 Manandyan 1956, 298.5–7 and 10–11; quotation in this letter as identified by Muradyan in her edition (p. 291 n. 10); the one in the letter on faith (no. 26) appears on pp. 223–224 of her edition (Muradyan 2012b) and is referenced by her in Muradyan 2013, 47 n. 51.


209

§ 11.39–42 The bond of philosophy is even stronger than that of blood; if the latter would be, however unjustifiably, attacked, adversaries will not be able to break the ties that philosophy creates Comment. Grigor plays upon the result of wisdom: it brings people together, like blood-ties, but while these may be severed or denied, those of kinship through philosophy cannot. § 12.43–47 Grigor has chosen only now to answer Amir Abrehim’s requests, at a time suitable for Ibrahim Comment. While many ask for insight in philosophy from a young age, it may well be too early to satisfy that hunger, as it must yield the right result. § 13.48–52 About the composition of poetry for Amir Abrehim Comment. Grigor mingles mythology from different backgrounds, quoting Clement of Alexandria’s Protreptics on Orpheus and in abbreviated form the story of the poet-performer Eunomos, whose string broke, the place of which was taken by a grasshopper. The latter then became renowned for its praise of God³². Grigor, through his “performance”, of poetry and musical play for Amir Abrehim, in all his imperfection as poet and musician, thus seeks the glory of God in his conveying of philosophical insights. § 14.53–57 The letter must be read out by Anania, and must be preserved well Comment. Grigor gives detailed information about the importance of the proper declamation of the letter. He here adduces the precepts given in Dionysius Thrax’s Grammar. Anania is careful with letters he transports, and reads them out in the right way. Grigor states that he will add information as well, thus accompanying the written word with an oral communication, that may be as momentous as what the letter has to say. § 15.58–60 Coda: an exercise in geometry ending with the hexagon as perfect form created by God Comment. Grigor engages in numerical and geometrical symbolism, involving the ternary, pentagons, hexagons, the binary, the seven, the monad, and ends mentioning that it was all called into being by God. The symbolism may carry different types of Christian arithmetic theology, containing Pythagorean and Platonic elements; in seeking for an explanation one may look to the

32 Grigor alludes to Clement of Alexandria once more, Muradyan 2012b, Greek Authors, 50–51 (II.1.h).

210


Timaeus, to the works of Philo of Alexandria, and once again to Clement of Alexandria.

7 Conclusion Grigor’s letter shows his vivid reception and absorption of Greek philosophy, mythology and grammatical thinking, expressed in a genre modelled on late antique Christian letter writing, all yoked together to serve Christian philosophy – not much of which is actually transmitted – and fusing it with theology through the concept of Christ the Logos, apologetically and deliberately applied here to persuade a Muslim Armenian to adopt Christianity.

Bibliography Adontz, N. (1970), Denys de Thrace et les commentateurs arméniens (translated from the original Russian edition of 1915 by René Hotterbeex), Louvain. Akinean, H. N. (1960), Ewkłideay girk‘ yałags tarerac’ erkrač‘ap‘ut‘ean, Vienna. Alpi, F. (2010), Le lettere 67, 68 e 69 di Grigor Magistros (XI sec.) contro i Tondrachiani: Traduzione e commento, Tesi di laurea in Lingua e Letteratura armena, Università degli Studi di Bologna. Arevšatyan, A. (2014), Anii eražstakan mšakuyt‘ĕ [The Musical Culture of Ani], Erevan. Calzolari, V. / Barnes, J. (eds.) (2009), L’oeuvre de David l’Invincible, Leiden-Boston. Cheynet, J.-C. / Dédéyan, G. (2007), “Vocation impériale ou fatalité diasporique: les Arméniens a Byzance (IVe-XIe siècle)”, in: G. Dédéyan (ed.), Histoire du peuple arménien, Toulouse, 297–326. Clackson, J. P. (1995), “The Techne in Armenian”, in: V. Law / I. Sluiter (eds.), Dionysius Thrax and the Techne Grammatike, Münster, 121–133. Cowe, S. P. (ed.) (1991), Commentary on the Divine Liturgy by Khosrov Anjewac‘i, New York. Cowe, S. P. (2005), “The Politics of Poetics: Islamic Influence on Armenian Verse”, in: J. J. van Ginkel / H. Murre-van den Berg / T. M. van Lint (eds.), Redefining Christian Identity. Cultural Interaction in the Middle East since the Rise of Islam, Leuven, 379–403. Č‘ugaszyan, B. L. (1987), “Noruyt‘ner Grigor Magistrosi ‘T‘łt‘erum’” [Novelties in Grigor Magistros’ “Letters”], in: Lraber hasarakakan gitut‘yunneri, 10, 46–53. Dadoyan, S. B. (2011–2014), The Armenians in the Islamic World. Paradigms of Interaction. Seventh to Fourteenth Centuries, 3 vols., New Brunswick-London. Dédéyan, G. (2003), Les Arméniens entre Grecs, Musulmans et Croisés. Étude sur les pouvoirs arméniens dans le proche-orient méditerranéen (1068–1150), 2 vols., Lisbonne. Der Nersessian, S. (1978), Armenian Art, London.


211

Donabédian, P. / Thierry, J.-M. (1987), Les arts arméniens, Paris. Dorfmann-Lazareff, I. (2004), Arméniens et Byzantins à l’époque de Photius: deux débats théologiques après le triomphe de l’orthodoxie, Louvain. Ervine, R. R. (1988), Yovhannēs Erznkac‘i Pluz’s ‘Compilation of Commentary on Grammar’, PhD Dissertation, Columbia University. Ervine, R. R. (1995), “Yovhannēs Erznkac‘ i Pluz’s ‘Compilation of Commentary on Grammar’ as a Starting Point for the Study of Medieval Grammars”, in: J. J. S. Weitenberg (ed.), New Approaches to Medieval Armenian Language and Literature, Amsterdam, 149–165. Gallay, P. (ed.) (1964–1967), Saint Grégoire de Nazianze, Lettres, 2 vols., Paris. Garsoïan, N. G. (1997a), “The Byzantine Annexation of the Armenian Kingdoms in the Eleventh Century”, in: R. G. Hovannisian (ed.), The Armenian People from Ancient to Modern Times, 1, London, 186–196. Garsoïan, N. G. (1997b), “The Independent Kingdoms of Medieval Armenia”, in: R. G. Hovannisian (ed.), The Armenian People from Ancient to Modern Times, 1, London, 143–185. Garsoïan, N. G. (1998), “The Problem of Armenian Integration into the Byzantine Empire”, in: H. Ahrweiler / A. E. Laiou (eds.), Studies on the Internal Diaspora of the Byzantine Empire, Washington (D. C.), 53–129. Garsoïan, N. G. / Thierry, J.-M. (2007), “L’indépendence retrouvée: Royaume du Nord et Royaume du Sud (IXe-XIe siècle)”, in: G. Dédéyan (ed.), Histoire du peuple arménien, Toulouse, 243–296. Greenwood, T. (2006), “Failure of a Mission? Photius and the Armenian Church”, in: Le Muséon 119, 123–167. Greenwood, T. (2008), “Armenian Neighbours (600–1045)”, in J. Shepard (ed.), The Cambridge History of the Byzantine Empire, c. 500–1492, Cambridge, 333–364. Greenwood, T. (2010), “Patterns of Contact and Communication: Constantinople and Armenia, 860–976”, in: R. G. Hovannisian / S. Payaslian (eds.), Armenian Constantinople, Costa Mesa, 73–99. Greenwood, T. (2011), “A Reassessment of the Life and Mathematical Problems of Anania Širakac‘i”, in: Revue des Études Arméniennes 33, 131–186. Hewsen, R. H. (2001), Armenia. A Historical Atlas, Chicago-London. Jeffreys, E. M. / Kazhdan, A. (1991), “Epistolography”, in: A. Kazhdan (ed.), The Oxford Dictionary of Byzantium, Oxford, 718–720. Kévorkian, R. H. (ed.) (2001), Ani. Capitale de l’Arménie en l’an mil, Paris. Kostaneanc‘, K. (ed.) (1910), Grigor Magistrosi t‘łt‘erĕ [Grigor Magistros’ Letters], Ałek‘sandrapol. Laourdas, B. / Westerink, L. G. (eds.) (1985), Photii Patriarchae Constantinopolitani Epistulae et Amphilochia, 3, Leipzig. Leroy, M. (1936), “La traduction arménienne d’Euclide”, in: Annuaire de l’Institut de philologie et d’histoire orientales et slaves 4, 785–816. Mahé, A. / Mahé, J.-P. (2000), Grégoire de Narek Tragédie. Matean Ołbergut‘ean. Le Livre de Lamentation, Louvain. Mahé, J.-P. (1987), “Quadrivium et cursus d’études au 7e siècle en Arménie et dans l’Empire Byzantin d’après le ‘K‘nnikon’ d’Anania Shirakac‘i”, in: Travaux et Mémoires 10, 159–206. Mahé, J.-P. (1996a), “Connaître la Sagesse: le programme des anciens traducteurs arméniens”, in: R. H. Kévorkian (ed.), Arménie entre Orient et Occident. Trois mille ans de civilization, Paris, 40–61.

212


Mahé, J.-P. (1996b), “Entre Moïse et Mahomet: réflexions sur l’historiographie arménienne”, Revue des Études Arméniennes 23, 121–153. Mahé, J.-P. / Rouhana, P. / Zekiyan, B. L. (eds.) (2010), Saint Grégoire de Narek et la liturgie de l’Église. Colloque international organisé par le Patriarcat Armenien Catholique a l’Universite Saint-Esprit de Kaslik (USEK), Liban, Kaslik. Mahé, J.-P. / Zekiyan, B. L. (eds.) (2006), Saint Grégoire de Narek théologien et mystique, Roma. Manandyan, H. (ed.) (1956), “Hermeay Eṙameci aṙ Askłepios sahmank‘” [Hermes Trismegistos’ Definitions to Asclepius], in: Banber Matenadarani 3, 287–314. Mat‘evosyan, K. A. (1986–1987), “Scriptoria et bibliothèques d’Ani”, in: Revue des Études Arméniennes 20, 209–221. Mat‘evosyan, K. A. (1997), Ani. Ekełec‘akan kyankĕ ev jeṙagrakan žaṙangut‘yunĕ [Ani. Ecclesiastical Life and Manuscript Legacy], Vałaršapat. Mkrtč‘yan, M. (ed.) (1981), Nersēs Šnorhali. Vipasanut‘yun, Erevan. Muradyan, G. (2012a), Grecisms in Ancient Armenian, Louvain. Muradyan, G. (2012b), Grigor Magistrosi Matenagrut‘yun (Matenagirk‘ Hayoc‘. Armenian Classical Authors, 16, 11th Century), Erevan. Muradyan, G. (2013), “Greek Authors and Subject Matters in the Letters of Grigor Magistros”, in: Revue des Études Arméniennes 35, 29–77. Mutafian, C. (2012), L’Armenie du Levant (XIe–XIVe siècle), 2 vols., Paris. Mutafian, C. / van Lauwe, É. (2001), Atlas historique de l’Arménie. Proche-Orient et Sud Caucase du VIIIe siècle av. J.-C. au XXIe siècle, Paris. Pĕltean (Beledian), K. (2008), Erkxosut‘iwn Narekac‘ii het [Dialogue with Narekac‘i], Erevan. Redgate, A. E. (1998), The Armenians, Oxford, 197–254. Russell, J. R. (2004), Armenian and Iranian Studies, Cambridge (Mass.). Samuelian, T. J. (2002), St. Grigor Narekatsi. Speaking with God from the Depths of the Heart, Erevan. Sanjian, A. K. (1993), “Gregory Magistros: An Armenian Hellenist”, in: J. S. Allen / C. P. Ioannides / J. S. Langdon / S. W. Reinert / M. V. Anastos (eds.), Τὸ Ἑλληνικόν. Studies in Honor of Speros Vryonis, Jr. 2, Byzantinoslavica, Armeniaca, Islamica, the Balkans and Modern Greece, New Rochelle-New York, 1993, 131–158. Sanjian, A. K. / Terian, A. (1985–1986), ‘An Enigmatic Letter of Grigor Magistros’, in: Journal of the Society for Armenian Studies 2, 85–95. Settipani, C. (2006), Continuité des élites à Byzance durant les siècles obscurs. Les princes caucasiennes et l’empire du VIe au IXe siècle, Paris. T‘amrazyan, H. (1986), Anania Narekc‘i. Kyank‘ĕ ev matenagrut‘yunĕ [Anania Narekac‘i. Life and Work], Erevan. T‘amrazyan, H. (1999), Narekyan dproc‘ĕ [The School of Narek], Erevan. T‘amrazyan, H. (2004), Grigor Narekac‘in ev norplatonakanut‘yunĕ [Grigor Narekac‘i and Neoplatonism], Erevan. T‘amrazyan, H. (2013), Grigor Narekac‘in ev narekyan dproc‘ĕ [Grigor Narekac‘i and the School of Narek], Erevan. Terian, A. (ed.) (2012), Magnalia Dei. Biblical History in Epic Verse by Grigor Magistros. The First Literary Epic in Medieval Armenian. Critical Edition with Introduction, Translation and Commentary, Louvain. Thomson, R. W. (1995), A Bibliography of Classical Armenian Literature to 1500 AD, Turnhout.


213

Thomson, R. W. (1997), “Armenian Literary Culture Through the Eleventh Century”, in: R. G. Hovannisian (ed.), The Armenian People from Ancient to Modern Times, 1, London, 199–239. Thomson, R. W. (ed.) (2005), Hamam, Commentary on the Book of Proverbs. Edition of the Armenian Text, English Translation, Notes and Introduction, Louvain. Thomson, R. W. (2007), “Supplement to A Bibliography of Classical Armenian Literature to 1500 AD: Publications 1993–2005”, in: Le Muséon 120, 163–223. Tinti, I. (2012), “On the Chronology and Attribution of the Old Armenian Timaeus: A Status Quaestionis and New Perspectives”, in: Egitto e Vicino Oriente 35, 219–282. Uluhogian, G. (2009), Gli Armeni, Bologna. Van Lint, T. M. (2010), “Grigor Magistros, Letters to Amir Ibrahim, and Magnalia Dei, The Mighty Acts of God”, in: D. Thomas / A. Mallet (eds.), Christian-Muslim Relations. A Bibliographical History. Volume 2 (950–1100), Leiden, 703–713. Van Lint, T. M. (2014), “La cultura armena nella visione del mondo di Grigor Magistros Pahlawuni”, in: C. Baffioni / R. B. Finazzi / A. Passoni Dell’Acqua / E. Vergani (eds.), Storia e pensiero religioso nel Vicino Oriente. L’Età Bagratide – Maimonide – Afraate. III Dies Academicus, Milano, 3–22. Van Lint, T. M. (forthcoming), “A Funeral Lament for Sparapet Vahram Pahlawuni. Myth, Folklore, Epic and Remembrance in Grigor Magistros’ Work”. Xač‘erean, L. (1987), Grigor Pahlawuni Magistros (985–1058). Keank‘n u gorcunēut‘iwnĕ [Grigor Pahlawuni Magistros (985–1058). Life and Works], Los Angeles. Zekiyan, B. L. (1999), La spiritualità armena. Il libro della lamentazione di Gregorio di Narek. Roma.

Armenuhi Drost-Abgaryan

The Reception of Eusebius of Caesarea (ca. 264–339) in Armenia Abstract: The contribution deals with the reception of the literary work of Eusebius of Caesarea in classical Armenian historiography (particularly in the formation of the text type of the Chronicon) as well as in the medieval translated literature and the manuscript tradition. In addition, the paper focuses on the current state of Eusebian research in Armenia.

Eusebius of Caesarea, the father of ecclesiastical history and the author of the first Christian Chronicle, was highly revered in Armenia, where his works served Armenian authors as an original source from the fifth until the seventeenth century. He was the author of some forty works on biblical exegesis, dogmatics and apologetics, but his most widely consulted works in Armenia are the Church history and the Chronicle (K‘ronikon or žamanakagrowt‘iwn). Both works were translated into Armenian during the so called Golden Age of Armenian literature, in the fifth century: the ecclesiastical history was translated from the Syriac version and the chronicle from the Greek original, although with traces of Syriac. However, the chronicle is now extant only in the Armenian translation, as the Greek original has been lost¹. These sources were adopted by the Armenian historiographer Łazar P‘arpec‘i (fifth century), Movsēs Xorenac‘i, Asołik (tenth century), Samuēl Anec‘i, Yovhannēs Sarkawag (eleventh century), a philosopher, mathematician and poet of the tenth–eleventh centuries; and additionally, by the thirteenth century historians Yovhannēs Awagerec‘, Kirakos Ganjakec‘i and Vardan Arewelc‘i; the fourteenth century historians Step‘anos Orbelean and Nersēs Palienc‘i or Palianenc‘ as well as David Bališec‘i, an author of the seventeenth century. The German linguist Rüdiger Schmitt reproached Armenologists for neglecting one of the important constitutive texts of Christian Oriental historiography, the Chronicle by the famous church historian Eusebius², which survived only in the old fifth century Armenian translation. I informed the experts at the conference on Julius Africanus, organized by Martin Wallraff (Munich) at the Friedrich Schiller University Jena (Eisenach 2005), of the longstanding research work by the

1 Cf. Thomson 1995, 51–52. 2 Schmitt 2001, 273–281.

216


leading Armenian historian and philologist Gevorg Abgaryan (1920–1998) from the Matenadaran institute for ancient manuscripts in Erevan. About 30 years ago Abgaryan discovered previously unknown traces of the Eusebius transmission in some further Armenian manuscripts and aimed to prepare a new edition of Eusebius’ Chronicle³. In the mid 1980’s, a project was set up, under his direction, to prepare a new edition of the Armenian text based on recent discoveries relating to the chronicle of Eusebius in the manuscripts of the medieval Armenian historians. Simultaneously the GCS series deputed Gevorg Abgaryan to update the German translation of Karst by inclusion of material from the oldest paper-manuscript in the Matenadaran manuscript collection (ms. Maten. 2679, dating from the year 981) containing a compilation of the chronicle. Unfortunately he was not able to bring this intensive task to completion. After his death in 1998, I took on this task. Because of the challenging obligations of a single chair of Christian Oriental Studies in Germany, it has up to now been impossible for me as an official standing alone in this academic field to fulfil the legacy of my father. Accordingly, I am grateful to the organizers of this outstanding conference at the University of Genoa for inviting me and helping me to cultivate this project and resist any temptation to abandon this lifelong endeavour. Below I will shortly introduce the most important new discoveries of the last few decades relating to the manuscript tradition of the Armenian version of this chronicle. I will also report on the results of the previous unpublished work, and outline the state of progress of the new edition. In particular, today I would like to give account of the reception of the literary work of Eusebius of Caesarea not only in classical Armenian historiography (particularly in the development of the Chronicle as a textual form), but also in medieval translated literature, miniature painting and the manuscript tradition.

1 Eusebius in Armenian historiography The Armenian historiographer Kirakos Ganjakec‘i (1201/1203–1271), the famous witness and reporter of the first Mongolian Campaigns in Asia and Europe as well as of information on the language and social life of the Mongols, mentioned the “Great Eusebios” in his History of Armenia among the historians who have left living monuments for the following generations.

3 Drost-Abgarjan 2006, 255–262. This article is recorded electronically in the English translation by Roger Pearse. Some updated excerpts of it are available in the present article as well.

The Reception of Eusebius of Caesarea (ca. 264–339) in Armenia

217

For that is a divine law: The fathers should show (this / the Book of Law, divine words and sayings of old) to the(ir) sons, so that they should make them known (to their children) and that another generation might know them as the prophet David demands⁴. For so instructed Moses, the Greatest among the prophets, the sons of Israel and said: You shall meditate on them day and night when you sit in your house, when you walk by the way, when you lie down, and when you rise up⁵. And to these who help us and extend their hand to us, join the circles of divine men who bequeathed their writings as living monuments for all nations / generations: so the Great Eusebios (mecn Ewsebios), who left two writings: the Chronicle (K‘ronikon), starting with forefather Adam … unto the Advent of Christ and the Church History, starting with the rising of the Sun of Justice … unto the days of pious Constantine⁶.

Historiography is, along with literary translation, the most extended genre of old and medieval Armenian literature. It provides early sources of information on the history of neighboring nations (lost or living), and it offers important insight into ethnic groups such as Arabs, Mongols, Tatars, Seljuk Turks, Caucasian Albans. Furthermore, knowledge on the Crusades can be gained from the Histories and Chronicles of Sebēos and Movsēs Kałankatowac‘i (seventh century), Łewond Vardapet (seventh/eighth century), Yovhannēs Drasxanakertc‘i, Stepannos Asołik, Uxtanēs and Aristakēs Lastivertc‘i (tenth/eleventh century) as well as Mattheos Uṙhayec‘i and Samuēl Anec‘i (twelfth century). The literary models of Armenian historians were the works of Old Greek historiographers and the Christian chronographers, particularly those of Eusebius of Caesarea. In the Armenian manuscript tradition, Eusebius is regarded as an inventor of a new literary text form, that of the “Chronicle” as well as that of “Tetraevangelia” (hamabarbaṙ), i.e. the Synopsis of four Gospels through ten canon tables. This concept of harmonisation of the Gospel was named after him, as the “Eusebian Canon Tables” or “Eusebian Chronographical Canon”. The most frequent epitheta of Eusebius in the manuscript colophons are “The Great Eusebius the Historiographer” (mec Ewsebios Patmagir) and “Eusebius the Chronographer” (Ewsebios žamanakagir). These descriptive terms highlight the importance of Eusebius for Armenians as an expert in Christian universal history and the creator of a new text form, namely the “Chronicle”, which is structured not only as a narrative text but also as a sequential record of events. Thus the focus of this new genre was no longer the historical event in itself but the time axes, the timeline in tabular form.

4 Pss. 77.5–6. 5 Cf. Dt. 6.7, 11.9; Jos. 1.8. 6 Kirakos 1961, 4–5.

218


These denominations are marked by titles “Eusebius the Patriarch” (Ewsebios hayrapet), “Eusebius the Bishop of Caesarea” (Ewsebios episkopos Kesarow) or Eusebius of Cesarea (Ewsebios Kesarac‘î) and “Eusebius of Pamphylos” (Ewsebios Pamp‘iwlay), in honour of his teacher and probably uncle, the Bible commentator Pamphylos. The usual designations “saint” (sowrb) or “blessed” (eraneli) are reserved in the Armenian colophons for his namesakes, the church fathers Eusebius of Emesa († 359), Eusebius of Jerusalem and Eusebius of Alexandria (apparently a fictitious name). Eusebius of Cesarea is, for the Armenians, first and foremost the Great Chronographer⁷. After the annalistic works, the Homily on the Assumption of Christ (Homilia in assumptionem Christi) by Eusebius is the most widely copied work in the Armenian liturgical collections (lectionaries, homiliaries and synaxaries, allocated on the 2nd of May). Furthermore, while the Church History was translated undoubtedly from the Syriac version into Armenian, as Paul Vetter concluded in his contribution “Über die armenische Übersetzung der Kirchengeschichte von Eusebius” (1881), the Chronicle is a translation from the Greek original despite its Syriacisms. The fifth and the sixth books are lacking in the Syriac version of the Church History translated from the Greek original in the fourth century, and consisting of ten books as a whole. These lost chapters survived in the Armenian version published in Venice in 1877. After this edition many other fragments were discovered in the Armenian liturgical manuscripts, leading to the need for a new edition of this Eusebian work. In 1782 the Constantinopolitan Armenologist, Orientalist and translator George Dpir Tēr Yovhannisean (1737–1811), also known with the surname Dpir Palatec‘i, discovered the Armenian version of the Chronicle of Eusebius, which became a useful source for the study of early Christian history in the orient and the west. This ensured that even though the Greek original is lost, the material remained available in a more genuine form to the following generations. George Dpir’s biographer, Ayvazean⁸, relates that George’s functions included liaising between the Constantinople Patriarchate and the Armenian Congregation of Mechitharists in Venice, who had emigrated from Constantinople and united with Rome. During the course of such functions, George happened to make a trip to Šamaxi in eastern Armenia, where, after a lavish meal with his

7 Cf. Thomson 1995, 50. 8 Cf. Ayvazean 1893, 11–73. The German Orientalist Julius Heinrich Petermann speaks about “Lector Georg, a learned Armenian from Constantinople” (Petermann 1865, 457–458).


219

host, with plenty of the local Madras wine, the guest became thirsty during the night. While searching for a water jug, he found a parchment manuscript with a tight leather cover which served in his host’s household as a lid for the water jug. This Codex, via Jerusalem, ultimately arrived in Constantinople and then continued on to Erevan, where it has been held since 1939 in the Erevan manuscripts institute, the Matenadaran, under the shelfmark ms. Maten. 1904. The manuscript became the basis for the edition of Mkrtič‘ Avgerean, 1762–1854 (who is known in Europe under the Latinized form of his name, Baptista Aucher), until recently the only complete publication of the Armenian text⁹. It was copied in 1793 by George Dpir, who has been called the forerunner of Armenian classical studies; the copy was given to the cloister library of the Venetian Mechitharists at San Lazzaro. The manuscript arrived there at Christmas 1794 and was catalogued under the number 931. Adolf von Harnack (1851–1930), the head of the Patristic commission of the Royal Prussian Academy of Science in Berlin, acting through his student Karapet Ter-Mkrtč‘ean (1866–1915), the discoverer and publisher of the Armenian version of the Proof of the Apostolic Preaching of Irenaeus of Lyons, made a photographic reproduction of the Eusebius-model of Aucher. This copy is stored today in the archive of the Berlin Academy of Science. Yovhannes Vardapet Zohrabean or Zohrab (1758–1929), also a Mechitharist and the editor of the famous Zohrab Bible – the first attempt at a critical edition of the Armenian Bible, which appeared in 1805 in Venice – brought the copy prepared in 1793 to Venice. Together with Angelo Mai (1782–1854), an Italian Cardinal and philologist, he translated the Armenian text via Italian into Latin and published it for the first time in 1818 in Milan¹⁰, shortly before the appearance of the Armenian-Latin Edition of Aucher. The Orientalist Julius Heinrich Petermann (1801–1876) prepared a new careful critical Latin translation, which appeared in 1875/1876 in the Berlin Eusebius edition of Alfred Schoene (1836–1918). The German translation of the Armenian version by Josef Karst (1871–1962) came out in 1911 in Leipzig. We will now present some observations on the complete Armenian version of the chronicle, which is found in the above-mentioned ms. Maten. 1904. This manuscript was dated to the twelfth century by Mkrtič‘ Awgerean (Aucher, see above), the editor of the Armenian text, and by the monk and scholar Galust TerMkrtč‘ean (1860–1918) who worked partly in Paris and Munich. It seems fairly

9 Aucher 1818. 10 Mai/Zohrab 1818 and Mai 1833, 1–406 as well as in Patrologia Graeca 19.99–598.

220


certain that the Armenian translation of the chronicle dates from the fifth century and was made directly from the Greek. When Petermann presented a report on the 17 August 1865 in the “Gesamtsitzung der Königlichen Akademie der Wissenschaften zu Berlin” concerning the Armenian manuscripts of the chronicle, this itinérant Codex was known under a quite different name. Petermann himself was unable to access the codex, which he was familiar with under the name of the “Jerusalem Codex”; it “was … highly desirable to examine the codex once again”, we read in his correspondence published in the “Monatsberichten der Berliner Akademie der Wissenschaften”, “this however is still missing, and was probably concealed there in Constantinople out of fear of removal. I too did not succeed in the previous autumn [i.e. 1864] in examining it there”¹¹. Based on this report Josef Karst also wrote in the introduction to his German translation: “The journey of P[etermann] to C[onstantinople] was unsuccessful; confessional jealousy obstructed liberal sight of the Codex”¹². But neither Petermann nor Karst nor others realised that the “Jerusalem Codex” in 1864 was not “still missing” and not concealed “in Constantinople out of fear of removal”, but was in the monastic library at Etschmiadzin¹³. Its description can be found in the Etschmiadzin library catalogue printed at Tiflis in 1863 (not 1865 as given by Petermann and Karst). Here we read: “Codex no. 1684 [now ms. Maten. 1904]: Chronographical history, or Chronicon … written 1695”¹⁴. This date, repeated by Petermann and Karst, is not correct, for, as Theodor Mommsen (1817–1903) also noted, “the statement of the catalogue, that it was written in the year 1144 of the Armenian calendar = 1695 A. D., has reference to n. 1683 [now ms. Maten. 1725], a copy of Eusebius’ Church History, and was transferred by confusion to the following number”¹⁵. Karst himself introduces three Codices in the introduction to his German translation: – “a Jerusalem … which probably dates to 1294–1307 A. D.” (which he designates as G, after the owner of the manuscript Grigor, probably the Cilician Catholicos Grigor Pahlavuni, 1113, or the Catholicos Grigor IV. 1173–1190); – an “Etschmiadzin codex … which is stored at the Etschmiadzin Metropolitan library”(which he designated as E); – a codex produced in 1696 in Tokat which he designates with N (Nersesian).

11 Petermann 1865, 459 ff.; cf. also Schöne/Petermann 1866–1875, 2, XLVIII. 12 Karst 1911, XVI. 13 Others have written on the “concealing” of the manuscript, cf. Rozanov 1881, 16. 14 Mommsen 1895, 322. 15 Mommsen 1895, 321.


221

Karst believes that the codices G (Jerusalem) and E (Ēǰmiacin) derive from an older archetype, while N (Tokat ms.) represents a “partial entirely neglected copy of E”¹⁶. Karst even used the 1898 photographic reproduction ordered by the “Berliner Königlichen Akademie der Wissenschaften” (ms. E). By 1895 Theodor Mommsen had already compared all these manuscripts and had concluded that the chronicle was transmitted via the manuscript E (= ms. Maten. 1904) and that both N (Tokat ms. = Venice, ms. 302) and G (Jerusalem ms. = Venice, ms. 931, the copy made by Dpir in the year 1793) were copied from this¹⁷. Unlike Karst, Mommsen designates by “G” not a “Jerusalem codex”, but rather “the manuscript copy prepared ca. 1790 on behalf of the Venetian Mechitharists. Nominally from Jerusalem, it had been brought to the library of the Armenian seminary in Constantinople”; now in “Venice, in the Mechitharist monastery”, it was “carefully compared there by Petermann”¹⁸. Mommsen noted that the so-called “Jerusalem Codex” (G in Karst) was at that time in Etschmiadzin (today in Erevan) and is the same one that was in Tokat in 1696¹⁹.

16 Karst 1911, X and XIII. According to Karst, manuscript N was acquired in 1856 by father Nerses Sargisean for the Venetian Mechitharist library (p. XIII) and represents a copy of E. According to Mommsen, manuscript E was in Tokat in 1696 (Mommsen 1895, 335). For Mommsen, there was only one ms. (E), of which Georg’s copy (G) and the Tokat copy (N) were merely apographa. For Karst on the other hand, E and G (Jerusalem) are two different mss., independent of one another, which both derive (Karst 1911, XIII) from an earlier archetype (Urtext). 17 Mommsen 1895, 335 f. Ms. no. 931 (G in Mommsen, copied by Lector Georg), in the manuscripts catalogue of the Venice Mechitharists by Čemčemean 1998, is itemized under catalogue number 1548. According to Petermann, a second copy of the chronicle arrived in Venice (probably N), in 1855, no. 302. According to the manuscript itself, it was a gift from the Archbishop of Amid, Minas, who was later the patriarch of Jerusalem, to Archbishop Sahak of Tigranakert, and it was transmitted via the “Superior of the cloister of Johannes of the Täufers, Vardapet Zacharias” through Father Nerses to Venice. The scribe of the manuscript was Michael of Tokat. According to Petermann this manuscript (N in Mommsen and Karst) is similar as regards the number of lines on the page (36), the number of pages (230) and the type of bookhand (minuscule/bolorgir), to the “Jerusalem” Codex (E in Mommsen, E + G in Karst). See Petermann 1865, 461. 18 Mommsen 1895, 321. Mommsen uses the same sigla as Karst (E, N, G); however, he argues that G is not the original Jerusalem codex, which was copied twice by Georg Dpir in 1790 and 1793 (thus Karst 1911, XIV), but rather the co p y of E made by Georg Dpir (according to Karst and Mommsen), the single original manuscript, and sent to Venice (Mommsen 1895, 335). 19 Mommsen 1895, 335. Unfortunately Karst did not know of Mommsen’s valuable investigation. Both Karst and Mommsen believed that N represents a copy of E. But according to Mommsen, E is the manuscript from which G was copied. According to Mommsen, G and N are not merely “closely related” (335). They are two different copies of the same ms. E. Unlike Karst (p. XIII), Mommsen suggests that N is an older and E a more recent copy. In addition, Manandyan 1905 escaped the attention of the author of the German translation.

222


The exciting odyssey of the Armenian chronicle manuscript causes confusion even today. Faulty statements about the manuscripts are also found in recent research. For example, this is shown by the title page of the 1970 monograph by Molly Miller, The Sicilian Colony Dates: Studies in Chronography, which includes a page from the Armenian chronicle-manuscript. The manuscript’s number (now 1904) and the date of the translation (fifth/sixth century) of the chronicle are not in line with the latest research²⁰. This means that the future edition of the Armenian text of the chronicle should be based only on the ms. Maten. 1904. Naturally the critical apparatus should also indicate next to this the complete or fragmentary copies in the different libraries. These codices are also to be used in the future edition of the Armenian text of the chronicle as auxiliary materials. In addition, Aucher’s edition was not correctly assessed by his European colleagues. The distinguished Armenian scholar Aucher/Avgerean, an outstanding connoisseur of ancient Armenian, Latin and Greek, created an edition which endured for a prolonged period of time; it contains simultaneously the original Armenian text, a Latin translation of the Armenian text and the Greek fragments (Syncellus etc). Unfortunately the introduction to the German translation does not express a corresponding appreciation. Josef Karst writes: “the first and previously only original edition, that of Aucher in 1818 … was published, [but] cannot be considered a true critical edition, because it did not consider … the better codex E. As a result of this, the work of Aucher suffers from … unfortunate weaknesses …, so that its translation by the Armenologist Saint-Martin (1791–1832) and later by Niebuhr (1776–1831) must bear the reproach of superficiality and unreliability”²¹. This unfavourable judgment of the Aucher edition is unfair since it does indeed also use codex E (= ms. Maten. No. 1904). As regards the Latin translation, the judgements of Saint-Martin and Niebuhr²² are unconvincing. Niebuhr reviews the edition of Aucher without being able to compare the translation with the text. The Armenologist Saint-Martin, intending to surpass one of the most thorough connoisseurs of the Armenian language, who

20 Miller 1970, 12. Gevorg Abgaryan showed definitively that the Eusebian chronicle in the Armenian version is transmitted through a codex unicus (ms. Maten. 1904, formerly no. 1724 in the cloister library of Etschmiadzin; in Mommsen E, in Karst E + G), reproduced in the copies in Venice from Constantinople (2 copies by Georg Dpir, of which the most exact and diplomatic copy is numbered 931) and Tokat (N). Apparently this codex of Jerusalem arrived in Etschmiadzin after travelling via Tokat, Šamaxi and Constantinople. 21 Karst 1911, XIV. 22 Karst refers to the “attacks” of St.-Martin 1820, 97–112 and Niebuhr 1822 and 1828, 179–304, especially 180 (Karst 1911, XIV).


223

had himself committed various gross sins of grammar, did not recognise that a principle of translation was at stake. The Latin translation of Zohrab-Mai was made from an Italian translation of the Armenian by Zohrab, and polished; i.e. the Italian text was wrapped in a good form of Latin, while for Aucher the language of the source text was decisive. He thus sought to represent the Armenian text as closely as possible, without consideration of the style of the Latinity. A comparison of Aucher edition with the Erevan ms. Maten. 1904 shows only a couple of small lacunae which Karst filled in the German translation. Now let me turn to the promised new discovery. As mentioned above, Gevorg Abgaryan discovered, about 30 years ago, that in the famous oldest paper manuscript of the Matenadaran, whose millennium was celebrated in 1981 (ms. Maten. 2679), there is a compilation from the Eusebius-chronicle. This manuscript of the cloister library of Etschmiadzin was known to German philology as ms. Etschmiadzin 102, through the description in the introduction of the chronicle of Hippolytus, found in Etschmiadzin until the removal of this library to the Erevan Matenadaran. In Erevan, it was then given the inventory number 2679. As well as the chronicle-compilation, this manuscript contains works of Armenian and Greek and Christian Oriental authors and church fathers, as well as Philo of Alexandria, Hippolytus, Eusebius of Caesarea, Epiphanius, Ephrem the Syrian, John Chrysostom, Cyril of Alexandria, Timothy of Alexandria, Gregory of Nyssa, Gregory the Great, Eusebius of Emesa and others. It is not known by whom and when the compilation was assembled. Based on the character of the orthographical mistakes, one can assume that the compiler used an old, and almost indecipherable manuscript. Especially distorted are the names of people and places unknown to the scribe. Nevertheless some passages that are missing in the complete version of the chronicle are found in the compilation. That these passages do genuinely belong to the Eusebian chronicle can be shown by comparison with authentic passages in the Latin translation of Jerome and in the chronicle of Samuēl Anec‘i (twelfth century). For example, Aucher supplements a section of the chronicle missing in ms. 1904 by using the corresponding passages from the Latin translation of Hieronymus²³ and from the chronicle of Samuēl Anec‘i. This passage is also available in the 300 years older compilation in ms. 2679, pp. 201ab-202a (however Josef Karst has not translated this addition of Aucher’s). Therefore, at one and the same time, the gap in the Armenian version can be filled and the authenticity of the corresponding passages in Jerome is proven.

23 E.g. Eus. Chron. 2.62–88 Aucher; cf. Hieron. Chron. 20–33 Helm; Karst 1911, XL.

224


The compilation is not in tabular form, but rather is written consecutively. It begins with the information on the ten Chaldaean kings (in Aucher, 1, 14–16) and ends with the Romans (Aucher, 1, 392), whereby the compiler, excerptor or epitomator simultaneously uses both the first, narrative part of the chronicle and the second, tabular part (Chronicon-canon) for his abridged version. This summary is important to supplement the gaps in the previously known Armenian version and for the correction of some numerals, persons and place names. Josef Karst was obliged to create a hypothesis for the defective reproduction of names occurring in the Armenian translation, which would not have been necessary if he had had the text of the new discovery available to him. In the introduction of his German translation, he wrote: “Also there is an attempt throughout in the transcription of the proper names to make the strange sounding words form in bite-sized pieces, to adjust them to the laws of sound of Armenian, to a certain extent to Armenianize it”²⁴. As an example of the Armenization he adduces the surname dictator, that is represented in Aucher edition as “Dikator” and sees therein a “dissimilation or removal of sound”²⁵. But this example emphasized by Karst is only a mistake in simple transliteration of the name, as it is given correctly (“Diktator”) in the new find. Moreover the form “Dikator” also appears in the Greek in Syncellus and Hesychius²⁶. Furthermore, the new discovery gives us criteria to determine the date when the Armenian version of the chronicle was made, and to address the question of which language was the source from which the Armenian version was translated. Petermann gave the opinion, “that the Armenian translation of the Eusebian chronicle, and consequently also the second part of it … without doubt comes from the 5th century A. D.”. This, he wrote, is supported by “both editors, Zohrab and Aucher, and the evidence itself from the same century. Both scholars agree that the Armenian version was translated directly from the Greek text and represented the original text usually so literally that, wherever this is lost, it can easily be recreated from the Armenian”²⁷. Josef Karst disagreed because he dated the two first authors who quote the Eusebius-chronicle, Movsēs Xorenac‘i and Łazar P‘arpec‘i, to the seventh-eighth centuries. Therefore he was certain that “the age of the Armenian version is not much older than the VIIth century” and supposed the last decades of the sixth

24 Karst 1911, XL. 25 Karst 1911, XL. 26 For this reference I would like to thank Prof. William Adler who approached me after my presentation at the conference and placed at my disposal the “Dikator”-passages in Syncellus: cf. Sync. 284.22; 298.4; 301.22; 310.21 Mosshammer. 27 Petermann 1865, 458 ff.


225

century as a terminus post quem for the origin of the translation²⁸. While there are still some doubts as to the date of Movsēs Xorenac‘i (usually fifth century), no-one today doubts the date of Łazar P‘arpec‘i. This historian lived in the fifth century AD and knew the Armenian translation of the chronicle. As for the Syriacisms, which are especially noticeable in the second part of the chronicle (in the canon), it should not be forgotten that many Syriacisms in the old Armenian translations had become rooted in and a component part of the Armenian language itself. The presence of these cannot be used as a specific feature of the translation of the chronicle²⁹. In my opinion the new, updated Eusebius edition in the GCS series should be presented as a parallel German-Armenian edition, with the updated Armenian critical text of the chronicle and the revised translation of Josef Karst. In the critical apparatus of the Armenian text, the witness of the new discovery and the fragments in other Armenian authors should be included. Gevorg Abgaryan located further traces of the Eusebius transmission in Armenian manuscripts: in the theologian, philosopher and educator Esayi Nč‘ec‘i, chancellor of Glajor-university (thirteenth century, mss. Maten. 1241, 5566), in Vanakan Vardapet (Codices Maten. 3074, 1254) as well as in a further anonymous compilation (ms. Maten. 5254), whose chronicle is ascribed to Philo of Alexandria, and which is identical to that found in ms. 2679, in addition to the new discovery of our Eusebius compilation. This chronicle was partly published in 1929 in Leipzig and in 1944 in Erevan as the work of the seventh-century universal scholar, Ananias of Shirak³⁰. In the meantime Gevorg Abgaryan also determined that the real chronicle of Ananias, which he himself mentioned in his work on Easter and which is also referred to in later sources, is lost. The author of the work published under this name is another historian of the seventh century. This is the third early historian, after Łazar P‘arpec‘i and Movsēs Xorenac‘i, known to have used Eusebius. The chronicle of Eusebius served Armenian authors as an original source from the fifth until the seventeenth century. Sometimes these authors used more complete copies of the chronicle than we have today. The codex Matenadaran 5254, mentioned above for the year 1280, states: “and after him his son Asardon – 8 years”. In the published Armenian version the name of Asardon is missing (Karst 1911, 14). Moreover, the Armenians not only used the tabular form of the Eusebian chronicle and the label “Chronographical canon”, but also, in Cilician-Armenian

28 Karst 1911, XXXVII. 29 See Lyonnet 1950, 111–114; Leloir 1972, 302–304; Ter-Petrosian 1981, 42–48. 30 Bauer/Helm 1929, 394–396 (not in the 2nd ed. 1955); Abrahamyan 1944.

226


of the twelfth century, they placed relevant excerpts from Eusebius at the top of each book of the Old Testament. These Eusebius excerpts should be used in the new complete edition. This was and remains the long-term program of the edition. In the short-term we have decided, together with the editorship council of the GCS series, to publish first of all a German-Armenian edition of the new discovery (the compilation or shorter edition of the chronicle) including the updated introduction and commentaries of Gevorg Abgaryan, in order to disseminate knowledge on the research into the new discovery.

