Ērān ud Anērān
Webfestschrift Marshak 2003


Agathias and the date of Zoroaster

Gherardo Gnoli

There is one good reason for having decided to present this short contribution of mine among the miscellany of studies in honour of B.I. Marshak. In the course of his calendar studies, the distinguished scholar did not fail to concern himself with the date of Zoroaster, even though he obtained results that I consider can no longer be agreed with. In his article on the historico-cultural significance of the Sogdian calendar, he proposed dating the Iranian prophet "not later than the eighth century B.C., and possibly even earlier"1, in declared agreement with my preceding position2, from which I have progressively distanced myself3, to the extent of fully rehabilitating W.B. Henning's thesis in my Yarshater Lectures in 19974.

Since my new position, which substantially corresponds to the one the late great Cambridge scholar brilliantly stated in a contribution presented at a colloquium held at Croydon under the auspices of the World Zoroastrian Organisation in late 19935, has been subjected to a detailed criticism by Professor A.Sh. Shahbazi6, in order to further clarify my thinking on the matter I thought the person we are honouring might enjoy having published in his Festschrift the first of a series of notes dedicated to a no less detailed defence of the historical exactitude of the traditional date of Zoroaster and thus of the results progressively obtained, after the discovery by H. Lewy7 regarding the drastic reduction of the Arsacid period under the first Sassanians, by S.H. Taqizadeh8 and by W.B. Henning9 in the 'forties and recently reproposed by I. Gershevitch and by the present author.

I must confess however that in delivering these pages to the press I also cherished a special hope - that Professor Marshak, as he is no stranger to the non-trivial problems presented by ancient Iran in the complex studies on calendars and chronology, after reading them will refrain from joining the camp of those who, probably to avoid tackling such a difficult topic or that they erroneously consider insignificant, in practice hide behind an a priori non liquet on the fundamental issue of the date of Zoroaster, at best making mention of their generic and substantially unfounded preferences10.

In his detailed article on what he calls "recent speculations" concerning the traditional date of Zoroaster (these "speculations" are allegedly those of my Zoroaster in history11, as well as of those of the late-lamented I. Gershevitch12 and, for the hypothesis of a Greek origin of that date, of P. Kingsley13) Professor Shahbazi makes several references to Agathias' Histories 14, albeit implicitly15.

The passages from Agathias mentioned in J.D. Frendo's translation16 are respectively II 24.617, II 25.7-818 and IV 24.119. Prof. Shahbazi also mentions IV 30.2-4 for the use the historian makes of the Persian sources20.

The first of these passages, II 24.6, is reported by Professor Shahbazi as follows:

"It is not possible to fix with any precision the dates of his floruit ... The Persians simply say that he lived in the reign of Hystaspes without making it clear whether they mean the father of Darius or some other monarch of the same name"21.

The text is thus clear and this translation does not differ appreciably, in our opinion, from that of Averil Cameron, who made a thorough comment of this and the two other passages22. What is instead astonishing is the interpretation Professor Shahbazi gives of it. Since it underpins his critique of the thesis of the historical exactitude of the traditional date of Zoroaster, it is worth quoting verbatim:

"Shortly after the middle of the sixth century, Agathias, who used Sasanian official records on history, wrote on Zoroaster's date ... [then follows the above-mentioned passage, and then:] This historical testimony makes it clear that in the sixth-century Persia, when Zoroastrianism flourished as the state religion, Zoroaster's followers did not know an exact date for him. Several 'traditions' were no doubt current, giving different speculative datings23. As far as the Zoroastrians were concerned, all that could be said was that 'Zoroaster lived under King Vištāsp'"24.

