Article

Syriac Text

The Syriac version of the Alexander Romance is preserved in five manuscripts, all in Nestorian script and of recent date (the oldest, held by the British Museum, was compiled in 1708-09), and was edited in 1889 by Ernest A. Wallis Budge (The History of Alexander The Great, being the Syriac Version, edited from five manuscripts, of the PseudoCallisthenes, Cambridge, 1889, reprint. Amsterdam, 1976 = BUDGE, History). The editor has adopted the criterion of the codex optimus: this edition is mainly based on the text of the oldest codex, known as A, while the variants in the other four codices (known respectively as B, C, D and E) are noted in the apparatus. The Syriac text belongs to branch a of the Pseudo-Callisthenes tradition, in other words the ancient recension, mainly represented by the Greek codex A (Paris. 1711), together with the Armenian version (5th century AD), the Latin text by Julius Valerius (4th century AD) and the Historia de proeliis by Leo the Archipresbyter (10th century AD). However, the Syriac text does not seem to be a pure and simple translation of any of the texts of the Pseudo-Callisthenes that have come down to us, both because of the different order in which certain subjects are dealt with, and, above all, because of the inclusion of a certain number of episodes that are not recorded in any of the Greek versions known to us, for example, Alexander’s journey to China (text and trans.: BUDGE, History, p. 195-201, p. 109-113). Furthermore, the Syriac version contains a considerable number of slight variations on the original Greek, which include some modifications that can definitely be attributed to the translator, who is assumed to have been a Nestorian Christian.

1. These discrepancies with the Greek Pseudo-Callisthenes, combined with a certain number of errors in the Syriac translation of Greek proper names, led Budge to suspect that the Syriac text was the translation, completed between the 7th and 9th century AD, of an Arabic version of the original Greek (BUDGE, History, p. lxi-lxii).
2. Th. Nöldeke rejects this thesis in a lengthy and erudite study entitled Beiträge zur Geschichte des Alexanderromans (Vienna, 1890 = NÖLDEKE, Beiträge). Instead he claims that the Syriac version must have been based on a lost intermediary Pahlavi translation of the original Greek, dating from the late Sasanian era (circa 6th-7th century AD). With the exception of some cursory disagreements (For example, Samuel Fraenkel’s recension of BUDGE, History, in Z.D.M.G., 45 (1891), p. 309-330, especially p. 313-322 (= FRAENKEL, Recension) and R.N. FRYE, Two Iranian Notes, in Papers in Honour of Professor Mary Boyce, vol. I, Leiden, 1985, p. 185-188 (= FRYE, Two Iranian Notes), confined to Orientalist circles, Nöldeke’s idea was accepted almost unanimously by scholars (for example, K. BROCKELMANN (et al.), Geschichte der christlichen Litteraturen des Orients, Leipzig, 1909, p. 45; A. BAUMSTARK, Geschichte der syrischen Literatur, Bonn, 1922, p. 125 (= BAUMSTARK, Geschichte); C.A. NALLINO, Tracce di opere greche giunte agli Arabi per trafila pehlevica, in A Volume of Oriental Studies presented to E.G. Browne, Cambridge, 1922, p. 345-363 (= NALLINO, Tracce di opere greche); E. YARSHATER, Iranian National History, in The Cambridge History of Iran, III/1, Cambridge, 1983, p. 472; S. BROCK, Syriac Perspective on Late Antiquity, London, 1984, II, p. 8 (= BROCK, Syriac Perspective); M. BOYCE – F. GRENET, A History of Zoroastrianism, vol. III, Leiden-Köln, 1991, p. 60 n. 40 (= BOYCE – GRENET, A History); P. ZIEME, Alexander According to an Old Turkish Legend, in La Persia e l’Asia centrale da Alessandro al X secolo. Atti del convegno dei Lincei, Roma 9-12 Novembre 1994, Rome, 1996, p. 25 and n. 2, etc).