2 The Canon Tables of Eusebius in the Armenian book art: origins of the art form “Khoran” The principle of tabular synchronisation was applied as mentioned above in the genre of Tetraevangelia. In the Armenian manuscript illuminations, the canon tables and the Epistula ad Carpianum (explaining the use of these) are shaped as tempiettos or portals introducing the recipients into a literary sacred space in order to enable them to participate in the Word/Logos of God and his good news as well as his plan of salvation, oikonomia. This new art form, which was highly productive in Armenian miniature painting, experienced a remarkable phase of development in shaping the Eusebian tables. The terminus technicus denominating this new design is Khoran (Xoran). The tradition of naming the Gospels and other liturgical books as a memory book (Hypomnema), a place which preserves remembrance of the good deeds of God and their confession by the first witnesses, clearly highlights the in-depth comprehension of sacred and profane history as a compendium of data to be perpetuated through the generations. Not only did the historians have a responsibility to save this collective memory, but each Christian family was also called upon to continue the preservation of memory by inscribing their own family history and the important events in the life of the family in the family Bibles (dates of births, marriages and deaths). Thus they could partake in the universal history of Christianity and extend it via Eusebian tables. Each family history was considered as a part of the commemoration of the life of Christ³¹. A series of Armenian Commentaries of Canon Tables (Step‘anos Siwnec‘i, Nersēs Šnorhali, Grigor Tat‘ewac‘i, Vanakan Vardapet, Grigor Xlat‘ec‘i) inter-

31 Drost-Abgarjan 2012, 95–110.


227

preted the symbolic and iconographic aspects of the canon-tables, mentioning the meanings of colours, of the floral, animal and anthropomorphic figures and motifs as well as the special symbolism of numbers³². For example, there are many interpretations on the Pythagorean number 10, a total of the numbers one to four, in the Armenian colophons which symbolize the universe in connection with the systematization of Gospels as a unity that promotes the healing and redemption of mankind. Another example is a fine metric colophon in a fourteenth-century New Julfa Gospel created in Tat‘ew monastery (1300 / no. 35), interpreting the symbolic character of illuminations (tempiettos) of Eusebian canon tables. The cock, for example, symbolizes the prophets who announce the light of God, Christ. The intertwined necks of the birds designate the unity of the Old and the New Testaments³³.

3 Codex Ēǰmiacin as a model of the Greek Gospel Book that Eusebios was commissioned to draw up by Emperor Costantine the Great The famous Codex Ēǰmiacin (989, Noravank/Błen, scribe Yovhannēs) with a luxurious ivory cover is an outstanding example of a classical Bible text and the oldest version of an Eusebian synopsis of the Gospels. It belongs to the genre of Tetraevangelia containing the four Gospels for the lections in the Armenian church service. The prototype of this Gospel is presumed to have been composed in the fifth century in the circle of Mesrop and Sahak. This Gospel is not only the oldest Armenian text of a Tetraevangelia attributed to the Holy Translators but also the oldest version of the Greek Prototype of the Gospel that was created by Eusebius in the fourth century by order of Emperor Constantine³⁴. *** The reception of Eusebius in the Armenian tradition has not been systematically studied. A Corpus Eusebianus comparable to the Corpus Nazianzenus in Louvainla-Neuve or the Corpus Athanasianus in Erlangen remains a desideratum.

32 Łazaryan 1995 and 2004. 33 Ter-Awetisean 1970, 54–56. 34 Buschhausen 2001.

228


Bibliography Abrahamyan, A. G. (1944), Anania Širakac‘ow matenagrowt‘yowně [The Written Work of Anania Širakac‘i], Erevan. Aucher (ed.) (1818), Ewsebi Pamp‘ileay Kesarats‘woy zhamanakakan’ erkmasneay [Eusebii Pamphili Caesariensis episcopi Chronicon bipartitum, nunc primum ex Armeniaco texto in Latinum conversum et ad notationibus auctum, Graecis fragmentis exornatum, opera P. Joannis Baptistae Aucher (alias Mkrtič‘ Awgerean) Ancyrani monachi Armeni et doctoris Mechitarista, Venetiis typis coenobii PP. Armenorum in insula Sancti Lazari], 2 vols., Venetiis. Ayvazean, A. (1893), Šar hay kensagrowt‘eanc’ [Series of Armenian Biographies], Constantinople. Bauer, A. / Helm, R. (eds.) (1929), Hippolytus, Werke. 4, Die Chronik, Leipzig. Karst, J. (ed.) (1911), Die Chronik des Eusebius aus dem Armenischen übersetzt, Leipzig. Ganjakec‘i, K. (1961), Patmowt‘iwn Hayoc‘ (ed. Karapet Melik‘-Ohanǰanyan), Erevan. Buschhausen, H. / Buschhausen, H. (2001), Faksimile-Ausgabe des Codex 2374 des MesropMashtots-Matenadaran in Erevan, Kommentarband (Bd. 2) zum Codex Etschmiadzin, Graz. Čemčemean, S. (1998), Mayr C‘owc‘ak hayeren jeragrac‘ Matenadaranin Mxit‘areanc‘ [Main Catalogue of Manuscripts of the Mkhitarean Library], 8, Venice. Drost-Abgarjan, A. (2006), “Ein neuer Fund zur armenischen Version der Eusebios-Chronik”, in: M. Wallraff (ed.), Julius Africanus und die christliche Weltchronistik, Berlin-New York, 255–262. Drost-Abgarjan, A. (2012), “Die Sprache des Gebetes und der Šarakane”, in: R. F. Taft, S. J. (ed.), The Armenian Surb Patarag or Eucharisti Holy Sacrifice. Scholarly Symposium in Honor of the Visit to the Pontifical Oriental Institute, Rome, of His Holiness Karekin II, Supreme Patriarch and Catholicos of All Armenians. May 8, 2008, Roma, 95–110. Helm, R. (ed.) (1913–1926), Eusebius, Werke. 7, Die Chronik des Hieronymos, 2 vols., Leipzig. Łazaryan, V. (ed.) (1995), Xoranneri meknut‘yunner [Commentaries of Xoran-s], Erevan. Łazaryan, V. (ed.) (2004), Meknowt‘iwnk‘ xoranac‘: Hetazotut‘yown ew Bnagrer [Commentaries of Xoran-s. Considerations and Texts], Ēǰmiacin. Leloir, L. (1972), “La version arménienne du Nouveau Testament”, in: K. Aland (ed.), Die alten Übersetzungen des Neuen Testaments, die Kirchenväterzitate und Lektionare, Berlin-New York, 302–304. Lyonnet, S. (1950), “Les origines de la version arménienne et le Diatessaron”, in: Biblica et Orientalia 13, 111–114. Mai, A. (ed.) (1833), Scriptorum veterum nova collectio, e Vaticanis codicibus edita, 8, Romae. Mai, A. / Zohrab, J. (eds.) (1818), Eusebii Pamphili Caesarensis Episcopi Chronicorum canonum libri duos, opus ex Haicano codice a Johanne Zohrabo expressum et castigatum; Angelus Maius et Johannes Zohrabus nunc primum coniunctis curis Latinitate donatum notisque illustratum additis Graecis reliquiis ediderunt, Mediolani. Manandean, Y. (1905), K‘ałuack’ner Evsebiosi Kesarac‘u K‘ronikonic‘: mi hin jeṙagir [Notes on the Chronicle of Eusebius of Caesarea: an Old Manuscript], Wien, 1–15. Miller, M. (1970), The Sicilian Colony Dates. Studies in Chronography, New York. Mommsen, T. (1895), “Die armenischen Handschriften der Chronik des Eusebios”, in: Hermes 30, 322–338.


229

Niebuhr, B. G. (1822), “Über den historischen Gewinn aus der armenischen Übersetzung der Chronik des Eusebius”, in: Abhandlungen der hist.-phil. Classe der königlichen Pr. Akademie der Wissenschaften 1820–1821, repr. in: Kleine historische und philologische Schriften, 1, Bonn 1828, 179–304. Petermann, J. H. (1865), “Über die bis jetzt vorhandenen Texte und Übersetzungen der armenischen Chronik des Eusebios”, in: Monatsberichte der königlichen Akademie der Wissenschaften zu Berlin, 457–462. Rozanov, N. (1881), Evsefij Pamfil, Moskva. Saint Martin, J.-A. (1820), Review of Aucher 1818, in: Journal des Savants, 1820, 97–112. Schmitt, R. (2001), “The Chronicle of Eusebius: a Text Neglected by Armenologists”, in: G. Fiaccadori (ed.), Autori classici in lingue del Vicino e Medio Oriente, Roma, 273–281. Schöne, A. / Petermann, H. (eds.) (1866–1875), Eusebii Chronicorum libri duo, Armenicam versionem Latine factam ad libros manuscriptos ed. H. Petermann; Graeca fragmenta colligit et recognovit, appendices chronographicas sex adjecit A. Schoene, 2 vols., Berolini. Ter-Petrosyan, L. (1981), “Kiwreł Erowsałemac‘oy “Koč‘umn ěncayowt‘ean” erki hayeren t‘argmanut‘ean naxōrinaki harc‘i šurǰ” [On the Prototype of the Armenian Translation of the Κατηχήσεις of Kyrill of Jerusalem], in: Ēǰmiacin 11/12, 42–48. Ter-Awetisean, S. (ed.) (1970), Katalog der armenischen Handschriften in der Bibliothek des Klosters in Neu-Djoulfa, 1, Wien, 54–56. Thomson, R. W. (1995), A Bibliography of Classical Armenian Literature to 1500 AD, Turnhout.

Alessandro Capone

Pseudo-Athanasius, De incarnatione Christi contra Apollinarium: Some Critical Remarks Abstract: The object of this study regards the three sections of De incarnatione Christi contra Apollinarium (= De incarnatione contra Apollinarium liber I [CPG 2231]), in which the Greek text can be compared with the Latin and Armenian versions. In this perspective, firstly, the Greek manuscript tradition will be re-examined and, secondly, the data offered by the ancient translations will be considered, in order to reconstruct, as far as possible, an overall outline of the textual tradition.

1 Introduction Before approaching the specific argument of the present contribution, which aims to discuss three passages of the pseudo-Athanasian De incarnatione Christi contra Apollinarium, which is more known like De incarnatione contra Apollinarium liber I (CPG 2231), it is timely to start by briefly noting some fundamental information¹. The text, handed down together with De salutari epiphania contra Apollinarium (= Contra Apollinarium liber II), which probably is an independent work and by a different author, is transmitted in 28 manuscripts, dated between the tenth and the sixteenth century. A = Ambrosianus I 59 sup. (464), thirteenth-fourteenth centuries B = Basiliensis gr. 32 (A. III. 4), thirteenth century 3 b = Genevensis gr. 29 tom. III (892), sixteenth century F = Florentinus gr. 695 (S. Marco), twelfth century G = Laurentianus Plut. 4,23, tenth century H = Laurentianus Plut. 4,20, twelfth century h (= 74) = Oxoniensis Holkhamicus gr. 28 (olim 51), fourteenth-fifteenth centuries K = Athous Vatopediou 5, fourteenth century L = Londinensis Burneianus 46, eleventh-twelfth centuries

1 For a more detailed analysis on the data here presented see my paper Capone 2015a, 167–184.

232

M N O Q S U V W Y 44 45 46 47 48 50 56 57 76 93

Alessandro Capone

= Marcianus gr. 49 (coll. 351), thirteenth century = Marcianus gr. 50 (coll. 369), eleventh century ex. (ff. 96–415) = Scorialensis X II 11 (371), thirteenth century = Laurentianus Riccardianus 4, fifteenth century = Parisinus Coislinianus gr. 45, twelfth century = Patmiacus A 3, eleventh century = Vaticanus gr. 400, thirteenth-fourteenth centuries² = Athous Vatopediou 7, eleventh century = Mosquensis Synod. gr. 115 (243), fifteenth century = Cantabrigiensis gr. 203 (Trinity College B 9.7), sixteenth century = Amstelodamus gr. 12 (I A 14), a. 1548 = Monacensis gr. 26, a. 1548 = Cantabrigiensis gr. 204 (Trinity College B 9.8), sixteenth century = Vindobonensis theol. gr. 2, fourteenth century = Oxoniensis Roe 29, a. 1410 = Londinensis Harleianus 5579, a. 1320/1 = Parisinus gr. 475, sixteenth century = Marcianus gr. 502 (coll. 804), thirteenth-fourteenth centuries = Vallicellianus 30 (C 4), fourteenth century³

Although the manuscripts of De incarnatione Christi contra Apollinarium have not been collated completely, nevertheless it is now possible to make a selection of the manuscripts which will be used in the present study⁴. Therefore, once the codices descripti have been eliminated, the other manuscripts are useful for the constitutio textus; they can be gathered together in the following way according to Opitz’s reconstruction and the most recent editions of Athanasius’ writings: (M L U W) = W-Sammlung (A B K F 76) = Ausgabe des Doxopatres 2 Text is held in ff. 131.135. 137–151: “Perditis vv. ἵνα καὶ πάθῃ – ἀπιστοῦμεν, ἐκεῖνος (PG 1104B-1108B) ex defectu duorum foliorum post 139v”; see Devreesse 1937, 99 f. 3 F. 419r-v holds only an excerptum of the text: λέγετε ὅτι ἡμεῖς (PG 26.1101C)-θεωρεῖται (PG 26.1104B). 4 De incarnatione Christi contra Apollinarium is handed down mainly in the manuscript tradition of Athanasius’ corpora, on whose composition and reconstruction it remains fundamental Opitz 1935. In recent years Athanasian studies have learned a new flowering time (see www.athanasius.theologie.uni-erlangen.de/): they have specified Opitz’s researches and reconstructed the relationships among the manuscripts of Athanasius’ corpora. On the basis of these researches it is possible to choose some manuscripts in order to discuss the loci critici and to set aside the complete collation of the codices for another moment.

Pseudo-Athanasius, De incarnatione Christi contra Apollinarium

233

These two groups form the so-called x-Tradition, in which De incarnatione Christi contra Apollinarium (= Contra Apollinarium liber I) is written after De salutari epiphania contra Apollinarium (= Contra Apollinarium liber II). (H O S G V) They form the so-called y-Tradition, in which all manuscripts, except V, hand down De incarnatione Christi contra Apollinarium (= Contra Apollinarium liber I) before De salutari epiphania contra Apollinarium (= Contra Apollinarium liber II). This group seems to date back to a more recent phase of the tradition (seventh-eighth centuries) compared with that testified by the x-Tradition. The edition now available for us is that of the Patrologia Graeca (26.1093– 1132), which reprints Montfaucon’s text⁵. He did not re-collate the manuscripts (S B 56 44 b3)⁶ used by Felckmann in his editio princeps⁷, but limited himself to prefer S, while Felckmann chose B. Now, in order to prepare a new critical edition of the text, it will be possible to benefit from some ancient translations too, in addition to the more copious number of manuscripts. A Latin translation can be read in Laurentianus San Marco 584, a manuscript of ninth century which contains a group of texts assembled by one or more dyophysitist theologians, probably of Alexandrine leanings, at the end of fifth or at the beginning of the sixth century⁸. We know also three Syriac translations (the third one is fragmentary) of De incarnatione Christi contra Apollinarium, which can be dated between the sixth and seventh century⁹. Finally in Armenian three excerpts of our text (from chapters 7, 17 and 15, according to the Armenian translation order) are quoted in Sigillum fidei, a group of texts which can be dated to the beginning of the seventh century. This collection is handed down by a manuscript, now untraceable, of the thirteenth century, partly recopied in 1629, and published in 1914¹⁰.

5 Montfaucon 1698, 735–762. 6 Note that recent studies have demonstrated that b3 is a copy of 56 and the latter depends on S, so that in the passages here discussed we will quote only S; it has been also shown that 44 is a copy of N, which is copy of U. 7 Felckmann 1601, 479–507. The correspondences between the names of the manuscripts used by Felckmann and modern sigla are: Seguerianus = S, Basiliensis = B, Goblerianus = 56, Anglicanus = 44 and Anonymous I = b3. 8 See Gain 1994, 368. The Latin translation has been published for the first time by Costa 1991, 490–506, who used only the manuscript Laurentianus. 9 See Thomson 1972a. 10 See Lebon 1929, 5 f. The critical edition of Sigillum fidei is that by Ter Merkertschian 1914. On this collection see briefly Driessen 1953, 95; Lafontaine/Coulie, 93 ff.; Capone 2011, 49 f. and 54.

234

Alessandro Capone

The object of the present study regards just these three sections of De incarnatione Christi contra Apollinarium, in which it is possible to compare the Greek text with the Latin and Armenian versions. In this perspective, firstly, Greek manuscript tradition will be re-examined and, secondly, data offered by the ancient translations will be considered¹¹, in order to reconstruct, as far as possible, an overall outline of the textual tradition. However, it is important to specify that during this preparation phase of the new critical edition only some loci critici will be discussed, for which the preferable interpretation will be proposed each time.

2 Chapter 7 Εἰ μὴ τὸ τοῦ Ἀδὰμ σύμπτωμα εἰς ἀσύγκριτον ἀνάστημα Χριστὸς ἀνεστήσατο ἐν ὁμοιώματι σαρκὸς ἁμαρτίας ὀφθείς, καὶ κατακρίνας τὴν ἁμαρτίαν ἐν τῇ σαρκί· ὥστε καὶ σαρκὶ ἐπὶ γῆς πολιτεύσασθαι, καὶ ἀνεπίδεκτον ἁμαρτίας δεῖξαι τὴν σάρκα, ἣν ἐκ τῆς πρώτης πλάσεως ὁ Ἀδὰμ ἀναμάρτητον ἐσχηκώς, ἐκ παραβάσεως δεκτικὴν ἁμαρτίας πεποίηκε, καὶ εἰς διαφθορὰν καὶ θάνατον καταπέπτωκε (PG 26.1104.34–41). Si non Adae casum ad incomparabilem restauratione Christus restauravit, in similitudine carnis peccati apparens et adiudicans peccatum in carne, quatenus et carne super terram conversari et insusceptibilem peccati ostenderet carnem, quam ex prima plasmatione sine peccato habens Adam, per praevaricationem susceptibilem peccati fecit et in omnimodam corruptionem et mortem deiectus est (Costa 494.168–173). zAdam yanc‘ansn yanhawasareali kangnowmn K‘ristos kangmēac‘ nmanowt’ean marmnoy mełac‘ ereweal, ew datapararteac‘ zmełs i marmni, orpēs zi marmnov i veray erkri aṙak‘inaszi ew anǝndownak mełac‘ c‘owc‘c‘ē zmarminn, zor yaṙaǰin stełcowacin mnmeł ownēr Adam, isk yanc‘anok‘n ǝndownak mełaz arar ew yapakanowt‘iwn ew i mah korcaneac‘ (Sigillum fidei 252.13–19).

Ὁ Ἀδάμ has been handed down in the position of Montfaucon’s text only by the manuscripts of the y-Tradition, while the witnesses of the x-Tradition have ὁ Ἀδάμ after ἐσχηκώς, but the editor does not inform us of this difference. The latter is also the reading that was possibly found in the Greek models of the Latin translation (habens Adam) and of the Armenian version (ownēr Adam). In this case the choice seems difficult, nevertheless it would be preferable to accept the reading handed down by the x-Tradition, with which furthermore the Latin and Armenian versions also agree. On the other hand it is easy to note that, here and in the other textual passages, the two translations follow most accurately the Greek

11 The Syriac text will be quoted according to the English translation of Thomson 1972b.


235

text’s ordo verborum. Therefore, also in this case they offer a further persuasive argument in favor of the text handed down by the manuscripts of the x-Tradition.

3 Chapter 17 Ἐν μὲν γὰρ τῇ ἡμετέρᾳ φύσει καὶ τῆς εὑρέσεως ἡ μάχη ἔτι γίνεται, καὶ τῆς ἐπισπορᾶς ἡ προσχώρησις διὰ τὸ ἡμῶν ἀσθενές· ἡ δὲ τοῦ Λόγου σάρκωσις κατὰ φύσιν Θεοῦ γενομένη, ἀνεπίδεκτος γέγονε τῶν ἐν ἡμῖν ἐκ τῆς παλαιότητος ἔτι πολιτευομένων πραγμάτων· καὶ διὰ τοῦτο διδασκόμεθα ἀπεκδύσασθαι τὸν παλαιὸν καὶ ἐπενδύσασθαι τὸν νέον. […] Καὶ ὅσα μὲν αὐτὸς συνέταξε θελήσας τῇ φύσει, εἰς ἑαυτὸν ἀνεδέξατο, ὅσα ἠθέλησε, τουτέστι γεννήσεως τῆς ἐκ γυναικὸς, αὐξήσεώς τε ἡλικίας, ἐτῶν ἀριθμήσεως, κόπου, καὶ πείνης, καὶ δίψης, καὶ ὕπνου, καὶ λύπης, καὶ θανάτου, καὶ ἀναστάσεως (PG 26.1124.17–24, 27–31). In nostra natura adinventionis pugna adhuc fit et superseminationis processu, propter id quod infirmum est in nobis, Verbi autem incarnatio, Dei secundum naturam facta, insusceptibilis facta est earum, quae in nobis per vetustatem adhuc militant, rerum, et propter hoc docemur expoliare nos vetere et induere novum. […] Et quaecumque quidem ipse constituit naturae sponte, in semet ipso suscepit, quaecumque voluit, id est nativitatem ex muliere, crementum etiam et aetatis annorum numerum, laborem, esuritionem, sitim et somnum et maerorem et mortem et resurrectionem (Costa 502.460–464, 466–470). K‘anzi merov bnowt‘eanb ew gtanelov paterazm ews lini, ew serman tanowmn i jeṙn meroy tkarowt‘ean: isk Banin marmnaworowt‘iwn ǝst bnowt‘ean Astowacowt‘eann ełeal anndownak ełew or i mez hnowt‘eamb sovorakan irk‘n ēin, ew vasn aysorik owsanimk‘ merkanal zhinn ew zgenowl znorn. […] ew ork‘an ink‘n hramaeac‘ bnowt‘eann kamelov yink‘ǝn ǝnkalaw ork‘an kamezaw, ays ē cnndeann, or i knoǰē, ačman hasaki, t‘uoz amaz ew ašxatowt‘ean, ew carawoy ew k‘ałzoy, ew aramowt‘ean ew mahow, ew yarowt‘ean (Sigillum fidei 259.7–12, 14–18).

Προσχώρησις is the reading accepted by the editor on the basis of two manuscripts (56 and b3), the second of which, as has been said, is a copy of the first-one and this is a copy of S, which however has προχώρησις. Nevertheless, Montfaucon did not reconstruct in such a way the relationships between the manuscripts which – it is useful to repeat – he did not re-collate, so that his choice seems to be based on independent manuscripts. Let us add that προσχώρησις is in F and L too, while προχώρησις is in B G Η O V U: it is a matter of two readings which, because of the easy paleographical confusion, cannot be brought back very clearly into the two great families of direct tradition. Therefore, it appears necessary to better understand the context in which these two variants are placed. The sowing which it refers to is that which is ascribed to the enemy in the parable of the darnel (Mt. 13.25). Now, from the context it seems that the author does not want to underline the approach of the second sowing, but the passing, the new operation, the progress of the enemy’s sowing which

236

Alessandro Capone

follows that of the man who sowed good seed in his field. In this perspective it appears preferable to accept the reading προχώρησις than προσχώρησις. Also the Latin (superseminationis processus), Armenian (serman tanowmn) and Syriac (the introduction of the evil seed … the passing of seed …) translations support this choice. But there is another element that is useful to clarify this textual problem. In chapter 10 of De salutari epiphania contra Apollinarium (= Contra Apollinarium liber II) we can read a passage very close to that discussed: Διὰ τοῦτο γὰρ προσῄει ὁ διάβολος τῷ Ἰησοῦ, ὡς ἀνθρώπῳ, μὴ εὑρίσκων δὲ ἐν αὐτῷ τῆς παλαιᾶς αὐτοῦ ἐπισπορᾶς γνώρισμα, μηδὲ τῆς πρὸς τὸ παρὸν ἐπιχειρήσεως προχώρησιν. (PG 26.1148.43-1149.1) Although De salutari epiphania contra Apollinarium is an independent work, as has been said, and the manuscripts offer the variants (προχώρησιν and προσχώρησιν) in this case too¹², nevertheless it is noteworthy that Montfaucon made a different choice in this case from that made in the first passage and he accepted the reading προχώρησιν. Also in this situation the comparison with the Latin version, which has progressum¹³, and with the Armenian translation (ew oč‘ aṙǝnt‘ēr bowṙn harkanelwoyn jeṙnarkowmn), is useful. It is necessary to note that the passage just considered is quoted in the documents of the third ecumenical Constantinopolitan Council (680–681), which testify the reading προχώρησιν: Διὰ τοῦτο γὰρ προσῄει ὁ διάβολος τῷ Ἰησοῦ ὡς ἀνθρώπῳ, μὴ εὑρίσκων δὲ ἐν αὐτῷ τῆς παλαιᾶς αὐτοῦ ἐπισπορᾶς γνώρισμα μήτε τῆς πρὸς τὸ παρὸν ἐπιχειρήσεως προχώρησιν¹⁴. In conclusion, all elements so far considered persuade us to choose the variant προχώρησις in place of προσχώρησις¹⁵. Finally, let us to focus attention on Montfaucon’s text θελήσας τῇ φύσει: the manuscripts offer the reading τῇ φύσει θελήσας, which is supported by the Latin (naturae sponte), Armenian (bnowt‘eann kamelov) and the second Syriac translation (and all the things that he commanded to nature, these by his own will he

12 From the note ad loc. we learn that, as in the previous case, manuscripts 56 e b3 have προσχώρησιν, while the others offer προχώρησιν. 13 We have to consider that two different translators might have translated De salutari epiphania contra Apollinarium (= Contra Apollinarium liber II) and De incarnatione Christi contra Apollinarium (= Contra Apollinarium liber I); see Costa 1991, 461 f., who probably did not fully understand the observations of Altaner 1941, 55 f. Nevertheless, this possibility does not invalidate our argumentation, rather it seems to make it stronger. 14 ACO 2.2.1.240.3-5. 15 Note that the two variants are competing in Pl. Ti. 40c too; see Taylor 1962, 241 f., where it is shown that, on the basis of Cicero’s Latin version (antecessiones) and of Calcidius’ witness (progressus), it is preferable to accept the reading προχώρησις against προσχώρησις, which is though to be the variant handed down by the authoritative codices.


237

took upon himself), but not by the first one (and all the things that he ordered, he willed by nature and took upon himself). In this case too, it appears worthwhile to accept the reading handed down by the manuscripts and confirmed by the ancient versions.

4 Chapter 15 Ὅτε τὸν Ἀδὰμ ἀρχῆθεν ἔπλασεν ὁ Θεὸς, μήτιγε σύμφυτον αὐτῷ δέδωκε τὴν ἁμαρτίαν; τίς οὖν ἔτι χρεία ἦν τῆς ἐντολῆς; πῶς δὲ αὐτὸν κατεδίκασεν ἁμαρτήσαντα; πῶς δὲ καὶ πρὸ τῆς παρακοῆς οὐκ ἐγίνωσκε καλὸν καὶ πονηρὸν ὁ Ἀδάμ; Ὃν ἔπλασεν ὁ Θεὸς ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνι τῆς ἰδίας ἀϊδιότητος, ἐποίησεν αὐτὸν φύσιν ἀναμάρτητον, καὶ θέλησιν αὐτεξούσιον· Φθόνῳ δὲ διαβόλου θάνατος εἰσῆλθεν εἰς τὸν κόσμον, εὑραμένου τῆς παραβάσεως τὴν ἐπίνοιαν (PG 26.1120.21–30). Quando primum plasmavit Adam, Deusne simul exortum ei dedit peccatum? Quae ergo iam necessitas erat mandati? Quomodo autem eum adiudicavit peccantem? Quomodo etiam et ante inoboedentiam non cognoscebat bonum et malum Adam? Quem plasmavit Deus ad incorruptionem et ad imaginem propriae sempiternitatis fecit illum natura impeccabilem et, voluntate, liberi arbitrii: “invidia autem diaboli mors intravit in orbem terrarum”, inveniente trasgressionis excogitationem (Costa 501.401–408). Yayžam zAdam yiskzbann stełc Astowac, mi ardewk‘ tnkakiz et nma zmełs. ew zinč‘ ews pētk‘ ein patowiranin ziard ews zna dataparteaz zmełuc‘ealn, orpēs ew nax k‘an zyanc‘aneln oč‘ čanač‘ēr zbari ew zč‘ar Adam. ayl zor stełc Astowac, yanapakanowt‘iwn ew i patker ǝnk‘eann nmanowt‘ean arar zma bnowt‘iwn anmełowt‘ean ew kams ink‘niǰxanakans: Maxanawk‘ bansarkowin emowt mah yaǰxarhs, gtanelov zyanc‘anac‘n zimastn (Sigillum fidei 285.23–286.2).

Ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνι is the reading of S, wherefrom it is accepted by Montfaucon, and of F V, while the other manuscripts offer ἐπὶ ἀφθαρσίαν καὶ εἰκόνα. Furthermore, Montfaucon notes that this textual section is twice quoted in the documents of the third ecumenical Constantinopolitan Council, where the reading ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνι occurs¹⁶. 16 Cf. ACO 2.2.1.246.17–21: Καὶ προήγαγε δύο χρήσεις, μίαν μὲν τοῦ ἁγίου Ἀθανασίου ἀρχιεπισκόπου Ἀλεξανδρείας τοῦ περὶ ἐνανθρωπήσεως λόγου καὶ κατὰ Ἀπολλιναρίου ἔχουσαν ἐπὶ λέξεως οὕτως· Ὅτε τὸν Ἀδὰμ ἀρχῆθεν ἔπλασεν ὁ θεός, μήτοιγε σύμφυτον αὐτῷ δέδωκε τὴν ἁμαρτίαν; τίς οὖν ἦν ἔτι χρεία τῆς ἐντολῆς; πῶς δὲ αὐτὸν κατεδίκασεν ἁμαρτήσαντα; πῶς δὲ καὶ πρὸ τῆς παρακοῆς οὐκ ἐγίνωσκε καλὸν καὶ πονηρὸν ὁ Ἀδάμ, ὃν ἔπλασεν ὁ θεὸς ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνι τῆς ἰδίας ἀϊδιότητος, ἐποίησεν αὐτὸν φύσιν ἀναμάρτητον καὶ θέλησιν αὐτεξούσιον; 2.2.1.336.4–8: Ἔτι ἀνεγνώσθη ἐκ τοῦ αὐτοῦ κωδικίου χρῆσις τοῦ ἁγίου Ἀθανασίου ἀρχιεπισκόπου Ἀλεξανδρείας ἐκ τοῦ περὶ τῆς ἐνανθρωπήσεως λόγου κατὰ Ἀπολιναριανῶν, οὗ ἡ ἀρχή· Ὁ μὲν τρόπος τοῦ εὐσεβοῦς, ἔχουσα οὕτως· Ὅτε τὸν Ἀδὰν ἀρχῆθεν ἔπλασεν ὁ θεός, μή τί γε σύμφυτον αὐτῷ δέδωκε τὴν ἁμαρτίαν; τίς οὖν ἦν ἔτι χρεία τῆς ἐντολῆς; πῶς δὲ αὐτὸν κατεδίκασεν ἁμαρτήσαντα; πῶς δὲ καὶ πρὸ τῆς παρακοῆς οὐκ ἐγίνωσκε καλὸν καὶ πονηρὸν ὁ Ἀδάμ, ὃν ἔπλασεν

238

Alessandro Capone

There is here a problem of syntax: in the manuscripts S F V and in the indirect tradition, followed by Montfaucon, ἀφθαρσίᾳ καὶ εἰκόνι depends on ἐπί, which in this case governs the two datives, while the Scorialensis (O) offers ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνα and finally the rest of tradition, followed by Felkmann, has the two accusatives governed by ἐπί. The latter reading is testified also by the ancient translations: the Latin one has ad incorruptionem et ad imaginem and the Armenian one offers yanapakanowt‘iwn ew i patker. At this point it is timely to note that in the passage under examination the author draws almost literarily on the text of Sap. 2.23 (Ὁ θεὸς ἔκτισεν τὸν ἄνθρωπον ἐπ’ ἀφθαρσίᾳ καὶ εἰκόνα τῆς ἰδίας ἀϊδιότητος ἐποίησεν αὐτόν). Now, the author of De incarnatione Christi contra Apollinarium quoted the same passage of Sap. just before in chapter 7: Ὅτι ἔκτισεν ὁ Θεὸς τὸν ἄνθρωπον ἐπὶ ἀφθαρσίᾳ, καὶ εἰκόνα τῆς ἰδίας ἀϊδιότητος ἐποίησεν αὐτόν (PG 26.1105.10 ff.), in a context very close to that under examination, where the drawing on Sap. 2.23 is followed by the explicit quotation of Sap. 2.24. Now, in the passage from chapter 7 the manuscripts offer the reading ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνα, but, what it is important to underline is that in the two cases the editor, as has already been seen, opted for different choices¹⁷, probably because he considered the latter a textual quotation and the former a kind of paraphrase of the biblical text, although in neither passages does Montfaucon indicate the reference to Sap. 2.23. In the light of these data and particularly on the basis of the reading of O (ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνα) it is possible to try an interpretation of the passage in question different from that proposed by Montfaucon and more coherent with the quotation of the biblical text. In this sense it appears opportune to keep the dative ἀφθαρσίᾳ as dependent on ἐπί, while the accusative εἰκόνα has to be considered as a predicative of αὐτὸν and as such the following ἀναμάρτητον and αὐτεξούσιον. Therefore, by accepting the reading of O, the passage can be understood in this way: Ὃν ἔπλασεν ὁ Θεὸς ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνα τῆς ἰδίας ἀϊδιότητος ἐποίησεν αὐτὸν φύσιν ἀναμάρτητον καὶ θέλησιν αὐτεξούσιον (God created him in a condition of incorruptibility and he made him as image of his own eternity, by nature sinless and by will possessing free will). On the other hand this interpretation is confirmed by the Latin version: Quem plasmavit Deus ad incorruptionem et ad

ὁ θεὸς ἐπὶ ἀφθαρσίᾳ καὶ εἰκόνι τῆς ἰδίας ἀϊδιότητος; ἐποίησεν αὐτὸν φύσιν ἀναμάρτητον καὶ θέλησιν αὐτεξούσιον. 17 Ad loc. Montfaucon notes: “Seg. Gobl. Felc. 1 et Anglic., ἐπὶ ἀφθαρσίᾳ. Editi vero, ἐπὶ ἀφθαρσίαν. Ibid. Anglic. Gobler. et Felc. 1 καὶ εἰκόνι, male”; Felckmann’s text is in fact ἐπὶ ἀφθαρσίαν καὶ εἰκόνα.


239

imaginem propriae sempiternitatis fecit illum natura impeccabilem et, voluntate, liberi arbitrii¹⁸.

5 Final considerations The shortness of the textual sections analyzed above does not permit of course the drawing of long-range conclusions, nevertheless, on first sight, it is possible to note that the ancient translations (Latin and Armenian) are characterized by a deep literality, which from the stylistic point of view does not offer a specimen of special value, but from the philological point of view represents a contribution of remarkable importance in terms of reconstructing the history of the textual tradition and settling some textual problems. The contribution of the ancient versions highlighted in the present study regards the ordo verborum, the meaning of some words and finally the interpretation of the text. Now, beginning from the re-examination of the direct tradition, the contribution of the versions has been looked over each time, without taking their correctness for granted. This perspective, testified by the studies dealing with the Greek tradition and the ancient translations¹⁹, has two important consequences: 1) the possibility to go back with a greater likelihood, or in a less hypothetical way, to an older stage of the tradition; 2) the possibility to take advantage, in the constitutio textus, of the text handed down directly and of the interpretation offered by the versions²⁰: as far as it is literary, in fact, a translation always presents a minimum of interpretation of the text and this contribution, which has however to be valued with the utmost caution, remains as a most useful support for the philologist who is called to choose in problematic circumstances²¹. Nevertheless, these considerations, even if they have been deduced only from the passages just discussed, would deserve further consideration and investigation which lie outside the scope of this contribution and need a more complex overall study which could open the dialogue and build on the acquisitions of classical philology and of the philology of the Oriental languages.

18 The second Syriac version too would seem to confirm this interpretation: Just as he created (him) incorruptible and as his own eternal image, he made him by nature sinless and with free-will in his essence. 19 See e.g. the studies in Bolognesi 2000 and also Scala 2005; Capone 2008, 69–81. 20 See Capone 2015b, 381-403. 21 See Canfora 2002, 43–46.

240

Alessandro Capone

Bibliography Altaner, B. (1941), “Altlateinische Übersetzungen von Schriften des Athanasius von Alexandreia”, in: Byzantinische Zeitschrift 41, 45–59. Bolognesi, G. (2000), Studi e ricerche sulle antiche traduzioni armene di testi greci, Alessandria. Canfora, L. (2002), Il copista come autore, Palermo. Capone, A. (2008), “De sancta Trinitate Dialogus IV: note critiche alla versione latina”, in: Classica et Christiana 3, 69–81. Capone, A. (ed.) (2011), Pseudo-Atanasio, Dialoghi IV e V sulla santa Trinità, testo greco con trad. italiana, versione latina e armena, Lovanii. Capone, A. (2015a), “Pseudo-Athanasius, De incarnatione contra Apollinarium: Einleitende Bemerkungen zur handschriftlichen Überlieferung”, in: S.-P. Bergjan / B. Gleede / M. Heimgartner (eds.), Apollinarius und seine Folgen, Tübingen, 167–184. Capone, A. (2015b), “Sulla versione latina delle Epistole a Cledonio di Gregorio di Nazianzo”, in: Augustinianum 55, 381–403. Costa, I. (1991), “Opere di Atanasio in una traduzione latina inedita”, in: Atti dell’Accademia Pontaniana n.s. 39, 459–506. Devreesse, R. (1937), Codices Vaticani Graeci (330–603), Città del Vaticano. Driessen, I. W. (1953), “Les recueils manuscrits arméniennes de saint Basile”, in: Le Muséon 66, 65–95. Felckmann, P. (ed.) (1601), Operum sancti patris nostri Athanasii archiepiscopi Alexandrini, tomus secundus, continens ea, quae a diversis interpretibus ex Graeco in Latinum sermonem conversa sunt: inter quae non pauca, quae tam Gr. quam Latine nunc primum eduntur: Adiectis etiam iis quae Graece haberi non potuerunt. Subiecta ad finem veterum de Athanasio Elogia, Fragmenta operum amissorum: tum Varians lectio ex vetust. Mss. Codd. Palat. Basil. Angl. et aliis. Additus index novus, copiosus, triplex, Heidelberg. Gain, B. (1994), Traductions latines de Pères grecs: la collection du manuscrit Laurentianus San Marcus 584, Bern-Berlin. Lafontaine, G. / Coulie, B. (1983), La version arménienne des discours de Grégoire de Nazianze: tradition manuscrite et histoire du texte, Lovanii. Lebon, J. (1929), “Les citations patristiques grecques de «Sceau de la foi»”, in: Revue d’Historie Ecclésiastique 25, 5–32. Montfaucon de, B. (ed.) (1698), Sancti patris nostri Athanasii archiep. Alexandrini Opera omnia quae extant vel quae eius nomine circumferuntur, ad mss. codices Gallicanos, Vaticanos &c. necnon ad Commelinianas lectiones castigata, multis aucta: nova interpretatione, praefationibus, notis, variis lectionibus illustrata: nova sancti doctoris vita, onomastico & copiosissimis indicibus locupletata. Opera & studio monachorum Ordinis S. Benedicti e Congregatione Sancti Mauri, Parisiis. Opitz, H.-G. (1935), Untersuchungen zur Überlieferung der Schriften des Athanasius, BerlinLeipzig 1935. Scala, A. (2005), L’antica traduzione armena della Lettera a Teodoro di Giovanni Crisostomo: ricerche linguistiche e filologiche, Alessandria. Taylor, A. E. (1962), A Commentary in Plato’s Timaeus, Oxford. Thomson, R. W. (ed.) (1972a), Athanasiana Syriaca. 1 (edition), Louvain. Thomson, R. W. (1972b), Athanasiana Syriaca. 2 (translation), Louvain.

Lia Raffaella Cresci

The Hexaemeron of George of Pisidia and the Armenian Version: Textual Investigations* Abstract: The Armenian translation of George of Pisidia’s Hexaemeron, dating back to the seventh–ninth centuries, is much older than the Greek codes which contain the poem (and which generally date back to the thirteenth-fourteenth centuries). This paper analyses a number of excerpts from the Armenian translation, in order to identify elements useful to the reconstruction of the ancient phases of the Greek version of the Hexaemeron.

It is well known, as demonstrated by many of the in-depth contributions published here, that Armenian literary culture played a role of fundamental importance in stimulating the circulation of Byzantine literary works. Indeed, at times the Armenian tradition was instrumental in assuring their very transmission, for instance in cases where no manuscript tradition has come down to us in Greek. Furthermore, even when a literary work has indeed been handed down in Greek, in some cases with a substantial number of Greek witnesses, analysis of the Armenian tradition may be of considerable interest for the insight it can afford from a number of perspectives. Here the focus of attention will concern a work that is of notable significance within the Byzantine literary panorama, because it stands at the intersection of numerous strands of a tradition of poetry that was undergoing an evolutionary process affecting content, style and metre: the Hexaemeron of George of Pisidia¹. This short poem in iambic trimeters, of approximately the length of a tragedy, can be regarded as a blend of at least two genres. On the one hand it reflects the hexaemeral genre properly speaking, which is mainly in prose and which, on the basis of the story in the Scriptures, describes the six days of creation, using as its reference model the Hexaemeron of Basil of Caesarea; on the other, it is modelled on the hymnodic/homiletic genre, with its pattern of praise of God the Creator as found in Psalm 103. The fluctuation between the two

* I wish to thank Professors Moreno Morani and Giusto Traina for their valuable advice and assistance. 1 It is no coincidence that George of Pisidia is regarded by Kazhdan (Kazdhan 1999, esp. 254– 256) as the first Byzantine poet and is identified as the starting point for the analysis by Lauxtermann 2003.