Indeed, had Professor Shahbazi stuck to the latter phrase alone, he would have avoided the serious objection that it is quite false that Agathias's evidence shows that in sixth-century Iran the Zoroastrians did not have an exact date for Zoroaster and that "several traditions" existed in this regard. All this is the result of a totally arbitrary inference that he draws. As an author who certainly had a solid classical background25, in full awareness of the existence of another Hystaspes, Agathias must logically have considered the problem of the identity of the Hystaspes during whose reign the Persians situated Zoroaster; but this was his problem, not theirs26. Only the Greeks were familiar with a Hystaspes who was the father of Darius I: the Persians were unaware of both father and son, as is clear from the list of kings reigning during the 258 years between Zoroaster and Alexander27. There is therefore no reason why, in the Persian material on which Agathias certainly drew through the interpreter Sergius there should be any indication of a possible relationship between the King Wištāsp of Zoroastrian tradition, who reigned at the beginning of the 10th millennium, and the Achaemenid Vištāspa, the father of Darius.

Unlike Lactantius28 and Ammianus Marcellinus29, who respectively denied and confirmed the identity of the two Hystaspes, Agathias remained doubtful. His was a reasonable doubt as, knowing that the Achaemenids had reigned over Asia for 228 years30, until the death of Darius III -and thus starting from 558 B.C. (see below)-, he probably also calculated that there were no chronological obstacles to this identification. Indeed, it cannot therefore be ruled out that he was aware of the traditional date of Zoroaster (558 B.C.) as "258 years before Alexander", which could equally well fit the King Vištāspa of Zoroastrian tradition and the Vištāspa "son of Arsames" and father of Darius, some thirty years before the establishment of Persian dominion. Therefore, the interpretation that Professor Shahbazi reiterates later31 and that, as we have said, represents a fundamental point in his whole argument, must therefore be certainly rejected.

However, by far the simplest and most probable explanation of this passage is that Agathias must have had before him the evidence of the Sassanian chronology in which, it is true, the tenth millennium, the millennium of Religion, had been identified as corresponding to the beginning of the Seleucid era, interpreted as the era of Zoroaster, beginning under the reign of Wištāsp32. Moreover, there is no doubt that the Byzantine historian was aware of the "faulty Persian chronology". This emerges from a comparison of this passage (II 24.6) with two other passages from the Histories: II 26.1 e IV 24.1. The first passage states that the Parthian period, from Arsaces I to the last Artabanus, lasted 270 years -a length of time quite comparable with that of the 266 years known from other sources33 and that in any case reduces the entire period by nearly half- and the second places the advent of Ardašīr "in the fourth year of the reign of Severus Alexander, five hundred and thirty-eight years after Alexander the Great", thanks to a western source, probably Syrian, as already proposed by Averil Cameron, referring to Jacob of Edessa, Elias of Nisibis and Bar Hebraeus34.

Knowing that Wištāsp was the king under whose reign Zoroaster had lived, that the Parthian period had lasted 270 years, that the advent of Ardašīr was to be dated as 538 of the Seleucid era, Agathias was therefore aware of a good number of the ingredients making up the traditional date of Zoroaster. Averil Cameron, pointing out how it is not surprising that "Agathias' source evidently said nothing about the traditional date for Zoroaster in Persian reckoning (258 years before Alexander)", proposed two alternative solutions for this silence -one certainly misleading (the data was allegedly "not a Sassanian date but a later calculation"), based on an incorrect reference to J. Duchesne-Guillemin35-, the other quite plausible: "even if it had been available to Sergius, this is just the sort of detail which he might have omitted"36.

Agathias, on closer scrutiny, does not seem to offer any further indications regarding the problem of the traditional date of Zoroaster. Nevertheless Prof. Shahbazi takes one further passage from the Byzantine historian, to which he attributes great importance, including it among the evidence that, in his opinion, proves that the beginning of the Seleucid era, "in chronological discussions", was known as "Alexander" in Sassanian Persia37. It would thus not be true that "before Alexander", explained as "before the Seleucid era", is "based only on a mistake of Bīrūnī", as Henning believed38.