1. Point of the paper: Reexamine dating and Syriac version of the Alexander Romance, and put forward some arguments which, in our opinion, demonstrate that the Syriac version was based directly on a Greek text.
2. Nöldeke’s hypothesis and errors
3. Nöldeke puts forward evidence to demonstrate that the Syriac text of the Pseudo-Callisthenes is not based directly on the original Greek, but on a Pahlavi version of the Greek text, written, at the latest, in the 7th century AD and subsequently lost. The most interesting evidence is of a philological-linguistic nature and can be subdivided into two groups: on the one hand, in the Syriac translation of Greek proper nouns there are a number of phonological-graphemic phenomena which, according to the German scholar, cannot be explained if the thesis of the existence of an intermediary written Pahlavi text is not accepted. On the other hand, there are predominantly lexical phenomena, including idioms, loan words, locutions, names of people and places, which reveal the Middle Persian origin of the text and also demonstrate an accurate knowledge of Persian geography and the history of the Sasanian era, a knowledge that can only be attributed to the compiler of the presumed intermediary Pahlavi version.

The first group includes variations such as the confusion between r and l in the Syriac translation of Greek proper nouns: e.g. l instead of r is found in the spelling of the Greek noun Gránikov, in Syriac «glnyqws» (BUDGE, History, p. 253, 6)4 ; while we find r where we would expect l e.g. in «’qrydys» (BUDGE, History, p. 127, 3) for the Greek Eûkleídjv. Nöldeke stresses that this confusion cannot be explained by considering the graphemic symbols of Arabic and Syriac, both scripts that distinguish between r and l, but becomes understand- able if we presume the existence of an intermediary Pahlavi text, whose script contains a single sign for both l and r (NÖLDEKE, Beiträge, p. 14). Response:

This argument, which appears to be so incontrovertible, proves, in actual fact, to be rather weak. First and foremost, there are relatively few instances of graphemic confusion between r and l: FRAENKEL (Recension, p. 316) points out that in this Syriac text l for the Greek r is found 17 times; r for the Greek l 14 times; but there are 57 instances, in the translation of Greek proper nouns, in which l is used correctly for the Greek l and 83 in which r corresponds to the Greek r (there are, however, also 14 doubtful cases in the use of the Syriac l and an equal number in the use of the r). Moreover, Greek proper nouns are subject to all kinds of distortions, and in a completely unsystematic way. The same noun may appear in different forms within the space of a few lines: e.g. the Greek Sesógxwsiv is found in Syriac in virtually all its possible spellings, from Sisiqosas («sysyqwss», BUDGE, History, p. 70, 14) to Sisiqonos («sysyqwnws», BUDGE, History, p. 71, 12), from Sisniqos («sysnyqws», BUDGE, History, p. 76, 5) to Siusiniqos («sywsynyqws», BUDGE, History, 173, 16), even to Sisanqos or Sisnaqos («sysnqws», BUDGE, History, p. 225, 11; 226, 1 and 4; 252, 2); the Greek Parmeníwn appears sometimes as Phormion («pwrmywn», BUDGE, History, p. 187, 1), sometimes Parmaon («prm’wn», BUDGE, History, p. 188, 16), and sometimes Plimtion («plymtywn», BUDGE, History, p. 137, 8). Furthermore, it is to be noted that these graphemic confusions are only to be found in proper nouns and hardly ever appear in loan words from the Greek. Therefore, we would have to imagine that the Syriac translator was capable in nearly all the cases of common nouns borrowed from the Greek of tracing them back to their original model, despite the ambiguities of the Pahlavi script.

On the contrary, we consider it more likely that the Syriac translator, though he knew Greek, did not have a great knowledge of Greek culture. In fact, if we presume that the first translation of the Greek text into Syriac was done by two translators (similar procedures were not infrequent in translations into and from exotic languages) (M. MANCINI, L’esotismo nel lessico italiano, Viterbo, 1992, p. 71-75), the first of whom read and translated the text out loud from Greek into Syriac and the second transcribed it, and we also assume that both of them knew the language, but were not well acquainted with Greek culture, it is understandable that proper nouns, being devoid of formal-semantic links in the two translators’ linguistic competence, might be subject to all kinds of distortions.