242


genres is mirrored in the two titles transmitted to us by the manuscript tradition: εἰς τὴν ἑξαήμερον² and εἰς τὸ ὡς ἐμεγαλύνθη τὰ ἔργα σου κύριε³. The tradition of this text is characterized by the wealth of witnesses (48 extant manuscripts)⁴, which is unusual for a Byzantine work of poetry. In fact it is remarkable even in comparison with the number of codices that preserve the encomiastic production of George of Pisidia himself⁵. The manner in which the text was published also raises a number of intriguing issues. It was very likely performed on a public occasion, probably also in order to draw benefit from the potential propagandistic appeal of the work in the context of Heraclius’ political leanings⁶. This first public execution was probably followed by a redefinition of the text, carried out by the poet himself, thereby giving rise to two different but chronologically contiguous redactions and ultimately resulting in far-reaching and extensive contamination. Accordingly, the publisher Fabrizio Gonnelli wisely chose not to close the recension but, rather, to seek to identify clusters of manuscripts and, insofar as possible, to draw up the profile of their respective ancestors⁷. Faced with the circumstance that the majority of the manuscripts stem from a fairly late date (mid thirteenth–mid fourteenth century) in comparison to the actual date of composition of the text (seventh century), one can reconstruct a quite early period in which variae lectiones – often adiaphora – became widespread. Thus the existence of an archetype with authorial variants cannot be ruled out. Such a complex tradition, which encompasses not only several distinct anthological collections of George’s poetic works⁸ but also the presence and absence – or different placement – of groups of lines, results in marked fluidity of the text. This circumstance makes the Armenian translation particularly noteworthy. Like its Slavic counterpart, not only does it testify to the fairly considerable and early literary fortune of this short work, and not only is it highly promising with regard to the study of translation techniques⁹, but it also represents a prime source for reconstructing non-negligible elements in the tradition of this text.

2 Attested for ex. in Vat. gr. 121 (thirteenth century). On the possibility that the two forms of the title may be alternative and/or may complement each other, cf. Gonnelli 1998, 87–88, 104, 114–115. 3 Attested for ex. in Roman. Vallic. gr. E 22 (twelfth century) and in Laur. V10 (fourteenth century). Cf. Gonnelli 1990, 411–413. 4 Listed and described in Gonnelli 1998, 17–35 and Tartaglia 1998, 42–43. 5 Listed and described in Tartaglia 1998, 41. 6 Cf. Gonnelli 1995, 113–142; Whitby 1995, 115–129; for a “political” assessment of the literary production of the age of Heraclius, cf. Frendo 1988, 143–156. 7 Gonnelli 1995, 103–108. 8 Gonnelli 1998, 56–58. 9 Classified as forming part of the Hellenising School, although with significant deviations as

The Hexaemeron of George of Pisidia and the Armenian Version

243

In this perspective, the – extremely limited – aim of this study is to make an at least partial contribution to a possible new line of research that could be rewarding for the broader community of experts in this field. It will thus be suggested that scholarly investigations on the text should not be conducted in a sporadic manner, but should instead involve a systematic comparison between the text of the Hexaemeron, of which we finally have a reliable critical edition¹⁰, and the Armenian translation. Two preliminary observations will highlight the limits and risks inherent in a comparison between the two texts, but also the strong potential of the investigation: firstly, the difficulty in reconstructing the original text of the translation, which has come down to us in fairly late manuscripts (thirteenth–fourteenth centuries)¹¹, and secondly, the amazing closeness of the target language to the source language. The technique of translation used in both the Slavic and the Armenian versions involves a procedure closely modeled on the Greek original, not only in terms of lexical calques, but also on the syntactic level. Faithfulness to the Greek text goes so far as to introduce syntactic structures that were non-existent in Armenian, such as the construction orpēs zi with the present infinitive to render the Greek consecutive form ὡς, ὥστε with the infinitive¹². Although this technique – testified not merely by the translation of the Pisidian poem but also elsewhere¹³ – impaired the clarity of meaning and the reception of the text in the target language, it provides an interesting starting point for an enquiry into the configuration of the original text as it was when work was begun on its translation. In effect, the extreme closeness of the Armenian version to the Greek text does to a certain extent make it possible to put forward hypotheses concerning the conditions of George’s text in the Greek antigraph used by the Armenian translator. Naturally, the investigation is subject to very many caveats, due to the possibility that the Armenian reading may be the outcome of a misunderstanding of the Greek antigraph, or the result of uncertainty or genuine errors arising from circum-

regards the rendering of the prefixes: Uluhogian 1957, 362. On the Hellenising School, cf. Mercier 1978; Calzolari, 1989; Coulie 1994–1995, 43–62; Sirinian, 1997, 199–210. 10 Gonnelli 1998, basically taken up again (no differences in textual choices are observable) by Tartaglia 1998. 11 Uluhogian 1959, 22–24 analyses cod. C of the Armenian Church of Cyprus (Cypr. arm. 16), now at Antilias in Lebanon, dated to 1307, and compares it with M (cod. 1104 of the Library of the PP. Mechitarists of Venice) and A 8 (cod. Constantinopolitanus Antonianus). The text of the Armenian translation was published in Venice in 1900 by Tiroyean, who based himself on codices M and A. Cf. also Bolognesi 1958, 19–25; Bolognesi 1969, 236. 12 Cf. Fermeglia 1964, 232; Gonnelli 1998, 38. 13 Cf. the extensive analysis carried out on the translation of the Προγυμνάσματα of Aelius Theon by Bolognesi, 1962.

244


stances typically associated with text transmission (such as misinterpretation of abbreviations, haplography, dittography). This notwithstanding, the hypotheses that will be outlined here may not be devoid of relevance for the tradition of the Hexaemeron, given that the scholars who have carefully studied the Armenian translation, for instance Uluhogian, date the Armenian version to a chronologically rather early period (eight–ninth centuries)¹⁴. Thus the Armenian version documents – indirectly, and with all the above mentioned caveats – a phase of the tradition of the Hexaemeron that is considerably closer to the seventh century than is attested by the tradition that has come down to us from the Greek text. We will put forward a few suggestions for further thought on two factors that seem to have characterized the transmission of the Pisidian text during the early period, i.e. during the eighth–ninth centuries: the presence of numerous explanatory glosses, and the existence of ancient variants. In ll. 48–50 the Greek text has: ἐπεὶ δὲ φεύγεις τοὺς ἐπαίνους ὡς ψόγους, πρὸς τὸν σκοπὸν τὸ νεῦμα συντείνας ἔχω, τοῖς σοῖς πεποιθὼς μυστικοῖς ἀκοντίοις.

This short sequence plays a fundamental role in the narrative framework of the poem, because it signals the closure of the proemial section of the dedication and marks the transition to the declaration of the reference sources, among which the Psalms of David take pride of place¹⁵. L. 50 expresses the poet’s confidence that he will succeed in fulfilling the proposal he has set himself – that of singing the wonders of creation – thanks to encouragement from the poem’s dedicatee, most likely the patriarch Sergius. The term ἀκοντίον clearly has a metaphorical value, but defining its precise meaning is far from simple. The two most recent Italian translations render the line with expressions meaning “trusting in the help of your mystical spur”¹⁶ and “placing faith in your mystical arrows”¹⁷. The Armenian text i k‘oysd yusac‘eal aławt‘ič‘ tēgs can be rendered as “placing trust in your prayerful arrows”. The translator renders μυστικοῖς with aławt‘ič‘, “prayerful”¹⁸. It was suggested that the translator had perhaps read, in the Greek text or as a gloss to the latter, the term παρακλήσεσιν, which has a very wide

14 Uluhogian 1991, 108–109. The vexata quaestio of the (completely hypothetical) identity of the translator is not addressed here. 15 Cf. Geo. Pis. Hex. 51–53. 16 Tartaglia, 1998, 313 (“fidando nell’aiuto del tuo mistico sprone”). 17 Gonnelli 1998, 119 (“confidando nei tuoi mistici dardi”). 18 The Slavic version (povelĕnie) also seems to be traceable back to a similar Greek text: cf. Fermeglia 1964, 248.


245

semantic range encompassing both the value of “prayer” and that of “exhortation, stimulus”¹⁹. In actual fact the two manuscripts of the Hexaemeron, namely Par. Suppl. gr. 139 dating from the fourteenth century, a veritable reservoir of variae lectiones, and Vat. gr. 1871 of the twelfth–thirteenth centuries, show a reading preceded by the abbreviation γρ., the signal of a variant, which replaces μυστικοῖς with εὐκτικοῖς. The Armenian translation, and the Slavic version as well, are thus based not so much on a gloss as, rather, on a fairly ancient variant, which had perhaps arisen with the intention of making the metaphor μυστικοῖς ἀκοντίοις less dense and therefore more easily comprehensible. But even if one were to judge εὐκτικοῖς as a lectio facilior, the Armenian translation would seem suggest that its circulation dates back to a substantially earlier chronological horizon than is attested, for instance, by the Vatican codex²⁰. This is a further reason why it would perhaps be helpful to integrate the critical apparatus of the Gonnelli edition with an indication of the presence of the reading εὐκτικοῖς not only in the manuscripts mentioned, but also in the Armenian version and, additionally, in the Slavic text. In contrast, it may have been a rather more mundane reason that prompted the Armenian translator to have recourse to a gloss at l. 931²¹: ποίους ὁρισμοὺς ἐκμαθὼν ῾Ιπποκράτους

and at l. 934²²: ποῖος Γαληνὸς τῶν ἱεράκων τὴν νόσον.

George is aiming to demonstrate that creations bears within itself a wisdom that originates from the divine action, so much so that animals display spontaneous medical knowledge in treating their ailments, without turning to the doctrine of the great physicians of the Greek tradition as a means of finding the remedies consisting of medicinal herbs. In the case of the mention of Hippocrates, George points out that if a buck is wounded by poisoned arrows, it neutralizes the toxin by eating dittany²³. This subject is extensively attested starting from Aristotle and Theophrastus, and it entered into popular lore via Aelianus; George himself exploited it, manipulating it in such a manner as to convey an anti-pagan impli-

19 Fermeglia 1964, 249. 20 Gonnelli 1998, 95–96 does not observe the coincidence between K P and the Armenian translation. 21 Cf. l. 1539 of the Armenian translation. 22 Cf. l. 1544 of the Armenian translation. 23 Cf. Hex. 931–933.

246


cation, by means of the final polemical mention of Asclepius²⁴, who was even raised to the rank of a god (l. 947): ᾽Ασκληπιοῦ δὲ τοῦ νόθου θεοῦ δίχα.

On the occasions when Hippocrates and Galen are mentioned, the Armenian tradition has only bzisk, “physician”. One may perhaps assume that the translator restricted himself to rendering in Armenian a marginal gloss, and that his intention was to clarify, for a reading public who lacked a basic background in the ancient medical tradition²⁵, the significance that should be associated with the mention of proper names which presumably no longer stood for the quintessence of the art of medicine. But on the other hand, neither can it be ruled out that replacement of the proper name with the indication of the τέχνη in which Hippocrates and Galen excelled can be traced back not to a marginal Greek gloss, but to a sort of Greek-Armenian glossary. The possible existence of such a glossary, as well as its conceivable utilization and structure, call for specific analyses based not merely on the translation of the Hexaemeron. In this sense, any analogies that may be detected in the Armenian rendering of words, constructions or specific iuncturae dating from different cultural contexts and periods could open up investigative paths that would allow the hypothesis or, in the best-case scenario, the reconstruction, at least in its overall form, of a tool of prime importance in the circulation of Greek and Byzantine texts in an Armenian environment. The translation of a gloss has been hypothesized in the Armenian version of l. 172 of the Greek text: ἢ πλῆθος αὐτοῖς ἐξανοίγων ὀμμάτων, ὡς καὶ τὸ σὸν φῶς ἐμφανέστερον βλέπειν, καὶ μὴ λαθεῖν τι τῶν κάτω κινουμένων.

George is describing the creation of the angels, specifying their qualities and forms, one of which is the peculiarity of possessing an extremely high number of eyes to be used for two purposes: contemplation of the divine light, and the possibility of observing anything that is moving on a lower level than that of the angels. The Armenian text renders τῶν κάτω with i storins. The presence of a gloss ἐπὶ γῆς, which in itself is not cogent on the basis of the Armenian translation – since i storins with the value of “lower” can easily render κάτω – could

24 Careful indication of the sources in Tartaglia 1998, 367. 25 Knowledge of which characterises George’s cultural heritage: cf. Bianchi 1965–1966, 137–143; Frendo 1975, 49–56.


247

receive confirmation from the Slavic translation po zemle²⁶. That is to say, this is one of the cases in which the pieces of evidence convalidate one other thanks to the closeness of the Armenian and Slavic translations²⁷. L. 480 ἰσαγγέλους δὲ προξενοῦντας ἀξίας brings to a conclusion a textual sequence found only in a few Greek witnesses²⁸. It consists of ll. 476–480: ὑπὲρ δὲ τὸν σμάραγδον ἐστιλβωμένους, προὔθηκε τοῖς θέλουσιν ὀρφανὰ βρέφη γέροντας ἄλλους καὶ νοσοῦντας καὶ ξένους, οὐ πορφύραν καὶ βύσσον ἐστολισμένους ἰσαγγέλους δὲ προξενοῦντας ἀξίας.

These lines, together with ll. 472–473, constitute one of the notable features of part of the tradition and they are ordered in a distinctive manner that sets them apart from any other grouping of lines²⁹. They are linked by strong text-internal coherence as they extol the value that the Creator awards to orphans, the poor, the aged, or foreigners – in other words, to all those who should be looked after by the Orphanage or the Care Home. The insertion of these lines into the narrative framework of the short poem may have been due to the author himself on the occasion of his appointment to the position of ὀρφανοτρόφος (head of the orphanage of Costantinople), and the lines constitute one of the pieces of evidence in favor of a second textual redaction traceable back to George himself³⁰. The text draws an analogy between the value of gemstones such as emerald or the splendor of crimson robes or pure white fine linen vestments, and the above mentioned categories of the most miserable beings who do not wear brilliant raiments but who “procure honours equal to those of the angels” in Gonnelli’s translation³¹ or “who lead to the same dignities as those of the angels” according to Tartaglia’s interpretation³². The Armenian version has ayl hawasar hrestakac‘ patwoy p‘aylic‘s, “but shining with honour no less than the angels”, in which ayl hawasar makes it possible to infer that the antigraph presented the reading ἰσαγγέλους of the majority of the witnesses rather than εἰς ἀγγέλους of mss. Vat. Gr. 166 of the fourteenth

26 Fermeglia 1964, 255. 27 Fermeglia 1964, 231–236, 268. 28 B (Oxon. Bodl. Clark), D (Paris. gr. 1302), I (Paris. gr. 1277), J (Berol. gr. Qu 49), Q (Vat. gr. 166), Smg (Roman. Vallic. gr. E 22), At (Athen. n. 329), Mr (Marc. gr. 620): cf. Gonnelli 1998, 148. 29 Gonnelli 1998, 48, 88. 30 Gonnelli 1998, 106. 31 Gonnelli 1998, 149 (“ma procurano onori pari a quelli degli angeli”). 32 Tartaglia 1998, 341 (“ma che introducono alle stesse dignità degli angeli”).

248


century and Roman. Vallic. Br. E 22 of the twelfth century (a typical itacistic error). But the rendering of προξενοῦντας by p‘aylic‘s (p‘ayl means “bright, shining”) seems to suggest that the translator had a Greek gloss he intended to explain, and furthermore, that he wanted to simplify the meaning of προξενέω, “to procure” which, in l. 480, clearly takes up again the term ξένους at the end of l. 478. The latter, including foreigners, prove to be capable of being the πρόξενοι who allow comparison with the angels, since those who are looking after them will have the honours that are reserved to angels. The play on words, which no doubt was challenging for the Greek reader too, may have led to the need for an explanation consisting of a gloss such as προφαίνοντας³³, which simplified the dense meaning of the verb προξενέω, awarding predominance to the value of “showing”. But προφαίνω also has the meaning of “making [something] shine”³⁴, to which the translator has given priority, very probably in relation to light, which since time immemorial has been associated with the angels. Ll. 1399–1400: ἡμεῖς γὰρ ἴσμεν ὡς ἐναλλαγὴν δέον τὸ θνητὸν ἡμῶν τοῦτο προσλαβεῖν δέρας

constitute the gloss of the long-drawn-out polemic against Mani, who was accused of not believing in the resurrection of the body. They represent the beginning of the definition of orthodox belief on this question. It should be noted that the word δέρας, a poetic and rather infrequent form for δέρμα³⁵, may have been somewhat difficult to understand. The Armenian translation³⁶ renders the end of l. 1400 with ayl yark, “other roof”: one may hypothesise that here ayl has an adverbial value corresponding to the Greek ἄλλως, which introduces a gloss or a variant, and that yark, “roof”, but also “tent”, is the inappropriate rendering of a term like σκῆνος, which has the twofold value of “tent” but also of “temporary dwelling”. This second meaning was designed to call to mind the passage from St. Paul (2 Cor. 5.1) which starts out with an οἴδαμεν γὰρ ὅτι³⁷ that clearly echoes George’s Atticistic beginning: ἡμεῖς γὰρ ἴσμεν. The scriptural and patristic value³⁸ of σκῆνος, which – above all in relation to the Pauline occurrence – likens the

33 Fermeglia 1964, 291–292. 34 LSJ, s.v. προφαίνω III. 35 Eur. Med. 480 (varia lectio), Bacch. 835. 36 Ll. 1142–1143. 37 Οἴδαμεν γὰρ ὅτι ἐὰν ἡ ἐπίγειος ἡμῶν οἰκία τοῦ σκήνους καταλυθῇ … 38 Cf. Clem. Al. Strom. 4.26: τὴν σάρκα τὸ σκῆνος τῆς ψυχῆς; Meth. Res. 1.53 τὸ ἑκάστου σκῆνος ἤτοι τὸ σῶμα.


249

body to a tent, i.e. the temporary abode of the soul, and which lay at the origin of the glossator’s intention, was not grasped by the translator. Instead, the latter opted for a word like yark, which was fairly distant from the Greek δέρας and was incapable of rendering the twofold semantic value of σκῆνος. Thus it is highly probable that the manuscript on which the Armenian translation was based also included an intricate apparatus of glosses designed to assist the reader in making use of and understanding a text that had a highly complex structure, with a dense web of often interwoven metaphors and rich in scriptural references that were not always made explicit. Accordingly, the content was only fully comprehensible to those who had an in-depth knowledge of patristics and of the literary, scientific and mythic heritage of Hellenic antiquity. In some cases, the Armenian translator, faced with the difficulty of penetrating the meaning of specific passages of the Greek text or of identifying the exact σῆμα of certain words, preferred to resort to the gloss, and not merely for exegetic reasons, but even for the actual purpose of translating: in other words, he translated the gloss rather than the Greek text. Thus as early as the seventh–ninth centuries, the text of the Hexaemeron had already given rise to the prolonged exegetic activity that was to persist for many centuries and which is testified by an authoritative witness such as the manuscript Paris. Suppl. gr. 139 of the fourteenth century³⁹, characterized by a wealth of glosses and variae lectiones (some of which may be authorial variants)⁴⁰, which were already present in the manuscript used as the model for the Armenian translation. The very likelihood that the Armenian translation constitutes evidence, albeit indirectly, of the presence of valuable readings already present in a fairly ancient phase of the tradition of the Hexaemeron represents a rather promising starting point for further analysis. Here we will limit the enquiry to an indication of some of the passages which deserve in-depth examination, inasmuch as they either document the care and attention devoted to this text in the Byzantine era by figures of great cultural prestige, or make it possible to put forward an at least approximate date for some of the variants already in circulation. The Codex Ottobonianus 342 (thirteenth–fourteenth centuries) contains, on f. 186 in the right-hand margin, a note explicitly attributed to John Tzetzes: τζέτζου. μηδέν τι μηδεὶς ἐνθαδὶ παραξέσει. / χρῆται γὰρ οὗτος καὶ ποσὶ τρισυλλάβοις. With this iambic couplet distico Tzetzes, adopting his usual brusque and icastic

39 Gonnelli 1998, 27, 52–53. 40 Pertusi 1959, 61.

250


style⁴¹, intended to issue a severe warning against any attempt to eliminate, in l. 40 (σπείρεις γὰρ αὐτὸς καὶ κατὰ πέτρας πολλάκις) the correction καί which introduces a trisyllabic foot in the fourth position and therefore does not allow the line to scan as a dodecasyllable. In effect the tradition overwhelmingly does not transmit καί, and this implies that the first α of κατά has to be considered as long. Tzetzes may have made the correction on the basis of his metrical skill and his knowledge of the Pisidian trimeter, in which trisyllabic substitutions of the iamb were still frequent⁴², but it is more likely that he found the reading καὶ κατά in some ancient codex, on the basis of which he opposed the normalising tendency – I would say, the anachronistically normalising tendency – of the copyists and grammarians of his day. Hence his peremptory and polemical tone, testified also in some of the marginal notes on the text of Thucydides transmitted by cod. Heidelberg. Palat. gr. 252, especially in the cases where Tzetzes noticed that the readings found in ancient codices were disregarded in favour of normalizing trends⁴³. Now, apart from the correction in the Ottobonianus, and in the absence of close examination of this specific point of Paris. Gr. 1302, the testimony of the Armenian translation, which effectively has ew, “and”⁴⁴, becomes of considerable interest. Therefore the codex utilized as a model by the Armenian translator had the correct reading, which has completely disappeared from the Greek tradition that has come down to us and would thus be unknown, were it not for the note by Tzetzes, which we may assume to have made use of a witness, now lost, which transmitted the correct reading. That the tradition of the text of the Hexaemeron raised doubts not only among the Byzantine γραμματικοί but also among a cultured public interested in studying the work is testified by a letter in which Constantine Acropolites, in the thirteenth century, asked a correspondent of his to send him a copy of the Hexaemeron so that he could annotate its readings and correct the text he had at hand, which was marred by a wide array of mistakes⁴⁵. Within this notably contaminated state of the tradition one finds a number of variae lectiones which, whenever they are accredited by the Armenian version, deserve careful examination, since they may well reflect an early state of the tradition of the Pisidian text. They may have been ancient variants or even authorial variants, given the possibility of the existence of more than one redaction traceable to George.

41 Luzzatto 1999, 27–28, 31–34, 38, 39–41. 42 Romano 1985, 1–22. 43 Luzzatto 1999, 49–59. 44 Gonnelli 1998, 49–50. 45 Ep. 78 Romano. Gonnelli 1998, 37, 50–51.


251

Here we will restrict the question to mentioning a few that may be of interest for analysis by scholars. Ll. 46–47: ἴσως γένοιτο τέκνον ἐκ στείρας πάλιν καὶ καρπὸς ἔμφρων ἐξ ἀκάρπου καρδίας.

In l. 47 some parts of the tradition, among which the Armenian translation, has the variant εὔφρων instead of ἔμφρων⁴⁶. The context is that of a prayer addressed to the dedicatee, containing the plea that by the support of his tears he irrigate the arid and barren land represented by the soul and the mind of the poet, in order that it may miraculously produce the fruit of the description of the wonders of creation. The fertilizing action of the patriarch Sergius’ tears is mentioned in connection with a reference to the miraculous conception by a barren woman. The reference, pointed out by Gonnelli⁴⁷, could be to Luke 1.36–37, i.e. to Elisabeth and her son John, and the use of the term ἔμφρων for the fruit sprung from a sterile land, i.e. from a barren heart, may be a reference to the wisdom necessary to sing the creative action of God. However, it cannot be excluded that the variant εὔφρων, shown by the Armenian tradition to be ancient, is not the outcome of a transcription error, but rather that it intends to underline the joyous action of the desired fruit, i.e. of the song itself, which aims to sing the wonders of the divine ἔργα. Seen in this perspective, the variant would be continuing to develop the theme of the barren mother’s joy at the fruit of conception, according to a scriptural typology which, before Elisabeth, involves figures such as Sarah, the mother of Isaac, and Anna, the mother of Samuel⁴⁸. In ll. 129–130 George seeks to explain why the pagans invented the myth of Atlas holding up the sky, amazed as they were at the lofty height of the sky itself and incapable of understanding how it could support itself unless it was held up by invisible columns. ἴσως τὰ μακρὰ τῆς ἀδηλίας βάθη στύλων ἀβάθρων ἐκτυποῦντες ἐμφάσεις

“perhaps imagining the great abysses of darkness in the appearance of columns lacking a base”⁴⁹. A minority strand of the text tradition, represented by the codex

46 Gonnelli 1998, 118. 47 Gonnelli 1998, 119. 48 1 Sam 2.1. 49 Gonnelli 1998, 125 (“forse immaginando i grandi abissi dell’oscurità nell’aspetto di grandi colonne senza base”); Tartaglia 1998, 319: “perhaps seeking to portray, through the image of

252


Paris. Gr. 1302 of the thirteenth century and by the Armenian tradition anyayt, has the reading ἀδήλων instead of ἀβάθρων. The reading ἀδήλων should be considered carefully, since its textual closeness to ἀδηλίας of the previous line is consistent with George’s usus scribendi, as he does indulge in anaphora of words having the same stem⁵⁰ and shows a particular predilection for the broad semantic range of the noun ἀδηλία and the adjective ἄδηλος⁵¹. The existence and persistence, in the tradition, of this passage containing the epithet ἄδηλος linked to στῦλος have left a trace in mss. Vatic. Ottob. Gr. 324 of the fourteenth century and Roman. Vallic. Gr. E 22 of the thirteenth century, both of which add a line after l. 130, namely στύλοις ἀδήλοις ἐκτυποῦντες πλασμάτων and στύλοις ἀδήλοις ἐκτυποῦντες ἐμφάσει, respectively⁵². The element George emphasises is precisely the invisibility of the columns that support the sky, which gave rise to the myth of Atlas. L. 1455 forms part of a complex explanation of the hierarchies of angels, and specifically of the Cherubim, endowed with innumerable eyes, intent on the contemplation of God and compelled to squint as a result of the radiance itself. The dualism between the joy at the sight of fire and the need to avert one’s eyes when drawing close to the fire, a common experience familiar to many a person, may provide the analogic basis for understanding the angels’ actions and reactions. L. 1455 is transmitted as follows in most of the tradition: εἰ γὰρ τὸ τῆς γῆς τοῦτο πῦρ ὁ προσβλέπων.

In contrast, a different version is handed down by codices Paris. Gr. 1302 of the thirteenth century, Paris. Gr. 1277, dating from the same period, and Berol. Gr. Qu 49 of the fourteenth century and, additionally, by the Armenian translation, namely: εἰ γάρ⁵³ τις ἡμῶν πρὸς τὸ πῦρ ἀποβλέπων⁵⁴.

The replacement of the connotation of fire as terrestrial by the syntagm τις ἡμῶν introduces an analogical comparison whereby, instead of examining the celestial

pillars lacking a base, the enormous depth of such an obscure and unknown element” (“volendo forse raffigurare nell’immagine di pilastri privi di base l’enorme profondità di un elemento tanto oscuro e sconosciuto”). 50 Purely as an example: Hex. 146–147: κόλπος δὲ παντὸς γίνεται κινουμένου, / φθάνει δὲ πᾶν κίνημα συγκινῶν ὅλα. 51 For a list of Armenian lexical calques from ἄδηλος and ἀδηλία cf. Fermeglia 1964, 315. 52 Gonnelli 1998, 124. 53 Omitted in Paris. Gr. 1302. 54 Berol. Gr. Qu 49 has ἀποβλέπει.


253

condition in relation to the earthly situation, the experience of humans is compared with that of the angels. This is a feature that characterizes and is constantly shared by the above cited group of codices and the Armenian translation, thus allowing reconstruction of an ancestor a, rich in variae lectiones, which, precisely because of the early age of the Armenian translation, should be assigned to the eighth century⁵⁵. It would be beyond the scope of available space in this paper to address the complex problem of the transmission, position and ordering of individual lines or of whole sections of the text. But what is clear is that it provides further evidence of the fluidity of the text as early as the eighth–ninth centuries. We need only cite the group of lines devoted to the catoblepa⁵⁶, noting that here too the Armenian translation makes a crucial contribution to defining ancestor a. Thus the testimony of the Armenian translation allows the conjecture that a series of variants, some of which may be the outcome of the multiple redactions of the short work, date back to a rather early age of the tradition of the Hexaemeron. In effect, precisely on account of the propagandistic elements that were among its characteristics, this composition was executed several times; consequently it underwent a number of manipulations, which to some extent arose from personal circumstances involving the author. In addition, the possibility of hypothesizing the presence of a rich apparatus of glosses in the antigraph of the Armenian translation is an element of prime interest in reconstructing the history of the comprehension and utilization of a text that displays such a dense and challenging form of expressive and stylistic configuration.

Bibliography Bianchi, G. (1965–1966), “Note sulla cultura a Bisanzio all’inizio del VII secolo in rapporto all’Hexaemeron di Giorgio di Pisidia”, in: Rivista di Studi Bizantini e Neoellenici 2–3, 137–143. Bolognesi, G. (1958), “Sulla traduzione armena dello HEXAEMERON di Giorgio Pisida”, in: Atti del Sodalizio Glottologico Milanese 11, 19–25. Bolognesi, G. (1962), La traduzione armena dei Progymnasmata di Elio Teone, Roma. Bolognesi, G. (1969), “Traduzioni armene di testi greci. Problemi di critica testuale e di interpretazione linguistica”, in: Studia classica et orientalia Antonino Pagliaro oblata, Roma, 1, 219–291.

55 Gonnelli 1998, 77–80. 56 Uluhogian 1991, 91–109; Gonnelli 1991, 355–363.

254


Calzolari, V. (1989), “L’école hellénisante”, in: Mikael Nichanian, Ages et usages de la langue arménienne, Paris, 110–130. Coulie, B. (1994–1995),“Style et traduction: réflexions sur les versions arméniennes de textes grecques”, in: Revue des Études Armeniennes 25, 43–62. Tiroyean, A. (ed.) (1900), Vecoreayk‘ Georgay Pisideay imastasiri, Venezia. Fermeglia, G. (1964), “Studi sul testo delle due versioni (slava ed armena) dello HEXAEMERON di Giorgio Pisida”, in: Memorie dell’Istituto Lombardo, Accademia di Scienze e Lettere, classe di Lettere 28.2, 227–333. Frendo, J. D. (1975), “Special Aspects of the Use of Medical Vocabulary in the Poems of George of Pisidia”, in: Orpheus 22, 49–56. Frendo, J. D. (1988), “History and Panegyric in the Age of Heraclius: the Literary Background to the Composition of the Histories of Theophylact Simocatta”, in: Dumbarton Oaks Papers 42, 143–156. Gonnelli, F. (1990), “Le parole del cosmo: osservazioni sull’Hexaemeron di Giorgio Pisida”, in: Byzantinische Zeitschrift 83, 411–422. Gonnelli, F. (1991), “Notizia sul catoblepa (Giorgio di Pisidia, Hexaemeron v. 959 ss.)”, in: Atti dell’Accademia Pontaniana n.s. 40, 355–363. Gonnelli, F. (1995), “Sulla datazione dell’Hexaemeron di Giorgio di Pisidia”, in: U. Criscuolo / R. Maisano (eds.), La poesia bizantina, Napoli, 113–142. Gonnelli, F. (ed.) (1998), Giorgio di Pisidia, Hexaemeron, Pisa. Kazhdan, A. P. (1999), A History of Byzantine Literature (650–850), Athens. Lauxtermann, M. (2003), Byzantine Poetry from Pisides to Geometres: Texts and Contexts, 1, Wien. Luzzatto, M. J. 1999, Tzetzes lettore di Tucidide. Note autografe sul Codice Heidelberg Palatino Greco 252, Bari. Mercier, C. (1978), “L’école hellénistique dans la littérature arménienne”, in: Revue des Études Arméniennes 15, 59–75 Pertusi, A. (ed.) (1959), Giorgio di Pisidia, Poemi. 1, Panegirici epici, Ettal. Romano, R. (1985), “Teoria e prassi della versificazione. Il dodecasillabo nei Panegirici epici di Giorgio di Pisidia”, in: Byzantinische Zeitschrift 78, 196–201. Sirinian, A. (1997), “La traduzione dei composti verbali greci nelle versioni armene delle orazioni di Gregorio di Nazianzo e nelle Regole di Basilio di Cesarea”, in: A. Valvo (ed.), La diffusione dell’eredità classica nell’età tardoantica e medievale, Alessandria, 199–210. Tartaglia, L. (ed.) (1998), Carmi di Giorgio di Pisidia, Torino. Uluhogian, G. (1962), “Contributi allo studio della traduzione armena dell’Hexaemeron di Giorgio di Pisidia”, in: Atti del Sodalizio Glottologico Milanese 11, 19–27. Uluhogian, G. (1991), “In margine alla versione armena dello Hexaemeron di Giorgio di Pisidia”, in: Rivista di Bizantinistica 1, 91–109. Whitby, M. (1995), “The Devil in Disguise: the End of George of Pisidia’s Hexaemeron reconsidered”, in: Journal of Hellenic Studies 115, 115–129.

IV. Philological, Linguistic and Lexical Problems

Giulia D’Alessandro and Lara Pagani

Homer in the Armenian Tradition* Abstract: 1) Contrary to the claims of a tradition which arose from the cataloguing of the content of ms. Paris, BNF, 260, fol. 32v–147v by the abbot of Villefroy (1739), the existence of an ancient Armenian translation of the Homeric poems is not attested. 2) The mythical material of the Trojan cycle can be detected, with relevant differences compared to the mainstream version, in a narrative in Armenian on the myth of the Trojan War and the founding of Rome preserved in ms. Erevan, Matenadaran, 437. An Appendix, by Chiara Aimi, is dedicated to ms. Erevan, Matenadaran, 437.

1 Introduction Scholars preparing to engage in research on Homer from any perspective – interpretive, philological, linguistic-expressive, or in relation to the history of tradition – are accustomed to finding themselves facing a virtually boundless array of data, material and bibliography. This is quite natural given that the subject of their investigation represents the pivotal focus of centuries and centuries of Greek literature and, more broadly, culture: moreover the literary heritage of the whole of the western world throughout the subsequent eras has been inescapably affected by the enduring presence of the Homeric poems. In contrast, a radically different situation is encountered when one approaches the situation in ancient Armenian literature. The latter, characterized as it was by the phenomenon of the translation of Greek literary works – which played a major role in the germinal phases of Armenian literature¹ – seems nevertheless to have reserved a relatively marginal role to Homeric poetry. Yet there is nothing surprising about this apparently unprecedented scenario, if it is viewed in the proper perspective. The early history of Armenian literature in the fifth century was marked by translation of the Bible and also by the rendering of other religious works, such as those of the

* The text of §§ 1 and 2 was drawn up by Lara Pagani (within the project “Omero, Esiodo, Pindaro, Eschilo: forme e trasmissione dell’esegesi antica”, in the program FIRB – Futuro in Ricerca 2012 of the Italian Ministero dell’Istruzione, dell’Università, della Ricerca), that of § 3 by Giulia D’Alessandro. 1 See e.g. the contributions by M. Morani, V. Calzolari and T. M. van Lint (esp. § 1) in this volume.

258


Fathers of the Church. This activity was prompted by the aspiration to acquire accurate knowledge of Christian beliefs. Alongside this range of texts, attention also turned, not only but principally, to a typology of texts that would provide linguistic-rhetorical, logical-philosophical, medical-scientific and historiographic-chronological tools², which involved literary spheres that had little in common with the Homeric poems. To this should be added that Armenian culture possessed an epic oral poetry of its own, of which we have knowledge through the testimony of the historiographer Moses of Khoren³. Thus the scanty presence of Homer is readily understandable, however much this may disconcert scholars of Ancient Greek. When we affirm that the Homeric poems had a secondary role in ancient Armenian literature, what we have in mind is that they did not receive an independent translation, at least as far as can be determined from current knowledge of the situation, to which we will return below. On the other hand, this by no means implies the mistaken idea that Homer was unknown in the Armenian world of antiquity⁴. On the contrary, it is quite possible that Homer was read in the original, at least by an educated circle; furthermore, one need only think of the mass of Homeric citations and references Armenian scholars must surely have encountered in the Greek works they translated: for instance, in the Techne grammatike attributed to Dionysius Thrax (for linguistic reasons), in Aelius Theon (for rhetorical observations), in historical or para-historical works such as The Alexander Romance by Pseudo-Callisthenes, and so forth⁵. Thus knowledge of the facts and characters of the stories in the Homeric poems was undoubtedly widespread in educated circles of ancient Armenia: for instance, the above-mentioned Moses of Khoren deals with questions involving the Trojan war and proposes a novel account in which the events of Troy intersect with Armenian history. Moses’ version features an Armenian chieftain who conducts an Ethiopic contingent that has arrived in Troy as a reinforcement and who is subsequently killed by Achilles – a sort of re-elaboration in an Armenian perspective of the events of the cyclic poem known as the Aethiopis (MX 1.32). There also exist hitherto unpublished Armenian texts referring to the poems in various ways, which are briefly men-

2 In addition to the contributions in this volume, see e.g. Hairapetian 1995; Bolognesi 2000, 1–28; Uluhogian 2009, 113–118. 3 Hairapetian 1995, 37 ff.; Uluhogian 2009, 119. 4 On the methodological error of assuming that an author was unknown in the Armenian context because no translations of the author are documented, cf., in this volume, the paper by V. Calzolari, esp. § 2. 5 Cowe 2010, esp. 3–6.

Homer in the Armenian Tradition

259

tioned by Peter Cowe in a recent article; they are transmitted in fifteenth and seventeenth-century manuscripts, but their original date is unknown⁶. What is certain is that as far as we know to date, the first recognized Armenian version of the Iliad dates from the mid nineteenth century, namely the translation by Elia T‘omačean (Venice, 1843), followed shortly afterwards by that of Arsen Bagratuni (Venice, 1864), who was also a poet in his own right and composed a poem in which the Homeric heritage exerted a certain influence⁷.

2 A Homeric glossary in Armenian? The above described situation notwithstanding, the scholarly tradition bears witness to the persistence over the centuries of a vulgata on the existence of an ancient Armenian translation of Homer. This assumption was based on what was believed to be a Homeric glossary in Armenian, reproduced on a manuscript held in the Bibliothèque Nationale in Paris and dating back to 1597, listed as number 260 of the Armenian collection in its most recent cataloguing⁸. It was recorded for the first time in 1739 by Abbot de Villefroy in his Catalogus of the manuscripts of the King’s Library, where it was described as a “lexicon displaying the more difficult terms to understand in the text of Homer which was once upon a time translated into the Armenian language, as well as in certain other writers”⁹. We also owe to Villefroy the entire section on the Armenian manuscripts of Paris in the monumental inventory of the holdings of numerous European libraries drawn up by Bernard de Montfaucon, likewise published in 1739¹⁰: here the glossary in question is recorded as number 70, together with the other works of a similar nature which are transmitted in the codex. Collectively, these works are presented as follows: “very short Armenian glossaries, with the exception of the one composed to facilitate an understanding of the translation of Homer into Armenian verse” (the exceptional nature of the Homeric glossary in comparison to the other works resides in the fact that – the description specifies – it contains about two

6 Cf. Cowe 2010, esp. 3 n. 15. 7 Cowe 2010, 3; cf. Hairapetian 1995, 522–523. 8 Kévorkian/Ter-Stépanian 1998, 831–832. 9 Villefroy 1739, 97 (Codices Armeni, no. CXXVI): “Lexicon quo Homeri in Armenam linguam olim conversi, ut & aliorum quorundam scriptorum voces obscuriores exponuntur”. 10 Catalogue des manuscrits arméniens de la Bibliothèque du Roy. Dressé en 1735, in Montfaucon 1739, 1015 ff. This section of Montfaucon’s Bibliotheca opens with the following notice to the reader: “Ce catalogue & les remarques qui l’accompagnent sont de M. l’Abbé de Villefroy, très habile dans les Langues Orientales. C’est lui qui me l’a remis pour l’inserer dans cet Ouvrage”.

260


thousand words)¹¹. The conviction of the existence of an Armenian translation of Homer also emerges in the introductory preface to the catalogue, in which Villefroy dwells at length on the Armenian nation’s predilection for the products of the spirit and for knowledge, as well as on the importance the people of Armenia awarded to the instrument of translation, so sadly scorned elsewhere. Citing an example of the acquisitions this cultural preference had assured for Armenia, Villefroy mentions not only “their Eusebius more complete than ours” and the historical accounts of the Emperors, but also, precisely, “their Homer in hexametric verse”¹². The bases had thus been laid for the postulate of an Armenian version of Homer, and the conviction was handed down over the centuries among the community of scholars without any further appraisal of its accuracy. The postulate was further endorsed shortly after Villefroy by Jean Baptiste de Villoison in the Prolegomena to his epoch-making edition of Homer (1788), in which, as is known, he also published the scholia of manuscripts Venetus A and Venetus B, containing the fruit of the ancient Greeks’ exegetic activity on the texts of the poems. Reflecting on the task undertaken by the ancient scholars, Villoison remarked that the study of Homer had absorbed the attention not only of the Greeks, but also of the “barbarae gentes”, and here he mentioned traditions concerning the existence of versions in Indian, Persian, Arabic and, specifically, Armenian¹³. On the latter point, he referred to the authoritative work of Villefroy, and underlined a shortcoming in the work by Fabricius, namely the absence of any mention of the presumed Armenian translation in the section of his Bibliotheca reserved to translations of Homer¹⁴. This was evidently due to the fact that Fabricius compiled his colossal history of Greek literature before Villefroy brought out his inventory of Armenian manuscripts.

11 Guillame de Villefroy, in Montfaucon 1739, 1021, no. 70: “Glossaires Arméniens fort courts, à la réserve de celui qui a été composé pour l’intelligence de la version d’Homère en vers Arméniens; car il contient près de deux mille mots”. 12 Guillame de Villefroy, in Montfaucon 1739, 1016: “Ce goût de toute la nation pour les productions de l’esprit & pour toutes le belles connoissances, lui a fait placer les Traducteurs mêmes parmi les hommes de Lettres & parmi les Patriarches. C’est par cette ressource si mal à propos dedaignée par quelques Sçavans, que les Armeniens ont trouvé le secret de devenir les dépositaires des sciences de l’Europe et de l’Asie. Leur Homère en vers hexamètres, leur Eusèbe plus entier que le nôtre, leur histoire des Empereurs […] justifierons l’estime de cette nation pour les habiles inteprétes, à qui souvent elle donnoit le nom de lumière du monde”. 13 Villoison 1788, XLIII n. 1. 14 Villoison 1788, XLIII n. 1: “Confer Fabricius […] quem Homeri Armenia Versio latuit” (Fabricius 1705, 300).


261

A significant figure whose work came out at a later date than the appearance of Villefroy’s catalogue, but was in no way influenced by the version of the facts proposed in the catalogue itself, was an Armenian scholar, Jacques Chahan de Cirbied, who taught his language in Paris and, among other things, published the Armenian version of the Techne grammatike attributed to Dionysius Thrax, with a French translation (1830). What is of interest in the present context is his book on Armenian grammar (which appeared in 1823), which mentions, in the preface, the set of Armenian glossaries of the King’s Library in Paris, adding, with regard to the one that concerns us here, that “it contains roughly two thousand poetic words”; significantly, however, de Cirbied included no reference to Homer or to a translation of Homeric verse into Armenian¹⁵. At the end of the route we are trying to reconstruct, which threads its way through the centuries, we will see that this is the position most closely corresponding to reality. But it long remained an isolated viewpoint. The dissertation by Johann Georg Wenrich on the translation of Greek authors into oriental languages (1842) repeated Villefroy’s conclusion regarding an ancient Armenian version of Homer, basing this deduction on the Paris glossary. With a further addition to his line of reasoning, he deduced that the poems must have been not only translated but also “diligently scrutinized” in the Armenian world¹⁶: otherwise, why would there be a glossary? A little later, in 1867, Victor Langlois supplemented the by now familiar inference on the meaning to be assigned to the Paris glossary with the further conjecture that it was “a dictionary for use in understanding the Iliad”¹⁷. However, this somewhat restrictive speculation may simply have reflected an unconscious identification of Homer with the first of his poems.