Indeed Agathias, "who used Sasanian royal documents"39, allegedly referred to the Seleucid era through "Alexander" as a chronological point in II 25.8, just as in IV 24.1: "Agathias adds (IV, 24) a summary of the Sasanian history from Persian official records and starts with (IV, 24, 1) the rise of Ardašīr which was 'five hundred and thirty-eight years after Alexander the Great'"40. Here, if Professor Shahbazi had quoted the chronological reference in full, he would have added some further useful information to what has been mentioned above: namely, the synchronization, between 538th year of the Seleucid era and the 4th year of the reign of Severus Alexander, "the Roman Emperor Alexander the son of Mamaea", mentioned by Agathias also in another chronological indication referring to the transition from the Arsacids to the Sassanians in passage II 26.1. In this case, this element clearly suggests that the source was not Persian, as pointed out above41.

Let us now examine the passage in question, II 25.7-8, in the translation by J.D. Frendo:

The Persian kings ruled for two hundred and twenty-eight years but their empire disintegrated completely when it was overrun by the forces of a foreign king. Alexander, the son of Philip slew their king Darius the son of Arsames, annexed the whole of Persia ...42

Anyone reading this passage, even without studying it in any detail will immediately realize that, also in this case, Agathias is drawing on Greek, not Persian, sources: the general tone until the end of the paragraph and the references to Alexander son of Philip and to Darius "son of Arsames" are decisive in this regard. As far as the chronology is concerned, Averil Cameron has clearly demonstrated how "Agathias now gives us Africanus' figure for the Achaemenids" and how "the reckoning is from 559 to 330, i.e., from Cyrus' first year ... to Alexander's first Babylonian year, i.e., the death of Darius III"43. The fact is that, even though Agathias uses Persian material, as in the case of the 270 years for the Parthian period in II 26.1, for the chronographical section he is largely dependent also on Greek and Syrian sources44.

As mentioned earlier, Professor Shahbazi attaches great importance to Agathias' evidence as far as the problem of the traditional date of Zoroaster is concerned, to the point of including "Agathias on Zoroaster's date" among the keywords of his article (the others being simply: "Zoroaster's date"; "Alexander's Era"; "Seleucid Era"; "Ardašīr's chronological fraud"; "traditional date of Zoroaster"45).

For him the evidence produced by the Byzantine historian is particularly important, as it allegedly proves that the Persians of the sixth century: a) did not conserve the date of their prophet; b) they had several traditions in this regard "that cannot be reconciled, and dismissed by the qualification of a date in the neighborhood of 600 B.C."46; c) they knew the Seleucid era as the "Era of Alexander"; they believed that 258 years passed between the beginning of the Achaemenian period ("Zoroaster") and the beginning of the Seleucid era ("Alexander"). He believes that his conclusion, which he based on his interpretation of passage II 25.7-8, confirms his previous hypothesis47, according to which the "traditional date" was "a scholastic fabrication which has no claim to historical authenticity nor an origin in a Greek fantasy"48, a scholastic fabrication "based on the reinterpretation, in the Hellenistic period, of the date of Cyrus the Great's Babylonian accession in 539 B.C.". Indeed, from this event to 312/11 B.C., i.e., to the beginning of the Seleucid era, 228 years are believed to have passed and the Magi allegedly identified the beginning of the Persian period with the Coming of Religion, "when Zoroaster was 30", consequently dating the prophet "(228 + 30 =) 258 years before the Seleucid Era, which came to be known as 'Alexander's Era' or simply 'till Alexander'."49.

However, all of these conclusions are baseless. For Persian history, Agathias could draw both on Greek and Syrian sources, as is shown by passages II 25.7-8 e IV 24.1 and on Persian material, in the first instance transmitted to him by the interpreter Sergius50, thanks to whom, in all likelihood, he learned about the drastic curtailment of the duration of the Parthian period, characteristic of the faulty Persian chronology (II 26.1) and of the fundamental fact that Zoroaster had lived at the time of Vištāspa (II 24.6).