Among the many lexical arguments Nöldeke advances to demonstrate the existence of an intermediary Pahlavi text, we should consider the cases in which the Syriac text contains nouns that are evidently Persian, but that are not found, or found in a different form, in the Greek original, such as, for example, Pariok, the name of a brigand (Syriac «prywg», BUDGE, History, p. 207, 6, 8, 12 and 14; 208, 6, 8 and 9)7 ; or Greek geographical names of regions, cities and rivers in Iran translated into Syriac with their corresponding Persian names (NÖLDEKE, Beiträge, p. 15; cfr FRAENKEL, Recension, p. 318). However, if we exclude the cases in which glosses of the type «the word x, which is y in Persian» (on which see below) and a certain number of Persian proper nouns, the number of loan words is very small, or at least not bigger than the number of Iranian loan words found in any Syriac text. Therefore, the presence of a certain number of Iranian loan words in the Syriac Alexander Romance do not provide reliable evidence to support the hypothesis of an intermediary Middle Persian version. Most of these loan words are in everyday use in Syriac: cfr, e.g., !”#$%& «nÌsyr’» naÌsira ‘hunt’8 ; ‘(#) «pyg’» paiga ‘infantry’9 ; instead +,-&+, «qwndqwr» qundaqor is a hapax legomenon (0 BUDGE, History, p. 203, 10). This is a Persian title pointed out by FRAENKEL (Recension, p. 319) and glossed in Syriac manuscript C with the Persian sardar ‘head, commander’. F. PENNACCHIETTI, Qundâkôr: un hapax siriaco del Romanzo di Alessandro tra filologia e archeologia, in M. LAMBERTI – L. TONELLI (eds.), Afroasiatica Tergestina. Contributi presentati al IX Incontro di Linguistica afroasiatica (camito-semitica), Trieste, 22-24 Aprile 1998, Padua, 1999, p. 71-82, believes that Syriac qundaqor derives from a Middle Persiankondak-avar ‘he who bears the sceptre’, in its turn a loan word from the Greek kontákion, literally ‘stick, staff’.

Nöldeke also cites a couple of cases in which the polysemy of the Syriac term can only be explained by its being a semantic calque from the corresponding Persian term. The most convincing of these cases would seem to be the use of the Syriac verb ’ekal ‘to eat’ in the sense of ‘to drink’, like the Persian xvardan, which has both meanings. However, Fraenkel (Recension, p. 315) questions whether these are really semantic calques from the Persian. Particularly, as far as the Syriac !”0%1 234!56 5#7 4! 23+87! (BUDGE, History, p. 237, 10-11) is concerned, corresponding in Greek to the sentence Ωn kaì êzemetrßsamen < ên > t¬ç megálwç deípnwç (p. 130, 12 KROLL; MÜLLER ed., 142a, l. 6, reads: oÃv êmetrßsamen ên t¬ç deípnwç), referring to two kraters whose capacity Alexander wishes to gauge by filling them with wine, FRAENKEL (Recension, p. 315 ff.) suggests changing the form of the verb ’ekal ‘to eat’ that appears in the Syriac text, namely ’ekalu(h)y «’klwhy» 23+87! into another verbal form, that is graphemically very similar, and derives from the verb ’akil ‘to measure’ (BROCKELMANN, Lexicon Syriacum, p. 325b), in other words, ’akilu(h)y «’kylwhy» 23+8#7!. In this way the Syriac text would correspond perfectly to the Greek text and this would rule out the possibility of a semantic calque from the Persian. Another argument used by Nöldeke concerns the many cases in which a Syriac term, more or less faithfully translated from the Greek, is glossed in the text by referring to Persian linguistic usage, in sentences of the type «x, which in Persian is y» (NÖLDEKE, Beiträge, p. 16; cfr FRAENKEL, Recension, p. 314). It is often a question of expressions the translator uses to create equivalences that are not merely linguistic between Greek, Persian, and sometimes even Egyptian names of deities and geographical names.