15 Cirbied 1823, XXXVI: “Dans les manuscrits arméniens de la Bibliothèque du Roi, au n° 126, nous avons un recueil de plusieurs petits glossaires […]. Le second glossaire renferme environ deux mille mots poétiques”. 16 Wenrich 1842, 76: “In Armeniacum sermonem Homeri carmina conversa fuisse discimus e catalogo codicum manuscriptorum bibliothecae regiae Parisiensis. Memoratur enim in catalogi hujus tomo I, p. 97, nro. CXXVI, inter codices Armeniacos, lexicon, quo Homeri in Armeniacam lingua olim conversi, ut et aliorum quorundam scriptorum voces obscuriores exponuntur. De eodem lexico Abbas de Villefroy in Montfauconii Bibliotheca bibliothecarum manuscriptorum nova, partis II, p. 1021, nro. 70 haec refert: […] Ex his colligas, Homeri carmina in linguam Armeniacam non tantum conversa, verum etiam diligenter lectitata fuisse”. 17 Langlois 1867, XXV: “Il paraît même qu’ils (sc. les Arméniens) traduisirent Homère, car la Bibliothèque impériale de Paris possède un vocabulaire pour servir à l’intelligence du texte de l’Iliade”.

262


In 1908, the new catalogue of Armenian and Georgian manuscripts of the Bibliothèque Nationale compiled by Frédéric Macler once again recorded this document as a “dictionary of the poetic words of Homer”¹⁸. These progressive steps of deduction eventually culminated in the synthesis drawn up by C. Frick in 1910, which put forward several conclusions regarded as “fairly certain”: namely, that there did exist an Armenian version of Homer, either of both poems or only of the Iliad; that such a version was composed in hexameters; and that many of its fragments survived in the Paris glossary. Various speculations also surrounded the chronology of the operation: although Frick felt that more precise information on this point would hopefully be forthcoming from a future linguistic investigation which, he believed, it was vital to carry out, he felt sufficiently confident to suggest that the Armenian translation of Homer was to be ascribed to a period that should neither be too close to the golden era of the fifth century, nor stray into the phase of decline that began in the twelfth century¹⁹. It was not until roughly ninety years or so later that the situation became a little clearer, with the publication of the last catalogue of Armenian manuscripts of the Bibliothèque Nationale, edited by Raymond Kévorkian and Armèn Ter-Stépanian (1998). Here the catalogue indicates the first part of the title, the incipit and the explicit of the supposed Homeric glossary. This set of data made it possible to “explode the chimaera”, to use an expression of Peter Cowe’s²⁰, inasmuch as the title of the glossary, in a Latin translation drawn up for us by M. Morani, reads “Haec vocabula sunt poetica propter metrum Homericorum carminum”, and the first entries are: “Deo similis, a Deo ornatus, a Deo factus, Deum vincere”. Thus what the glossary contains is a collection of poetic words, in which the Homeric poems appear only as a term of comparison in relation to the metre, rather than a genuine Homeric glossary from which the existence of an ancient Armenian version of the poems could have been inferred. Certainly, a title of this

18 Macler 1908, 137: “Fol. 1. Vocabulaire de mots hébreux. Fol. 32v. Dictionnaire des mots poètiques d’Homère. Fol. 147v […]” (the text in question is listed as no. 260 among the “Dictionnaires”). 19 Frick 1910, esp. 447: “Aus den mitgeteilten Zeugnissen darf man wohl mit ziemlicher Sicherheit so viel folgern, daß es eine armenische Homerübersetzung gegeben hat, und zwar entweder von der Ilias und Odyssee oder von der Ilias allein, ferner daß diese Übersetzung in Hexametern abgefaßt war und von derselben in dem besprochenen Glossar des cod. Parisinus zahlreiche Bruchstücke erhalten sind. Wann diese Homerübersetzung entstanden ist, wird sich vorläufig nicht feststellen lassen. Doch dürfte sie schwerlich dem goldenen Zeitalter (5. Jahrh.) der armenischen Literatur zuzuweisen sei. […] Anderseits aber wird man auch kaum über das 12. Jahrh. hinabgehen dürfen, weil von da an der Verfall der amenischen Literatur beginnt. Genaueres wird sich vermutlich durch eine sprachliche Untersuchung ermitteln lassen”. 20 Cowe 2010, 2.


263

kind suggests that Homer was a well-known author in the Armenian world, but such a fact, as mentioned earlier, is hardly surprising. Before concluding this first section of our paper, it may be worth adding a final piece to the puzzle of the tradition, content and aim of this collection of words, considered for centuries to be a putative “Homeric glossary”. As early as 1918 Franz Wutz spoke of an Armenian “Homerlexikon” (in quotes), maintaining that even at first glance it in no way represented a genuine Homeric dictionary: rather, it was a “medley” deriving from the most disparate sources, in which a considerable portion of the Armenian words were identifiable as biblical terminology, while the remainder consisted of a series of etymologies, apparently derived from onomastic lists²¹. Wutz bewailed the fact of having access to only three witnesses of this work: two seventeenth-century manuscripts held in Berlin and one held in Munich dating from the eighteenth century. The state of affairs at that time (1918) in history undoubtedly played a role in this limitation. Wutz also cited a manuscript held in Vienna as an additional source of transmission of the alleged Homeric lexicon, but he made no mention of the Paris manuscript, although the title of the latter coincides – at least with regard to the part made known by the catalogue – with the beginning of the title present in the manuscripts studied by Wutz²². Moreover, the same heading appears in other manuscripts as well, including a further manuscript held in Paris and dating from the thirteenth century²³, and one in Erevan dating from the fourteenth century. The latter has been interpreted, in a recent study on medieval Armenian poetry, as an example of one of the thesauri of special words that were used as teaching aids in the historical-cultural environment of the area in question, where they were intended provide help and practice in stylistic variation for beginners who were taking their first steps in the composition of poetry²⁴. Thus the presumed Homeric glossary in Armenian, held in Paris, which has been the object of so much speculation or indeed pure guesswork, seems to have several other manuscript witnesses, or to have undergone a number of different

21 Wutz 1918, esp. 253, 263. 22 Wutz 1918, 252–268. The manuscripts in question are indicated by Wutz (252) as: Berol. Petermann 145 (1602), ff. 59a-147b; Berol. Minutius 273 (1618), unspecified sheets; Monac. Arm. 22 (eighteenth century); Wien, Hofbibl. (1683), ff. 118r ff. The title of the lexicon is reproduced on p. 252 n. 1. 23 Paris, BNF, 302, ff. 93v–140v: Macler 1908, 154–155 (“glossaire de mots poétiques figurant dans les oeuvres d’Homère”); Kévorkian/Ter-Stépanian 1998, 875–876. This is a codex of miscellanea, shown by the records of the Bibliothèque Nationale to have been purchased by the library in 1838; therefore it is not mentioned in the catalogues or in the previous studies examined here. I am grateful to A. Orengo for this information. 24 Erevan, Matenadaran, 8198. I draw this data from Cowe 1995, esp. 39 n. 45.

264


redactions. Further details in this regard have yet to be established and await an examination of the contents of the codices in question. But this is beyond our sphere of expertise and also beyond the scope of this paper.

3 An Armenian text on the destruction of Troy and the construction of Rome One of the tangible pieces of evidence suggesting that the mythic tradition presupposed by the Homeric epic was indeed known in the Armenian world is represented by a text preserved in Manuscript 437 of the Matenadaran Library of Erevan, entitled Saks awerman Ilioni yev šineloyn Hṙoma (On the destruction of Troy and the construction of Rome). This text has come to attention recently thanks to the cataloguing of the Matenadaran material, and it has been studied by Peter Cowe, from whose article I have derived the data pertaining to the codex that transmits it and, more generally, the documentary basis for the reflections I put forward in the section here below²⁵. Traceable back to the thirteenth century, the manuscript probably constituted a textbook, in bolorgir script (a minuscule widely used between the thirteenth and sixteenth centuries); it contained a series of introductions and commentaries on the writings of the Fathers of the Church. The spaces left free between the main texts were used by the three scribes for the purposes of copying materials of different types and origins: an excerpt from Plato’s Timaeus (f. 253v), a few of Aesop’s fables (f. 293v) and the narration of Trojan material that we will describe here below (f. 116r). The Trojan material, like the other two passages used to fill up the empty spaces in the manuscript, is judged by Cowe to be an excerptum from a larger work, on account of its size and structure²⁶. It is opened and closed by specifications of the temporal framework in which the events are set, an indication of the time frames of the events being given by the sequences of institutional posts pertaining to different locations, based probably on Eusebius’ Chronicon. A similar characteristic is also shared by another text on the Trojan war, namely a section of an anonymous Syriac chronicle²⁷ in which the narrative portion is preceded and followed by chronological linkages determined on the basis of the succes-

25 Eganyan 2004, 675–690; Cowe 2010, who offers a translation and a commentary on the text of the mythic tale in question. For an overall description which explores the ms. in greater depth, see the Appendix by C. Aimi, infra. 26 Cowe 2010, 8. 27 Nau 1908; Cowe 2010, 8.


265

sion of the judges of Israel. However, while the account of the facts given in the Syriac text concerning the Trojan war follows the data in the Greek tradition fairly closely – with the exception of placing Menelaos in Rhodes rather than in Sparta, and the modification of a few proper names – the Armenian text differs substantially from the Greek mythic version. The deviations are pervasive and concern the geographic setting, the chronology, the characters and the plot²⁸. As far as the localization of Troy is concerned, instead of placing it in Anatolia, at the entry to the Hellespont, the Armenian excerptum situates it “to the west of the Macedonians in the land of Europe”, which suggests that in all probability it should be identified with the Ilium situated in Epirus, close to Chaonia, on the river Thyamis²⁹. Moreover, the Ilium of the Armenian text is situated “in the land of the Achaeans”. Such an expression in all likelihood alluded to the whole of Greece, as was already the case in the Homeric poems, where the term Ἀχαιοί was utilized to indicate the Greeks in general, foreshadowing an identification that became commonplace later, when the Greek province of the Roman empire was called by the name of Achaia³⁰. Moreover in the Armenian text the scene of the unfolding events that led up to the outbreak of war, namely the act committed by a prince of Ilium, who abducts a woman belonging to an important family of the city where he is hosted, is not Sparta but Thessaloniki, whose geographic position is specified as being “to the east of Macedon”. Therefore the conflict is presented as being internal to Greece. However, if the Trojan events have their origin in Thessaloniki, then a certain chronological incoherence arises, given that the founding of this city is dated, albeit with uncertainty as regards the exact year, to the end of the fourth century BC³¹, whereas, as we will shortly see, our text presupposes that the foundation of Rome – which is subsequent to the return from Troy – takes place in the eighth century BC, i.e. four centuries earlier. In this case, the Armenian text not only departs considerably from the original version, but also becomes enmeshed in a veritable internal contradiction. Additionally, although the Armenian text supplies precise references for the places forming part of the story, it never cites the names of the characters of the Trojan saga: there is no mention of Helen, Paris, Menelaos or Agamemnon. What is more, even the relations among the characters are completely different from

28 Cf. Cowe 2010, 9. 29 Bürchner 1914, 1065; cf. Cowe 2010, 9. 30 Cowe 2010, 9 notes only the second aspect. 31 The ancient sources provide no information on the chronology of the founding of the city, but the most widely accepted hypothesis is that the city was founded by Cassander one year after his marriage to Thessalonika, and therefore in 316/315: see Oberhummer 1936, 143–164, esp. 145 and Errington 2002, 451–453, esp. 451.

266


those known in the Greek tradition: the woman ravished by the prince of Ilium is not the wife but rather the daughter of a dignitary of the city where the prince is hosted. Consequently, the offensive against Ilium does not – in contrast to the Greek myth – stem from an oath of earlier days, according to which all of Helen’s pretenders had sworn to protect both Helen herself and her future husband – that is to say, whichever one of the pretenders would eventually be chosen as her husband – in the case of need³²; rather, it arises as a reaction against Ilium’s brutal refusal to free the maiden, in response to a letter from her parents. A further discrepancy, which is of particular interest, can be seen in the figure responsible for the sack of Ilium, identified as Heracles in the Armenian excerptum. Even in the Greek tradition, the name of this hero is linked to a destruction of Troy, though not in relation to the war that became legendary through the Homeric poem, but rather in an earlier conflict that had broken out because Laomedon, Priam’s father, had not paid Heracles the agreed sum for freeing his daughter, Hesione. This section of the Trojan mythic cycle remains external to the episode on which the Iliadic narration focuses. It nevertheless can be perceived within the plot of the poem and is mentioned several times, according to a mode of composition that is common in the Iliad, whose narrative fabric can often be seen to presuppose a broader mythic canvas, known and shared by poet and public alike³³. Thus the indication of Heracles as the conqueror of Ilium in the Armenian text could represent a trace of the Greek tradition on the “first war of Troy”. The differences compared to the well-known Greek tradition also involve size. In the Armenian narration, many elements are, as it were, “extra-large”, in accordance with the hyperbolic nature that characterizes the text³⁴. The duration of the war is extended from the canonical ten years to a period of fifteen years, and the tale of the stratagem involving the horse portrays the Trojans as being persuaded to accept a gift of two thousand wooden horses from their enemies, a gift that is actually a deceitful sign of peace, as the horses, each twenty cubits high, have secretly been filled with forty thousand warriors. As a result of the disproportionate size of the pack of horses, a part of the city walls has to be knocked down so that they can be transported into the city, and it is precisely through

32 Cf. e.g. Paus. 3.20.9. 33 In the Iliad a few mentions of the “first Trojan war” can be found, namely in 5.638–642, 14.250–256, 15.24–28 and 20.145–148. In this regard cf. Kullmann 1960, 227–302; Fehling 1991, 10–11; Burgess 2001, 47–49; West 2011, 30–32. 34 Cf. e.g. the description of Ilium as a great city that had no equal and the characterization of the troops that laid siege to it and destroyed it as innumerable.


267

these gaps that the greater part of the army invades the city, giving rise to the massacre carried out during the night and thus bringing about the fall of Ilium. The heroes’ return from war likewise shows some discrepancies as compared to the predominant Greek tradition, or rather, in this case, the Graeco-Roman tradition. Thus rather than the surviving Trojans led by Aeneas, it is the victorious princes who, during their return from Ilium, are blown off course by a storm (which leads them 2.500 miles astray, according to the by now familiar tendency to exaggeration that characterizes this text), and they end up on the shores of Italy. Here the captured Trojan women, whom they have brought with them as prisoners of war, excogitate a ploy to avoid being shipped back to the kingdom of the victorious forces, where they would be enslaved: they set fire to the ships and marry the Greek princes, thereby giving rise to a new community in Italy. Here too, as in the case of the identification of Heracles as the conqueror of Ilium, a precedent in the Greek world can be pinpointed: thus according to the evidence of Dionysius of Halicarnassus, this version of the story was related by Aristotle³⁵. In the Armenian text, the narrative continues: 45 years later, Hṙomelos, a head of the community set up by the princes who had returned from Ilium, founded a city at a distance of 27 miles from the sea, to which he gives his own name. Once again, parallels can be detected in the Greek literary tradition, specifically in Dionysius of Halicarnassus – who places Lavinium in more or less the same position, namely at 24 stadia from the sea – and in Strabo, who reports that Aeneas built a city, though in this case not identified as Lavinium but stated to be Rome itself, at a distance of 24 stadia from the sea³⁶. Finally, the Armenian excerptum allows a glance towards the later part of the story in Rome, drawing attention to the celebrations for the millennium of the city and highlighting Constantine’s transfer of the capital of the Empire to Byzantium. But once again, the text allows error to creep into the chronology, with a mistake of four years in the date of the millennium of Rome. The Armenian text states that at the age of forty-four Philip of Arabia came to the throne, precisely when Rome was celebrating the millennium of its own foundation; in actual fact, however, Philip of Arabia became emperor in 244 AD, whereas the celebration of the millennium of Rome took place in 248 AD, i.e. four years later. As suggested by G. Traina, a four-year misalignment could be due to the fact that the probable

35 Dion. Hal. 1.72.3–4 (= Arist. fr. 609 Rose). On the city of Lavinium in Dionysius of Halicarnassus cf. Dubourdieu 1993, 71–82; cf. also Dion. Hal. 1.34.2, which relates that some soldiers returning from Heracles’ expedition against Laomedon, above all soldiers originating from the Peleponnesus and prisoners of war captured by Heracles himself, settled on the Capitoline Hill. 36 Dion. Hal. 1.56.2 (on wh. see Vanotti 1995, 205–210); Strab. 5.3.2 (on wh. see Radt 2007, 61–63).

268


chronological reference point for the temporal framework, Eusebius’ Chronicon, was computed on the basis of the four-yearly Olympiads. The Armenian text closes by offering a sort of comparative chronology of the facts of Ilium, on the basis of various systems: the succession of the rulers of Israel, Assyria, Sicyon, Athens and Egypt and, finally, the Olympic system. While, as mentioned above, the chronological elements found in the Armenian excerptum on the Trojan war and the founding of Rome probably derive from the work of Eusebius³⁷, it nevertheless cannot be postulated that the entire text transmitted by the Matenadaran ms. was strongly dependent on the Armenian version of Eusebius’ chronicle, because a series of discrepancies can be detected³⁸. In Cowe’s view, it is far more probable that the Armenian text is a translation of another text, that was likewise written in Greek, as suggested by certain formal aspects such as the frequent use of hypotaxis and the widespread presence of the genitive absolute, both of which are more characteristic of Greek rather than Armenian syntax³⁹. Moreover, several elements analysed by Cowe as secondary corruptions, which may have arisen from successive misunderstandings of the translation from Greek, seem to suggest the existence of a certain number of intermediate copies between the exemplar in question here (which dates back, as mentioned, to the thirteenth century) and the translator’s autograph; the latter would thus have been composed one or two centuries earlier⁴⁰. Any attempt to identify the Greek source of the story handed down in Armenian by the Matenadaran manuscript poses an extremely tricky problem, although it could perhaps be suggested that the most likely path would have involved the main routes of transmission of the Trojan cycle in the Byzantine world, that is to say, the tradition of the Byzantine chronicles and erudite treatises⁴¹. In the latter field, what may come to mind is work such as that undertaken in the eleventh and twelfth centuries by Michael Psellos and John Tzetzes, who recapitulated, retold, reformulated and popularized the traditional material of the Trojan legend in order to “complete” Homer’s achievement and make it accessible to the public of their day. In the sphere of chronicle literature, one need only recall that John Malalas made extensive use, in his work on the Trojan saga, of the work by Dictys

37 Cf. Helm 1984, 217–218; Cowe 2010, 6, 8, 11, 17–18. 38 Cowe 2010, 12–13. An example of a discrepancy between the Armenian excerptum and Eusebius can be seen in the fact that according to the Armenian version there existed no kings in Greece, but only princes, at the time of the Trojan war, whereas Eusebius used the form βασιλεύς as the designation indicating both the ancient Athenian and Argive rulers. 39 Cowe 2010, 12–13. 40 Cowe 2010, 14. 41 Beck 1971, 167–169.


269

Cretensis (who is datable in the imperial era and whose work is known to us in the Latin translation): significantly, Dictys’ writings, together with those of Dares Phrygius (likewise in Latin), show the first signs of the rise of narrative and fictional elements that would later become typical of mythic cycle story-telling, with differentiations and transformations as compared to the canonical version. In the western medieval world, Dictys and Dares formed the basis of the Roman de Troie by Benoît de Sainte-More (twelfth century), which, after having been translated into Greek, was later reintroduced in the east⁴². The tendency to award priority to fictional aspects with elements that diverged from the original myth would later become a distinctive feature of the subsequent Byzantine verse novel. Striking examples of this new approach can be found in the Iliad by Constantine Hermoniacos, or in the Byzantine Achilleid (a fanciful biography of Achilles in which the Homeric hero takes on the features of the medieval knight and dies after being betrayed by Paris, who tricks him into entering a church in Troy), but also in the translation of the Roman de Troie and in the Byzantine Iliad of ms. Par. Suppl. Gr. 926⁴³ (in which Paris comes to Menelaos’ royal palace dressed as a monk and enters into the King’s service, only to run away with Helen who is expecting a baby and is wearing men’s clothes as a disguise). Thus one can unreservedly agree with Renata Lavagnini’s statement that what we find here is a series of “vistose deviazioni dalla leggenda tràdita” and that “gli eroi delle vicende troiane siano andati assumendo per via una fisionomia propria, che ne attenua le caratteristiche omeriche facendone dei semplici protagonisti del romanzo”⁴⁴. The features displayed by the text of the Matenadaran manuscript are fully in line with this evolution of the Trojan material; accordingly, this context may have constituted the background of a putative Greek original that gave rise to the Armenian version.

42 On the transmission of the Trojan material in the medieval period, cf. Montanari 2008, 259– 264. 43 Made known by Nørgaard/Smith 1975; see also the review by Montanari 1977. 44 Lavagnini 1997, esp. 61, points out that we have, in this new approach, a series of “conspicuous deviations from the legend that has been handed down” and that the “heroes of the Trojan exploits increasingly assumed a physiognomy of their own, which attenuated their Homeric characteristics and turned them into the typical protagonists of a romance”.

270


Bibliography Beck, H.-G. (1971), Handbuch der Altertumswissenschaft. 12, Byzantinisches Handbuch. 2.3, Geschichte der byzantinischen Volksliteratur, München. Bolognesi, G. (2000), Studi e ricerche sulle antiche traduzioni armene di testi greci, Alessandria. Bürchner, L. (1914), “Ilion (no. 7)”, in: PWRE 9.1, Stuttgart, 1065. Burgess, J. S. (2001), The Tradition of the Trojan War in Homer and the Epic Cycle, Baltimore-London. Cirbied, J. C. de (1823), Grammaire de la langue Arménienne, Paris. Cowe, S. P.(1995), “Models for the Interpretation of Medieval Armenian Poetry”, in: J. J. Sicco Weitenberg (ed.), New approaches to Medieval Armenian language and literature, Amsterdam, 29–45. Cowe, S. P. (2010), “A Unique Variant on the Trojan War and Founding of Rome According to a Recently-discovered Armenian Epitome”, in: Revue des études arméniennes 32, 1–24. Dubourdieu, A. (1993), “Denys d’Halicarnasse et Lavinium”, in: Pallas 39, 71–82. Eganyan, Ō. et al. (eds.) (2004), Mayr c‘uc‘ak Hayerēn jeṙagrac‘ Maštoc‘i anuan Matenadarani [Grand catalogue of the Armenian manuscripts of the Mastoc‘ Matenadaran], 2, Erevan 2004. Errington, R. M. (2002), “Thessalonike (no. 1)”, in: DNP 12.1, Stuttgart, 451–453. Fabricius, J. A. (1705), Bibliotheca Graeca, sive Notitia scriptorum veterum Graecorum, 1, Hamburgi. Fehling, D. (1991), Die ursprüngliche Geschichte vom Fall Trojas, oder: Interpretationen zur Troja-Geschichte, Innsbruck. Frick, C. (1910), “Die syrische, die armenische und die georgische Übersetzung der Homerischen Gedichten”, in: Berliner philologische Wochenschrift 14, 444–447. Hairapetian, S. (1995), A History of Armenian Literature. From Ancient Times to the Nineteenth Century, Delmar-New York. Kévorkian, R. H. / Ter-Stépanian, A. (avec le concours de B. Outtier et de G. Ter-Vardanian) (1998), Manuscrits arméniens de la Bibliothèque nationale de France: catalogue, Paris. Kullmann, W. (1960), Die Quellen der Ilias (troischer Sagenkreis), Wiesbaden. Langlois, V. (1867), Collection des historiens anciens et modernes de l’Arménie, 1, Paris. Lavagnini, R., “Storie troiane in greco volgare”, in: F. Montanari / S. Pittaluga (eds.), Posthomerica. 1, Tradizioni omeriche dall’Antichità al Rinascimento, Genova, 49–62. Macler, F. (1908), Catalogue des manuscrits arméniens et géorgiens de la Bibliothèque nationale, Paris. Montfaucon, B. de (1739), Bibliotheca bibliothecarum manuscriptorum nova, 2, Parisiis. Montanari, F. (1977), “Review of Nørgaard/Smith 1975”, in: Athenaeum 55, 480–484. Montanari, F. (2008), “Die Rezeption der homerischen Dichtung im lateinischen Mittelalter”, in: Homer. Der Mythos von Troia in Dichtung und Kunst, München, 259–264. Nau, F. (1908), “Traduction de la chronique syriaque anonyme éditée par Sa Béatitude Mgr Rahmani, patriarche des Syriens catholiques”, in: Revue de l’Orient Chrétien 13, 90–99. Nørgaard, L. / Smith, O. L. (eds.) (1975), A Byzantine Iliad. The Text of Par. Suppl. Gr. 926, edited with critical apparatus, introduction and indexes, Copenaghen. Radt, S. (ed.) (2007), Strabons Geographika. 6, Buch V-VIII: Kommentar, Göttingen. Schöne, A. (ed.) (1900), Die Weltchronik des Eusebius in ihrer Bearbeitung durch Hieronymus, Berlin.


271

Uluhogian, G. (2009), Gli Armeni, Bologna. Vanotti, G. (1995), L’altro Enea. La testimonianza di Dionigi di Alicarnasso, Roma. Villefroy, G. de (1739), Catalogus codicum manuscriptorum Bibliothecae Regiae, 1, Parisiis. Villoison, J. B. G. d’Ansse de (1788), “Prolegomena”, in: Homeri Ilias ad veteris codicis Veneti fidem recensita. Scholia in eam antiquissima ex eodem codice aliisque nunc primum edidit cum asteriscis, obeliscis, aliisque signis criticis, Venetiis. Wenrich, J. G. (1842), De auctorum Graecorum versionibus et commentariis Syriacis, Arabicis, Armeniacis Persicisque commentatio, Lipsiae. West, M. L. (2011), The Making of the Iliad. Disquisition and Analytical Commentary, Oxford. Wutz, F. (1918), “Onomastisches in einem armenischen Homerlkexikon?”, in: Orientalistische Studien: Fritz Hommel zum sechzigsten Geburtstag am 31. Juli 1914 gewidmet von Freunde, Kollegen und Schülern, 2, Leipzig, 252–268.

Chiara Aimi

Appendix – Codex M 437 Glajor, thirteenth century ex. fourteenth century in. Paper, mm 248 × 167, ff. 438 + 2 (390 and 402 are repeated); quires: [1], 1–39; sexternions, with exceptions (5 is repeated; [1], 1–3, 31, 33–36: 10 ff.; 9: 14 ff.; 20: 8 ff.; 23: 16 ff.); blind ruling, generally indiscernible; pricking is sometimes visible in the lower and outer margins; writing frame: mm 214 × 136, 43–47 lines, on two columns; unit of ruling: mm 4,5; guard leaves: two bifolia of a parchemin missal in gothic script with music notation (Germany, fifteenth century); pastedown: paper with a summary of the content of the codex, dated October 15th, 1611; binding: dark brown leather, with quadruple frame and diagonals engraved. Three holes on the front and back cover, plausibly due to the lost clasps. Bibliography: Eganyan 2004, 675–690; Yovsēp‘ean 1942–1943, 156–157; Kiseleva 1980, 92; Abrahamyan 1983, 133–134 no. 6; Łazarosyan 1984, 56–59; Xač‘erean 1988, 370–371 no. 5; Mathews/Sanjian 1991, 198 no. 6; Cowe 2010, 1–24.

The codex is written on thick oriental polished paper, with no signs of chainlines, wire-lines and watermarks. Black ink was employed predominantly, alternating with a brown to reddish ink. The majority of the quires are sexternions, put together by means of grecquage. The original signatures in capital letters, placed on the recto of the first sheet and on the verso of the last one, in the middle of the lower margin, are rarely visible due to a trimming; seventeenth century marks, displayed on almost every leaf of the first half of each quire, compensate for them. The codex shows noticeable signs of wear, especially on the initial and final sheets of each quire; during a restoration in 1611, some of them were repaired, while others were replaced. The five initial quires of the codex were also replaced¹. The writing frame, divided into two columns, leaves a narrow margin and intercolumn; the ruling is not visible to the naked eye, while the pricking holes in the lower and outer margins, seldom if ever visible, were trimmed. The page design is steady throughout the codex, except for the number of lines, ranging from 43 to 47.

1 The following are modern sheets: 1–67, 112, 186, 254, 266–267, 294, 341–343, 366, 387–93, 396– 438; two of them are opening sheets (267, 343), one is second to last (341), seven are final (112, 186, 254, 266, 294, 342, 366). The first five quires, entirely replaced, are written in a calligraphic notrgir with black ink. They contain a comment on Narek and the discourse on the opening of St. John’s Gospel by Yovhannēs Orotnec‘i. There is no way to determine whether this was the content of the original quires or not; nevertheless, the date of the last author (1315–1387) would rule out the generally accepted dating of the manuscript (70s-80s of the thirteenth century: see below).

Appendix – Codex M 437

273

The codex is entirely laid out in a hasty and minute bolorgir that allows little scope for calligraphy. The chief scribe, Esayi, is generally identified as Esayi Nč‘ec‘i (mid thirteenth century-† 1338), a great scholar and leading figure of the school of Glajor in its prime². He coordinated the work of the other scribes: Simon, who wrote from f. 116r (col. II l.17) to f. 126v, completing the 9th quire started by Esayi, and Sargis, who supplanted Esayi from f. 153v (col. I l. 38) and wrote up to 253v, thus probably concluding the 20th quire (the last sheet, f. 254, was substituted in the seventeenth century)³. A note at f. 126v sheds light on their collaboration: “Honorable brother Esayi, you have earned my contribution in doing these forty columns; now pardon my inability and the effort involved: be indulgent and do not forget in the prayers: if I can be of any help, I am ready⁴”. Decoration is virtually absent; the beginning of each text is characterized mostly by the first letter in erkat‘agir, the typeface being slightly smaller, rarely in red ink. Although the decoration is almost absent, one can find inner references to other works and schemes, functional to understanding the text, such as the renowned “Tree of Philosophy” by Aristotle (f. 122v): Ašot Abrahamyan, in his study on the “University” of Glajor, made reference to this diagram listing theoretical and practical sciences in order to define the backbone and framework of University teaching in Glajor⁵, where the manuscript is assumed to have been copied (see below).

2 “Esayi” is actually the name of both the scribe and the commissioner of the manuscript. They are often regarded as the same person (Mathews/Sanjian 1991, 198 – “probably”; Eganyan 2004, col. 675; Cowe 2010, 7; in contrast to Xač‘erean 1988, who considers Nersēs Mšec‘i as the receiver), but the way they are addressed in the notes gives cause for doubt. Scribe Simon calls the Esayi with whom he is sharing the labour of copying “brother Esayi” (f. 126v), while it is scribe Esayi himself who begs the reader “to remember the receiver of the codex Esayi vardapet” (f. 293r). Sargis, on the other hand, while asking us to commemorate “the blessed rabuni Esayi” (f. 189v), does not refer explicitly to the commissioner or the scribe. An autoptic comparison with other manuscripts supposedly written by Esayi Nč‘ec‘i (e.g. M 1097 and M 1422) and their colophons would probably help clarify the issue. 3 The autoptic examination suggests that some portions of the text ascribed by the catalogue to Esayi (ff. 110v col. I l. 29–116r col. II l. 16 and 152v–153r) do in fact have some distinctive features (the shape of ž, h and c‘, the ligature of lc, the lower horizontal strokes lengthened downward and generally more curved shapes) that would suggest the intervention of a fourth hand. The scribe’s notes, slightly different from the others by Esayi, unfortunately leave no clues as to his identity. 4 Simon effectively wrote 43 columns. 5 Abrahamyan 1983, 60–64. The scheme taken as a model by the author, fairly similar to ours, is actually the one shown by ms. M 353, f. 297r. Historical sources do not allow us to consider Glajor as a kind of lay university comparable to those in medieval Europe, as Xač‘erean claimed: it was a monastic school. This is confirmed by the content of the corpus of 69 extant manuscripts

274

Chiara Aimi

As far as the content is concerned, ms. 437 is primarily constituted by patčaṙk‘, propaedeutic explanations on the writings of Gregory of Nazianzus, Philo, Basil of Caesarea, David the Invincible, Evagrius and Pseudo-Dionysius; it is therefore a comment on the nurb greank‘, i.e. “patristic writings, as well as certain works of ancient Greek authors of significance in reconciling scriptural revelation with the insights of philosophical logic and metaphysics”⁶. As Peter Cowe observed, the content may be similar to the Book of Causes (Girk‘ Patčaṙac‘) written by Grigor Abasean (†1221), abbot of the monastery of Sanahin, which differs from our manuscript inasmuch as ms. 437 overlooks the comments on the Holy Scriptures, presenting instead a wider selection of commentaries on patristic texts⁷. The texts listed, which occupy most part of the manuscript, are interspersed with shorter texts, often as a completion of the quires: “On the Distruction of Ilion and Building of Rome”⁸; writings on Gregory of Nyssa by Vardan Hałpatec‘i; David the Invincible, On grammar; John the Philosopher, On triangles; an excerptum of Plato’s Timaeus⁹; some fables under the name of Ołonpianos (Ulpianus/ Olympianus). As to the time of the writing, although the manuscript is not explicitly dated, a few clues can be drawn from the copyists’ notes. Esayi (Nč‘ec‘i) is referred to as rabuni, “master” (f. 189v) and vardapet, “head teacher” (f. 293r). His academic

copied there, mainly dealing with theology rather than philosophy and showing an exegetical approach. See Mathews/Sanjian 1991, 22–26 and the appendix at 197–205, where the codices are listed and briefly described, thus rectifying the previous lists in Abrahamyan 1983, 130–182 (with the text of colophons), Łazarosyan 1984, 56–59, and Xač‘erean 1988, 367–416. The precise location of Glajor, also known as Ałberc‘, and its foundation are still debated by scholars. It has been identified with T‘anahat, in Vayoc‘ Jor, but neither historical sources nor archaeological finds have provided reliable evidence: see Mathews/Sanjian 1991, 17–21. 6 Širinian 2011, 161. For definitions of the different types of commentaries in Armenia, see Širinian 2000, 36–64, and Širinian 2005, 140–155. 7 Cowe 2010, 6–7. 8 For this precious witness of classical culture within the Armenian tradition, see the contribution by Pagani-D’Alessandro here above. 9 The excerpt partially covers the second column of f. 253v (ll. 16–43); the following sheet was replaced in the seventeenth century. Therefore we can speculate that the extract was originally longer, up to seven times longer than it is now. M 437 is actually the most ancient direct witness of an Armenian translation of Plato. The text (27d 6 [τί τὸ ὂν ἀεί]-28c [γενομένῳ]) coincides with the translation extant in manuscript V 1123, that preserved the five known dialogues (Timaeus, Minos, Laws, Euthyphro, Apology of Socrates), differing only in minor details (reference is made to the Mekhitarist edition: Suk‘rean 1877, 91, ll. 11–32): l. 11 zinč‘ ē or: zinč‘ or M 437; l. 14 t‘uic‘i: t‘uic‘in M 437; l. 16 umemnē: umemnē ē M 437; ll. 19–20 vareal: varel M 437; l. 27 zi: t‘ē M 437; l. 31 zgalik‘n: sgalik‘n M 437; l. 31 zgaworut‘eamb: sgaworut‘eamb M 437. For bibliographic references see the contribution by Irene Tinti in this volume, in particular n. 2.

Appendix – Codex M 437

Fig. 1: M 437, f. 122v (Courtesy of “Matenadaran” Mesrop Mashtots Institute of Ancient Manuscripts).

275

276

Chiara Aimi

status allows us to set a likely terminus post quem in the late 70s of the thirteenth century. To set a teminus ante quem, Cowe suggests that the attitude of reverence with which the scribe Esayi recalls his master Nersēs (Mšec‘i, † 1284) in the notes may signify that the latter was still alive when the manuscript was written¹⁰. If we restrict the act of copying to the late 70s and the early 80s, we can presumptively infer that it took place either at Aṙak‘eloc‘ Vank‘, near the city of Muš, where Esayi and Nersēs were present until 1280, or at Glajor/Ałberc‘, in the Ełegec‘ valley, where the master and his pupil moved in that year.

Bibliography Abrahamyan, A. (1983), Glajori hamalsaranə [The University of Glajor], Erevan, 133–134. Cowe, S. P. (2010), “A Unique Variant on the Trojan War and Founding of Rome According to a Recently-discovered Armenian Epitome”, in: Revue des Études Arméniennes 32, 1–24. Eganyan, Ō. et al. (eds.) (2004), Mayr c‘uc‘ak Hayerēn jeṙagrac‘ Maštoc‘i anuan Matenadarani [Grand catalogue of the Armenian manuscripts of the Mastoc‘ Matenadaran], 2, Erevan, 675–690. Kiseleva, L. I. (1980), Katalog rukopisej i fragmentov latinskogo alfavita hranjaščiesja v Matenadarane [Catalogue of Manuscripts and Fragments in Latin Alphabet Preserved in the Matenadaran], Erevan, 92. Łazarosyan, A. (1984), “Glajori hamalsaranum əndōrinakvac jeṙagrer” [The Manuscripts Executed at Glajor], in: Ēǰmiacin 11–12, 56–59. Mathews, T. F. / Sanjian, A. K. (1991), Armenian Gospel Iconography. The Tradition of the Glajor Gospel, Washington (D. C.), 198 no. 6. Širinian [Širinyan], M. E. (2000), “Meknołakan žanri kazmaworumə ev zargac‘umə Hayastanum” [The formation and Development of the Genre of Commentaries in Armenia], in: Aštanak 3, 36–64. Širinian [Širinyan], M. E. (2005), K‘ristoneakan vardapetut‘yan antik ev hellenistakan tarrerə [Antique and Hellenistic Elements of Christian Teaching], Erevan. Širinian, M. E. (2011), “Philo and the Book of Causes by Grigor Abasean”, in: S. Mancini Lombardi / P. Pontani (eds.), Studies on the Ancient Armenian Version of Philo’s Work, Leiden-Boston, 155–189. Suk‘rean, A. (ed.) (1877), Płatoni Imastasiri Tramaxōsut‘iwnk‘. Ewt‘ip‘ṙon, Paštpanut‘iwn Sokratay ew Timēos, Venice. Xač‘erean, L. G. (1988), Esayi Nč‘ec‘i ew Glajori hamalsaranə (1280–1340 t‘t‘.) [Esayi Nč‘ec‘i and the University of Glajor (1280–1340)], Los Angeles, 370–371, no. 5. Yovsēp‘ean, G. K. (1942–1943), Xałbakeank‘ kam Pṙošeank‘ Hayoc‘ patmut‘ean mēǰ [Xałbakeank‘ or Pṙošeank‘ within Armenian History], 3, New York, 156–157.

10 Cowe 2010, 7. The most significant note in this regard is the following: “Commemorate the […] man of God, the great rhetorician Nersēs, teacher of my most disquieted self” (f. 88r; translated by Cowe 2010, 7 n. 45).

Irene Tinti

Grecisms in the Ancient Armenian Timaeus Abstract: The article presents the results of a linguistic and traductological analysis of the ancient Armenian version of the Timaeus as preserved in the primary witnesses (chiefly, ms. V 1123, San Lazzaro, Venice), in relation with its Greek source. The author addresses three linguistic features influenced by Greek (non-Classical compound verbs, genitive absolutes and circumstantial infinitives in the instrumental case), examining their distribution and degree of correspondence with their Greek equivalents. The data suggest that the translator did not operate mechanically. He was clearly able to recognise morphological or syntactical parallelisms between Greek and Armenian elements, but did not feel compelled to apply any biunique correspondences between them. Grecisms clearly coexisted alongside more genuinely Armenian options in his linguistic competence, as they were not confined to passages in which the corresponding Greek features were attested. Once combined with the results acquired through a previous analysis focused on the semantic field of “being” and “becoming”, the new data concur in drawing the picture of a skilled translator (or team), who had a good – although by no means perfect ‒ understanding of the linguistically and philosophically complex source text, and was able to make creative use of linguistic features of different origins in order to convey its perceived meaning.

This paper will relate the results of a linguistic and traductological analysis of the ancient Armenian version of the Timaeus in relation with its Greek source¹, conducted as part of a broader research project devoted to re-examining the complex matter of its dating and attribution through a comparison with the language of Grigor Magistros Pahlawowni’s letters².

1 The research was carried out during the author’s tenure as a Fellow at the Center for Eastern Mediterranean Studies, Central European University, Budapest, and subsequently as a Calouste Gulbenkian Post-Doctoral Fellow at the Oriental Institute, University of Oxford. The theses exposed herein are those of the author and do not necessarily reflect the opinion of the Central European University Foundation (CEUBPF) or the Gulbenkian Foundation. Sincere thanks are due to Prof. Theo van Lint (Oxford), Dr Alessando Orengo (Pisa), Dr Federico Alpi and Dr Chiara Aimi (Bologna). 2 The five extant ancient Armenian versions of Platonic, or Pseudo-Platonic, dialogues (Timaeus, Euthyphro, Apology of Socrates, Laws, and Minos) are anonymous and undated, although a sure terminus ante quem for the Timaeus (the late twelfth century) is provided by two quotations in Nersēs Lambronac‘i. The matter of the translations’ authorship and chronological collocation

278

Irene Tinti

Until very recently³, the language and translation technique of the Armenian Timaeus had never been examined in any detail: indeed, the only studies specifically devoted to the text were two articles by Mariapina Dragonetti, chiefly philological in scope⁴. However, a few preliminary opinions had been tangentially expressed in contributions devoted to other texts or topics. In his first article on the Platonic dossier, Frederick C. Conybeare vaguely mentioned a stylistic difference between the Timaeus and the other dialogues, without clarifying whether the supposed dissimilarity concerned the language, the degree of faithfulness to the Greek source, or both⁵. More precisely, Rosa Bianca Finazzi, in an article devoted to the Armenian version of the Laws, suggested that the translation of the Timaeus seemed less literal than the others, since it included some expansions and explanations of the text⁶. In any case, it should be pointed out that, although previous articles have often underlined the extreme literalness of the Platonic translations (minus the Timaeus), this general assumption has been partially challenged by Chiara Aimi, at least as far as the Apology of Socrates is concerned⁷. A similar judgement concerning the Armenian translation of the Timaeus was expressed by Louis Mariès. Having analysed a few passages while conducting a study on the semantic field of “being” in Eznik Kołbac‘i (fifth century), he pointed out that ‒ despite including some unusual features and obscure passages ‒ it

has been much debated, with proposed dates as early as the fifth and as late as the eleventh century, but none of the competing proposals has been proved conclusively yet (cf. Tinti 2012b for an analysis of the available evidence and of the relevant scholarly debate). In a future publication, we will re-examine the clues that can be gathered from Magistros’ letter to the vardapet Sargis (n. 50 in Muradyan 2012b; n. 21 in Kostaneanc‘ 1910) and relate the results of the comparison between the version of the Timaeus and Grigor Magistros’ language. We will also present three quotations from the Armenian Timaeus, respectively detected in Vahram Vardapet Rabown (thirteenth century), Grigor Tat‘ewac‘i (fourteenth–fifteenth centuries), and a miscellaneous compilation (possibly later than the twelfth century). 3 Tinti 2012a: cf. infra. 4 Dragonetti 1986 is devoted to comparing the text of the only extant complete manuscript, V 1123, with the printed edition by Sowk‘rean 1877 (cf. infra); Dragonetti 1988 addresses the relations between the Armenian translation and different branches of the Greek manuscript tradition. 5 Conybeare 1889, 341. 6 Finazzi 1990a, 65. 7 Aimi 2008–2009, 23 ff. and 2011, 20. The mutual relations between the five extant Armenian Platonic versions still need to be clarified. Since no comparative linguistic analysis has ever been conducted, at present it is not known whether they were rendered into Armenian by the same person or at least within the same environment, and whether the conclusions drawn from analysing one version can also be applied to the others.