The translation of Agathias' Histories that J.D. Frendo published in 197551 without doubt imposes itself as authoritative, based as it is on the critical edition of R. Keydell52. However, to take it alone, or any other translation, as a basis, without making any thorough analysis of the sources, Greek or Oriental, cannot suffice to solve the often difficult problems related to the encounter between Persia and Byzantium. In addressing the topics treated herein we have for some time been able to make use of the excellent support represented by Averil Cameron's detailed studies of the Byzantine historian53, both on his sources54, and above all on Agathias himself and Sassanian history55. They should not be neglected.


1 B.I. Marshak, The historico-cultural significance of the Sogdian calendar, "Iran", XXX, 1992, pp. 145-154: p. 153 note 36.

2 G. Gnoli, Politica religiosa e concezione della regalità sotto i Sassanidi, in La Persia nel Medioevo, Rome 1971, pp. 225-251: pp. 233-238; Id., Zoroaster's time and homeland, Naples 1980, pp. 166-174; Id., De Zoroastre à Mani. Quatre leçons au Collège de France, Paris 1985, pp. 36-39.

3 G. Gnoli, The idea of Iran. An essay on its origin, Rome 1989, p. 64 note 53 and p. 66; Id., "East and West", XLII, 1992, pp. 519-527 (review of H. Humbach, The Gāthās of Zarathushtra and the Other Old Avestan texts): p. 520; Id., Once more Zoroaster's time: a Manichaean dating, "East and West", XLV, 1995, pp. 313-319: pp. 313-314; Id., Sulla data di Zoroastro nel proemio di Diogene Laerzio, in Mousa. Scritti in onore di Giuseppe Morelli, Bologna 1997, pp. 179-195: pp. 193-195; Id., Further considerations on a Manichaean dating of Zoroaster, in Proceedings of the Third European Conference of Iranian studies held in Cambridge, 11th to 15th September 1995, Part 1: Old and Middle Iranian studies, ed. N. Sims-Williams, Wiesbaden 1998, pp. 13-20: p. 20.

4 G. Gnoli, Zoroaster in history, New York 2000.

5 I. Gershevitch, Approaches to Zoroaster's Gathas, "Iran", XXXIII, 1995, pp. 1-29. Without footnotes, appendices and bibliography, this paper was published also in the Proceedings of the First Gāthā Colloquium, ed. F. Vajifdar, Whiteleaf, Surrey 1998, pp. 11-26, with the title: Dissent and consensus on the Gathas.

6 A.Sh. Shahbazi, Recent speculations on the "traditional date of Zoroaster", "Studia Iranica", XXXI, 2002, pp. 7-45.

7 H. Lewy, The genesis of the faulty Persian chronology, "Journal of the American Oriental Society", LXIV, 1944, pp. 197-214.

8 S.H. Taqizadeh, The "Era of Zoroaster", "Journal of the Royal Asiatic Society", 1947, pp. 33-40.

9 W.B. Henning, Zoroaster. Politician or witch-doctor?, Oxford 1951.

10 Close to this kind of position is a scholar like Sh. Shaked, Scripture and exegesis in Zoroastrianism, in Homer, the Bible, and beyond. Literary and religious canons in the ancient world, ed. M. Finkelberg and G.G. Stroumsa, Leiden-Boston 2003, pp. 63-74: pp. 63-64, while, although with diametrically opposite views, different opinions are held by M.A. Dandamaev, "VDI", 2002/2, pp. 216-218 (review of G. Gnoli, Zoroaster in history), e di J.R. Russell, The place and time of Zarathushtra, in A Zoroastrian tapestry. Art, religion and culture, ed. Ph.J. Godrej and F.P. Mistree, Usmanpura Ahmedabad - Middletown, N.J. 2002, pp. 29-39: p. 29.