1. Grecisms
2. Nöldeke (Beiträge, p. 16) adds that the great abundance of Grecisms in the Syriac text does not suffice to challenge the validity of his conclusion, considering the large number of borrowings from the Greek that Syriac had assimilated and that, therefore were to be regarded as in common usage. It is advisable to point out immediately that the Grecisms that appear in the Syriac text are not only loan words, though the loan words constitute the largest category. For an exhaustive analysis of Greek loans in Syriac see: A. SCHALL, Studien über griechischen Fremdwörter im Syrischen, Darmstadt, 1960. They also include those Greek terms that were part of the everyday language in Syriac, like «lqn’» lekánj and «klmys» xlámuv, pointed out by Nöldeke (Beiträge, p. 16 n. 11) as being among the Greek elements of no importance.

1. Calques from the Greek:
2. The calques from the Greek to be found in the Syriac PseudoCallisthenes are almost exclusively structural calques, in other words, compound words that reproduce the components of the term taken as a model in the indigenous language. This implies that the person who makes a calque has a good enough linguistic competence in the model language to be able to analyze the parts of the compounds. Moreover, the presence of calques, in the Syriac Romance, albeit not a large quantity, is particularly revealing because they are coined formations, not to be found anywhere else in Syriac literature. Nöldeke (Beiträge, p. 12-13) notices the presence of some of these new formations based on the Greek, but he uses them merely to demonstrate, in opposition to Budge, that it is impossible for this text to have been based on an Arabic version17 . One example of a structural calque is the Syriac translation of the Greek word Åinokérwtev as «qrny nÌyr’» (BUDGE, History, p. 211, 15), mentioned above (cfr §3); it is to be noted that in another passage in which rhinoceroses are referred to (p. 109, 18 KROLL), the Syriac translator resorts to a paraphrase: ‘beasts with horns on their noses’ (BUDGE, History, p. 174, 10-11; trans. 97). In addition to this, we also think the following cases are of interest:

1. • 1) «plg sny’» (BUDGE, History, p. 10, 17; 14, 15; 16, 5), calque from the Greek mes±liz ‘middle-aged’ (p. 6, 5 and 6 KROLL); in the feminine in BUDGE, History, 209, 13 «plgt snyh hwt», which in Greek corresponds to the syntagm mésjv ™likíav tugxánousa (cfr NÖLDEKE, Beiträge, p. 12).
   • 2) ‘8#) 2*4F «twry pyl’» (BUDGE, History, p. 174, 14; trans. p. 98) that literally translates taureléfantev in the Greek text (p. 109, 21 KROLL); these are fantastic animals that Alexander encounters during his journey through India and that he writes about in his letter to Aristotle.
   • 3) ‘8A* C,”< «‘rqy rgl’», literally ‘with twisted legs’ (BUDGE, History, p. 174, 15; trans. 98) which corresponds to the Greek ïmantópodev; in the same context as the previous one, this compound adjective refers to human beings with six hands, teeth like dogs and faces like women. At this point the manuscripts read «‘qrby rgl’» ‘with scorpions’ feet’ (cfr BROCKELMANN, Lexicon Syriacum, p. 544b), corrected by Budge as «‘rqy rgl’»; the latter also appears in BUDGE, History, p. 177, 16 (trans. p. 99: ‘whose feet are twisted’); cfr BROCKELMANN, Lexicon Syriacum, p. 551a.
   • 4) ?#81 C8<F «t‘ly lly’» (BUDGE, History, p. 175, 17; trans. p. 98 ‘night-foxes’), based on the Greek nuktalÉpekev handed down in manuscript A (p. 110, 1 KROLL, apparatus; in the text Kroll prefers nuktálwpev; the same reading as manuscript A is to be found in the Armenian version and in Julius Valerius); the term is found in the same context as the two previous calques; it refers to reptiles that the Indians call ‘night-foxes’.
2. [12:52 PM]
   • 5) «‘mry lly’» (BUDGE, History, p. 176, 4; trans. p. 98): this form, whose first component is the verb ‘mr ‘habitavit, vixit’ (cfr BROCKELMANN, Lexicon Syriacum, p. 532a) is corrected by Budge as «‘wrby lly’» (p. 176, apparatus), in order to obtain a first component that means ‘raven’ (cfr BROCKELMANN, Lexicon Syriacum, p. 546b); this correction, if acceptable, returns to a structural calque from the Greek nuktikórakev (p. 110, 4 KROLL), namely, ‘night ravens’.
   • 6) «dwq’ d-s‘’» ‘horoscope’ (BUDGE, History, p. 8, 12 and 9, 6; trans. p. 5 and p. 6 ‘watcher of the hours’), a structural calque from the Greek Üroskópon (p. 5, 3 and 7 KROLL); the first occurrence of this word is preceded by the explanatory gloss ‘which they call in Greek…’; the ‘normal’ word for the ‘horoscope’ in Syriac is «byt yld’» (BUDGE, History, p. 9, 9 and 10; p. 10, 10; p. 27, 2), or malasa (cfr BROCKELMANN, Lexicon Syriacum, p. 390b), maray sa‘e (BROCKELMANN, Lexicon Syriacum, p. 401b), or also only sa‘ta, literally ‘hour’ (BROCKELMANN, Lexicon Syriacum, p. 765a).
   • 7) «dmwt qrnt’» (BUDGE, History, p. 20, 8; trans. p. 12 ‘horned’, literally ‘image of horns’), corresponding to the Greek adjective kerasfórov (p. 12, 13 KROLL, attribute of Mßnj).