Grecisms in the Ancient Armenian Timaeus

279

seemed to be conducted less mechanically than other translations from Greek (such as that of Aristotle’s Categories). In his opinion, the translator still retained some awareness of what was permissible in the earliest, “most Classical” stages of the language⁸. Unfortunately, Mariès’ assessment cannot necessarily be relied upon, because it was based on the text as printed in the old Mekhitarist edition by Arsēn Sowk‘rean⁹, who significantly and often arbitrarily altered the text in order to make it adhere more closely to the Greek source that was available to him¹⁰, and normalised aberrant forms¹¹. Given the edition’s scarce reliability, Dragonetti proposed several emendations to the edited text of the Armenian Timaeus¹², restoring the readings of the only complete manuscript (V 1123, dating back to the sixteenth or seventeenth century)¹³ whenever she deemed it necessary. However, since she did not list all the relevant changes made by Sowk‘rean, any linguistic analysis still needs to be based on the text as preserved in the primary direct witnesses. Besides the complete codex kept in the Mekhitarist library of San Lazzaro, Venice (in which the text of the Timaeus encompasses pages 1 to 91, according to the numbering present in the codex itself), excerpts from the Timaeus are preserved in earlier manuscripts. That some passages might circulate independently from the complete translation is hardly surprising, given the particular prestige enjoyed by the Timaeus even in comparison with other Platonic works¹⁴. A passage On Colours (Vasn gownoc‘) has been handed down, in part of the tradition, as an appendix to the Book on Nature (Girk‘i veray bnowt‘ean) by Išox, a Syrian working in Cilicia in the thirteenth century. The chapter diverges from the corresponding section of the Timaeus (59.17–60.21 in ms. V 1123)¹⁵ only in minor details, and has

8 Mariès 1928, 90. In the Armenian context, “Classical” usually refers to texts that do not stray significantly from the norms adopted in the earliest documented stages of the language (fifth century); in any case, the label “Classical Armenian” can be somewhat misleading, insofar as it implies a degree of standardisation that the earliest available texts do not show (cf. e.g. Weitenberg 1993; Belardi 2004; Tinti, forthcoming). 9 Sowk‘rean 1877, 75–174. Besides the Timaeus, this edition includes the Apology and the Euthyphro, whereas the Laws and the Minos were edited in Zarbhanalean 1890. 10 On the Greek text that was available to Sowk‘rean, cf. Aimi 2008–2009, 19 and 2011, 17–18. 11 On the scarce reliability of the Mekhitarist editions of the Armenian Platonic dialogues, cf. e.g. Solari 1969; Rossi 1982–1983; Dragonetti 1986 and 1988; Finazzi 1974 and 1977; Scala 1999; 2000; 2001; 2002; Bolognesi 2000; Aimi 2008–2009 and 2011. 12 Dragonetti 1986, 28 ff. 13 Cf. Čemčemean 1998, 556 and Aimi 2008–2009, 13 ff. 14 On the surviving traces of indirect tradition, cf. supra, n. 2 and Tinti 2012b. 15 Cf. Sowk‘rean 1877, 142.24–143.27. The corresponding section in the Greek text is 67d 5–68d 2.

280

Irene Tinti

been critically edited by Stella Vardanyan¹⁶, on the basis of several manuscripts from the Matenadaran of Erevan, the most ancient of which (M 4268) dates back to the fifteenth century. Another passage (corresponding to section 13.31–14.15 in ms. V 1123¹⁷, but probably originally longer) is preserved in manuscript M 437 (folio 253v)¹⁸, dating back to the thirteenth or fourteenth century¹⁹. The accuracy of the preliminary statements on the Timaeus listed above was verified for the first time by us in a study in Italian. It included an in-depth linguistic and traductological analysis of the Armenian Timaeus, albeit focused on a specific semantic field, namely, that of “being” and “becoming”, which features prominently in Plato’s philosophical discourse²⁰. Plato uses the semantic and actional opposition between the Greek stative verb εἰμί (“to be”) and its dynamic counterpart, the achievement verb γίγνομαι (“to become”, “to be born”, “to happen” and also, in the perfect, “to be”), and between different forms of γίγνομαι itself, in order to convey ontological distinctions that prove to be a challenge even for modern interpreters and translators. Verifying which of these nuances the translator could recognise and render can be a good indicator of his linguistic competence²¹, and provide a meaningful sample of his translation technique, as well as showcase the lexical and terminological issues which arise when a technical text is transposed into a different language. That is especially true as the target language did not possess an equal set of mor-

16 Vardanyan 1979, 104–105. 17 Cf. Sowk‘rean 1877, 91.11–91.32. The corresponding section in the Greek text is 27d 6–28c 2. 18 Cf. Eganyan 2004, 675–690; Cowe 2010, 8; Aimi 2014, 298 and the appendix by C. Aimi in this volume. 19 All three sources have been consulted while preparing this article: the analysis of the text as preserved in manuscript V 1123 has been based on digitised searchable versions, and on colour photographs taken from the original in 2010. Colour photographs of ms. M 437 have been kindly provided by Chiara Aimi. For the passage On Colours, the analysis has been based on Vardanyan 1979. Scribal errors have been corrected whenever necessary, and the emendations proposed by Sowk‘rean 1877 and Dragonetti 1986 have been taken into account. The analysis of the Greek text, which dates back to the last phase of Plato’s production (between 360 and 350 BC according to Fronterotta 2006, 18 ff.), has been based on the critical editions by Burnet 1902 and Rivaud 2011. Since apparent divergences between the source text and the Armenian version may actually depend on textual variants attested in the Greek manuscript tradition, the variant readings listed in Jonkers 1989 have also been checked systematically (although, in some cases, the Armenian text may also reflect a Greek form that is no longer preserved in any of the extant witnesses). 20 Tinti 2012a: the analysis of the Armenian text was conducted according to the criteria and based on the sources listed in the previous note (minus the passage preserved in M 437, which was not yet available to us at the time). 21 Here and below, we speak of “a / the translator” in the singular, but that does not necessarily imply than more than one person could not have been involved in the enterprise.


281

phologically and semantically differentiated verbal forms (e.g. infinitives and participles). Concerning the degree of structural adherence to the source, it is particularly important to check whether the translator mechanically preserved biunique correspondences between Greek and Armenian elements, going so far as to adopt linguistic structures modelled on their Greek counterparts, or was more concerned with preserving the global sense of the original (thus essentially adopting the text itself as his translation unit)²². Thus, this sample analysis allowed us to develop an initial assessment, to be further verified with a different approach (cf. infra). As concerns “being” and “becoming”, despite a few instances of neutralisation in specific tenses or moods²³, the semantic and actional opposition is generally maintained, especially in philosophically relevant contexts, since Armenian possesses lexical items that are suitable to convey this basic dichotomy (chiefly, em and goy for “being” and linim/ełē for “becoming”)²⁴. Moreover, this version reflects a remarkable awareness of the opposition between different forms of γίγνομαι: thus, for instance, in order to convey the crucial distinction between several Greek infinitives or participles, the translator employs complex strategies involving finite verbs, periphrases, and even artificial forms²⁵. In any case, no biunique lexical correspondences are attested²⁶, although some of them are very consistent (e.g. γίγνομαι-linim). Multiple occurrences of the same philosophically charged elements are rendered differently whenever the translator perceived that they were used with a different meaning and/or in a different context (e.g. γίγνομαι is occasionally rendered with cnanim, “to be born”). A few Armenian lexical items have undergone a semantic evolution (broadening or specialising their scope) under the influence of the corresponding Greek terms²⁷: for instance, lineli, “future”, is used in the Timaeus, and elsewhere in philosophical Armenian translations²⁸, as an equivalent of γε(ν)νητός, “gener-

22 Cf. Sgarbi 2001, 588. 23 E.g. in the future, which is unsurprising from a typological point of view: cf. Dahl 2000. 24 The interlingual correspondences between these Armenian verbs and εἰμί and γίγνομαι respectively were already strikingly regular in the translation of the Gospels (cf. Coigneallaig 1968), although in that case the opposition was neutralized to linel in the infinitive. In the Timaeus, on the other hand, the infinitive gol, which is very scarcely if ever attested in Classical Armenian, is used to render εἶναι; clearly, using linel as a suppletive infinitive of the verb of “being”, according to the Classical usage, would have created a terminological problem, neutralising the opposition between εἶναι and γίγνεσθαι. 25 Cf. Tinti 2012a, 206 ff. 26 On the coexistence of different renderings of the same lexical item in Armenian translations from Greek, cf. e.g. Sgarbi 1999, 146. 27 On the contrary, so far we have not detected any Greek loanwords in the Armenian Timaeus. 28 E.g. in Philo, De Providentia: cf. Olivieri 1999–2000, 304.

282

Irene Tinti

ated” or “connected with the realm of becoming”. Clearly, these semantic calques did not necessarily take place in the Timaeus for the first time: more likely, the translator adopted elements of a pre- or coexisting philosophical vocabulary influenced by Greek. Thus, far from being examples of mechanical lexical substitution, these instances represent the answer to a terminological issue. Further clues suggest that the translator was not slavishly dependent on his model. Although it is generally possible to understand which segment of the Armenian text translates a specific segment of the Greek one – and this suggests that the sentence, or even the phrase, rather than the text, is the basic translation unit – the word order is almost never preserved, and new elements (especially pronouns and copulas) are often introduced in order to clarify the text. Besides, prepositional phrases can be rendered with adjectives, and substantivated participles and pronouns with nouns; idiomatic periphrases (e.g. ἔχω + adverb as an expression of state) can be mechanically reproduced, but also rendered with a suitable Armenian equivalent; syntactic dependencies are often altered, etc. This happens too often to think that each and every difference between the Armenian text and the Greek source can be due to the presence in the source itself of a variant reading which has not been preserved in the Greek manuscript tradition²⁹. These structural alterations do not significantly change the original text, but rather contribute to preserve and convey its meaning, which is generally understood, especially when the Greek syntax is not excessively complex and the deeper philosophical implications not too obscure³⁰. Some small expansions might also be due to the presence of a scholion in the underlying text, or to the influence of the interpretamentum provided for a certain word in a lexicon used by the translator³¹. Others are probably due to the translator’s wish to clarify the text, and are generally very small and prompted by the immediate context (e.g. a specific noun may be used instead of a generic pronoun, referents which were merely implied in the source text might be made explicit, etc.). On the whole, this version shows a respectful and yet dynamic approach to the text: the translator seems to have analysed each passage on its own – and tried to accurately convey its perceived meaning – without striving to achieve a general terminological coherence in the entire dialogue. He was clearly able to recognise the philosophical relevance of certain elements, and tried to preserve the meaning of the source text with the linguistic tools at his disposal, although

29 Cf. supra, n. 19. 30 On the linguistic and philosophical complexity of the Timaeus, cf., among others, Lozza 1994, VII; Jowett, 2004. 31 Cf. Aimi 2008–2009, 42 ff.; 2011, 20; 2014. See also Finazzi 1990b, 177; Pontani 1997; Scala 1999; Olivieri 2000; Tosi 2003.


283

he was certainly not exempt from occasional blunders. Particularly complex passages can be rendered very literally (which may suggest that he did not feel confident enough to stray from the letter), or even downright misinterpreted; in other instances, an interpretation may be adopted that is admissible from a linguistic point of view, but does not agree with the currently accepted meaning of the passage in question. That could be seen as a clue of a less than perfect understanding of the source language as used in the given text, or of imperfect bilingualism. However, it does not necessarily imply that the translator did not know contemporary Greek well, but simply that he occasionally had trouble understanding the language of Plato. That is not surprising, since even Greek interpreters of the Hellenistic period, who were chronologically and culturally closer to the text than the Armenian translator would have been, did not always understand the philosopher’s intended meaning³². That is not only due to Plato’s intentional ambiguity, connected with his philosophical method³³, but also to the natural diachronic evolution of Greek: this involved, among other things, a restructuring of the verbal system and thus neutralised the semantic and functional distinction between certain forms of the same verb (e.g. the perfect and the aorist) and made subtle aspectual oppositions increasingly obscure to the readers. As for the linguistic features of the Armenian Timaeus, the data are generally consistent with those provided by Aimi for the Apology³⁴. The language is undoubtedly heavily influenced by Greek³⁵. At the morphological level, there are instances of structural calque: for example, the sequence oč ew mi, literally “not even one”, is used to reproduce the internal structure of the indefinite pronouns οὐδεμία, “no one”, and μηδέν, “nothing”. However, different strategies are used to render Greek compound verbs (cf. infra), and even multiple occurrences of the same verb can be rendered differently. The Greek influence is especially clear at the syntactic level, where numerous calques are attested. Thus, Greek constructions such as genitive absolutes are reproduced with Armenian material (cf. infra); infinitive clauses may have a subject in accusative rather than in dative, as would have been expected; the z- preposition is not used as a mark of the accusative, as in Classical Armenian, but basically as an equivalent of the Greek article, also with a nominative (and that is consistent with a general uncertainty in distinguishing nominative and

32 Cf. Sedley 1997. 33 Cf. Baltes 1999. 34 Cf. supra, n. 7. 35 As is well known, language contact can take place through the medium of written texts: in that sense, translations constitute an ideal context for linguistic interference: cf. Kranich/ Becher/Höder 2011.

284

Irene Tinti

accusative forms)³⁶; infinitives in the instrumental case, with a subject in nominative or genitive, are used to render Greek circumstantial participles, with an overextension deviating from the Classical Armenian usage; the artificial construction orpēs … etʿē, imperfectly modeled on the Greek πότερον … ἤ, is used to render disjunctive interrogatives; etc. However, the frequent Grecisms (at different levels of the linguistic analysis) do not appear to be used in order to slavishly reproduce the source text. Indeed, there are no biunique correspondences between Greek and Armenian elements: in other words, even when a typically Greek structure is used, that does not necessarily imply that the same structure was used in that passage in the underlying text (cf. infra). On the whole, the translator seems to have considered linguistic elements modeled on Greek as viable options in the register of Armenian he was using – which, in itself, shows significant similarities with the language of other Hellenising texts³⁷ – and introduced them alongside more genuinely Armenian ones, such as infinitives with a subject in the dative case, undeclined participles with a subject in the genitive case, etc. In order to verify these assessments, we will now focus on three specific Hellenising features, drawing examples from the entire text and systematically comparing it with its Greek source (including the relevant variant readings that might justify apparent divergences). First of all, we will address the rendering of Greek verbal compounds, and verify to what extent the translator employed Armenian structural equivalents precisely calqued on Greek. Our previous analysis highlighted that, out of 35 instances of compounds of εἰμί and γίγνομαι, only 3 are rendered in Armenian with a structural equivalent³⁸, that is, with a corresponding simple verb associated with a preverb (namely, ἔνειμι with nergoy, ἐπιγίγνομαι with nerlinim, προσγίγνομαι with aṙlinim). It should be pointed out that these three instances occur within a short distance from one another³⁹, in a section of the text devoted to physical and psychological illnesses: therefore, although it might have been

36 The merging of the two cases in the plural is widespread in medieval Cilician Armenian: cf. Karst 1901, 208 ff. 37 Cf. e.g. Mercier 1978–1979; Lafontaine 1983; Weitenberg 2001–2002; Muradyan 2012a. With “Hellenising texts”, we do not necessarily refer here to the so-called Hellenising School, to which the rendering into Armenian of several Greek technical texts has been traditionally ascribed (cf. Tinti, forthcoming), but rather to any Armenian text that displays a significant degree of Greek influence. 38 Tinti 2012a, 194 ff. 39 In 79.19 (ms. V 1123) = 162.27 (Sowk‘rean 1877) = 83c 3 (Greek); 81.19 (ms. V 1123) = 164.29 (Sowk‘rean 1877) = 84 e 10–85a 1 (Greek), and 83.33 (ms. V 1123) = 167.7 (Sowk‘rean 1877) = 86 e 3 (Greek), respectively.


285

tempting to attribute the choice of a structurally faithful rendering to a desire for greater terminological accuracy, the verbs are actually not used here in a philosophically relevant meaning with reference to the general distinction between “being” and “becoming”⁴⁰. Elsewhere, the verbal base is rendered with a suitable Armenian counterpart, whereas the additional information conveyed by the preverb is expressed through an adverb, adjective, noun or prepositional phrase (e.g. merj em, “to be near” for πάρειμι, “to be present”)⁴¹. In addition, a compound can be rendered with a simple Armenian verb corresponding to its verbal base⁴² (e.g. linim, “to come to be”, and oč‘ linim, its negative form, for προσγίγνομαι, “to come to be”, and ἀπογίγνομαι, “to be absent”, “to be taken away”)⁴³, a compound with a different verbal base (e.g. šaramerjim, “to be united”, built on merjim, “to come close”, for συγγίγνομαι, “to be together”)⁴⁴, or even a totally unrelated verb, as long as it is semantically and/or contextually suitable (e.g. pakasim, “to be missing”, for ἄπειμι, “to be absent”)⁴⁵. While broadening the scope of our inquiry, we identified all the verbal compounds (roughly 1300)⁴⁶ that are present within the Greek Timaeus, and examined the rendering of numerous occurrences, in order to get a general picture of the different techniques adopted. For the purpose of the present article, we are going to exemplify our findings by addressing in detail a manageable yet meaningful sample, namely, the compounds built on λαμβάνω, which as a simple verb basically means “to take” and “to receive”, but also “to seize”, “to detect”, “to understand”, “to admit”, etc. The sample has been selected for several reasons: first of all, λαμβάνω is one of few verbs which occur in the Timaeus in composition with a double prefix⁴⁷. Moreover, since the simple verb covers a rather large

40 There is no reason to believe that this section of the text might be attributable to a different translator than the rest, either: the coexistence of different techniques in rendering compound verbs is well attested elsewhere in Armenian translations from Greek (cf. e.g. Sirinian 1993). 41 In 2.16 (ms. V 1123) = 76.17 (Sowk‘rean 1877) = 17b 7 (Greek). 42 Cf. Aimi 2008‒2009, 23. 43 In 78.2 (ms. V 1123) = 161.6‒7 (Sowk‘rean 1877) = 82b 4 (Greek). 44 In 19.16 (ms. V 1123) = 97.27 (Sowk‘rean 1877) = 34b 6 (Greek). 45 In 2.10 (ms. V 1123) = 76.8–9 (Sowk‘rean 1877) = 17a 6 (Greek). 46 The exact number varies according to the criteria adopted for inclusion (e.g. whether forms such as ἐγκωμιάζω – which is not etymologically a compound but forms its preterites like one – or καθαίρω – which may look like a compound but does not behave like one from a morphological point of view – are included or not). The rough total given above includes participial forms, but not verbal adjectives. 47 That happens in 8 instances: 27b 7 (Greek) = 13.17 (ms. V 1123) = 90.27 (Sowk‘rean 1877); 39c 4 (Greek) = 24.19 (ms. V 1123) = 103.27 (Sowk‘rean 1877); 44b 8 (Greek) = 30.26–27 (ms. V 1123)

286

Irene Tinti

semantic field, and can be associated with numerous preverbs, the compounds can differ greatly in meaning, and that in turn can potentially favour a creative and not standardised rendering. Finally, compounds of λαμβάνω occasionally occur multiple times in close proximity, a clearly intentional reiteration, and it is therefore interesting to ascertain whether the translator maintained a lexical repetition in Armenian as well. We analysed the 32 occurrences of the simple verb in comparison with those of its compounds (39), and then cross-referenced them with their Armenian renderings, verifying whether any biunique correspondences between Greek and Armenian verbs could be detected. Here, we will mostly focus on the relevant verbal forms as if they were isolated elements, for the purpose of argumentation; however, in all cases, our considerations are based on a systematic examination of the sentences and broader passages in which each of them occurs. This is because, on the one hand, the context can clearly affect the chosen rendering, and on the other, we needed to verify whether any additional element had been inserted into the target text in order to convey more precisely the meaning of the Greek compound⁴⁸. Overall, the data from the new sample agree with our previous assessment: the translator does not adopt a mechanical approach or pursue a slavish adherence to its source, but uses instead different strategies, even while rendering the same compound, thus obtaining different degrees of structural parallelism. None of the compounds of λαμβάνω with a double preverb is translated with a corresponding Armenian compound with a double prefix⁴⁹. In a single instance, namely, when συμπεριλαμβάνω, “to enclose”, is rendered with šowrǰ əmbṙnem⁵⁰, the first Greek prefix seems to be rendered with the preverb ənd, “with” (here associated with the verbal base bṙnem, “to take”, “to keep”), whereas the other is clearly translated with the separate adverb šowrǰ, “around”. In another occur-

= 111.30 (Sowk‘rean 1877); 58a 5 (Greek) = 47.9 (ms. V 1123) = 129.32 (Sowk‘rean 1877); 65d 6 (Greek) = 56.31 (ms. V 1123) = 139.29 (Sowk‘rean 1877); 74d 6 (Greek) = 68.20 (ms. V 1123) = 151.22–23 (Sowk‘rean 1877); 83d 1 (Greek) = 79.27 (ms. V 1123) = 162.35–36 (Sowk‘rean 1877), and 84e 1 (Greek) = 81.8 (ms. V 1123) = 164.16 (Sowk‘rean 1877). Λαμβάνω has been selected over εἶμι, “to go”, which is also attested 8 times in composition with a double prefix, because choosing the latter would have involved an additional set of philological issues, since some of its forms are prone to be confused with similar forms of εἰμί, “to be” (as is well known, Greek manuscripts did not note diacritics before the ninth century AD). 48 In the course of this traductological analysis, we detected the same kind of adjustments to the source text that we had previously noticed in different sections of the dialogue: changes in word order, small insertions, altered syntactic dependencies, etc. 49 Such as those listed by Muradyan 2012a, 39. 50 74d 6 (Greek) = 68.20 (ms. V 1123) = 151.22–23 (Sowk‘rean 1877).


287

rence without any appreciable difference in meaning⁵¹, συμπεριλαμβάνω is rendered with i miasin əmbṙnem: here, σύν seems to be rendered twice, while περί does not seem to have an exact correspondent. Ənd might actually correspond to περί in the meaning of “about”⁵², but, in that case, we should assume that the translator adopted different lexical equivalences while rendering the same preposition in the same compound (which would be in itself significant). However, since əmbṙnem, “to take”, “to reach”, “to comprehend”, is clearly considered elsewhere as a suitable semantic equivalent for the simple verb λαμβάνω (cf. infra), i miasin likely renders here the additional notion expressed by σύν, while περί is simply not translated as such. Thus, we could also question the assumption that, in the previous example (šowrǰ əmbṙnem), both preverbs are actually rendered: if šowrǰ = περί and əmbṙnem = λαμβάνω, then σύν = ø. Moreover, συμπεριλαμβάνω with the meaning of “to include” is translated elsewhere⁵³ with paraṙem (cf. infra), in which σύν is not represented as a separate element. Finally, in other instances, double compounds are rendered with a contextually suitable Armenian verb, with no degree of structural similarity whatsoever. For example, ἥλιον ἐπικαταλάβῃ, “catches up with the sun”, is suitably translated as hasani aṙ aregakn⁵⁴, “reaches the sun”: the notion conveyed by ἐπί is rendered with aṙ in a prepositional phrase, but the verb itself does not mirror the Greek one. Taking into consideration single preverb-compounds as well, the different translation strategies adopted can be ideally placed on a scale of structural adherence⁵⁵. At one end of it, there are instances in which a Greek compound is rendered with its exact Armenian equivalent, a non-Classical Armenian compound calqued on Greek. However, that happens only in 4 instances out of 39, and always with the same verb, namely, περιλαμβάνω, which is rendered once with bakaṙem⁵⁶, “to contain”, “to comprehend”, and 3 times with paraṙem⁵⁷, “to include”, “to com-

51 83d 1 (Greek) = 79.27 (ms. V 1123) = 162.35–36 (Sowk‘rean 1877). 52 However, this interlingual correspondence is not listed in NB, s.v. ənd, in the bilingual lists in the database LALT, nor in Muradyan 2012a, 29–30. 53 58a 5 (Greek) = 47.9 (ms. V 1123) = 129.32 (Sowk‘rean 1877). 54 39c 4 (Greek) = 24.19 (ms. V 1123) = 103.27 (Sowk‘rean 1877). 55 As mentioned above, our evaluation of the degree of structural adherence displayed by each rendering (which is not always easily determined) is based on a careful consideration of the context in which the verbal forms occur. 56 33b 3 (Greek) = 18.11 (ms. V 1123) = 96.19 (Sowk‘rean 1877). 57 30c 8 (Greek) = 16.11 (ms. V 1123) = 94.9 (Sowk‘rean 1877); 39 e 5 (Greek) = 25.3 (ms. V 1123) = 104.19 (Sowk‘rean 1877); 57a 7-b 1 (Greek) = 46.5 (ms. V 1123) = 128.25 (Sowk‘rean 1877). The cor-

288

Irene Tinti

prehend”. Both verbs include the root of aṙnowm, “to receive”, “to take”, a well attested correspondent of λαμβάνω⁵⁸, and are likely backformations on participial forms ultimately going back to Classical Armenian collocations (bakaṙem < bakaṙeal < bak aṙeal < bak aṙnowm⁵⁹; paraṙem < paraṙeal < par aṙeal < par aṙnowm⁶⁰). The correspondence between bak, par and περί is also attested in Hellenising translations⁶¹. There are no other occurrences of bakaṙem in the dialogue, whereas paraṙem renders, in one instance, περιέχω, “to surround”⁶², and in another, unsurprisingly, πέριξ […] λαμβάνω⁶³. Since aṙnowm is also a suitable equivalent of ἔχω, “to have”, “to keep”, paraṙem is only used in the Timaeus to render its Greek structural correspondents. At the other end of the scale, there are semantically and contextually suitable Armenian renderings with no structural similarity with the Greek element they are translating. This solution is actually quite frequent (15 instances out of 39, including the double compounds). For instance, περιλαμβάνω is also rendered with galarem⁶⁴, “to wing around”, and patatem⁶⁵, “to embrace”. Ἀναλαμβάνω, meaning “to restore”, “to repair” and also “to run over in one’s mind”, is rendered in one instance with hastatem⁶⁶, meaning “to confirm”, “to consolidate”, “to restore”, “to refresh”.

respondence is attested 4 times if we consider the double compound mentioned above, in which case the structural parallelism is, however, not exact. 58 The correspondence is attested 11 times in the Timaeus, e.g. in 35a 6 (Greek) = 19.31 (ms. V 1123) = 98.15 (Sowk‘rean 1877). Sowk‘rean (1877: 97.6) is probably right in introducing aṙnowl as the counterpart of Greek λαβεῖν (33d 3), whereas ms. V 1123 (18.30) reads aṙowl. Aṙnowl renders λαβεῖν also in 69a 4 (Greek) = 61.6 (ms. V 1123) = 144.9 (Sowk‘rean 1877). 59 Bak aṙnowm is already attested in the Bible with the meaning of “to encompass”, “to surround”; according to the information provided in NB, s.v., bakaṙem, with a similar meaning, is attested in forms other than the participle from the sixth century at least. Unsurprisingly, Sowk‘rean 1877 reintroduces bak aṙeal, whereas the manuscript reads bakaṙeal. 60 Par aṙnowm is already attested in the Bible with the meaning of “to dance”, and in the sixth century with the meaning of “to surround”, “to comprehend” (cf. NB, s.v.), probably because of a semantic calque on Greek compounds like περιλαμβάνω or περιέχω, which were also structurally similar. According to the information provided in NB, s.v., paraṙem is attested from the sixth century at least. Sowk‘rean 1877 reintroduces analytic forms in all the occurrences of the verb but two (104.19 and 128.25), whereas ms. V 1123 consistently reads paraṙeal or paraṙin (in 46.5). 61 Cf. Adontz 1970, CLXXI and Muradyan 2012a, 29 ff. 62 60c 1 (Greek) = 50.6 (ms. V 1123) = 132.29–30 (Sowk‘rean 1877). 63 36c 3 (Greek) = 20.29 (ms. V 1123) = 99.15 (Sowk‘rean 1877). 64 81a 7 (Greek) = 76.22 (ms. V 1123) = 159.24 (Sowk‘rean 1877). 65 77e 4–5 (Greek) = 73.23 (ms. V 1123) = 155.24–25 (Sowk‘rean 1877). 66 26a 3 (Greek) = 12.1 (ms. V 1123) = 88.31 (Sowk‘rean 1877).


289

To the semantically suitable renderings could also be added the 3 instances in which different compounds of λαμβάνω are rendered with an equivalent of the simple verb, namely aṙnowm⁶⁷. Between these two opposites ‒ extreme adherence on the one hand and “free” rendering on the other ‒ are some intermediate instances whose categorisation is not so clear-cut. As seen above, a Greek compound can be rendered with a corresponding verbal base associated with a separate element, like an adverb or a prepositional phrase. In these cases, on the one hand, each part of the Greek compound finds an equivalent in the Armenian text; on the other, the final structure is undoubtedly different. Besides, the association between a verb and an adverb is clearly not a Hellenising feature in itself⁶⁸. In other instances, an idiomatic expression involving the corresponding Armenian verbal base can be used⁶⁹: cf., for instance, the well established collocution yanjn aṙnowm, “to take upon oneself”, which is used once to render διαλαμβάνω⁷⁰, “to receive each his own share (of the task)”. Once again, the Armenian element is not artificially modeled on Greek, but it has likely been selected also because it partially mirrors the Greek one. Again, compounds of λαμβάνω may be rendered with a Classical Armenian verb, semantically and contextually suitable, which also happens to be a compound, but not necessarily a perfect structural equivalent of the Greek one. Thus, when šrǰap‘akem, “to surround”, is used to render περιλαμβάνω⁷¹, šrǰ- (<šowrǰ, “around”) is actually a suitable equivalent of περί, but the simple verb p‘akem, “to close”, is not an equivalent of λαμβάνω. Elsewhere, the verbal base may correspond, but the preverb does not. On that account, it is worth noticing that περιλαμβάνω⁷², “to contain”, μεταλαμβάνω⁷³,

67 Παραλαμβάνω in 30a 4 (Greek) = 15.24 (ms. V 1123) = 93.17 (Sowk‘rean 1877) and in 69c 5 (Greek) = 61.27 (ms. V 1123) = 144.30 (Sowk‘rean 1877); ἀπολαμβάνω in 59a 4 (Greek) = 48.15 (ms. V 1123) = 131.2–3 (Sowk‘rean 1877). 68 Cf., for partially similar remarks, Muradyan 2012a, 41–42 and 46–47. 69 As seen above, bakaṙem and paraṙem likely developed respectively from the collocations bak aṙnowm and par aṙnowm, which came to be used as suitable correspondents of certain Greek compound verbs. 70 26d 5 (Greek) = 12.25 (ms. V 1123) = 89.28 (Sowk‘rean 1877). 71 56e 3 (Greek) = 45.24 (ms. V 1123) = 128.10 (Sowk‘rean 1877). 72 53 c 7 (Greek) = 41.16 (ms. V 1123) = 123.26 (Sowk‘rean 1877). 73 26e 2–3 (Greek) = 12.30 (ms. V 1123) = 90.2 (Sowk‘rean 1877).

290

Irene Tinti

“to take instead”, διαλαμβάνω⁷⁴, “to divide”, “to grasp separately”⁷⁵, and the double compound ἀνταπολαμβάνω⁷⁶, “to receive in return”, can all be rendered with ǝndownim/ǝnkalay⁷⁷, “to receive”, “to take”. Similarly, παραλαμβάνω⁷⁸, “to receive”, καταλαμβάνω⁷⁹, “to overtake”, περιλαμβάνω⁸⁰, “to include”, and προσλαμβάνω⁸¹, “to receive”, can all be rendered with ǝmbṙnem, “to take”, “to reach”, “to comprehend” (which is also used twice while rendering compounds with double preverbs: cf. supra). Not all of these preverbs are usual Greek correspondents of ənd⁸², although, with a bit of flexibility, all of them could probably be rendered with it in specific circumstances. However, both ǝndownim and ǝmbṙnem are frequently used to render the simple verb λαμβάνω in the Timaeus (11 and 5 times respectively)⁸³. Thus, rather than being selected as precise structural equivalents of so many different Greek compounds, they likely would have been considered as suitable semantic renderings in their respective contexts, possibly with the added bonus of sharing with the Greek compounds a certain degree of internal motivation, thus being potentially analysable in their inner components. On that account, it should be pointed out that, although that is not the case with the aforementioned bakaṙem and paraṙem, other post-Classical compounds in the Timaeus render not only their exact Greek counterparts (which seems to be the most frequent case

74 62c 5 (Greek) = 52.32 (ms. V 1123) = 135.24 (Sowk‘rean 1877). 75 In this case, the notion of distribution/separation is made clear by the context. 76 27b 7 (Greek) = 13.17 (ms. V 1123) = 90.27 (Sowk‘rean 1877). 77 In 44e 4 (Greek) = 31.13 (ms. V 1123) = 112.22 (Sowk‘rean 1877) ǝndownim is used where the Greek text had ἀντιλαμβάνω. However, in this case the rendering is freer and also involves a noun: οἷς ἀντιλαμβανόμενον, “taking hold of which” = yoroc‘ ǝnkaleal zōgnakanowt‘iwn, “receiving support from which”. 78 68e 3 (Greek) = 60.30 (ms. V 1123) = 143.37 (Sowk‘rean 1877). 79 Twice in 80b 1–2 (Greek) = 75.16 (ms. V 1123) = 158.21 (Sowk‘rean 1877): the lexical repetition is maintained in Armenian. 80 81c 2 (Greek) = 76.33 (ms. V 1123) = 160.1 (Sowk‘rean 1877). 81 82a 6 (Greek) = 77.27–28 (ms. V 1123) = 160.33 (Sowk‘rean 1877). 82 Cf. LALT, and NB, s.v. 83 In fact, əndownim renders λαμβάνω as frequently as aṙnowm does (cf. supra). For λαμβάνω / əndownim, cf. e.g. 24a 2 (Greek) = 9.30 (ms. V 1123) = 86.8 (Sowk‘rean 1877) and 86e 7 (Greek) = 84.5 (ms. V 1123) = 167.13 (Sowk‘rean 1877). For λαμβάνω / ǝmbṙnem, cf. e.g. 68d 3 (Greek) = 60.22 (ms. V 1123) = 143.27 (Sowk‘rean 1877) and 92c 5 (Greek) = 91.11 (ms. V 1123) = 174.24 (Sowk‘rean 1877).


291

overall), but also other compounds⁸⁴, or even semantically analogous simple verbs⁸⁵. As a final note, it is worth remarking that – although biunique correspondences between Greek and Armenian compounds (as with all lexical units in general) are not preserved as a rule, but each case is rather examined and rendered on its own – the translator makes sure that the obviously intentional lexical repetitions are maintained, while keeping at the same time the necessary opposition between different forms. For example, in order to render the parallel pair καταλαμβάνουσίν τε καὶ καταλαμβάνονται⁸⁶, “they overtake and are overtaken”, the translator does not use the obvious solution, that is opposing hasanen to hasanin, probably because the verb is consistently used in the -im form in the Timaeus⁸⁷. Instead, the relevant forms are rendered with hasanin and hasealk‘ linin respectively: the same verb is used in both instances, and the formal difference of diathesis is somewhat artificially preserved by opposing a periphrasis involving the participle, to which a passive meaning can be ascribed, to the medial form, in this case active in meaning. Elsewhere⁸⁸, the sequence ἐφαίνετο καταλαμβάνοντα καταλαμβάνεσθαι, “they seem to be reached, while reaching”, is rendered with hasow ełealk‘ erewin ink‘eank‘ hasealk‘, “having reached them, they seem to be themselves reached”; the participle hasealk‘ is once again ascribed to the passive side of the opposition, while the active form is rendered with a collocation involving the related adjective hasow. Still, the re-occurrence of similar elements and the difference in diathesis are both preserved. Moving on to the syntactic level of the linguistic analysis, we will focus, albeit briefly, on two non-Classical Armenian structures influenced by Greek, namely the genitive absolute and the infinitive in the instrumental case used with a cir-

84 E.g. šaramerjem / -im, “to connect”, “to be connected”, renders not only συνάπτω (36c 1 Greek = 20.27 ms. V 1123 = 99.12–13 Sowk‘rean 1877), “to join together”, and συλλέγομαι (18e 2 Greek = 3.33 ms. V 1123 = 78.19–20 Sowk‘rean 1877), “to come together”, but also προσάπτω (33d 5 Greek = 18.32 ms. V 1123 = 97.8 Sowk‘rean 1877; 80b 4 Greek = 75.19 ms. V 1123 = 158.25 Sowk‘rean 1877) and συγγίγνομαι (cf. supra). 85 E.g. ǝnkalowm, “to receive”, renders, among other verbs, δέχομαι (51a 3 Greek = 38.20–21 ms. V 1123 = 120.11 Sowk‘rean 1877), “to accept”, αἱρέω (87b 8 Greek = 84.23 ms. V 1123 = 167.33 Sowk‘rean 1877), “to take”, and ἔχω (73c 4 Greek = 67.2 ms. V 1123 = 150.2 Sowk‘rean 1877), “to have”. 86 38d 4–5 (Greek) = 23.21 (ms. V 1123) = 102.26 (Sowk‘rean 1877). 87 The regular ending -el is usually attested in the infinitive, although, in one instance, the non-Classical ending -il is used instead: cf. 75.30 (ms. V 1123) = 158.36 (Sowk‘rean 1877) = 80c 6 (Greek). 88 39a 5 (Greek) = 24.5 (ms. V 1123) = 103.11 (Sowk‘rean 1877).

292

Irene Tinti

cumstantial meaning. Both can express the circumstances (time, motive, obstacles overcome, etc.) in which the event detailed by the main clause takes place. As is well known, in the Greek genitive absolute a participle in the genitive case is associated with a nominal element (noun, pronoun, substantival adjective), also in the genitive, which constitutes its subject. Hellenising translators, possibly favoured in their choice by options already present in the earliest documented stages of Armenian⁸⁹, often reproduced the Greek structure with indigenous lexical material⁹⁰. Following the same method adopted while analysing the compounds, we identified at first the genitive absolutes present in the Greek text of the Timaeus, and then set out to verify how they were rendered into Armenian, in order to ascertain whether the translator adopted any biunique or significantly consistent interlingual correspondence. Overall, that does not seem to be the case. Greek genitive absolutes can indeed be reproduced with Armenian material, through syntactic calque. Thus, τούτου δ’ ὑπάρχοντος⁹¹, “this being the situation”, can be rendered with ew sora eloy (this participle of the verb of “being” is in itself a non-Classical form)⁹². In the plural, τοιούτων δὲ ἀποτελεσθεισῶν τεττάρων⁹³, “four such (angles) being brought to completion”, can be faithfully translated with ew ayspiseac‘ č‘oric‘ katareloc‘.

89 As shown by Weitenberg 1997, even the most typically Hellenising features can be profitably analysed not as extraneous monstruosities completely severed from synchronic linguistic usage, but rather as choices made on the basis of actually available tendencies. In this case, although genitive absolutes are not present as such in prototypical fifth century texts, the existence of participles or perfects with a subject in genitive (clearly by then an integral part of the language, no matter their origin) might have favoured the spread of this Greek syntactic feature in translated (and even original) Armenian literature. 90 Cf. Muradyan 2012a, 161 ff. 91 30 c 2 (Greek) = 16.4 (ms. V 1123) = 94.1 (Sowk‘rean 1877). 92 According to NB, s.v. em, a participial form eal is attested as such in the twelfth century; declined forms such as eloy and elov, however, are common in earlier Hellesining texts. Strictly speaking, they can be ascribed either to the participle or to an infinitive el (for the latter explanation, cf. Muradyan 2012a, 118–119), and indeed, the right interpretation must be verified case by case. However, as far as Armenian genitive absolutes are concerned, the presence of plural forms such as eloc‘ (e.g. in 88a 8-b 1 Greek = 85.28 ms. V 1123 = 169.3 Sowk‘rean 1877), and the parallelism with the Greek structure, suggest that the forms are meant as suitable equivalents of Greek participles. Furthermore, examples such as ew soc‘a oč‘ ełeloc‘, which renders τούτων δὲ μὴ γενομένων, “these not being born” (41b 7-c 1 Greek = 27.4 ms. V 1123 = 107.9 Sowk‘rean 1877) confirm that the declined verbal forms in Armenian genitive absolutes are participles and not infinitives. 93 55a 2 (Greek) = 43.14 (ms. V 1123) = 125.29 (Sowk‘rean 1877).