11 Shahbazi, art. cit. pp. 10-36.

12 Ibid., pp. 36-39.

13 Ibid., pp. 39-43. Cf. P. Kingsley, The Greek origin of the sixth-century dating of Zoroaster, "Bulletin of the School of Oriental and African Studies", LIII, 1990, pp. 245-265.

14 Ibid., pp. 8, 21, 31, 34.

15 Ibid., p. 10.

16 Agathias. The Histories, transl. with an introd. and short explanatory notes by J.D. Frendo, Berlin 1975.

17 Shahbazi, art. cit., p. 8.

18 Ibid., p. 31. The passage is incorrectly cited as II 26.7-8. Also in a previous articles (On the Xwadāy-nāmag, in Iranica Varia. Papers in honor of Professor Ehsan Yarshater, Leiden 1990, pp. 208-229: p. 229 note 147) Professor Shahbazi confuses this passage with another one, namely IV 24.1, instead containing the synchronims between the fourth year of Severus Alexander's reign (222-235) and the year 538 of the Seleucid era.

19 Ibid., pp. 31-32.

20 Ibid., p. 8 note 1, with cross-reference to the translation by J.D. Frendo (see note 16), pp. 133-134. See already by Shahbazi himself, The "traditional date of Zoroaster" explained, "Bulletin of the School of Oriental and African Studies", XL, 1977, pp. 25-35: p. 29 note 33. For the first passage of the book IV, 30.2-4 referring to the Persian sources, see in particular Av. Cameron, Agathias on the Sassanians, "Dumbarton Oaks Papers", XXIII-XXIV, 1969-1970, pp. 67-184: pp. 134 (text), 135 (translation), 161-163 (comment).

21 Shahbazi, Recent speculations, p. 8, with reference to p. 58 of the translation by J.D. Frendo (see note 16).

22 Av. Cameron, art. cit., for II 24.6 and pp. 80 and 82 (text), 81 and 83 (translation), 94 (comment); for II 26.7-8 on pp. 84 (text), 85 (translation), 103 (comment); for IV 24.1 on pp. 120 (text), 121 (translation), 136-137 (comment). Passage II 24.6 is published also in J. Bidez and F. Cumont, Les Mages hellénisés. Zoroastre, Ostanès et Hystaspe d'après la tradition grecque, Paris 1938, vol. II, on p. 83 as fr. D 11 and, partially, on p. 34 as fr. B 22.

23 Professor Shahbazi refers here to paragraphs 15 and 19 of his article (pp. 20-21 and 23-25), in which he comprehensively expounds his ideas on the topic with a number of references.

24 Shahbazi, Recent speculations, p. 8.

25 Frendo, op. cit., pp. IX sq.

26 See the excellent comment by Av. Cameron (art. cit., p. 94) on this point.

27 Cf. A.V.W. Jackson, Zoroaster. The Prophet of Ancient Iran, New York 1901, p. 161.

28 Lactantius, Divinae Institutiones, VII 15.19. Cf. Bidez-Cumont, op. cit., II, p. 366.

29 Ammianus Marcellinus, XXIII 6. 32. Cf. Bidez-Cumont, op. cit., II, p. 359.

30 Agathias, Histories, II 25.7-8, for which see below.

31 Shahbazi, Recent speculations, p. 21.

32 This was the discovery of Hildegard Lewy, which S.H. Taqizadeh, who had independently reached the same solution, acknowledged as having priority: Taqizadeh, art. cit., p. 38 note 1. Cf. Henning, op. cit., p. 37.

33 For the evidence see Shahbazi, Arsacids, vi: Arsacid chronology in traditional history, in Encyclopaedia Iranica, vol. II, 1987, pp. 542-543. On the figure: Th. Nöldeke, Geschichte der Perser und Araber zur Zeit der Sasaniden aus der arabischen Chronik des Tabari übersetzt und mit ausführlichen Erläuterungen und Ergänzungen versehn, Leyden 1879, p. XVI note 3 and p. 1 with note 1; Lewy, art. cit.; Av. Cameron, art. cit., pp. 105-106; Gnoli, Zoroaster in history, pp. 143 sq.