‘Significant’ errors: misunderstandings of the Greek

(1) In an epistle from Darius in reply to two of his satraps who had asked for his advice and help in facing the danger represented by the arrival of Alexander, the Achaemenid sovereign exhorts the two dignitaries to resist him by doing battle, without yielding to the temptation of ignoble flight, which would lead them to ruin. In any event, as there is no loan word from the Greek in Syriac, we must suppose that here the presence of the Syriac †ir must be a misunderstanding due to the great phonic similarity between the Greek word (as pronounced in the late Hellenistic era) and the Persian name. Note that, both in this passage and in the gloss mentioned, the name of the deity appears in the form “#9, with an initial †e†, like the name Tiridates, translated as K/-/”9 (BUDGE, History, p. 43, 2); elsewhere in Syriac, however, the Persian name Tir, especially in compounds, is found with an initial taw (cfr R. PAYNE SMITH, Thesaurus Syriacus, Oxford, 1879-1901, reprint. Hildesheim, 1981, vol. II, c. 4429: «tyrdt» [= PAYNE SMITH, Thesaurus Syriacus]). For other cases with an initial †e†, cfr PAYNE SMITH, Thesaurus Syriacus, vol. I, col. 1421 («†’rd’†ys») and the Supplementum of MARGOLIOUTH (p. 143b: «†yrgwsnsp»; «†yrmh»). Furthermore, in Syriac the Greek qßr would be translated as !F+#: «Ìywt’» (cfr BROCKELMANN, Lexicon Syriacum, p. 229a): as in BUDGE, History, 234, 8, corresponding to the Greek qjría (p. 129, 6 KROLL).