293

Very rarely, a declined participle (infinitive?) in the singular form is associated with a nominal element in the plural form (e.g. soc‘a oč‘ eloy, which renders τούτων ἀπόντων, “these things being missing”)⁹⁴. However, in the Timaeus, the overwhelming majority of instances presents the expected association of two plural participial forms; therefore, it is reasonable to assume that the apparent exceptions are probably the result of textual corruption, given the graphic similarity of and , especially in capital script (<Յ>; <Ց>)⁹⁵. In any case, the translator also adopts different, less literal renderings, suitable to the specific context: thus, a Greek genitive absolute can be translated with an explicit subordinate clause (e.g. ew erēk yoržam asēir dow, “and yesterday when you were speaking”, renders λέγοντος δὲ δὴ χθὲς σοῦ, “while you were speaking yesterday”)⁹⁶, with an undeclined participle with a regular subject in genitive (e.g. im kamec‘eal suitably renders ἐμοῦ γε ἐθέλοντος, “if I consent to / wish it”)⁹⁷, or with a construction involving an infinitive in the instrumental case (see infra). On the other hand, at least in one instance, an Armenian genitive absolute renders a different Greek structure, namely a conjunct participle (σὺ παρών, “you, being present” = merj eloy k‘o)⁹⁸. As for the instrumental of the infinitive, it was already used in Classical Armenian to express the manner or means through which the event conveyed by the main verb comes about⁹⁹; however, in Hellenising texts, it came to be used, more broadly, with a circumstantial meaning, and as a fitting correspondent for Greek genitive absolutes¹⁰⁰. Through the usual cross-referencing method, we ascertained that, in the Timaeus, the correspondence is indeed well attested. Instrumentals of the infinitive, with a subject either in genitive (e.g. ew aynpēs mnalov sora, “and while he remained thus” = μένοντος δὲ [sc. αὐτοῦ])¹⁰¹ or, occasionally, in nominative¹⁰² (e.g. pēspēs goink‘ […] linelov, “while colours of all sorts are born” = παντοδαπῶν […]

94 57e 5–6 (Greek) = 47.2 (ms. V 1123) = 129.22 (Sowk‘rean 1877). 95 It is also worth mentioning that Muradyan 2012a, 161 ff., Mercier 1978–1979 and Lafontaine 1983 do not list any similar “anomaly” while discussing genitive absolutes. 96 25e 2 (Greek) = 11.28 (ms. V 1123) = 88.22–23 (Sowk‘rean 1877). 97 41a 8 (Greek) = 26.28 (ms. V 1123) = 106.26 (Sowk‘rean 1877). The rendering suggests that the Greek source text read ἐμοῦ γε ἐθέλοντος rather than ἐμοῦ γε μὴ ἐθέλοντος: cf. Burnet 1902 and Rivaud 2011, ad loc.; Jonkers 1989, 40, 86, 126, 139, 144, 320; Tinti 2012a, 226. 98 17b 7 (Greek) = 2.16–17 (ms. V 1123) = 76.17 (Sowk‘rean 1877). 99 Cf. Jensen 1959, 180; Minassian 1996, 203 ff. 100 Cf. Muradyan 2012a, 164 ff., with references. 101 42e 6 (Greek) = 28.29–30 (ms. V 1123) = 109.25 (Sowk‘rean 1877). 102 Cf. Mercier 1978–1979, 74.

294

Irene Tinti

γιγνομένων χρωμάτων)¹⁰³, can indeed render Greek genitive absolutes. It should be pointed out that, when the subject of the infinitive, in nominative, is not the same as the subject of the main verb, an anacoluthon arises. However, instrumental infinitives with a subject in nominative¹⁰⁴ can also render Greek conjunct participles (e.g. ink‘n yink‘ean daṙnalov = αὐτὴ ἐν αὑτῇ στρεφομένη, “herself revolving in herself”)¹⁰⁵. As a matter of fact, instances like these, where the instrumental of the infinitive stands in for a Greek conjunct participle in the nominative case, probably gave origin to the juxtaposition between the instrumental and a subject in nominative in Hellenising texts, which was then extendend to the rendering of genitive absolutes. In any case, unsurprisingly, conjunct participles are also rendered in the Timaeus by their Armenian counterparts (e.g. ew [mek‘] əst bnowt‘ean hatowc‘ealk‘, “and [we], giving according to nature” = καὶ [ἡμεῖς] κατὰ φύσιν δὴ δόντες)¹⁰⁶. To sum up, the analysis of the distribution of three specific features influenced by Greek (post-Classical compounds, genitive absolutes and circumstantial infinitives) confirms our initial assessment of the language of the version of the Timaeus: Grecisms coexisted alongside more genuinely Armenian options in the translator’s linguistic competence. Although there is a general preference for using Hellenising elements in order to render the corresponding items in the Greek text, that is not always the case: thus, at least to a certain extent, Hellenising and Armenian structures were perceived as semantically equivalent and fit to be used in analogous contexts. Furthermore, the absence of any biunique interlingual correspondences reinforces the notion that the translator did not operate mechanically on his source text: he was clearly able to recognise the morphological or syntactical correspondences between Greek and Armenian elements, but did not feel compelled to apply them on a systematic basis, because his primary aim was to reproduce the text’s meaning. All these elements concur in drawing

103 68a 5–6 (Greek) = 59.33–60.1 (ms. V 1123) = 143.3–4 (Sowk‘rean 1877). The text of the passage On Colours as edited by Vardanyan 1979, 104.22–105.1 reads šinelov, “building”, rather than linelov. 104 So far we have not detected in the Timaeus any example of an Armenian infinitive in the instrumental case, with a subject in genitive, used to render a Greek conjunct participle. On that account, it should be pointed out that the Greek conjunct participle appears by definition in the same case as the element with which it is associated, and is not as syntactically isolated as the genitive absolute: therefore, while rendering the nominal element into Armenian, assigning a different case to it could change the syntactical dependencies of the entire clause. 105 36e 3–4 (Greek) = 21.13–14 (ms. V 1123) = 100.6 (Sowk‘rean 1877). 106 17c 10 (Greek) = 2.28 (ms. V 1123) = 77.8 (Sowk‘rean 1877).


295

the picture of a skilled translator, who had a good (although by no means perfect) understanding of the source text, and was able to use creatively linguistic features of different origin.

Bibliography LALT = Weitenberg, J. et al. (eds.) (2003), Leiden Armenian Lexical Textbase, Leicester (www.sd-editions.com/LALT/access.html). NB = Awetik‘ean, G. / Siwrmēlean, X. / Awgerean, M. (1836–1837), Nor bargirk‘ haykazean lezowi, Venetik. Adontz, N. (1970), Denys de Thrace et les commentateurs arméniens, Louvain. Aimi, C. (ed.) (2008–2009), La traduzione armena dell’Apologia di Socrate di Platone: edizione critica e commento, Dissertation, Università di Bologna. Aimi, C. (2011), “Platone in Armenia. Osservazioni sulla traduzione dell’Apologia di Socrate”, in: Rassegna Armenisti Italiani 12, 15–21. Aimi, C. (2014), “Tracce di lessicografia greca nell’antica traduzione armena dell’Apologia di Platone”, in: Eikasmós 25, 295–312. Baltes, M. (1999), “Γέγονεν (Platon, Tim. 28 B 7). Ist die Welt real entstanden oder nicht?”, in: Id., ΔΙΑΝΟΗΜΑΤΑ, Kleine Schriften zu Platon und zum Platonismus, Stuttgart-Leipzig, 303–325. Belardi, W. (2004), “Del rapporto gerarchico tra grammatica e lessico nell’armeno più antico”, in: Atti della Accademia Nazionale dei Lincei. Classe di scienze morali, storiche e filologiche. Rendiconti s. 9, 15, 185–203. Bolognesi, G. (2000), “Note sulla traduzione armena delle Leggi di Platone”, in: Id., Studi e ricerche sulle antiche traduzioni armene di testi greci, Alessandria, 313–322. Burnet, J. (ed.) (1902), Platonis Opera: tomus IV Tetralogiam VIII continens, Oxford. Čemčemean, S. (1998), Mayr c‘owc‘ak Hayeren jeṙagrac‘ matenadaranin Mxitareanc‘ i Venetik, 8, Venetik. Conybeare, F. C. (1889), “On the Ancient Armenian Versions of Plato”, in: The Classical Review 3.8, 340–343. Coigneallaig, M. Ó. (1968), “On Verbs of Being in Classical Armenian”, in: J. W. M. Verhaar (ed.), The Verb ‘Be’ and its Synonyms: Philosophical and Grammatical Studies, 3, Dordrecht, 44–52. Cowe, P. (2010), “A Unique Variant on the Trojan War and Founding of Rome According to a Recently-Discovered Armenian Epitome”, in: Revue des Études Arméniennes 32, 1–24. Dahl, Ö. (2000), “Verbs of Becoming as Future Copulas”, in: Id. (ed.), Tense and Aspect in the Languages of Europe, Berlin-New York, 351–361. Dragonetti, M. (1986), “La traduzione armena del Timeo: rapporti tra il ms 1123 e l’edizione a stampa”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 120, 3–33. Dragonetti, M. (1988), “La traduzione armena del Timeo platonico e la tradizione del testo greco”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 122, 49–84.

296

Irene Tinti

Eganyan, Ō. et. al. (eds.) (2004), Mayr c‘owc‘ak hayerēn jeṙagrac‘ Maštoc‘i anowan matenadaranin, 2, Erevan. Finazzi, R. B. (1974), “Note sulla traduzione armena del V libro delle Leggi di Platone”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 108, 203–222. Finazzi, R. B. (1977), “La versione armena del Minosse”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 111, 27–34. Finazzi, R. B. (1990a), “Una traduzione armena di Platone”, in: G. Fiaccadori (ed.), Autori classici in lingue del vicino e medio Oriente. Atti del III, IV e V Seminario sul tema: “Recupero di testi classici attraverso recezioni in lingue del Vicino e Medio Oriente” (Brescia, 21 novembre 1984; Roma, 22–27 marzo 1985; Padova-Venezia, 15–16 aprile 1986), Roma, 65–75. Finazzi, R. B. (1990b), “Versioni armene di testi greci: problemi di lessicologia”, in: G. Fiaccadori (ed.), Autori classici in lingue del vicino e medio Oriente. Atti del III, IV e V Seminario sul tema: “Recupero di testi classici attraverso recezioni in lingue del Vicino e Medio Oriente” (Brescia, 21 novembre 1984; Roma, 22–27 marzo 1985; Padova-Venezia, 15–16 aprile 1986), Roma, 171–177. Fronterotta, F. (ed.) (2006), Platone, Timeo, Milano. Jensen, H. (1959), Altarmenische Grammatik, Heidelberg. Jonkers, G. (1989), The Manuscript Tradition of Plato’s Timaeus and Critias, Dissertation, Vrije Universiteit, Amsterdam. Jowett, B. (ed.) (2004), Timaeus by Plato, web edition (http://creativecommons.org/licenses/ by-nc-sa/2.5/au/) (reproduction of the section devoted to the Timaeus in: Id. [ed.] [1892], The Dialogues of Plato, 3, 3rd edition revised and corrected, Oxford). Karst, J. (1901), Historische Grammatik des Kilikisch-Armenischen, Strassburg. Kostaneanc‘, K. (ed.) (1910), Lettres de Grègoir Magistros, XIme s., Ałēk‘sandrapōl. Kranich, S. / Becher, V. / Höder, S. (2011), “A Tentative Typology of Translation-Induced Language Change”, in: S. Kranich / V. Becher / S. Höder / J. House (eds.), Multilingual Discourse Production. Dyachronic and Synchronic Perspectives, Amsterdam-Philadelphia, 9–44. Lafontaine, G. (1983), “La traduction arménienne des Catégories d’Aristote par David l’Invincible”, in: Le Muséon 96, 133–144. Lozza, G. (ed.) (1994), Platone, Timeo, Milano. Mariès, L. (1928), “Étude sur quelques noms et verbes d’existence chez Eznik, à propos de la page 241”, in: Revue des Études Arméniennes 8, 79–210. Mercier, C. (1978–1979), “L’école hellénistique dans la littérature arménienne”, in: Revue des Études Arméniennes 13, 59–75. Minassian, M. (1996), Grammaire descriptive de l’arménien classique, Genève. Muradyan, G. (2012a), Grecisms in Ancient Armenian, Leuven. Muradyan, G. (ed.) (2012b), “Grigori Magistrosi t‘owłt‘k‘ ew č‘ap‘aberakank‘”, in: Matenagirk‘ Hayoc‘. Žz hator. Ža dar. / Armenian Classical Authors. Volume 16. 11th Century, Erevan, 139–385. Olivieri, M. (1999–2000), Il Secondo Libro del De Providentia di Filone Alessandrino: i frammenti greci e la traduzione armena, Dissertation, Università di Bologna. Olivieri, M. (2000), “Influenze di lessici greci nelle traduzione armene di Filone”, in: Eikasmós 11, 235–247.


297

Pontani, P. (1997), “A Contribution to the Specification of the Greek Lexicons Used by the Translators of the Yownaban Dproc‘”, in: N. Awde (ed.), Armenian Perspectives, London, 191–199. Rivaud, A. (ed.) (2011), Platon, Oeuvres complètes. 10, Timée – Critias, Paris. Rossi, P. (1982–1983), Osservazioni linguistiche e filologiche sulla traduzione armena dell’Apologia di Socrate di Platone, Dissertation, Università Cattolica del Sacro Cuore di Milano. Scala, A. (1999), “Uno scolio al testo delle Leggi di Platone penetrato nell’antica traduzione armena”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 133, 303–309. Scala, A. (2000), “Contributo per un’edizione critica della traduzione armena delle Leggi di Platone: il testo del libro XI nel ms. 1123 di S. Lazzaro e nell’edizione veneziana del 1890”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 134, 225–240. Scala, A. (2001), “A proposito della traduzione armena di Platone, Leggi, XI, 921a 7: una piccola nota filologica”, in: R. B. Finazzi / A. Valvo (eds.), Pensiero e istituzioni del mondo classico nelle culture del Vicino Oriente, Alessandria, 257–258. Scala, A. (2002), “Per una migliore valutazione della traduzione armena delle Leggi di Platone. Considerazioni filologiche sul libro XI” in: V. De Angelis (ed.), Sviluppi recenti nella ricerca antichistica, Milano, 335–343. Sedley, D. (1997), “Plato’s Auctoritas and the Rebirth of the Commentary Tradition”, in: J. Barnes / M. Griffin (eds.), Philosophia Togata II. Plato and Aristotle at Rome, Oxford, 110–129. Sgarbi, R. (1999), “Gradi di congruenza lessicale nella recezione armena di antichi testi greci”, in: Aevum 73, 145–150. Sgarbi, R. (2001), “Il contributo dell’analisi interlinguistica greco-armena alla teoresi sulla traduzione”, in: R. B. Finazzi / P. Tornaghi (eds.), Cinquant’anni di ricerche linguistiche: problemi, risultati e prospettive per il terzo millennio, Atti del IX convegno internazionale di linguisti (Milano, 8–9–10 ottobre 1998), Alessandria, 585–589. Sirinian, A. (1993), “La traduzione armena dei composti verbali greci nell’Orazione 7 di Gregorio di Nazianzo”, in: Le Muséon 106, 89–95. Solari, R. (1969), “La traduzione armena dell’Eutifrone di Platone”, in: Rendiconti dell’Istituto Lombardo Accademia di Scienze e Lettere. Classe di Lettere e Scienze Morali e Storiche 103, 477–499. Sowk‘rean, A. (ed.) (1877), Płatoni Imastasiri Tramaxōsowt‘iwnk‘. Ewt‘ip‘ṙon, Paštpanowt‘iwn Sokratay ew Timēos, i Venetik. Tinti, I. (2012a), “Essere” e “divenire” nel Timeo greco e armeno. Studio terminologico e indagine traduttologica, Pisa. Tinti, I. (2012b), “On the Chronology and Attribution of the Old Armenian Timaeus: a Status Quaestionis and New Perspectives”, in: Egitto e Vicino Oriente 35, 219–282. Tinti, I. (forthcoming), “Problematising the Greek Influence on Armenian Texts”, in: Rhesis. International Journal of Linguistics, Philology and Literature 7.1. Tosi, R. (2003), “Tradizione esegetica nella tarda grecità e traduzioni della Scuola Ellenizzante”, in: A. Sirinian / S. Mancini Lombardi / L. D. Nocetti (eds.), Le scienze e le arti nell’Armenia medievale. Seminario internazionale (29–30 ottobre 2001): Atti, Bologna, 107–111. Vardanyan, S. (ed.) (1979), Išox, Girk‘ i veray bnowt‘ean, Erevan.

298

Irene Tinti

Weitenberg, J. (1993), “The Language of Mesrop: l’Arménien Classique pour lui-même?”, in: C. Burchard (ed.), Armenia and the Bible. Papers Presented to the International Symposium held at Heidelberg July 16–19, 1990, Atlanta, 221–231. Weitenberg, J. (1997), “Linguistic Continuity in Armenian Hellenizing Texts”, in: Le Muséon 110, 447–458. Weitenberg, J. (2001–2002), “Hellenophile Syntactic Elements in Armenian Texts”, in: SLOVO 26–27, 64–72. Zarbhanalean, G. (ed.) (1890), Płatoni Tramaxōsowt‘iwnk‘. Yałags Ōrinac‘ ew Minovs, i Venetik.

Andrea Scala

Greek, Syriac and Armenian in Contact: Lexical and Textual Outcomes Abstract: Plurilingualism is, and was also in the past, one of the most relevant feature of the Armenian culture. The first centuries of the Christian era were no exception: lay élites were plurilingual and spoke at least Armenian and Parthian until the fourth century, and later Armenian and Middle Persian. The Arsacid court – where, together with Armenian and Parthian, Greek also circulated – was likewise plurilingual. In addition to Armenian and Parthian, the high clergy had known Syriac since the fourth century and later, from the beginning of the fifth century, Greek as well. In such a complex landscape it is not surprising that a very large part of the Armenian lexicon consists of loanwords and that many old Armenian translations show features depending on their Syriac or Greek sources, and sometime on both. Such a situation suggests that in order to understand the history of the Armenian lexicon as a means to reconstruct the genesis of the textual redactions of some works, especially the books of the Bible, it is often necessary to adopt an approach that takes into consideration three or more languages. The article discusses some lexical and textual outcomes of the interaction between Armenian, Greek and Syriac in Late Antiquity, with particular focus on the phonetic characteristics of some Greek and Syriac loanwords in Armenian and on the text of Acts 22.27–28, in which the Armenian translation seems to be related in some manner to a Syriac model.

Over the centuries, one of the most important features of Armenian culture has always been its plurilingualism. The first centuries of the Christian era made no exception: lay elites were plurilingual and, until the fourth century, spoke at least Armenian and Parthian, while subsequent to this, both Armenian and Middle Persian. The Arsacid court was plurilingual as well where, together with Armenian and Parthian, Greek also circulated. In addition to Armenian and Parthian, the high clergy would have known Syriac since the fourth century and, from the beginning of the fifth century, also Greek. Moreover, Imperial Aramaic and Greek were present in the Armenian plateau as international languages respectively since the Achaemenid and Hellenistic time. In such a complex landscape it is not surprising that a very large part of the Armenian lexicon consists of loanwords. The corollary that results from this sociolinguistic situation is very simple: in order to understand and to reconstruct the history of the Armenian

300

Andrea Scala

lexicon it is often necessary to have a point of view that considers three or more languages. In his very accurate studies on the Armenian lexicon, Giancarlo Bolognesi¹ points out that for some Greek loanwords in Armenian one must admit a passage through Parthian. So what at a first glance seems to be a Greek loanword must sometimes be considered as an Iranian loanword of prior Greek origin. This is the case for instance with Arm. kałapar, “model” and lampar, “torch” where the presence of -r- instead of -d- that we find in the Greek primary lexical sources καλοπόδιον and λαμπάς, -άδος points to Parthian intermediation². In other cases it is irrefutable that some words of Greek origin enter Armenian through the intermediation of Syriac³: cfr. astratigay, “military chief” < Syr. esṭraṭīghā (from Gr. στρατηγός) and p‘ilisop‘ay, “philosopher” < Syr. pīlōsōphā, (from Gr. φιλόσοφος). This field of research has been less explored than it deserves. Of course the study of the phonological integration of the loanwords is crucial for establishing the path followed by a word before becoming established setting in the Armenian lexicon. In fact a close examination of the phonetic changes that occurred in the loanwords on account of their phonological integration permits the individuation of some criteria for evaluating the first and last origin of a foreign word. In addition to the lexical products of plurilingualism, other spaces of multilingual interference were active in ancient Armenia, especially in the redaction of texts. Some textual passages in Armenian biblical translation are difficult to justify by referring to merely one linguistic tradition and seem to imply a multiple Vorlage.

1 Lexical interference: the reflex of the Syriac voiceless stops in Armenian Limiting the discourse to the interference between Armenian and Greek via Syriac at the time of the first Armenian documents, let us begin from the lexicon and take into consideration the cases of Arm. k‘oṙ, “a grain measure” (e.g. Num. 11.13 and many other occurrences) and p‘egenay/p‘eganay, “rue” (Lc. 11.42). Both lexemes are listed by Birgit Olsen among the Greek loanwords in biblical Armenian⁴ and

1 On the work of G. Bolognesi, cf. the contribution by M. Morani, in this volume. 2 Bolognesi 1990, 291. 3 Bolognesi 1990, 292–293; Morani 2011, 132–133. 4 Olsen 1999, 928.

Greek, Syriac and Armenian in Contact: Lexical and Textual Outcomes

301

the Greek models for the loanwords respectively are κόρος e πήγανον, but it is reasonable to doubt their direct Greek origin. To some extent, Hübschmann was already of a different opinion about p‘egenay/p‘eganay and in his Armenische Etymologien included it among the syrische Wörter⁵, whereas in discussing the origin of k‘oṙ he proposed a Greek model χόρ⁶, attested to only twice in the whole of ̌ howkyan Greek literature. With regard to p‘egenay/p‘eganay, “rue” Bolognesi, Ja and Morani⁷ follow Hübschmann’s opinion. A fact that immediately attracts our attention is the presence in the alleged Greek models of initial voiceless stops, which are reflected in the Armenian loanwords by aspirate voiceless stops. This reflex is anomalous for the Greek loanwords, where Greek voiceless stops are usually reproduced through Armenian voiceless stops, remaining in fact unchanged. Some clear examples can be found in Arm. kēt, “sea monster” (Gen. 1.21; Esth. 14.5; Job 9.13; Dan. 3.77), krkēs, “circus” (2 Macc. 4.9), pandoki, “inn” (Lc. 10.34), pnak, “platter” (Lc. 11.39) from κῆτος, κιρκήσιον, πανδοκεῖον e πίναξ (acc. πίνακα). But in the case of loanwords from Syriac, which in the model have a labial or velar voiceless stop in word initial position, their reproduction in Armenian with an aspirated voiceless stop is quite normal: cfr. Arm. k‘ahanay, “priest” and p‘rkel, “to save” from Syriac kāhnā e pəraq, “he saved”. Moreover, in the final -ay p‘egenay/p‘eganay shows a trace of the Syriac emphatic state, which is well attested also in other words⁸. Thus it is more probable that k‘oṙ and p‘egenay/p‘eganay have been borrowed from Syriac (cf. Syr. kor/kur and peghanā) rather than from Greek. The form χόρ quoted by Hübschmann has only one occurrence outside Epiphanius Περὶ μέτρων; this single occurrence is in Heron De mensuris 61.1, where we read κόρος μόδιοι λ· παρ’ Ἑβραίοις δὲ χὸρ λέγεται. So Gr. χόρ can be considered as a mere transcription from a Semitic model. According to these criteria p‘egenay/p‘eganay has a prior Greek origin but comes to Armenian from Syriac and for the same reason k‘oṙ is unlikely as a direct loanword from Greek. Starting from these examples it can be useful to observe in detail how the voiceless stops in word initial position are reproduced in the Armenian loanwords from Syriac and Greek. Let us consider the graphic evidence:

5 Hübschmann 1897, 317. 6 Hübschmann 1897, 388. 7 Bolognesi 1990, 292; Jahowkyan 1987, 479; Morani 2011, 132. 8 Bolognesi 1990, 292; Morani 2011, 132.

302

Andrea Scala

Gr.

τ ↓ ‫ܛ‬ ↓

θ ↓ ‫ܬ‬ ↓

κ ↓ ‫ܩ‬ ↓

χ ↓ ‫ܟ‬ ↓

տ

թ

կ

ք

Syr. Arm.

π

φ ↓ ‫ܦ‬ ↓

փ

For the Greek loanwords in Syriac we can quote the following examples: dental stops: cfr. τέχνη, “art” > ‫ ܛܟܢܐ‬ṭekhnā; τρισκελής, “tripod” > ‫ ܛܪܝܣܩܐܠ‬ṭrīsqelā; θέατρον, “theater” > ‫ ܬܐܛܪܘܢ‬te‘aṭrōn; θήκη, “container” > ‫ ܬܝܩܐ‬tīqā; velar stops: cfr. καταδίκη, “judgment” > ‫ ܩܛܕܝܩܐ‬qaṭādhīqā; κορμός, “trunk” > ‫ ܩܘܪܡܐ‬qōrmā; χειμών, “storm” > ‫ ܟܝܡܘܢ‬kīmōn; χριστιανός “a Christian” > ‫ܟܪܝܣܛܝܢܐ‬ kərīsṭyanā; labial stops: cfr. ποιητής, “poet” > ‫ ܦܘܐܝܛܐ‬pō‘yeṭā, ποδάγρα, “gout” > ‫ܦܕܓܪܐ‬ podhaghrā; φιάλη, “cup” > ‫ ܦܝܐܠ‬piyalā; φιλόσοφος, “philosopher” > ‫ܦܝܠܘܣܘܦܐ‬ pīlōsōphā. For the Syriac loanwords in Armenian there are these instances: ‫ ܛܪܦܐ‬ṭarpā, “leaf” > տերեւ terew; ‫ ܬܓܪܐ‬taggara, “trader” > թանգար t‘angar; ‫ ܩܛܘ‬qeṭṭau, “flax” > կտաւ ktaw; ‫ ܟܗܢܐ‬kahnā, “priest” > քահանայ k‘ahanay; ‫ܦܪܩ‬ pəraq, “he saved” > փրկել p‘rkel. For the Greek loanwords in Armenian we find the following correspondences: Gr.

τ ↓

θ ↓

κ ↓

χ ↓

π ↓

φ ↓

Arm.

տ

թ

կ

ք

պ

փ

Cfr. τάρταρος, “hell” > տարտարոս tartaros; θέμα, “province, district” > թեմ t‘em; καρκίνος, “crab” > կարկինոս karkinos; χάραξ (acc. χάρακα), “pole” > քարակ k‘arak; παγκράτιον, “pancratium” > պանկրատիոն pankration; φώκη, “seal” > փոկ p‘ok. The comparison between the integration of Greek and Syriac word initial voiceless stops shows us that the intermediation of Syriac for Greek words cannot be established for dental and velar stops, where the outcomes are identical, whereas the labial voiceless stops permit a more solid hypothesis about the real origin


303

of a loanword. A loanword having a labial voiceless պ [p-] in word initial position cannot be traced back to Syriac while a loanword, lexically Greek, with a labial aspirated voiceless փ [ph-] in word initial position in Armenian has undergone a Syriac intermediation if in Greek we have a non-aspirate labial voiceless π [p-]. The phonetic content of these graphic correspondences is problematic and could perhaps be reconstructed as follows: Gr. Syr. Arm.

[t-] ↓ [tˤ] ↓ [t]

[th]/[θ-] ↓ [t(h)] ↓ [th]

[k-] ↓ [q] ↓ [k]

[kh]/[x] ↓ [k(h)] ↓ [kh]

[p] [ph]/[f] ↓ [p(h)] ↓ [ph]

[t] ↓ [t]

[th]/[θ] ↓ [th]

[k] ↓ [k]

[kh] /[x] ↓ [kh]

[p] [ph]/[f] ↓ ↓ [p] [ph]

and Gr. Arm.

A preliminary remark that must be made is that Gr. χ is never integrated through Syr. ‫ܚ‬, probably because this latter represented a voiceless pharyngeal fricative. In Armenian loanwords from Greek we find խ only in one case, moreover rather uncertain (i.e. Arm. տառեխ taṙex from Gr. τάριχος, “salted fish”), while the usual integration of χ is with the aspirate voiceless stop ք; this fact could mean that the pronunciation of Gr. χ in the Greek variety with which Armenian was in contact, was not fricative but affricate [kx-] or aspirate [kh-], perhaps according to a school pronunciation, or that Arm. խ did not represent a velar fricative but a pharyngeal fricative. However the rendering of θ, χ and φ with aspirated voiceless stops does not reassure us about the conservation of such a pronunciation for these Greek consonants, because a phonological adaptation of a voiceless fricative with aspirated voiceless stops would not surprise, and in fact can still be heard today when Armenians not very familiar with English pronunciation replace [θ] with [th], for instance pronouncing as [thɔt] instead of [θɔt]. Moreover the possible presence of the aspiration in Syriac dental and velar non-emphatic voiceless stops is a redundant feature from the phonological point of view, but the reproduction of these sounds in Armenian through aspirated stops prompts us to consider this feature as highly salient. The case of the labial voiceless stop is very eloquent in this regard: in Syriac it was the only labial voiceless stop, there was no emphatic counterpart, and in Armenian it is regularly

304

Andrea Scala

reproduced with an aspirated labial stop. Probably all non-emphatic occlusive segments in Syriac were aspirated. As for the loanwords in Armenian having a labial in word initial position, the correspondences presented above point to the necessity of reconsidering the origin of Arm. połotay, “broad street, avenue”. The loanword is listed among the Syriac loanwords by Hübschmann⁹, but accompanied by the question “aus dem Syr. oder Griech.?”, and also by Ačaṙean¹⁰. However, the Syriac model plāṭīā should have as its outcome a form with a labial aspirated voiceless stop in word initial position, so it is more probable that the word has come into the Armenian lexicon directly from the Gr. πλατεῖα. In this case the final -ay in Armenian does not compel us to suppose a Syriac model in emphatic state. With regard to the vocalism, it may be seen that the correspondence between gr. a and arm. o also appears in other loanwords such as bonosos, “aurochs” < Gr. βόνασος; korołiovn, “coral” < Gr. κοράλλιον; monozon, “monk” < Gr. μονάζων; siwłobay, “syllable” < Gr. συλλαβή. It has been supposed that this fact reflects a feature of the pronunciation of the Greek attested in the Syriac cultural area since the fourth century¹¹. Nor is the epenthesis unknown in some loanwords such as Hoṙom, “Roman”, cfr. Hṙom, “Rome”. Moreover, according to the correspondences shown by velar sounds in the Greek or Syriac model as well as in Armenian, some other loanwords such as Arm. k‘anown, “rule” and k‘oweay, “cube, die” cannot derive directly either from Gr. κανών and κυβεία or from Syr. qānūn and qubhāyā. The intermediation of an Iranian language is perhaps a reasonable way to explain the initial k‘- in these Armenian words.

2 Textual interference: the case of Acts 22.27–28 In addition to lexical outcomes the contact between Syriac, Greek and Armenian was also of relevance in the translation of the Bible. As for the history of the Armenian Bible one of the most controversial points is even the role of the Syriac Bible in the process of formation of the Armenian text¹². This role, actually acknowledged by most scholars, has often been minimized for some books of the Bible¹³. The final text of the Armenian Bible proves to be the result, in

9 Hübschmann 1987, 315. 10 Ačaṙean 1971–1979, s.v. 11 Bubeník 1989, 229 and 254 n. 23; Orengo 2010, 462 n. 57. 12 Lehmann 1982; Cowe 1990–1991; Id. 1993; Id. 1999; Lucca 2010, 48–53. 13 Cox 2006.


305

many cases, of an accurate revision on the Greek (Arm2) of a previous translation (Arm1), which often depended on the Syriac¹⁴. Of course this general idea should not be extended to every book indiscriminately, cf. e.g. the case of the two versions (Arm1 and Arm2) of the Chronicles, which both depend, in the opinion of Cowe and Lucca¹⁵, on a Greek Vorlage, which may perhaps be traced back to the Antiochian milieu¹⁶. However, every new investigation implying a systematic comparison of Syriac, Greek and Armenian texts seems to suggest that the Syriac model has left some linguistic, textual and cultural traces in the Armenian Bible. From this perspective we can take into consideration the Armenian translation of The Acts of the Apostles 22.27–28. In this passage the difference between the Greek text and the Armenian translation is considerable and may possibly be explained via Syriac. The facts told by Luke in chapter 22 of The Acts are very famous: Paul held a long discourse in the Temple of Jerusalem, narrating the most important moments of his conversion. The crowd begins to rail against him and the tribune who led the temple’s Roman garrison arrests Paul and decides to interrogate him. Paul is bound and a centurion threatens to beat him with a whip. At that point Paul reminds the centurion that it is not lawful to beat a Roman citizen prior to judging him. The centurion immediately calls the tribune and between him and Paul the following dialogue arises: Λέγε μοι, σὺ Ῥωμαῖος εἶ; ὁ δὲ ἔφη, Ναί. ἀπεκρίθη δὲ ὁ χιλίαρχος, Ἐγὼ πολλοῦ κεφαλαίου τὴν πολιτείαν ταύτην ἐκτησάμην. ὁ δὲ Παῦλος ἔφη, Ἐγὼ δὲ καὶ γεγέννημαι. “Tell me, are you a Roman citizen?” Paul answered, “Yes”. To this the tribune replied, “It cost me a large sum to acquire this citizenship.” “But I was born to it,” said Paul. (Acts 22.27–28).

The Armenian translation is quite different: Asa dow c‘is, Hṙovmayec‘i es? Ew na asē, ayo. Patasxani et hazarapetn, es bazowm glxoc‘ zzinoworowt‘iwns zays stac‘ay. Asē Pawłos, es cneal isk i dmin em. “Tell me, are you a Roman citizen?” Paul answered, “Yes”. To this the tribune replied, “It cost me a large sum to acquire this military service.” “But I was born to it,” said Paul.

The rendering of τὴν πολιτείαν ταύτην, “this citizenship” with zzinoworowt‘iwns zays “this military service, hanc militiam” is surprising and makes Paul’s reply es cneal isk i dmin em “but I was born to it”, rather obscure. Since the only possible antecedent for the anaphoric expression i dmin is zzinoworowt‘iwns “this mili-

14 Lucca 2010, 48–53. 15 Cowe 1990–1991; Lucca 2002 and 2007. 16 Cf. Lucca 2002, 183–185.

306

Andrea Scala

tary service” stated by the tribune, the result is that Paul is said to have been in the army since birth, which is inconsistent with the context and the cotext. The Alexanian edition of the Armenian Acts shows us that the Armenian manuscript tradition has no significant variants in this passage¹⁷ and the first hypothesis that comes to mind is that the Greek model could contain a reading τὴν ὁπλιτείαν ταύτην. Nevertheless such a variant is totally absent, to my knowledge, from the Greek manuscript tradition of The Acts and seems very improbable as a misreading of the Armenian translators/reviewers. The substantive ὁπλιτεία is a word unknown to the Septuagint and to the Greek New Testament, as well as being very rare in the Greek literature, where we find only two occurrences in works written before the Armenian translation of the Bible (Pl. Leg. 706c 8 and Hdn. Pros. cath. 291.12 Lentz). Such a word, morphologically transparent but very rare, is easier to understand than to produce as a misreading, whereas it is credible as the result of a conjecture on the part of the translators, based on the input of the Syriac or of a previous Armenian version having a Syriac Vorlage. The problematic character of the Armenian translation of this passage could depend on the difficulty in rendering the political and civil concept of πολιτεία, “citizenship”, which was probably extraneous to the Armenian world in the fifth century, as well as, in this case for religious reasons, to the Hebrew world, and consequently, to the Hebrew-Christian culture of Late Antiquity. It may be useful to compare the Armenian text of this passage with the Syriac Bible. As is known, the possible model for first Armenian translation of the Bible (Arm1) was a Syriac text, known as the Old Syriac version (or Curetonian text). There are extant manuscripts of the Old Syriac Version for the Gospels, but not of the Acts¹⁸. In this case we can tentatively use the Peshitta. The two Syriac versions are often very close. Here is the text of Acts 22.27–28 in the Peshitta: ᵓemmar lī ᵓa(n)t r(h)ūmōyō ᵓa(n)t we(ᵓ)mmar leh ᵓīn waᶜnō killīarkhō we(ᵓ)mmar ᵓennō bəkhespō saggīō(ᵓ) qənīthōh lər(h)ūmōyūthō ᵓāmar leh pawlōs ᵓennō dēn ᵓōp bōh ᵓethyaldeth “Tell me you, are you a Roman citizen?” Paul answered, “Yes”. The tribune replied and said, “It cost me a large sum to acquire this Romanity.” “But I was born to it,” said Paul.

In this Syriac version, made prior to the schism that followed the Council of Ephesus in 431, πολιτεία has been translated with r(h)ūmōyūthō an abstract noun (morphologically comparable with Lat. romanitas), derived through the derivative morpheme –ūth from r(h)ūmōyō “Roman”¹⁹. A look to the Syriac lexicog-

17 Alexanian 2012, 68. 18 Alexanian 2012, XXIV. 19 About the suffix –ūth and its functions cfr. Nöldeke 1880, 76–77.


307

raphy informs us that r(h)ūmōyō, in addition to “Roman”, also means “soldier, miles”²⁰, with a historical continuity that reaches the Neo-Aramaic dialects where we find rimōyō, “policeman”²¹. In accordance with this meaning of r(h)ūmōyō, it is not difficult to imagine that an abstract noun derived from it through the derivative morpheme –ūth could potentially be interpreted as “*soldierness”, that is, “military service”; for a morphological parallel derivation cfr. Lat. miles, “soldier”, and the abstract militia, “military service”. The Syriac lexicography confirms this possibility as historically documented and r(h)ūmōyūthō is glossed as “military service” and “militia”²². It is noteworthy that in Acts 23.23 and 23.27, i.e. in the chapter following our passage, Gr. στρατιώτης, “soldier” and στράτευμα, “army” are translated in the Peshitta with just r(h)ūmōyō “Roman(s)”. From this paradigmatic net arises, in my opinion, the zinoworowt‘iwn, “military service” of the Armenian Bible. The revision of the Armenian translation on the basis of the Greek text probably did not in this passage modify the translation that issued from a previous Syriac model. It is possible to suppose, as stated above, that the reviewers facing this discrepancy between Greek and Arm1 thought that πολιτεία should be corrected to ὁπλιτεία. The major problem is that the Armenian text makes no sense, and the idea that a previous translation based on Syriac has been accepted by the reviewers on the basis of metaphors such as Paul = soldier of Christ does not make the text more consistent. It is incontestably true that a rich metaphoric field in which the Christian faith is represented as a military service is well attested in Paul’s Epistles (1 Tim. 1.18; 2 Tim. 2.3–4; Philm. 1.2; Phil. 2.25; 1 Cor. 9.7; and cf. also the documentation provided by Bauernfeind²³). In any case Paul could not be considered as a soldier of Christ right from his birth. In whatever way we wish to reconstruct the unattainable thread of facts that occurred in the studia of the Armenian translators when they were translating or reviewing this passage, it is probable that the difficulty was the semantic interpretation of gr. πολιτεία. An exploration of all the biblical occurrences of πολιτεία and its paradigmatic cognates πολίτευμα and πολιτεύομαι shows us that they have a clear religious meaning (cfr. 2 Macc. 6.1, 4.11 and 11.25; 3 Macc. 3.4; 4 Macc. 2.8, 8.7, 17.9 and 2.23; Ac. 23.1; Eph. 2.12; Phil. 1.27 and 3.20). In particular the Greek lexeme πολιτεία receives in the Bible the new semantic acceptation of “all the religious precepts as a whole respected by the pious men”, precepts that were mainly based on the Mosaic law. This new meaning arose in Biblical

20 Brockelmann 1928; Payne Smith 1879–1901; Sokoloff 2009, s.v. 21 Brockelmann 1908, 391. 22 Brockelmann 1928; Payne Smith 1879–1901; Sokoloff 2009, s.v. 23 In TLNT 8, 701–714.

308

Andrea Scala

Greek from the classical meaning of “rights of a citizen, civil polity, constitution of a state”²⁴ through a process of interference with Jewish linguistic and cultural taxonomies, these latter pre-existing the Hellenization of the Jewish educated classes. A glance at the Armenian translation shows us that the renderings of πολιτεία with krawn(k‘), “religious customs” (cf. 2 Macc. 4.11) and with aṙak‘inowt‘iwn, “virtue” (cf. Eph. 2.12) are clearly oriented towards a religious acceptation. The classical civil and lay meaning of “citizenship” seems to occur in the Bible only in 3 Macc. 3.21 and 3.23 together with Ac. 22.28. In 3 Macc. 3.21 and 3.23 Ptolemy Philopator speaks about the concession of the Alexandrian citizenship to the Hebrews, who refused it disdainfully. Here the Armenian translation is particularly free in rendering ἐβουλήθημεν καὶ πολιτείας αὐτοὺς ᾿Αλεξανδρέων καταξιῶσαι (3 Macc. 3.21), “and we wanted to confer on them the Alexandrian citizenship” with hramayec‘ak‘ Ałēk‘sandrac‘woc‘ k‘ałak‘akic‘ aržanis aṙnel i hamari, “we ordered the Alexandrians to consider them as worth being their fellow citizens” and οὐ μόνον ἀπεστρέψαντο τὴν ἀτίμητον πολιτείαν (3 Macc. 3.23), “not only did they refuse the inestimable citizenship” with oč miain i bac‘ meržec‘in zereweli patiw mecaranac‘ ənkerowt‘eann, “not only did they refuse the magnificent honour of the friendship’s privileges”. In this case too the Armenian translation does not offer an equivalent to Gr. πολιτεία, which has a clear legal and civil meaning, even though in 3.21 the paraphrasis of the passage contains a word such as k‘ałak‘akic‘, “fellow citizen”, which is not too far from a lay dimension. In any case the cotext reinforces the idea that the πολιτεία does not pertain to the lay sphere: the Hebrews refuse the citizenship, affirming that their citizenship, their πολιτεία, is not based on civil requirements but on religious ones. So let us draw some conclusions: the rendering of πολιτεία by krawn(k‘), “religious customs” shows that the religious sense of πολιτεία present in Jewish Greek was known to the Armenian translators. In Ac. 22.27–28 the assumption of a religious meaning of πολιτεία would provide a great contrast with the context and the cotext. It is the tribune who claims to have acquired the πολιτεία through the payment of a large sum. In the Peshitta the translation of πολιτεία refers to the civil and legal dimension it had in the Roman Empire. The Armenian translators were probably unable to solve the enigmatic contradiction present in a text in which Paul and the tribune affirm that they have the same πολιτεία. Of course the contradiction exists if we consider πολιτεία only in its religious sense. The comparison with a possible Arm1 or with a Syriac version was not enough for the translators/reviewers to overcome the obstacle. The outcome is that Arm. zinoworowt‘iwn, which arose from Syr. r(h)ūmōyūthō, “Romanity” interpreted in

24 Biscardi 1984, 1203–1205.


309

its secondary meaning of “military service”, has been conserved in the revision (or Arm2), due perhaps to the view that τὴν πολιτείαν ταύτην should be amended to in τὴν ὁπλιτείαν ταύτην. In the light of this framework we can propose some final considerations of semantic and cultural nature: firstly, for the Armenian Bible translators/reviewers, who were the most refined Armenian-speaking intellectuals of the time, the semantics of Gr. πολιτεία depended substantially on Biblical Greek and in the cultural landscape of the Armenians the concept of civil citizenship seems to be rather exotic. In this respect Armenian culture proves to be closer to the categories of the Iranian or Jewish-Hellenistic world rather than to those of the Roman world. On the contrary, the Syriac translators recognized the polysemy of πολιτεία in the Greek of their time and were able to distinguish between the reference to a civil dimension and a religious one. Secondly, it is possible to maintain that the civil and legal conceptualization (and acceptation) of Gr. πολιτεία had undergone a crisis after the concession of citizenship to the freemen dwelling in the limes (edict of Caracalla, 212 AD, called Constitutio Antoniniana, that seems to exclude from the citizenship only the dediticii²⁵). In an empire where all inhabitants are citizens, the concept of πολιτεία must have been reshaped and so employed as to indicate the difference between being inside or outside of the imperial borders. Of course this distinction made no sense for those who were outside the borders, as was the majority of Armenians. Finally Greek, like every language, was not a monolithic system, with no inner variation. On the contrary, many varieties of Greek circulated in Late Antiquity, where this language presented a diasystem with diatopic, diamesic, diaphasic, diastratic and ethnic variation. For the first two generations of Armenian translators, knowledge of Greek was based mainly on Biblical and, perhaps, exegetical texts, a variety of Greek where the civil dimension of a term such as πολιτεία is almost inexistent. Syriac translators and Armenian translators faced with Ac. 22.27–28 read the same text but understood it differently.