34 In addition to Th. Nöldeke, loc. cit. in the preceding note: cf. Av. Cameron, art. cit., p. 136, with corresponding bibliographic references. She aptly points out that also the synchronization with Severus Alexander imposes the conclusion that Agathias "surely got his 538 years from a Syrian source, not from his Persian material where he would not have found the synchronization with Alexander Severus" (ibid.). Av. Cameron had already dealt with the topic in his important study on the Persian calendar, Le calendrier perse, "Orientalia", 1941, pp. 1-64: pp. 45-51, for the problem of the chronology of the early Sassanians.

35 J. Duchesne-Guillemin, La religion de l'Iran ancien, Paris 1962, p. 136, where however it is not stated that the traditional date of Zoroaster was not Sassanian but a later calculation.

36 Av. Cameron, Agathias on the Sassanians, p. 94.

37 To this evidence, and its interpretation by Professor Shahbazi, I devoted my contribution to the Fifth European Conference of Iranian studies to be held in Ravenna, October 2003.

38 Henning, op. cit., p. 41.

39 Shahbazi, Recent speculations, p. 31.

40 Ibid., pp. 31-32.

41 See above, note 34.

42 Frendo, op. cit., p. 60; cf. Shahbazi, Recent speculations, p. 31.

43 Av. Cameron, Agathias on the Sassanians, p. 103, with references to the sources (Eusebius, Syncellus, Pseudo-Leo Grammaticus).

44 Ibid., pp. 100-104.

45 Shahbazi, Recent speculations, p. 7.

46 Cf. Henning, op. cit., p. 35: "The Zoroastrian tradition has preserved a date which would put Zoroaster in the neighborhood of 600 B.C.", cited by Shahbazi, Recent speculations, p. 18.

47 Shahbazi, The "traditional date of Zoroaster" explained, cit. (note 20); cf. Id., art. cit. (see above, note 18), p. 218.

48 Shahbazi, Recent speculations, p. 8. The phrase "origin in a Greek fantasy" is a reference to P. Kingsley's thesis (see above, note 13), which he rejects when he discusses it in point VI of his article.

49 Shahbazi, ibid., pp. 9-10.

50 Regarding Agathias, in addition to the entries of the Pauly Wissowa (M. Hartmann, I, 1893, cols. 743-745) and of the Kleine Pauly (R. Keydell, I, 1969, cols. 116-117), it is useful to consult also the entry in the Encyclopaedia Iranica (M.-L. Chaumont, I, 1985, pp. 608-609), written from an Iranistic perspective. For the passages from the Histories II 23-25 see also A. de Jong, Traditions of the Magi. Zoroastrianism in Greek and Latin literature, Leiden 1997, pp. 229-250.

51 See above, note 16.

52 Agathiae Myrinaei Historiarum Libri Quinque, ed. R. Keydell, Berlin 1967.

53 Av. Cameron, Agathias, Oxford 1970.

54 In particular: Av. Cameron, Zonaras, Syncellus and Agathias - A note, "Classical Quarterly", N.S., XIV, 1964, pp. 82-84; Ead., Christianity and tradition in the historiography of the late Roman Empire, ibid., pp. 316-328; Ead., Herodotus and Thucydides in Agathias, "Byzantinische Zeitschrift", LVII, 1964, pp. 33-52.

55 Av. Cameron, Agathias on the Sassanians. Prior to this work, the only study devoted solely to the Persian sources of Agathias was that of J. Suolahti, On the Persian sources used by the Byzantine historian Agathias, "Studia Orientalia", XIII, 1947, pp. 1-13, which, according to Av. Cameron "does little more than scratch the surface" (art. cit., p. 112).

© The Author/s

Actualizado el 24/07/2004

Valid HTML 4.01!
Any Browser!
Creative Commons License
Como lo hacemos?