(2) If this were the only error of the kind to be found in the Syriac text, we could not exclude with reasonable certainty that it may be attributed to the translator of the hypothetical intermediary Pahlavi version, all the more so, since the Greek word is confused with a Persian name. The name Heracles can be recognized in the first form, but the two following forms do not have a plausible meaning in Syriac. Prof. Fabrizio Pennacchietti for pointing out the fact that the sequence in question makes sense in a dialect of north-eastern New Aramaic, where it means literally «Heracles who is rising in altitude», in the sense of ‘growing up’: this is, therefore, evidence of a later copy of the Syriac text, in which a copyist who was a New Aramaic speaker corrected the Syriac sequence that did not make sense, since he did not have the Greek text to hand, and used his own linguistic competence to interpret an ambiguous syntagm. Cfr now F. PENNACCHIETTI, Interventi in neoaramaico da parte di copisti della versione siriaca del Romanzo di Alessandro, in Bayn al nahrayn. Mesopotamia Quarterly (Baghdad), no. 97-100, vol. 25 (1997), p. 53-62 (p. 389-398) and F. PENNACCHIETTI, Una frase neoaramaica nella versione siriaca del Romanzo di Alessandro, in C. BAFFIONI (ed.), Atti del Convegno di Napoli-Sorrento, 28-31 ottobre 1998, Alessandria, 2000, p. 73-77.

(3) A slightly different case is a passage in which Alexander’s killing of a large stag (∂lafov megístj) assumes the significance of a prediction: on the spot (tò ∂dafov, literally ‘mound’) where the stag died, the Macedonian will found a city. At a certain point the Greek text reads: paragenámenov [i.e. ˆAlézandrov] dè êpì toútou toÕ êdáfou ör¢ç ktl. (p. 28, 12 KROLL); instead, the corresponding Syriac passage reads: ‘[…] he returned and came to the stag’ (‘8/! «’yl’»: BUDGE, History, p. 68, 16, trans. p. 38). As Budge suggests, it is very likely that in this case the Syriac translator confused ∂lafov ‘stag’ with ∂dafov ‘mound, spot’.

(4) The beginning of the second book of the Alexander Romance recounts how Alexander reaches Plataeae, a Greek city where Proserpine is worshipped: (p. 63, 14- 15 KROLL). Instead, the second book of the Syriac version begins thus: «Again Alexander set out from Corinth and came to Plataeae, a city of the Athenians, where they worship fire». The presence of a Zoroastrian cult of fire at Plataeae is clearly absurd and can only be explained by imagining that the Syriac graphemic sequence in question, namely,«l-nwr’ plÌyn» (BUDGE, History, p. 113, 2-3) ‘they worship fire’, is an accidental corruption, due to a later copyist, of «l-qwr’ plÌyn», the form nura, in fact, can easily be restored as qora, namely, the exact transcription of the Greek Kórj.

1. The historical and cultural context
2. The philological arguments advanced so far, and particularly the presence of errors resulting from misunderstandings of the Greek text, demonstrate — I believe — that the Syriac Alexander Romance is not based on an intermediary Middle Persian version, but is, on the contrary, a translation into Syriac of an original Greek version. This confirms the hypothesis briefly advanced by Frye (Two Iranian Notes, p. 186-187), according to whom the Syriac version of the PseudoCallisthenes could not have been based on a Pahlavi text, since it makes no mention of the legend, attributed by ™abari (Annales, I, 694-702) to the «wise ancients» and present in the Sahnama of Ferdousi and in the Eskandarnama of Nezami, which legitimated Alexander by making him the stepbrother of Darius III (Cfr also C.A. CIANCAGLINI, Alessandro e l’incendio di Persepoli nelle tradizioni greca e iranica, in A. VALVO (ed.), La diffusione dell’eredità classica nell’età tardo antica e medievale, Atti del Convegno di Trieste, 19-20 settembre 1996, Alessandria, 1997, p. 78 and n. 51 ( = CIANCAGLINI, Alessandro e l’incendio di Persepoli).