Bibliography TLNT = G. Kittel / G. Friedrich (eds.) (1933–1973), Teologisches Wörterbuch des Neuen Testament, 1–9, Stuttgart. Ačaṙean, H. (1971–1979), Hayerēn armatakan baṙaran, 1–4, Erevan. Alexanian, J. M. (2012), The Ancient Armenian Text of the Acts of the Apostles, Lovanii.

25 Cfr. Montevecchi 1998, 362.

310

Andrea Scala

Biscardi, A. (1984), “Polis politeia politeuma”, in: Atti del XVII congresso internazionale di papirologia, Napoli, 3, 1201–1215. Bolognesi, G. (1990), La tradizione culturale armena nelle sue relazioni col mondo persiano e col mondo Greco-romano, in: Id., Studi glottologici filologici orientali, Brescia, 271–318. Brockelmann, C. (1908), “Corp. script. Christ. Orient. Currant. J.-B.-Chabot, I. Guidi, H. Hyvernat, B. Carra de Vaux. Scriptores Syri, Series secunda, Tomus XXVIII. Philoxeni Mabbugensis tractatus de trinitate et incarnatione, ed. et interpretatus est A. Vaschalde – Series tertia, Tomus XXV. Vitae virorum apud Monophysitas celeberrimorum, ed. et interpret. est E. W. Brooks”, in: Zeitschrift der Deutschen Morgenländischen Gesellschaft 62, 388–397. Brockelmann, C. (1928), Lexicon Syriacum, 2nd ed., Halle. Bubeník, V. (1989), Hellenistic and Roman Greece as a Sociolinguistic Area, Amsterdam-Philadelphia. Cowe, P. (1990–1991), “The Two Armenian Versions of Chronicles, their Origin and Translation Technique”, in: Revue des Études Arméniennes 22, 53–96. Cowe, P. (1993), “Problematics of Edition of Armenian Biblical Texts”, in: H. Lehmann / J. J. Sicco Weitenberg (eds.), Armenian Texts Tasks and Tools, Aarhus, 26–37. Cowe, P. (1999), “Text Critical Investigation of the Armenian Version of Third Corinthians”, in: V. Calzolari Bouvier et al. (eds.), Apocryphes arméniens, Lausanne, 91–102. Cowe, P. (2014), “The Armenian Version of the New Testament”, in: B. D. Ehrman / M. W. Holmes (eds.), The Text of New Testament in Contemporary Research, Leiden-Boston, 251–292. Cox, C. (2006), Armenian Job, Leuven. Cox, C. (2014), “The Armenian Bible: Status Quaestionis”, in: V. Calzolari (ed.), Armenian Philology in the Modern Era, Leiden-Boston, 231–246. Hübschmann, H. (1897), Armenische Grammatik, Leipzig. Jǎ howkyan, G. B. (1987), Hayoc‘ lezvi patmowt‘yown, Erevan. Lehmann, H. J. (1982), “Some Questions Concerning the Armenian Version of the Epistle of James”, in: Acta Jutlandica 56, 57–82. Lucca, P. (2002), “1–2 Cronache nella versione armena della Bibbia: dipendenze testuali e tecniche di traduzione”, in: Bazmavēp 160, 150–187. Lucca, P. (ed.) (2007), La prima versione armena di 1–2 Cronache: edizione critica, Dissertation, Venezia. Lucca, P. (2010), “Versioni armene di testi siriaci. Breve panoramica sulle traduzioni armene dal siriaco nei secoli V-XIII e loro caratteristiche”, in: Annali di Ca’ Foscari 49/3, s. or. 41, 43–62. Montevecchi, O. (1998), “Note sull’applicazione della Constitutio Antoniniana in Egitto”, in: Ead., Scripta selecta, a cura di S. Daris, Milano, 355–369. Morani, M. (2011), “Prestiti siriaci in armeno. Alcune riflessioni”, in: L. Busetto / R. Sottile / L. Tonelli / M. Tosco (eds.), Studies in Language and African Linguistics in Honour of Marcello Lamberti, Milano, 123–142. Nöldeke, T. (1880), Kurzgefasste syrische Grammatik, Leipzig. Olsen, B. (1999), The Noun in Biblical Armenian, Berlin-New York. Orengo, A. (2010), “L’armeno del V secolo. Note per una storia della lingua armena”, in: R. Ajello / P. Berrettoni / F. Fanciullo / G. Marotta / F. Motta (eds.), Quae omnia bella devoratis. Studi in memoria di Edoardo Vineis, Pisa, 447–468. Payne Smith, R. (1879–1901), Thesaurus Syriacus, 1–2, Oxford. Sokoloff, M. (2009), A Syriac Lexicon. A Translation from the Latin, Correction, Expansion, and Update of C. Brockelmann’s Lexicon Syriacum, Winona Lake-Piscataway.

Rosa Ronzitti

Etymological Hypotheses Between Armenian and Indo-Iranian: New Proposals and Remarks Abstract: The publication of the Etymological Dictionary of the Armenian Inherited Lexicon by Hrach K. Martirosyan offers the opportunity to posit new etymologies with the aid of further and helpful investigations not only of linguistic but also of onomasiological and ethnological nature, e.g. astełkʻ molarkʻ, “wandering stars”, astucoy eznak, “ladybird”, babič‘, “witch-doctor”, cʻncam, “to rejoice”.

1. The publication of the Etymological Dictionary of the Armenian Inherited Lexicon by Hrach K. Martirosyan, which came out in 2010, offers the opportunity to reflect that even in the present day it is by no means unfeasible to embark on a wide-ranging etymological project, and that such an undertaking can be accomplished with the aid of further and helpful investigations not only of a linguistic but also onomasiological and ethnological nature. 2. Martirosyan has the great merit of presenting the headwords in transliteration, thereby enabling those who do not have a good command of the Armenian alphabet to benefit from direct access to the Lexicon, which has been broadened by the inclusion of dialect material. The work thus takes on a completely new aspect ̌ hukian, who focused as compared to the historical studies of Ačaryan and Ja mainly on the classical language. In the new dictionary, many of the entries provide a detailed description and a wealth of information, and some of them are actually the first scientifically acceptable and complete treatment of the word involved. The decision to illustrate only the inherited vocabulary inevitably implies a selection upstream, which means that the reader cannot become acquainted with the terms that have been excluded, namely the loanwords from Persian and Syriac. Furthermore, on closer inspection, it may seem that not all the entries are genuinely inherited (echtarmenisch) and, vice versa, that some inherited terms do not appear in the Lexicon because they are classified as loanwords. 2.1. First of all, one may have reservations concerning the inclusion of ampar, “planet” (pp. 51–52), a term which – on the assumption it is not a ghost-word – will be critically re-examined here. The word ampar appears only in an Armenian

312

Rosa Ronzitti

dictionary published by Łewond Ališan in 1910 in Venice and is drawn from an unspecified author, who could, in turn, have taken it from the Commentary on the Genesis by Ełišē. Since ampar forms part of the noun phrase ampar astełkʻ, it would seem logical to interpret it as a preposed adjective, without any plural marking, and qualifying astełkʻ, “stars”. After rejecting the earlier idea that ampar should be segmented as an-par, “motionless”, Martirosyan proposes a link with amp, “cloud”, thereby giving the term the meaning of “celestial body” (“the heavenly one”). However, if we take ampar as an Iranian loanword, we can formulate a less generic etymology, more appropriately aligned with the motivation underlying both the Greek πλάνης, “vagabond” and also another Armenian expression, astełkʻ molarkʻ, “wandering stars” = “planets”, which can be contrasted with astełkʻ anmolarkʻ, “non wandering stars”, i.e. the so-called “fixed” stars¹. Wandering is a characteristic of the planets, both properly speaking (their irregular and backwards-facing orbits give the impression of being random characteristics) and also in a figurative sense, since it can be seen as an anarchic violation of a celestial order which, on the other hand, is respected by the stars themselves. In Iranian thought of the Sasanid period, this fact led to the well-known demonization of planetary bodies. A glimpse of its origin can be perceived as early as in the Avesta, which manifests revulsion against shooting stars, described as “worm” stars (kǝrǝma-): pairikā̊ … yā̊ stārō kǝrǝmā̊ patanti antarə za̜m asmanəmča, “the witches that fall between heaven and earth like worms” (these are the words of the hymn to Syrius Yt. 8.8, cited in accordance with Bartholomae)². Panaino summarises the issue as follows: “The duel between the forces of good and the forces of evil, between cosmic order and disorder, between life and death is thus

1 Greek πλάνης is generic and means “wandering”. In Xen. Mem. 4.7.5 and Arist. Mete. 342b 28, one finds the syntagm πλάνητες ἀστέρες. Furthermore, Plato already used ἄστρα πλανητά in the neuter in Leg. 821b and Ti. 38c. Of these two passages, the first is extremely interesting because it highlights the motive underlying the term, in order to refute it: in fact, in the dialogue between Clinia and the Athenian, the latter shows the mistake inherent in the idea that some celestial bodies seem to wander across the sky without ever passing along the same route again. The passage from Timaeus is a meditation on the origin of time and the cosmos: the birth of time is set in relation with the generation of the sun, the moon and “five other stars that are called planets” (Mercury, Venus, Mars, Jupiter, Saturn). 2 Bartholomae 1904, 469. It should be noted that the word “worm” appears only in this one place in the Avesta that has come down to us. For the translation and the commentary of Yt. 8.8 cf. Panaino 1990, 34 and 97. On the battle of the stars hurled by the spirit of evil against Sirius and the fixed stars, cf. again Panaino 1995, 19 ff. (with a specific reversal: in the Qur’ān the shooting stars are, by contrast, hurled against the demons, cf. 67.5 and passim).

Etymological Hypotheses Between Armenian and Indo-Iranian

313

reproduced, in a manner that conforms to the typical framework of Zoroastrian dualism, on the astral plane as well, according to a cosmological conception that identifies the (apparent) regular movement of the fixed stars as a manner in which the cosmic order (aṣ̌ a) is made manifest (whereas it attributes a specifically demonic significance to the disordered movement of shooting stars and of all the astral bodies whose orbit is unpredictable). This conception would subsequently be re-elaborated in the Sasanid context, when the planets would take over from the shooting stars the role as antagonists of the fixed stars […]”³. In effect, Middle Persian āxtar, “star” is simply a back-formation from abāxtar, “planet”, i.e. “that which faces backwards”, with deglutination of ab-, which becomes reinterpreted as a negative prefix⁴. Thus, the planet is defined ex negativo: it is the non-star (variant: ne āxtar) or even, in that cultural framework, the gēg, the “thief, bandit”. The word abāxtar itself derives from Old Iranian *apāxtara-, “backwards-facing, backwards-looking”, which in turn comes from a common Indo-Aryan form *apāk- or *apānk-, “backwards-facing”. Precisely because it refers first and foremost to the spatial dimension, the name has also remained in the meaning of “north”⁵, suggesting that the framework of orientation had one’s back turned towards the north and the face turned towards the south⁶. As far as the term gēg is concerned, to our knowledge no etymology has so far been proposed, but we would argue that a connection should be made with Proto-Indo-Iranian *gāyaka- from the Proto-Indo-European stem *gu̯ eh2i̯oko-, in which the root for “go” can be identified as well as two derivative suffixes that form a conglomerate as in *martyaka-, “mortal”. Thus, gēg would be the “vagabond”, with a pejorative implication. Returning now to Armenian, a similar iconymic process can certainly be read in the above cited syntagm astełkʻ molarkʻ, given that the adjective molar, from *mol-ro-⁷, indicates an evil deviation from the straight path of righteousness and the verb molim means “to go mad”. The blurring of any distinction between the literal plane (straying from the marked path) and the figurative plane (ethically or psychologically deviant behaviour) is a recurrent metaphor in this linguistic family. In Christianised Armenia the Zoroastrian cult would later be defined as molutʻiwn, “heresy” (in Tʻovmay Arcruni, ninth–tenth centuries); furthermore, mol-i, literally “mad, furious”, is a type of witchdoctor in the thrall of ecstatic

3 Panaino 2002, 30. 4 Eilers 1967, 112 ff. and 1987. 5 MacKenzie 1971, 2. 6 Bartholomae 1904, 79. 7 Olsen 1999, 338.

314

Rosa Ronzitti

forces. Martirosyan⁸ believes that the adjective molar is related to ancient Indian malvá- and Dutch mal, both meaning “stupid”, whereas Mayrhofer⁹ remains more prudent. Even Latin malus could conceivably be compared with these items¹⁰. Since it would appear to be advisable also to seek an etymology for ampar that would be in line with such clearly consolidated values, which are in effect attested precisely in the peri-Armenian world, we propose to interpret it as a loanword based on an Iranian compound formed from am-, “around” and par-, “move through, pass”. The element am- is a preposition fairly well attested in various Indo-European languages, together with the more widespread *ambhi/ *m̥bhi¹¹. With respect to the latter terms, am- has an independent existence, though Untermann¹² believes that *ambhi may have originated from the union of *am- with *-bhi. Taken on its own, am-, “around”, is present in Osco-Umbrian and, probably, also in Greek and Sanskrit, if the two terms ἀμφορεύς, “amphora” and ambhr̥ ṇá-, “recipient” do not derive from haplologies of *ἀμφιφορεύς and *ambhibhr̥ ṇá- (they should also be compared with Oscan amfret, “he carried around”, 3. pl. preterite etc.). The Latin verb amicio, “throw a garment around” testifies to the preverb before an initial vowel¹³. The element paris the root of passage, of moving through space, well attested in the Avestan corpus (always preverbed) and in various phases and texts of Middle Persian. Bartholomae¹⁴ assigns to this root the value of “hindurch-, hinübergehen”, Cheung¹⁵ “to go over, cross over”.

8 Martirosyan 2010, 472. 9 Mayrhofer 1994, 334. 10 Cf. de Vaan 2008, 360. On the Latin vocalism, Schrijver 1991, 454–476, esp. 457 establishes the rule that */o/ becomes /a/ after */m/ or after any labiovelar, including */u̯ /. 11 Pokorny 1959, 34. 12 Untermann 2000, 83. 13 Opinions are divergent in this regard. Recently de Vaan 2008, 37 has suggested, instead, that amicio may derive from *amb-i̯acio, while the independent existence of am- in Osco-Umbrian is not open to debate; however, words such as ambiegnus, “having lambs on two sides” and ambiaxio, “in heaps” constitute potential counter-examples, inasmuch as /b/ does not fall before a semivowel. Also extremely interesting are the words of Paulus (ex Festo), which appear to testify an independent use of am for Latin as well: am praepositio loquelaris significat circum, to be compared with the glosses am fines, am segetes. De Vaan downplays the significance of this evidence without a real reason: “they may be based on a contemporary analysis of the compounds”. 14 Bartholomae 1904, 851. 15 Cheung 2007, 293–294.


315

A relation to astronomic movements can be noted in Manichean Middle Persian xwrpr’n (/xwarfarān/), “west”, literally “passage of the sun”. The same compound can be found preverbed with *au̯ a- in Pahlavi hwl-(’)wpl’n /xwarōfrān/. Pahlavi likewise has hnb’ršn /hampārišn/, “confluence”, with consonantism that preserves the voiceless plosive precisely after the nasal¹⁶. A stem in *-o- par, paroy, “the act of dancing” is attested by Ciakciak¹⁷: a stem with initial /p/ provides a better clue to its nature as a borrowing rather than as an unlikely root beginning with Indo-European */b/. The term ampar would thus derive from a Middle Persian adjective (probably Parthian) with the meaning of “wandering”, and the syntagm ampar astełkʻ would be the exact correspondent of the Greek type πλάνητες ἀστέρες. 3. Another very interesting aspect of the dictionary concerns its attention to folklore and to anthropological data, drawn both from Armenian works that are fairly inaccessible abroad and also from field research. The dialectal strata not only provide data useful for reconstruction, but also allow in-depth ethno-folkloristic investigation that reveals a capacity for conservation and, as it were, for embodying nuggets of more “archaic” aspects of language as compared to those reflected in the standardized national languages. Thus one can welcome the statement that it makes no sense to define dialectal words as nor baṙer, “new words”¹⁸ in cases where they are clearly just as ancient as those of the classical language. For instance, the name for a bat, gišer(n)uk, with an -uk formation identical to that of the Latin vesperūgō, is an isogloss that can be derived only by comparing Latin with two Armenian dialects that are not only peripheral but also distant from each other. And so on and so forth. 3.1. In this regard, the Appendix to the work (pp. 769–807) examines several semantic fields that form part of standard ethno-anthropological investigation, such as astronomy, the division of time, animals, plants. Here one finds full confirmation of well-trodden iconymic universals, for example the “ladybird” as the “Lord’s cow” (or “little cow”). See Łarabał astucoy kov or astucoy eznak, which not only recalls the Slavic and Romance denominations (e.g. Russian bož’ja korovka, Romanian vaca domnului etc.), but, as I have pointed out elsewhere, also the Sanskrit equivalent indragopaka-, “lady-bug”, “she who has Indra as her

16 With regard to the cluster /mp/ cf. Pahlavi hampursag, “consulting, taking counsel” in MacKenzie 1971, 41. 17 Ciakciak 1837, 2, 1182. Cf. also the verb parem, “to dance, to jump, to leap up and down or somersault” (ib. 1186). 18 Martirosyan 2010, 691.

316

Rosa Ronzitti

cowherd”¹⁹. As regards the entry ezn, “calf, ox” (p. 247), of which eznak represents the diminutive, the reason underlying the entomonym does not seem to have been fully grasped, as Martirosyan states that astucoy ezn, “lady-bug” seems to suggest “that ezn earlier had feminine (or generic) semantics”. But what we are dealing with here is not a generic female animal: rather, it is specifically a “cow”, given the conformity with the above-mentioned onomastic type and the abundance of comparisons, all gathered together under the entry “lady-bug” in the Atlas Linguarum Europae²⁰. 3.2. There is extensive presence of kinship zoonyms involving animals endowed with strong symbolic significance, to which Martirosyan devotes the attention these certainly deserve. Thus the “spider” is mamuk (dialectal), i.e. the “little old grandmother”²¹; the “snail” is anaxat‘um/n, “Lady mother” (p. 787); the “scorpion” is called tat, “grandmother” (p. 787). Ambivalent relations with the sacred are revealed by the depictions of the spider: if satanay (p. 788) is beyond doubt a demonic name, elsewhere the “spider” is identified with the Virgin Mary²². A similar reversal (or better, perhaps, an exchangeable relation between the two poles of the supernatural) seems also to involve the frog and the toad, which are demonic animals par excellence. Particularly revealing, in our view, are the names surb (dialectal), “a kind of small frog that lives in humid holes” and surp, “frog, toad” (p. 588). They could be regarded as antiphrastic, given that surb means “pure, holy”, unless we are dealing here with a homophone from a different root. In any case, ambivalence towards the sacred is a constant feature: a zoonym such as mariam-gort, “large frog”, literally “the frog of the Virgin Mary”, restores the accursed amphibian to the sphere of Christianity. 3.3. Another enormously widespread belief that should be included among popular lore is the theme of “milk-sucking” animals, i.e. creatures of various kinds (bats, owls, toads, lizards, butterflies) which were said to make their way

19 Ronzitti 2010, 42. 20 Barros Ferreira/Alinei 1990, esp. 117–119. 21 Martirosyan 2010, 446. 22 In a fourteenth–fifteenth century text the Blessed Virgin is compared to a spider that weaves an insubstantial web. On p. 797 there is also a mention of dialectal Persian denominations such as Xurāsānī šotor-e xodā, “camel of the Lord” and Lurī xar-e xodā, “the Lord’s donkey”. In addition, Arm. *ernǰak, “spider” once again seems to propose a relation with the world of large animals, given the relationship that appears to exist between *ernǰak and erinǰ, “filly, cow, young cow, bride”, hence the possible interpretations “little heifer, little filly, young bride”. This could be a case of a correspondence that arose independently of the historical etymological analysis, which sees *ernǰak as related to ἀράχνη (p. 270).


317

into barns at night to suck milk from the animals’ udders (sometimes, in a vampiresque version, even the blood)²³. The compound kovadiacʻ, which is present with several dialectal variants, indicates a type of lizard and is immediately interpretable as “milk-sucker” (pp. 372–373). This could be an extremely ancient formation, given that it corresponds member by member to both to Sanskr. godhā́, “lizard” and also to Lat. būfō, “toad” (or some other noxious/demonic animal, like small insects). Martirosyan appropriately posits that at the base of kovadiacʻ there lies the root compound *gu̯ oṷ-dheh1-eh2-²⁴. In many dialects one also finds the form kov(a)cuc: here what is interesting is the second member, of which Martirosyan offers no treatment. It seems evident that -cuc must mean “suck”, but a root *g̑ ug̑ - (*g̑ ugu̯ - or *g̑ ug-) is missing in Pokorny and in the dictionaries of Indo-European terms²⁵. 3.4. In this field that offers such rich potential for exploration, the term babič‘, “witch-doctor” (p. 162), attested only in medieval glossaries, deserves attention. Martirosyan is not familiar with any prior etymologies for this term and proposes to treat the lexical base as a reduplication of the verb bam, “say, speak”. The root *bheh2-, “speak” (*bhā- in Pokorny²⁶) has extensive confirmation in Indo-European contexts, as has long been known. Martirosyan quotes above all the Slavic family, which includes Serbo-Croatian bȁjati, “to engage in witchcraft, to exorcise”, Bulgarian báj, “to whisper magic spells”, Russian básnja, “fable” etc. We would argue that this list of possible cognates could be expanded by an even more cogent comparison with the Slavic term baba, “old woman, witch, sorceress” (cf. the Baba Jaga, a well-known witch of Russian fables). This word may constitute a significant Slavic-Armenian isogloss because of the identity of the root, of the morphological structure (broken reduplication) and of the semantic field (witchcraft). Armenian additionally shows a suffixation in *-iki̯ā, that is used to form nomina agentis²⁷. Furthermore, if we give the reduplication an iterative meaning, we would have on the one hand Proto-Armenian *bābiki̯ā, “he who speaks repeatedly” and on the other the Proto-Slavic term *bābā, “she who speaks repeatedly, woman, old woman, witch”. The hypothesis of a loanword

23 I take the liberty of referring the reader to Ronzitti 2011 for the abundant bibliography on the topic. 24 Cf. also Ronzitti 2011 for the details of the reconstruction. 25 The forms *g̑ ug- and *g̑ ugu̯ - are likewise possible because in any case the second dorsal becomes an affricate before /u/ on account of a phonetic law specific to Armenian, which we will mention again shortly. 26 Pokorny 1959, 105–106. 27 Olsen 1989, 474–480.

318

Rosa Ronzitti

taken from a Slavic masculine is also possible, though this would be in contrast with the Armenian final aspiration, which would seem to point instead to direct development of the sequence /iki̯ā/ rather than to a loanword taken from a hypothetical noun *babič‘, “that which pertains to the baba”²⁸. The direct route would imply a magic-oriented semantic development of the reduplicated root as early as in Proto-Indo-European, which could then easily give rise to the polysemy “old woman/witch”. In our view, the possibility of a common etymology between the Slavic baba and the Armenian babič‘ is preferable to the separation of the two terms, as would be inevitable if one were to follow Pokorny, who believes that the Slavic word goes back to a base *baba-, a Lallwort attested in the names of sounds, child language and so forth²⁹. The presence of a simple voiced sound in the protolanguage would exclude Armenian from the picture, or alternatively it would presuppose a loan occurring later than the consonant rotation (but it should be borne in mind that the suffix -ič‘ is rotated)³⁰. As far as the underlying motif of word is concerned, it may be useful to draw a broad comparison with the entry “Sorcière etc.” of the Atlas Linguarum Europae³¹, in particular with the iconymic group that involves the theme of “predicting”, “speaking”, “murmuring”, “praying”, “blessing”, “bewitching”. It is precisely the Slavic group that shows a clustering of these motifs: compare the oriental Slavic type koldun’ja, “she who speaks” or Ukrainian šeptuha (Lituanian šaptune), “she who murmurs”. Of interest is also Russian gadalka, “female soothsayer”, if it is related to the ancient Slavic gadati, “predicts, speaks”, which would bring the term into relation with logonyms of high sacrality such as Avestan gāϑā, “religious chanting” and Old Indic gāthā. Eastern Romance is also congruent, from the semantic point of view, with the geographic area in question as Roma-

28 In the large dictionary of Slavic dialects бабич is given with the meanings of “that which pertains to the baba” and “Pelecanus crispus” (ornithonym) (Filip 2002, 20). 29 Pokorny 1959, 91. 30 It could be objected that intervocalic /b/ (from Indo-European */bh/) undergoes fricativization, becoming /u̯ /, and that the expected outcome would therefore be **bawič‘. But if, on the other hand, we start out from the reduplicated base *bheh2-bhh2-, the second aspirated voice sound would not be between vowels, but rather between two laryngeals. The crucial point rests on whether the fricativization began when the laryngeals were still in existence or whether they had already disappeared, and the objection is valid only if this process began after the dephonologization of the laryngeals; if so, this fact would require us to consider any Armenian word with intervocalic /b/ as non-inherited. 31 Caprini/Alinei 2007.


319

nian has bosconitoare, derived from bosconi, “to murmur, to stammer”, a probable Graecism that once again points to the iconymy of “saying”³². 4. The dialect data can highlight new isoglosses. The entry *muž, “fog” (pp. 483–484), for which Martirosyan somewhat doubtfully puts forward an Iranian attribution, can also be explained as echtarmenisch and in any case should not be treated separately from the rich etymological family that ranges from English smoke to Sanskr. múhyati, “is confused, is blurred”³³. The phoneme written as <ž> would thus derive from an aspirated voiced labiovelar preceded by a vowel and followed by a palatal vowel, which would presuppose a stem in *-i- *mugu̯ hi-³⁴. This implies that the entry *(s)meu̯ kh-, *(s)meu̯ g-, *(s)meu̯ gh-, “to smoke, smoke” of Pokorny³⁵ should be supplemented with the reconstructed *(s)meu̯ gu̯ h-, a clear testimony of a root with the greatest variety of enlargements in dorsal. Armenian itself preserves the form from */kh/ mux, “smoke” (p. 484). In contrast to Martirosyan, I would take the connection of the latter term with *muž to be obvious and I would avoid resorting to the idea of an Indo-European substrate to account for the presumed phonetic difficulties of the etymon (i.e. the oscillations of the final consonantism), which can easily be overcome through the theory of enlargements. 5. One etymological enigma that can be solved easily involves the verb cʻncam, “to rejoice, to shine brightly” (the latter meaning is dialectal). It is usually, albeit with considerable uncertainty, equated with Sanskrit chand-, “to appear, to seem beautiful, to be pleasing” (pp. 629–630). However, this equation requires a complicated line of reasoning to explain the final consonantism, since one would

32 Caprini/Alinei 2007, 196. Among the various etymological hypotheses concerning bosconi (to which we would add boscorodi, “to engage in magical practices”, with the second member rod, “word” of Bulgarian origin), the hypothesis of a Paleobalkanism that passed into Romanian through a Slavic language seems plausible (see already Boisacq 1923, 116). For the Greek starting points βάσκανος, “magic spell” and the gloss βάσκειν· λέγειν, κακολογεῖν, it is hypothesized that they are in turn related to φάσκω, “to say” and Latin fascinum, “magic spell”. In this case the variant with initial /b/ would be a borrowing from a de-aspirating language, hence the Paleobalkan hypothesis put forward by Boisacq (taken up also by Frisk 1954, 223–224). According to this view, Slavic in turn took over from Paleobalkanic (= Thracian or Daco-Mysian) the base βάσκω-, rendering it with /boskō/. Uncertainty remains concerning Romanian boscar, “conjuror”, which Cioranescu 2007, 108 derives from the surname of the Italian conjuror Bosco (1793–1862), and this would give the whole family a decidedly recent origin. 33 Ronzitti 2011, 73 ff. 34 Schmitt 1981, 65 for the parallelism with iž ‘serpent’ from *ēgu̯ hi-. 35 Pokorny 1959, 971.

320

Rosa Ronzitti

expect *cʻ(i)nč- from *sk̑endi̯o-, as Martirosyan himself admits³⁶. He thus proposes a derivation from the stem of the sygmatic aorist *skend-s-³⁷. First of all, we give a more precise semantic description of the Armenian verb, by virtue of the text of Luke 1.47, Annunciation of the Virgin: Ew c’ncac’aw hogi im yAstuac P’rkič’ im καὶ ἠγαλλίασεν τὸ πνεῦμά μου ἐπὶ τῷ θεῷ τῷ σωτῆρί μου et exsultavit spiritus meus in Deo salutari meo ‘And my spirit exulted in God my saviour’.

Here c’ncac’aw is the third aorist middle (intransitive) of c‘ncam, “to exult, to make a celebration”. The Greek text has ἠγαλλίασεν: ἀγαλλιάω, which is regarded as a late form of ἀγάλλομαι, expresses an overwhelming joy. The etymology of ἀγάλλομαι is highly uncertain. On the other hand, the Latin exsultavi is traceable to the idea of movement. In effect the verb salto, of which exsulto is an apophonic compound, means “to jump, dance” and constitutes the iterative of salio, to which it adds a specific actional characterization. It should be noted that the sequence *sk̑- gives c’ in Armenian regardless of whatever is the following vowel, whereas *sk- certainly gives c’ if it is followed by a non-front vowel; but if, on the other hand, *sk- is followed by a front vowel, opinions are divided: Birgit Olsen argues that the outcome is š-, while Martirosyan admits the outcome c’, reserving the outcome š- to *skH- clusters³⁸. Taking all these factors into account, we thus reconstruct for Armenian a root *skV(n)g̑ - or *sk̑V(n)g̑ -, where the nasal is an infix, while the nature of the dorsal immediately after the sibilant can be either velar or palatal, since even in Olsen’s most restrictive hypothesis, namely that *sk̑V- gives š- (and not c’) before *e/i, it is sufficient to posit a non-palatalizing *o/u vocalism, whereas for Pisani, on the contrary, one would have to presuppose precisely a palatal vocalism. The final */g/, on the other hand, must be palatal in order to give the regular outcome ˂c˃ = /ts/³⁹. As far as the possible root vowels are concerned, */a/ should in any case be excluded, as it would not fall between consonants.

36 Godel 1975, 82, on the other hand, argues that the cluster *-di̯- gives Arm. c. If he is right, then the present stem would also be appropriate. 37 For a parallelism with anic, “nit” from *s(k)onid-s (Martirosyan 2010, 749). 38 Olsen 1999, 812–814; Martirosyan 2010, 719. Another different position is put forward by Pisani 1950, who contends that *sk- gives sk- when it is followed by a non-front vowel but c‘ or š when it is followed by a front vowel. In contrast *sk̑- would give c‘ or š regardless of the following vowel. 39 In actual fact, as pointed out earlier (n. 25), even a form with final velar would be feasible for Armenian, on condition that the root vowel was /u/, as this vowel caused palatalization.


321

Pokorny offers various possibilities. Insofar as they are compatible with the semantics, there would be two suitable bases. First and foremost, mention should be made of *skeg-, “eilen, springen, schütteln”⁴⁰, which appears in various languages, among which Old English. Here we find the verb sceacan (Proto-Germanic *skakanan, modern English shake), “to shake, to hasten, to run away” and also “to vibrate, to make something vibrate”⁴¹. The voiced velar plosive in final position seems to be reconstructed only due to the parallelism with the synonymous and homophonous root *(s)kek- (combined in the same lemma by Pokorny⁴²) and which could agree with the Old Indic outcomes khája- m., “commotion”, “hurly-burly” and khajā f., “stick for making butter” if one postulates an analogical extension of the /j/ from the verb khajati, “to stir”. It is undeniable, however, that an Indo-European palatal */g̑ / in *skeg̑ - would give a more regular explanation of the Sanskrit, and without any difficulty, of the Old English as well. And by taking the form *skeg̑ - as the starting point, one would have a justification precisely of the Armenian c’ncam, which can derive either from *skeng̑ -eh2-mi (if Martirosyan is right concerning the outcome of *ske-, see supra) or from *skong̑ -eh2-mi⁴³ (in Olsen’s more restrictive hypothesis). One should also take note of the existence of *skeu̯ g-⁴⁴, a root classified as a sub-entry of *skeu̯ b-, *skeu̯ bh-, “dahinschiessen (flink); schiessen, werfen, schieben”. Only Germanic continuators are known for the form ending in */g/, among which Middle Low German schocken, which means “to tremble”, and Middle High German schocken, “to swing, to wobble, to dance”; German Schock means “violent movement, a shudder”. There is nothing to rule out the possibility that the Indo-European root which lies at the base of the Germanic family may have ended in a palatal (the velar that can be read in Pokorny should be understood more precisely as an optional palatal). In contrast, the Old Church Slavonic palatal ščuti, “to hunt, to chase” excludes the palatal (as a root-final Indo-European palatal would have given †ščusti), but this does not prevent the reconstruction of *skeu̯ g̑- for Armenian, since the Slavic verb may derive from *skeu̯ tēi (formed directly on the root *skeu̯ -) as well as from *skeu̯ gtēi or *skeu̯ ktēi, as Pokorny would prefer.

However, since /u/ occurs here after the nasal infix, it is not in direct contact with the velar, and there is no guarantee that the palatalization occurred regardless of the surrounding conditions. 40 Pokorny 1959, 923. 41 Falk/Torp 1909, 447. 42 Pokorny 1959, 922–923. 43 Denominative from a noun with */o/ vocalism. 44 Pokorny 1959, 955.

322

Rosa Ronzitti

The concept of repeated bodily movement easily adapts to describing a psychic state of excitation/ being shaken, such as that which overwhelms Mary upon hearing the angel’s words. In effect, theologians use the phrase Conturbatio Virginis to describe the moment following the Annunciation. One may perhaps also sense a nuance of fear and bashfulness in this psychological state, if the family of the same root *skeu̯ g̑- does indeed include English shy and the corresponding German scheu (from *skuǥwa- < *skuk̑-u̯ ó-). The study by Falk and Torp⁴⁵ divides Protogermanic *skukk- (< *skug̑ -n-), “sich hin und her bewegen”, under which he includes Middle High German schoc, from *skuh- (< *skuk-), under which he lists Engl. shy. 6. The section reserved to toponyms (pp. 669–686) may be useful in that it allows the incorporation of Armenia into a broader sphere of general Indo-European toponomastics. 6.1. Ardean-kʻ (p. 669) is a village of the Ayrarat that can no longer be located today, and is mentioned only by the historian Pʻawstos Buzand (5.6) as a place where a treasure or a hoard of grain was stored. Since there are no etymological proposals for this hapax, which is attested by the accusative plural Ardeans, Martirosyan puts forward one of his own, appealing to the dialectal Armenian *ard(i)umn, “earth’s goods, harvest”. However, it would seem more immediate, or at least equally valid, to draw a comparison with the Gallic coronym Arduenna [silua] (from which the modern Ardennes). Both Celtic and Armenian could be traced back to a widely attested Proto-Indo-European adjective *h2erdh(u̯ )o-, “high, elevated”. For Protoceltic, Matasović⁴⁶ reconstructs *ardu̯ o-, to be compared with Avestan ərəδwā̊ and Latin arduus⁴⁷. An adjective ardean, “elevated” would derive in a regular manner from *h2erdhih3no-. The formant *-ih3no- gives a possessive value to the bases with which it aggregates, and this leads to the further suggestion that the name of the village was a collective plurale tantum indicating the group of dwellings and/or men that are in an elevated place.

45 Falk/Torp 1909, 467. 46 Matasović 2009, 40–41. 47 The initial /a/ of arduus could imply that Latin borrowed from Celtic. There is a complex relationship between all these forms and Old Indic ūrdhvá-, “tall”, Greek ỏρθός, “erect, straight”, as the latter could belong to a related family with initial */u̯ / (for further details see Ronzitti 2014, 196); for Armenian there is no reason to postulate a bilabial approximant before the thematic vowel.


323

6.2. Another interesting item is K‘ark‘ē, “N. of a mountain in the province of Turuberan”, in the districts of Tarōn and Palunikʻ (p. 685). No insuperable problems are raised by the toponomastic element *kar-k-, “a steep stony slope”, which is widely attested (above all, but not exclusively) in the Celtic and Germanic languages⁴⁸; but what remains to be solved is the anomaly of the sequence /rk/, which should give /rg/ in Armenian. The conjecture advanced by Martirosyan, namely that the lack of sonorization of /k/ after /r/ is due to the morphemic boundary or to a loanword from an unknown substrate, actually involves two different hypotheses. The first one introduces a sonorization rule applying to voiceless plosives after /r/⁴⁹, a restriction without counter-examples but in any case ad hoc (at least at the current state of research); the second, which implies a substratism borrowed after consonant rotation and assuming two voiceless aspirates present in the source language, postulates a whole unknown language purely for the sake of just one word. But such a hypothesis becomes pointless if there exist preferable internal explanations. An explanation can be given, in the case in question here, in terms of a laryngeal interposed between an initial syllable and the root, thus *kr̥ H-ko-. The formant *-ko- may represent a broken reduplication or, more simply, a suffix added to the root stem. Classical Sanskrit karkara-, “hard, solid, a stone”⁵⁰, is not mentioned, but should certainly be included in the series, especially since it too presents itself as a form having reduplication (*korHkorHo-). See also Lat. carcar (variant of carcer), “fenced enclosure, prison, barrier” and the Ligurian toponym Carcare, which could be a Paleo-Ligurian term⁵¹. 7. Such suggested alterations and additional details (the list of which could be extended) should be taken as further and on-going reflections pertaining to this impressive project, which differs from many etymological dictionaries published over the last decade in its originality and the care taken to contextualize all the lemmata, providing an in-depth investigation of the semantic nuances of each one. The dictionary is also noteworthy for its attention to the hitherto unexplored mine of dialects, which prove to be useful on both sides of the sign for the purposes of reconstruction. Martirosyan’s formal rigour is always accompanied by a detailed discussion that covers the history, culture and texts of the rich Armenian

48 See Pokorny 1959, 531: there the root *kar- is reconstructed with */a/, but a regular apophony is also admissible, even for Latin cancer, as */e/ preceded by a velar becomes /a/. 49 Passing through the pre-Proto-Armenian stage */gh/. 50 Mayrhofer 1997, 65. 51 I am grateful to Prof. Guido Borghi for this kind suggestion.

324

Rosa Ronzitti

tradition, and does not limit itself to a mere display of technical-reconstructive skill or an updating of already known etyma.

Bibliography Barros Ferreira, M. / Alinei, M. (1990), “Coccinelle. Cartes de motivations”, in: Atlas Linguarum Europae, 1.4, Commentaires, Gorcum, 99–199. Bartholomae, C. (1904), Altiranisches Wörterbuch, Strassburg. Boisacq, É. (1923), Dictionnaire étymologique de la langue grecque, Heidelberg-Paris. Caprini, R. / Alinei, M. (2007), “Sorcière, Колдунья, Witch, Hexe, Bruja, Strega”, in: Atlas Linguarum Europae, 1.7, Commentaires, Roma, 169–225. Cheung, J. (2007), Etymological Dictionary of the Iranian Verb, Leiden-Boston. Ciakciak, E. (1837), Dizionario Armeno-Italiano, Venezia. Cioranescu, A. (2007), Dicţionarul Etimolog̭ic al Limbii Române, Bucureşti. De Vaan, M. (2008), Etymological Dictionary of Latin and the other Italic Languages, Leiden-Boston. Eilers, W. (1967), “Stern-Planet-Regenbogen. Zur Nomenklatur der orientalischen Himmelskunde”, in: W. Hoenerbach (ed.), Der Orient in der Forschung. Festschrift für Otto Spies, Wiesbaden, 92–146. Eilers, W. (1987), “Āxtar”, in: Encyclopedia Iranica, online edition (www.iranicaonline.org/ articles/axtar-middle-and-new-persian-star-or-constellation). Falk, H. / Torp, A. (1909), Wortschatz der germanischen Spracheinheit. 5. unveränderte Auflage, Göttingen. Filip, F. P. (2002), Slovar’ Russkich Narodnych Govorov. Vypusk Vtoroj. Ба – блазниться. 2-e издание, Sankt-Peterburg. Frisk, H. (1954), Griechisches Etymologisches Wörterbuch. 1, Α-Κο, Heidelberg. Godel, R. (1975), An Introduction to the Study of Classical Armenian, Wiesbaden. MacKenzie, D. N. (1971), A Concise Pahlavi Dictionary, London. Martirosyan, H. K. (2010), Etymological Dictionary of the Armenian Inherited Lexicon, Leiden-Boston. Matasović, R. (2009), Proto-Celtic Etymological Dictionary, Leiden-Boston. Mayrhofer, M. (1994), Etymologisches Wörterbuch des Altindoarischen. 2, Heidelberg. Mayrhofer, M. (1997), Etymologisches Wörterbuch des Altindoarischen. 3, Heidelberg. Olsen, B. A. (1999), The Noun in Biblical Armenian, Berlin-New York. Panaino, A. (1990), Tištrya. Part I. The Avestan Hymn to Sirius, Roma. Panaino, A. (1995), Tištrya. Part II. The Iranian Myth of the Star Sirius, Roma. Panaino, A. (2002), “A proposito dell’uranografia mazdaica e del concetto di ‘sfericità del mondo’ nella cultura astrale iranica preislamica”, in: C. Dognini (ed.), Kosmos. La concezione del mondo nelle civiltà antiche, Alessandria, 27–39. Pisani, V. (1950), “Studi sulla fonetica dell’armeno”, in: Ricerche Linguistiche 1, 165–193. Pokorny, J. (1959), Indogermanische etymologisches Wörterbuch, Tübingen-Basel. Ronzitti, R. (2010), “Natura maligna: raffigurazioni degli Strigidi nella letteratura indiana antica”, in: Quaderni di Semantica 31.1, 41–62.