Finally, as far as the translation of Greek scientific works into Middle Persian during the late Sasanian era is concerned, despite the fact that none of these works has come down to us, it is virtually certain that they existed, as Pahlavi. For example, in the Denkart (p. 428, 15 ff. and p. 412, 17 ff. ed. Madan) we read that, at the time of Sahpuhr I, many Greek and Indian scientific texts were collected and added to the Avesta. On Greek texts translated into Syriac from Pahlavi, cfr NALLINO, Tracce di opere greche; W.H. BAILEY, Zoroastrian Problems in the Ninth-Century Books, Oxford, 1943, p. 158; A. PAGLIARO, Letteratura della Persia preislamica, in A. PAGLIARO – A. BAUSANI, Storia della letteratura persiana, Milan, 1960, p. 89-90; J.-F. DUNEAU, Quelques aspects de la pénétration de l’hellénisme dans l’empire perse sassanide (IVe -VII e siècle), in P. GALLAIS – Y.-J. RIOU (eds.), Mélanges offerts à René Crozet, Poitiers, 1966, vol. I, p. 13-22; M. ULLMANN, Die Medizin in Islam (Handbuch der Orientalistik, Ergänzungsband VI, 1), Leiden, 1970, p. 103-107. Byzantine also. Cfr Agathias, II 28, 1 on the translations of Greek works into Middle Persian that were commissioned by Chosroes I; cfr also D. GUTAS, Greek Thought, Arabic Culture, London-New York, 1998, p. 25. Arabic also. Artaxerxes I, Sahpuhr I and Chosroes I are the three Sasanian sovereigns who had scientific Greek and Indian works translated into Persian, according to the Kitab alFihrist of Ibn al-Nadim (238 f.); cfr NALLINO, Tracce di opere greche, p. 362-363.

1. There exists yet another shorter and secondary version of the Syriac Legend of Alexander in the western-Syriac Chronicle of the Pseudo-Dionysius (J.-B. CHABOT (ed. and trans.), Incerti autoris Chronicon Pseudo-Dionysianum vulgo dictum, I (Corpus Scriptorum Christianorum Orientalium, 91, p. 121; Script. Syri, 43, p. 66), Louvain, 1927, 1949, p. 41, 15-45, 10 (text), p. 33, 21-36, 20 (trans.); cfr REININK, Alexandre, p. 151 n. 8). Finally, there is also a brief biography of Alexander in Syriac (y P. DE LAGARDE, in Analecta Syriaca, London, 1855, p. 205-208; trans. in BUDGE, History, p. 159-161; cfr BAUMSTARK, Geschichte, p. 125 n. 4).
2. Neither the author, nor the precise date of the Syriac version of the Alexander Romance are known. But a series of features, particularly a certain number of biblical quotations (BUDGE, History, p. lix; cfr also FRAENKEL, Recension, p. 320), suggests that the Syriac translator was a Nestorian Christian, perhaps a Monophysite (Nöldeke has some doubts about this: cfr Beiträge, p. 17) and that he must have translated the text around the 7th century. As regards dating the work, in addition to the aspects pointed out by Nöldeke (Beiträge, p. 16), it is worth underlining a somewhat significant fact: every time we find the name of the Achaemenid king Zérzjv in the Greek text, in the Syriac text it is replaced by Kosro («kwsrw»). This occurs about a dozen times (Cfr BUDGE, History, p. 119, 2; 120, 16; 121, 8; 123, 4; 132, 16; 136, 9; 138, 8; 158, 9; 253, 11; 236, 4; 237, 13. See NÖLDEKE, Beiträge, p. 13 and n. 4) and is evidently intentional, though it is not very clear whether the allusion is to Chosroes I or Chosroes II. We have already mentioned the important role played by the Oriental Christians within the orbit of Sasanian power. Their knowledge of Syriac, Greek and Persian (Cfr BROCK, Syriac Perspective, VI, p. 17-18) and the geographical and cultural context in which they lived between the 5th and 7th centuries must certainly have favoured the use of various Persian terms in the text, and — moreover — explains the anonymous author’s familiarity with the history and geography of Iran. In any event, it must be remembered that not only is the number of Persian lexical items in the Syriac Alexander Romance relatively small (if the ethnonyms and anthroponyms are excluded), but it is not always easy to attribute them with any certainty to Middle or to New Persian.

In fact, the difference between Middle and New Persian — if only the lexicon is considered — is not always very perceptible, except in cases in which it is a question of words that are certainly Arabo-Persian. Consequently, in my opinion, it is possible to suggest that at least some of the Persian forms may have been added to the Syriac text a long time after it was first written.