Ronzitti, R. (2011), Quattro etimologie indoeuropee: lat. būfō, it. gufo, ingl. smog e drug, Innsbruck. Ronzitti, R. (2014), Due metafore del caso grammaticale, Innsbruck. Schmitt, R. (1981), Grammatik des Klassisch-Armenischen mit sprachvergleichenden Erläuterungen, Innsbruck. Schrijver, P. (1991), The Reflexes of the Proto-Indo-European Laryngeals in Latin, Amsterdam-Atlanta (GA). Untermann, J. (2000), Wörterbuch des Oskisch-Umbrischen, Heidelberg.

325

Index of passages

Greek Texts Aelianus Varia Historia 13.3: p. 94 n. 63 Aelius Herodianus De prosodia catholica 291.12 Lentz: p. 306 332.10 Lentz: p. 168 and n. 21 332.16 Lentz: 168 n. 28 Aeschylus Persae 745–750: 97 n. 83 864–866: 95 and n. 66 Anthologia Palatina 9.304: 97 Appianus Mithridatica 67.285: 181 and n. 79 Aristoteles Analitica Posteriora 2.89b 23: 55 n. 36 Fragmenta F 609 Rose: 267 and n. 35 Meteorologica 342b.28: 312 and n. 1 Rhetorica 1407a: 312 and n. 1 Aristoxenus F 50 Wehrli: 101 and n. 100 Arrianus (FGrHist 156) T 2; F 30 (= fr. 1, Parth., Roos): 165 n. 12 T 5 (= T 12 Roos): 177 n. 67, 187 n. 111 T 13 Roos: 178 n. 71 F 37 (= T 14 Roos): 178 n. 71 F 140 (= 53, Parth., Roos): 175 n. 60 F *50 (Parth., Roos): 176 n. 60 F *52 (Parth., Roos): 176 n. 60

Asinius Quadratus (FGrHist 97) T 1 (= FRH, 2, p. 1128, T 5): 165 n. 10 T 3 (= FRH, 2, p. 1128, T 3): 165 n. 12 FF 5–20, 27–30 (= FRH, pp. 1132, 1134, 1136, 1138, 1140, 1142, frs. 6, 12–31): 164 n. 5 F 5 (= FRH, 2, p. 1134, fr. 12): 172–174 F 6 (= FRH, 2, p. 1134, fr. 13): 172 and n. 45, 173–175 F 7 (= FRH, 2, p. 1136, fr. 14): 172 and n. 45, 173–174, 176 F 8a (= FRH, 2, p. 1136, fr. 15a): 164, 166–167, 167 n. 14, 171–174, 177, 180 F 8b (= FRH, 2, p. 1136, fr. 15b): 164, 166–167 and n. 14–15, 172, 177, 180 F 9 (= FRH, 2, p. 1136, fr. 16): 164–166, 169–171 F 16 (= FRH, 2, p. 1138, fr. 23): 164, 180 and n. 77 F 18 (= FRH, 2, p. 1140, fr. 26): 164, 166, 169, 180 and n. 76, 182 and n. 83 F 19 (= FRH, 2, p. 1140, fr. 25): 164 n. 5 F 20 (= FRH, 2, p. 1132, fr. 6): 164 and n. 5–6 F 21 (= FRH, 2, p. 1132, fr. 8): 180 n. 77 F 26 (= FRH, 2, p. 1132, fr. 7): 180 and n. 77 F 27 (= FRH, 2, p. 1140, fr. 27): 166 n. 13 F 28 (= FRH, 2, p. 1142, fr. 28): 172 n. 44 F 29a (= FRH, 2, p. 1142, fr. 29): 164 F 29b (= FRH, 2, p. 1142, fr. 30): 164 F 30 (= FRH, 2, p. 1142, fr. 31): 173 n. 46 [Athanasius] De incarnatione Christi contra Apollinarium (= De incarnatione contra Apollinarium liber I – CPG 2231) 7 (PG 26.1104.34–41): 234–235 7 (PG 26.1105.10–12): 238 15 (PG 26.1120.21–30): 237–239 17 (PG 26,1124,17–24.27–31): 235–237 De salutari epiphania contra Apollinarium (= De incarnatione contra Apollinarium liber II – CPG 2231) 10 (PG 26.1148.43–1149.1): 236

330

Index of Passages

Athenaeus 633 d–e: 128 n. 49 Bacchylides Epinicia 3.23–62: 88 n. 27 Basilius Magnus Caesariensis PG 29.23: 49 and n. 7 Berossus (FGrHist 680) F 11: 93 and n. 57 [Callisthenes] Historia Alexandri 1.1–14: 89 n. 33 1.13.4: 27 1.41.7: 95 n. 70 2.2.14: 96 n. 73 2.12.1: 25 2.22.12: 25 Carmen Byzantinum de Alexandro (Marcianus gr. 408) 4103: 27 and n. 19 Cedrenus: see Georgius Cedrenus Clemens Alexandrinus Protrepticus 4.47.8: p. 000 and nt. 54 5.65.2: p. 000 and nt. 57 Stromata 4.26 p. 7 nt. 39 Concilium Universale Constantinopolitanum Tertium (Acta Conciliorum Oecumenicorum, ed. E. Schwartz) 2.2.1.240.3–5: 236 and n. 14 2.2.1.246.17–21: 237 n. 16 2.2.1.336.4–8: 237 n. 16 Constantinus Acropolites Epistulae 78 Romano: 250 and n. 45

Ctesias Cnidius (FGrHist 688) F 1 (b; p δ ed. Lenfant): 44 and n. 5, 101 n. 99, 130 F 5: 101 n. 100, 130 F 8: 100 n. 101 F 8d: 130 F 9 § 5: 84 n. 5, 87 n. 22, 88 n. 24, 130 F 13b, 33: 94 n. 63 F 14: 44 n. 3 F 15: 44 n. 3 F 15, 47–50: 94 n. 63 F 15b: 118 n. 21 F 16: 75 n. 6, 118 n. 21 F 61 a: 44 and n. 4 F 61 b: 44 and n. 4 F 66: 129 n. 51 Dinon Colophonius (FGrHist 690) F 9:128 n. 49 Dio Cassius 57.5: 147 n. 31 68.20.3: 175 n. 55 68.32.4: 175 n. 58 69.15.1: 177 n. 67, 187 and n. 111 71.2.1: 174 n. 52 78.21.1: 183 78.21.2–4: 183 n. 87 79.27.4: 183 n. 90 Diodorus Siculus 2.11.4: 44 n. 5 2.32: 129 n. 51 9.2 (= F 4 Cohen-Skalli): 84 n. 5, 96 and n. 78 9.27 (= F 38 Cohen-Skalli): 84 n. 5 9.31 (= FF 41–45 Cohen-Skalli): 84 n. 5, 96 and n. 78, 97 n. 79 9.33 (= F 50–51 Cohen-Skalli): 84 n. 5, 87 n. 22 11.69: 94 n. 63 15.90–93: 89 n. 30 16.40–46: 89 n. 30 16.47–50: 94 n. 63 17.5: 94 n. 63 17.50.4: 95 n. 70 17.51.1: 95 n. 70

Greek Texts

Dionysius Areopagita De divinis nominibus p. 111.7 Suchla: 18 Dionysius Halicarnassensis Antiquitates Romanae 1.34.2: 267 n. 35 1.56.2: 267 n. 36 1.72.3–4: 267 and n. 35 De demonstratione, 7: 11 Euagrius Scholasticus Historia Ecclesiastica 5.24: 165 n. 12 Eunapius Vitae sophistarum 10: 49 and n. 5 Euripides Bacchae 835: 248 n. 35 Medea 480: 248 n. 35 Eusebius Caesariensis Praeparatio Evangelica 6.10.31: 168 and n. 21 Favorinus Fragmenta fr. 61 Amato: 28 n. 21 fr. 61a Amato: 28 Georgius Cedrenus Compendium Historiarum 1.564.7–10 Bekker: 93 and n. 55 Georgius Monachus Chronicon 1.12.18–20 De Boor-Wirth: 28 n. 63 Georgius Pisides Hexaemeron 40: 249 46–47: 251 48–50: 244

331

51–53: 244 n. 15 129–130: 251 146–147: 252 n. 50 170–172: 246 472–473: 247 476–480: 247 478: 248 480: 247–248 931: 245 931–933: 245 n. 23 934: 245 947: 246 1399–1400: 248 1455: 252 1625: 17 Georgius Syncellus Ecloga chronographica 282 Mosshammer: 122 and n. 28, 123 and n. 30 291 Mosshammer: 122 n. 28 Gregorius Nazianzenus Epitaphia 5 13A: 49 and n. 9 Funebris oratio in laudem Basilii Magni (orat. 43) 17: 49–50 and n. 11 Epistulae 43.17: 50 and n. 12 Heron Alexandrinus Mensurae 61.1: 301 Herodianus Historicus 3.9.2: 182 n. 84 6.12: 143 n. 12 Herodotus 1.28: 98 n. 86 1.29–33: 102 and n. 105 1.53: 97 n. 79 1.71–94: 86 n. 12 1.72: 73 n. 2 1.84–91: 87 and n. 18 1.86–92: 84 n. 4

332

Index of Passages

1.86: 97 n. 90 1.86.2: 87 1.86.3–6: 87 1.87: 87 1.88–89: 87 n. 20 1.95: 125–126 and n. 41 1.95.1–2: 101 and n. 96 1.102: 130 1.103–106: 133 n. 59 1.103.2: 101 n. 96 1.106: 130 1.107: 130 1.109–122: 130 1.126–127: 130 1.130.3: 87 n. 20, 101 n. 96, 130 1.132: 128 n. 49 1.155–156: 87 n. 20 1.180.1: 73 n. 2 1.194.2: 73 n. 2 1.194.4–5: 73 n. 2 1.204–214: 98 n. 87 1.207–208: 87 n. 20 1.214: 130 2.1: 130 3.14.11: 87 n. 20 3.17–26: 98 n. 87 3.34: 87 n. 20 3.36: 87 n. 20 3.93.1: 73 n. 1, 133 and n. 61 3.124–125: 87 n. 23 4.81.3: 92 n. 48 4.85–88: 91 n. 43 4.92: 91 n. 44 4.122–142: 98 n. 87 5.49.6: 73 n. 2 5.52.3–5: 73 n. 2 5.73.2: 95 and n. 68 6.48: 95 and n. 68 6.118.1–3: 93 and n. 58 7.22–24: 97 n. 83 7.32: 95 n. 69 7.33–36: 97 n. 83 7.43.1: 96 n. 74 7.44: 91 n. 43 7.45–46: 96 and n. 76 7.59–101: 91 n. 43 7.60: 92 n. 46

7.62.1: 80 and n. 23 7.73: 73 and n. 2 7.75: 98 n. 86 7.108: 96 n. 74 7.127.2: 96 n. 74 7.129–130: 98 n. 86 7.132: 98 n. 86 7.133: 95 n. 69 7.172–173: 8 n. 86 7.196: 96 n. 74 7.222–228: 98 n. 86 7.226: 95 and n. 71 8.26.2: 80 n. 23 8.32–33: 98 n. 86 8.66: 98 n. 86 8.97–117: 98 n. 87 8.118–120: 98 n. 87 9.96.2: 80 n. 23 9.102.4: 80 n. 23 Hesychius κ 2361, s.v. Κέρτα: 181 and n. 79 Isocrates 4.89: 97 Jamblichus Syrus Babyloniaca p. 66 Habrich: 179 n. 73 Johannes Laurentius Lydus De magistratibus reipublicae Romanae 3.53.4, p. 108 Schamp: 178 n. 71 Malalas Chronographia 6.5–11.119–122 Thurn: 100 and n. 95, 101 and n. 103 6.8.120–121 Thurn: 97 n. 79 6.10.122 Thurn: 87 n. 22 Johannes Stobaeus 2.31.115: 96 n. 72 3.7.45: 96 n. 72 Josephus Antiquitates Judaicae

Greek Texts

11.1–3: 100 n. 92 11.3–21: 116 n. 2 11.184: 121 n. 20, 123 n. 34 Bellum Judaicum 7.7.4: 177 n. 68 Julianus Imperator Epistulae 15: 36 and n. 11 Julius Africanus Chronographiae pp. 5–6 F 34 Wallraff-Roberto: 100 and n. 95 F 73 Wallraff-Roberto: 101 and n. 102 Libanius Epistulae 278: 49 n. 9 Lucianus Alexander 27: 174 n. 52 Juppiter Confutatus 14: 97 n. 79 Juppiter Tragoedus 20: 97 n. 79 43: 97 n. 79 Quomodo historia conscribenda sit 21: 174 n. 52 25: 174 n. 52 26: 174 n. 52 28: 173 n. 49 31: 173 n. 49 Rhetorum Praeceptor 18: 97 n. 84 Lydus: see Johannes Laurentius Lydus Lysias 2.29: p. 000 Manetho (FGrHist 609) T 8e: 89 n. 94 F 2: 89 n. 94 F 3: 89 n. 94

Maximus Tyrius 5.2a: 97 n. 79 13.5e: 97 n. 79 Methodius Olympius de Autexusio 3.1.5 (= p. 12–17 Vaillant): 38–41 16.3 (= p. 70–71 Vaillant): 42 de Resurrectione 1.53: 248 n. 38 Nicolaus Damascenus (FGrHist 90) F 22: 101 and n. 101 F 66: 101 and n. 101, 128 and n. 49 F 68: 84 n. 5, 87 n. 22 Novum Testamentum Actus Apostolorum 22.27–28: 305, 308–309 23.1: 307 23.23: 307 23.27: 307 1 Epistula Pauli ad Corinthios 9.7: 307 2 Epistula Pauli ad Corinthios 5.1: 248 Epistula Pauli ad Ephesios 2.12: 307 Epistula Pauli ad Philippenses 1.27: 307 2.25: 307 3.20: 307 Epistula Pauli ad Philemonem 1.2: 307 1 Epistula Pauli ad Timotheum 1.18: 307 3.7: 48 n. 1 2 Epistula Pauli ad Timotheum 2.3–4: 307 Evangelium secundum Lucam 1.36–37: 251 1.47: 320 10.34: 301 11.39: 301 11.42: 300

333

334

Index of Passages

Evangelium secundum Matthaeum 13.25: 235 Oenomaus Cynicus F 5 Hammerstaedt: 97 n. 79 Papyri P.Oxy. 5, 841 (frr. 126–128) (see Pindarus): 11 Pausanias 2.15.1: 92 and n. 53 2.22.5: 92 and n. 53 2.32.5: 92 and n. 53 3.12.7: 95 n. 69 3.17.6: 92 and n. 53 3.20.9: 266 n. 32 5.17.1: 92 and n. 53 6.19.14: 92 and n. 53 8.46.3: 93 n. 60 Phlegon Trallianus (FGrHist 257) F 34.17: 122 and n. 26 F 39.14–20: 122 n. 26 F 39: 94 and n. 65, 97 and n. 81 Photius Bibliotheca 58, 17a Bekker: 165 and n. 12 72, 43b 3 Bekker: 118 n. 21 94, 78a Bekker: 179 n. 73 Pindarus Paeanes, fr. 52k Maehler: 11 Plato Leges 706c 8: 306 821b: 321 n. 1 Timaeus 17a 6: 285 n. 45 17b 7: 285 n. 41, 293 n. 98 17c 10: 294 n. 106 18e 2: 291 n. 84 24a 2: 290 n. 83 25e 2: 293 n. 96 26a 3: 288 n. 66 26d 5: 289 n. 70

26e 2–3: 289 n. 73 27b 7: 285 n. 47, 290 n. 76 27d 6–28c 2: 280 n. 17 30a 4: 289 n. 67 30c 2: 292 n. 91 30c 8: 287 n. 57 33b 3: 287 n. 56 33d 3: 288 n. 58 33d 5: 291 n. 84 34b 6: 285 n. 44 35a 6: 288 n. 58 36c 1: 291 n. 84 36c 3: 288 n. 63 36e 3–4: 294 n. 105 38c: 312 n. 1 38d 4–5: 285 n. 86 39a 5: 291 n. 88 39c 4: 285 n. 47, 287 n. 54 39e 5: 287 n. 57 40c: 236 n. 15 41a 8: 293 n. 97 41b 7-c 1: 292 n. 92 42e 6: 299 n. 101 44b 8: 286 n. 47 44e 4: 290 n. 77 51a 3: 291 n. 85 53c 7: 289 n. 72 55a 2: 292 n. 93 56e 3: 289 n. 71 57a 7-b1: 287 n. 57 57e 5–6: 293 n. 94 58a 5: 286 n. 47, 287 n. 43 59a 4: 289 n. 67 60c 1: 288 n. 62 62c 5: 290 n. 74 65d 6: 286 n. 47 67d 5–68d 2: 279 n. 15 68a 5–6: 294 n. 103 68d 3: 290 n. 83 68e 3: 290 n. 78 69a 4: 288 n. 58 69c 5: 289 n. 67 73c 4: 291 n. 85 74d 6: 286 n. 47, 50 77e 4–5: 288 n. 65 80b 1–2: 290 n. 79 80b 4: 291 n. 84

Greek Texts

80c 6: 291 n. 87 81a 7: 288 n. 64 81c 2: 290 n. 80 82a 6: 290 n. 81 82b 4: 285 n. 43 83c 3: 284 n. 39 83d 1: 286 n. 47, 287 n. 51 84e 1: 286 n. 47 84e 10–85a 1: 284 n. 39 86e 3: 284 n. 39 86e 7: 290 n. 83 87b 8: 291 n. 85 88a 8-b 1: 292 n. 92 92c 5: 290 n. 83 Plutarchus Agesilaus 36–40: 89 n. 30 Apophthegmata Laconica 225b: 95 n. 72 Artoxerxes 1.4: 118 n. 21 7.3: 75 n. 6 Crassus 33.1.7: 143 n. 1 Lucullus 29.4: 143 n. 1 Solon 27.1: 102 Procopius Bella 1.18.52–53: 92 n. 49 1.18.54: 92 n. 50 Ptolemaeus Geographia 5.12.1 Nobbe (= Stückelberger-Graßhoff): 167 n. 16 5.13.1 Nobbe (= Stückelberger-Graßhoff): 167 n. 16 5.13.9 Nobbe (cf. 2, p. 550 StückelbergerGraßhoff; 5.12.4 Müller): 168 and n. 21 6.5.1 Nobbe (= Stückelberger-Graßhoff): 176 n. 63

335

Scholia in Aristophanis Plutum, 8b5 (III 2/3 Regtuit): 97 n. 79 Socrates Scholasticus Historia ecclesiastica 4.26: 32 n. 6 6.13: 32 Sozomenus Gazaeus Historia ecclesiastica 6.17: 32 n. 6 Stephanus Byzantinus β 112 Billerbeck, s.v. Βόγχαι: 166 n. 13 γ 72 Billerbeck, s.v. Γηλύς: 172 n. 43 γ 79 Billerbeck, s.v. Γίνδαρα: 172 n. 44 γ 122 Billerbeck, s.v. Γωγαρηνή: 168 n. 28 ζ 32 Billerbeck-Zubler, s.v. Ζωβίδαι: 172 n. 45 μ 101 Billerbeck, s.v. Μαυρούσιοι καὶ Μαῦροι: 172 n. 45 623.9–10 Meineke, s.v. Τιγρανόκερτα: 166 705.10–13 Meineke, s.v. Ὠβαρηνοί: 166 713.5–7 Meineke, s.v. Ὠτηνή: 166 Strabo 1.3.21: 167 n. 16 5.3.2: 267 and n. 367 11.3.2: 167 n. 16 11.13.5: 76 n. 9 11.14.4: 118 n. 5 11.14.4–5: 169 and n. 26, 28 11.14.6: 169 and nt. 31 11.14.15: 86 and n. 7, 117 and n. 4 15.3.18: 128 n. 499 Suda ε 147, s.v. ἐγνωσμένον: 175 n. 60 ε 2773, s.v. ἐπιχειρήσειν: 176 n. 60 κ 1905, s.v. Κοδράτος: 165 n. 10 μ 298, s.v. Μαυρούσιοι: 176 n. 60 Syncellus: see Georgius Syncellus Themistius Orationes 34.8: 178 n. 71

336

Index of Passages

Theon rhetor Progymnasmata p. 102, 25–26 Patillon-Bolognesi 12

Vita Aesopi 92–100: 89 and n. 37 101–123: 89 and n. 36

Theophanes Confessor Chronographia p. 315 De Boor (= 446 Engl. transl. Mango/ Scott): 185 n. 100

Xenophon Agesilaus 2.28–31: 89 n. 30 Anabasis 1.2.14–18: 91 n. 43 1.6.1–11: 75 n. 6 2.4.8–9: 75 n. 6 2.5.40: 75 n. 6 3.4.13: 75 n. 6 3.5.17: 75 n. 6 4.1.3: 75 n. 6 4.3.1: 75 n. 6 4.3.3: 75 n. 6 4.3.4: 75 n. 6 4.3.20: 75 n. 6 4.4.1: 75 n. 6 4.4.4: 75 n. 6 4.4.18: 75 n. 6 4.4.21: 75 n. 6 4.5.33–34: 75 n. 6 7.8.25: 75 n. 6 Cyropaedia 1.1.3: 129 n. 53 1.2.1: 129, n. 51, 131, 132 n. 55 1.2.2: 131 1.3.1–4.26: 131 1.3.9: 131 1.4.16–24: 77 n. 13 1.4.25: 132 n. 55 1.5.2: 77 n. 13, 131 2.1.5: 77 n. 13 2.1.6: 78 n. 15 2.4.1–9: 77 n. 14 2.4.7–8: 77 n. 13 2.4.9: 131 2.4.12: 77 n. 12 2.4.16: 78 n. 15 2.4.18: 77 n. 12 2.4.21: 78 n. 15 2.4.22: 77 n. 12, 131 2.4.30: 78 n. 15 2.4.31: 77 n. 12 3.1.1: 77 n. 12

Vetus Testamentum (LXX) Daniel 3.77: 301 [Esdras I] 2.12–30: p. 5 n. 29 Esdras II 4.7–24: p. 5 n. 29 4.6–7: p. 6 n. 34 Esther 14.5: 301 Genesis 1.21: 301 Job 9.13: 301 2 Macchabeorum 4.9: 301 5.21: 97 n. 85 6.1: 307 4.11: 307 11.25: 307 3 Macchabeorum 3.4: 307 3.21: 308 3.23: 308 4 Macchabeorum 2.8: 307 2.23: 307 8.7: 307 17.9: 307 Numeri 11.13: 300 Regnorum I (= Samuelis I) 2.1: 251 and n. 48 Sapientia 2.23: 238 2.24: 238

Latin Texts

3.1.1–43: 77 n. 12 3.1.14: 78 n. 16 3.1.14–15: 76 n. 8 3.1.2: 77 n. 11 3.1.7: 76 n. 8 3.1.20: 76 n. 8 3.1.23: 76 n. 8 3.1.27–28: 76 n. 8 3.1.36–41: 76 n. 8 3.1.37: 77 n. 11 3.1.38: 131 3.1.38–40: 78 n. 16 3.1.40: 131 3.1.41: 77 n. 11 3.1.43: 77 n. 11 3.2.1 –2: 134 n. 63 3.2.1–3: 76 n. 8 3.2.4–17: 78 n. 15 3.2.11: 76 n. 8 3.2.14–24: 77 n. 12 3.2.25–30: 77 n. 14 3.2.28–31: 77 n. 12 3.3.1–4: 77 n. 12 3.3.5: 76 n. 8 3.3.43: 77 n. 13 3.3.50: 77 n. 13 4.1.8: 77 n. 13 4.2.9: 76 n. 8 4.2.18: 76 n. 8

4.2.43: 76 n. 8 4.5.4: 76 n. 8 4.5.35: 76 n. 8 4.5.40: 77 n. 13 4.6.2–7: 77 n. 13 5.1.27: 76 n. 8 5.2.27–28: 77 n. 13 5.3.42: 76 n. 8 5.4.12: 77 n. 13 5.4.13: 78 n. 15 6.1.21: 76 n. 8 6.2.1–3: 77 n. 14 6.2.9–11: 77 n. 14 7.2.2–29: 84 n. 5, 87 n. 22 7.5.30–32: 77 n. 13 8.3.25: 76 n. 8 8.4.1: 76 n. 8 8.4.24: 76 n. 8, 77 n. 11 8.5.17: 131 8.7: 131 Hellenica 4.8.21: 80 n. 22 5.14: 89 n. 30 Memorabilia 4.7.5: 312 n. 1 Zonaras 11.23–24: 177 n. 67 12.2: 174 n. 52

Latin Texts Boëthius Consolatio philosophiae 2.2: 87 n. 22 Caesar De Bello Gallico 1.26.3: 156 n. 73 Cicero Epistulae ad Quintum fratrem 1.11.23: 129 n. 53 Tusculanae Disputationes 1.42.101: 95 n. 72

Curtius Rufus 3.2.2: 92 n. 47 3.2.8–9: 91 n. 42 4.12.9: 91 n. 42 4.15.32: 95 n. 70 Eutropius 8.3.2: 168 n. 21, 175 n. 55 8.6.2: 175 n. 55 8.10.2: 174 n. 52 Festus Breviarium 14: 175 n. 55

337

338

Index of Passages

15: 168 n. 21 20: 175 n. 55 Fronto Epistulae ad Verum Imperatorem 2.18: 171 n. 39 Principia Historiae 16: 174 n. 52 19: 174 n. 52 Hieronymus Stridonius Chronicon a. 113, 117 p. Chr., pp. 196–197 Helm: 175 n. 55 Julius Valerius Historia Alexandri Magni 1.13: 28 n. 21 Justinus Epitoma historiarum Philippicarum Pompei Trogi 1.7.9: 84 n. 5, 87 n. 22 3.1: 94 n. 63 Mythographus Vaticanus Primus 1.196 Zorzetti-Berlioz: 87 n. 23 Orosius Historiae adversus paganos 7.15.2: 174 n. 52 Plinius Maior Naturalis Historia 6.10.29: 167 n. 16

6.13.39: 167 n. 16 6.13.42: 167 n. 16 12.13.49: 167 n. 16 36.4.9: 92 and n. 52 Ravennatis Anonymi Cosmographia 2.12, p. 22 Schnetz: 168 n. 21 Scriptores Historiae Augustae Avidius Cassius 1.1–3: 164 n. 5 Hadrianus 5.8: 175 n. 56 7.1–2: 175 n. 57 21.11: 175 n. 55 Lucius Verus 6.9: 174 n. 52 8.1–4: 164 n. 5 Marcus Aurelius 8.6: 174 n. 52 Suetonius Tiberius 71: 147 n. 31 Tacitus Annales 2.56: 144 n. 10 12.45: 145 n. 12 Vitruvius De architectura 5.12.3: 151 n. 53

Armenian Texts¹ Translations into Armenian: [Callisthenes] Historia Alexandri Recensio A § 29: 27

§ 170: 25 § 206: 26 Recensio B § 29: 27 n. 2 § 206: 26 n. 16

1 The English titles of the original Armenian works follow the Bibliography of Thomson: Thomson, R. W. (1995), A Bibliography of Classical Armenian Literature to 1500 AD, Turnhout.

Armenian Texts

Dionysius Areopagita: see supra, Greek texts Eusebius Caesariensis Chronicon 1.103 : 86 and n. 15 2.104: 123 n. 30 Novum Testamentum Evangelium secundum Ioannem 6.6: 151 n. 55 12.6: 151 n. 55 13.29: 151 n. 55 29.29: 151 n. 55 Evangelium secundum Marcum 6.25: 155 n. 68 6.28: 155 n. 68 Evangelium secundum Matthaeum 14.8: 155 n. 68 14.11: p. 155 n. 68 Philo Iudaeus De providentia 2.97 Auch. = Pind. fr. 107 Maehler: 11 Plato Timaeus 2.10 (ms. V 1123) = 76.8–9: 285 n. 45 2.16–17 (ms. V 1123) = 76.17: 285 n. 41, 293 n. 98 2.28 (ms. V 1123) = 77.8: 294 n. 106 3.33 (ms. V 1123) = 78.19–20: 291 n. 84 9.30 (ms. V 1123) = 86.8: 290 n. 83 11.28 (ms. V 1123) = 88.22–23: 293 n. 96 12.1 (ms. V 1123) = 88.31: 288 n. 66 12.25 (ms. V 1123) = 89.28: 289 n. 70 12.30 (ms. V 1123) = 90.2: 289 n. 73 13.17 (ms. V 1123) = 90.27: 285 n. 47, 290 n. 76 13.31–14.15 (ms. V 1123) = 91.11–91.32: 280 15.24 (ms. V 1123) = 93.17: 289 n. 67 16.4 (ms. V 1123) = 94.1: 292 n. 91 16.11 (ms. V 1123) = 94.9: 287 n. 57 18.11 (ms. V 1123) = 96.19: 287 n. 56 18.30 (ms. V 1123) = 97.6: 288 n. 58 18.32 (ms. V 1123) = 97.8: 291 n. 84 19.16 (ms. V 1123) = 97.27: 285 n. 44 19.31 (ms. V 1123) = 98.15: 288 n. 58

339

20.27 (ms. V 1123) = 99.12–13: 291 n. 84 20.29 (ms. V 1123) = 99.15: 288 n. 63 21.13–14 (ms. V 1123) = 100.6: 294 n. 105 23.21 (ms. V 1123) = 102.26: 291 n. 86 24.5 (ms. V 1123) = 103.11: 291 n. 88 24.19 (ms. V 1123) = 103.27: 286 n. 47, 287 n. 54 25.3 (ms. V 1123) = 104.19: 287 n. 57 26.28 (ms. V 1123) = 106.26: 293 n. 97 27.4 (ms. V 1123) = 107.9: 292 n. 92 28.29–30 (ms. V 1123) = 109.25: 293 n. 101 30.26–27 (ms. V 1123) = 111.30: 286 n. 47 31.13 (ms. V 1123) = 112.22: 290 n. 77 38.21–21 ms. V 1123 = 120.11 Sowk‘rean 1877: 291 n. 85 41.16 (ms. V 1123) = 123.26: 289 n. 72 43.14 (ms. V 1123) = 125.29: 292 n. 93 45.24 (ms. V 1123) = 128.10: 289 n. 71 46.5 (ms. V 1123) = 128.25: 287 n. 57, 288 n. 60 47.2 (ms. V 1123) = 129.22: 293 n. 94 47.9 (ms. V 1123) = 129.32: 286 n. 47, 287 n. 53 48.15 (ms. V 1123) = 131.2–3: 289 n. 67 50.6 (ms. V 1123) = 132.29–30: 288 n. 62 52.32 (ms. V 1123) = 135.24: 290 n. 74 56.31 (ms. V 1123) = 139.29: 286 n. 47 59.17–60.21 (ms. V 1123) = 142.24–143.27: 279 59.33–60.1 (ms. V 1123) = 143.3–4: 294 n. 103 60.22 (ms. V 1123) = 143.27: 290 n. 83 60.30 (ms. V 1123) = 143.37: 290 n. 78 61.6 (ms. V 1123) = 144.9: 288 n. 58 61.27 (ms. V 1123) = 144.30: 289 n. 67 67.2 (ms. V 1123) = 150.2: 291 n. 85 68.20 (ms. V 1123) = 151.22–23: 286 nn. 47, 50 73.23 (ms. V 1123) = 155.24–25: 288 n. 65 75.16 (ms. V 1123) = 158.21: 290 n. 79 75.19 (ms. V 1123) = 158.25: 291 n. 84 75.30 (ms. V 1123) = 158.36: 291 n. 87 76.22 (ms. V 1123) = 159.24: 288 n. 64 76.33 (ms. V 1123) = 160.1: 290 n. 80 77.27–28 (ms. V 1123) = 160.33: 290 n. 81 78.2 (ms. V 1123) = 161.6–7: 285 n. 43 79.19 (ms. V 1123) = 162.27: 284 n. 39

340

Index of Passages

79.27 (ms. V 1123) = 162.35–36: 286 n. 47, 287 n. 51 81.8 (ms. V 1123) = 164.16: 286 n. 47 81.19 (ms. V 1123) = 164.29: 284 n. 39 83.33 (ms. V 1123) = 167.7: 284 n. 39 84.5 (ms. V 1123) = 167.13: 290 n. 83 84.23 (ms. V 1123) = 167.33: 291 n. 85 85.28 (ms. V 1123) = 169.3: 292 n. 92 91.11 (ms. V 1123) = 174.24: 290 n. 83 Theon rhetor see supra, Greek texts Vetus Testamentum Armenice Redditum 2 Chronica 3.11: 158 n. 81 24.8: 151 n. 55 Deuteronomium 6.7: 217 n. 5 11.9: 217 n. 5 Judith 14.2: 158 n. 81 Joshua 1.8: 217 n. 5 1 Macchabaeorum 3.23: 158 n. 81 9.1: 158 n. 81 9.12: 158 n. 81 9.14: 158 n. 81 9.16: 158 n. 81 Psalmi 77.5–6: 217 n. 4 3(1) Reges 6.24: 158 n. 81 6.27: 158 n. 81

Armenian authors: Agat‘angełos History of Armenia 760.3 (394.31): 151 n. 58 760.7 (394.31): 151 n. 58 764.2 (397.2): 151 n. 58 765.2 (397.2): 151 n. 58 765.5 (397.12): 151 n. 58 766.1 (397.16): 151 n. 58

Eznik of Kołb Ełc ałandoc‘ 1.4 (= par. 4 Mariès-Mercier): 38 n. 17, 39–41 1.11 (= par. 43 Mariès-Mercier): 42 and n. 22 Grigor Magistros Letters 2, § 1.1–3: 206 2, § 2.4–7: 206 2, § 3.8–9: 207 2, § 4.10–12: 207 2, § 5.13–15: 207 2, § 6.16–19: 207 2, § 7.20–24: 208 2, § 8.25–29: 208 2, § 9.30–34: 208 2, § 10.35–38: 208 2, § 11.39–42: 209 2, § 12.43–47: 209 2, § 13.48–52: 209 2, § 14.53–57: 209 2, § 15.58–60: 209 Kirakos Ganjakec‘i History of Armenia 4–5: 217 n. 6 Koriwn Life of Maštoc‘ 24: 151 n. 57 Movsēs Dasxuranc‘i (also known as Kałankatowac‘i) History of the Caucasian Albanians 2.14: 155 n. 69 Movsēs Xorenac‘i History of Armenia 1.2: 93 and n. 56 1.2–3: 124 n. 38 1.12: 85 n. 6 1.15: 74 n. 5 1.16: 74 n. 5 1.18: 74 n. 5 1.19: 125 n. 40 1.21: 125 n. 40 1.22: 80 n. 20, 119 n. 11, 131

Epigraphic Texts

1.24–31: p. 1; 79 n. 19, 115 1.24: 80 n. 20 1.25: 125 n. 40, 131 1.26: 125 n. 40, 131 1.27: 131 1.29: 125 n. 40, 131 1.30: 126 n. 43, 127 nn. 44, 45 1.31: 125 n. 39 1.32: 258 1.α: 119 n. 11, 125 n. 39 1.β: 119 n. 11, 124 n. 39, 126 n. 43 2.8: 85 n. 6 2.10: 90 n. 41, 120 n. 14, 122 n. 24 2.11: 85

341

2.11–13: 83 n. 2, 85, 115, 118 n. 5 2.12–13: 85–99, passim 2.13: 119 n. 13, 120 2.44–46: 118 n. 8 2.48: 185 n. 99 2.49: 169 n. 31 2.62.15: 155 n. 69 2.65: 185 n. 100 3.45.7: 151 n. 56 [P‘awstos Buzand] (Buzandaran Patmut‘iwnk‘) Epic Histories 3.7: 91 and n. 45

Babylonian Texts Nabonidus Chronicle 2.16: 86 n. 11

2.16–18: 86 n. 16, 88 n. 15

Egyptian Texts The “Aretalogy” of Imhotep-Asclepius: 89 n. 31 The Demotic Chronicle: 89 n. 31

The Dream of Nectanebos: 89 n. 31 The Prophecy of Petesis: 89 n. 31

Syriac Texts Chronicle of Arbela

chap. 3, pp. 8–10 Mingana (= 7–10 Kawerau = 30–31 Ital. transl. Ramelli): 179 n. 73

Epigraphic Texts AE 1910, 161 (ILS 9117 = AE 1911, 83): 144 n. 8 1956, 31: 145 n. 14 1968, 510: 144 n. 7 1968, 511: 144 n. 7 CIL III 1559 (cf. CIL III, 8009; AE 1960, 339; AE 2007, 36; 1205): 146 n. 16 6052 (cf. III, 13627 = ILS, 394): 144 n. 9 XIV 3608 (EE IX, 470 = ILS 986 = InscrIt IV.1, 125): 146 n. 228

IGRR I 222 (= CIG III, 6559 = OGIS 382 = IG 1472): 143 n. 2, 170 n. 38 IGUR II 415: 170 n. 38 SEG XV 838: 171 n. 38 XLV 1322 (cf. SEG XLVII, 1163; SEG LIII, 2228): 145 n. 13

List of Contributors Chiara Aimi holds a PhD in Literature, Philology and History at the University of Bologna. She is working on the ancient Armenian translation of Plato’s Laws, preparing a critical edition of this work (book V). Valentina Calzolari is Professor of Armenian Studies and President of the Association Internationale des Études Arméniennes. She is co-director of the Commentaria in Aristotelem Armeniaca. Davidis Opera (Philosophia Antiqua). In 2014, she published Armenian Philology in the Modern Era: From Manuscript to Digital Text (Handbuch der Orientalistik VIII, vol. 23/1). Alessandro Capone is Assistant Professor of Early Christian Literature at the University of Salento – Lecce (Italy). He works on the manuscript tradition of some Greek and Latin writers, the polemical literature of the fourth–fifth centuries and the history of classical scholarship. He published Pseudo-Atanasio. Dialoghi IV e V sulla santa Trinità, testo greco con traduzione italiana, versione latina e armena, Lovanii 2011 (Corpus Scriptorum Christianorum Orientalium 634, Subsidia 125). Furthermore, he recently published (with A. Beccarisi) Aliter. Controversie religiose e definizioni di identità tra Tardoantico e Medioevo, Roma 2015. Lia Raffaella Cresci is Full Professor of Byzantine Philology at the University of Genoa. She works on Byzantine poetry (sixth–twelfth centuries), historiography (sixth–thirteenth centuries) and rhetoric (twelfth century). Giulia D’Alessandro is a PhD student in Classical Philology at the University of Genoa and at the University of Cologne. Her interests concern Homeric epics and the Hellenistic Age. She is now working on the critical edition, with commentary, of the fragments of the grammarian Ammonius of Alexandria. Armenuhi Drost-Abgaryan is Extraordinary Professor of Armenian Studies at the University of Halle-Wittenberg (Germany). She studied Armenology, Classical Philology, Byzantine and Christian Oriental studies in Erevan, Moscow and Halle. Her research interests include the comparative study of Armenian Literature in the context of the Greek-Byzantine and Christian Oriental languages and cultures.

344

List of Contributors

Federico Frasson holds a PhD degree in Historical Sciences of Antiquity and is currently Research Fellow at the University of Genoa. His research focuses mainly on fragmentary Greek historiography, ancient warfare, Latin epigraphy and ancient topography. Gianfranco Gaggero is Associate Professor of History of the Ancient Near East at the University of Genoa. His main fields of research are Thrace, Roman Africa, Protobyzantine historiography, Suetonius. His work has focused in particular on the Syro-Palestinian area in the Hellenistic and Roman period, as well as the impact of Near Eastern figures such as Semiramis, Sesostris, Ramesses II in classical times. Francesca Gazzano is Associate Professor of Ancient Greek History at the University of Genoa, where she also teaches courses in Greek Epigraphy and Greek Historiography. Her major research fields include Greek political history, Greek historiography – especially Herodotus and the lost historians of Lydia – Oratory and Greek Interstate Relations and Diplomacy. She is joint-director (with L. Santi Amantini) of the series “RIS, Rapporti Interstatali nella Storia” (L’“Erma” di Bretschneider, Rome). Anahide Kéfélian holds a Research Fellowship of the Gulbenkian Foundation. She is writing a PhD dissertation at the University Paris-Sorbonne on the Roman presence in the kingdom of Greater Armenia. Theo Maarten van Lint has been Calouste Gulbenkian Professor of Armenian Studies at the University of Oxford and Fellow of Pembroke College, Oxford since 2002. His research interests include the development of the Armenian poetic tradition, medieval and early modern Armenian literature. He is preparing a translation and commentary of the epistolary of the eleventh-century erudite nobleman Grigor Pahlawuni Magistros, and works on the reception of Ezekiel’s Throne Vision in Armenian literature and art. He is joint-editor with Dr Emilio Bonfiglio of the volume on Literature in the Handbook of Armenian Studies of the Association Internationale des Études Arméniennes, published in the Handbuch der Orientalistik by Brill, Leiden. Together with Mr Robin Meyer he is the curator of the exhibition Armenia: Masterpieces from an Enduring Culture (Bodleian Library, Oxford, October 23, 2015–February 28, 2016), and co-author and editor of its catalogue.

List of Contributors

345

Francesco Mari holds a PhD in Greek History at the Universities of Strasbourg and Genoa. He studies ancient Greek politeness, with regard to conversation, gestural expressiveness and meeting occasions in the Archaic and Classical periods. He also works on the symbolic features of diplomatic encounters between the Greeks and the Achaemenid Persians. Moreno Morani is Full Professor of Linguistics at the University of Genoa. Alessandro Orengo is Assistant Professor in Linguistics at the University of Pisa. Lara Pagani is Associate Professor of Ancient Greek Literature at the University of Genoa. Her research concerns ancient Greek scholarship and grammar, Homeric studies, Greek lexicography and the reflection on language in ancient Greece. She is co-editor of the encyclopedia Lexicon of Greek Grammarians of Antiquity (LGGA), edited by Brill. Rosa Ronzitti is Assistant Professor of Historical and General Linguistics at the University of Genoa. She works on Etymology, Vedic Language and Indo-European Studies. She is the author of five monographs and forty articles published in national and internationals journals. Andrea Scala is Associate Professor of Linguistics at the State University of Milan. He works on Contact Linguistics, Sociolinguistics and Historical Linguistics with particular reference to Armenian, Romani and Italo-Romance varieties. Irene Tinti is currently a postdoctoral researcher in Armenian Studies at the University of Geneva (Département MESLO). She performs research on the Armenian Platonic dossier, its language, translation technique and attribution, presently focusing on the Laws. She is also assisting Prof. Alessandro Orengo (Pisa) in editing the forthcoming Brill volume on Armenian Linguistics. Giusto Traina, PhD, is Full Professor of Roman History at the Paris-Sorbonne and Senior Fellow of the Institut Universitaire de France. A specialist of ancient Armenia, his work focuses on Armenian historiographers (Movses Xorenac‘i and the Buzandaran Patmut‘iwnk‘). He also translated the Venice manuscript (San Lazzaro, 424) of the Armenian Pseudo-Callisthenes (Padova, 2003).

Greek Texts and Armenian Traditions an Interdisciplinary Approach

Recommend Documents