[JNES 64 no. 2 (2005)]ç 2005 by The University of Chicago.All rights reserved.0022–2968–2005/6402–0002$10.00.

101

WHEN DARIUS DEFEATED ALEXANDER: COMPOSITION AND REDACTION IN THE DYNASTIC PROPHECY

MATTHEW NEUJAHR, Yale University

In 1975 A. K. Grayson published a fascinating little text which he named “TheDynastic Prophecy” (BM 40623).1 The text reviews, in the form of a series of predictions,Mesopotamian history from the Neo-Assyrian period through the onset of the Greek pe-riod. One of the most interesting features of this text is that—according to the majority ofinterpreters—it reports that after having suffered an initial setback, Darius III returned torepulse Alexander the Great and instituted a glorious period in Babylon.2 There has beenno convincing explanation of why this text should include such an egregiously erroneousaccount of history. The lack of any plausible explanation is due in large part to the failureof commentators to adduce as comparative controls formally similar texts that includeprecisely the same sort of references, i.e., references to events which never transpired atthe end of a series of ex eventu predictions. Such comparison undertaken in the presentstudy will suggest that what we have in the Dynastic Prophecy is the redacted version ofan older composition that had concluded with an authentic attempt at predicting victoryand a glorious reign for Darius III.

The Dynastic Prophecy is, of course, the latest member of that corpus of texts variouslycalled Akkadian prophecies,3 Akkadian apocalypses,4 and Akkadian literary predictivetexts.5 All the texts of this category share as their most salient feature historical reviews inthe guise of future predictions—that is, vaticinia ex eventu.6 In the case of the DynasticProphecy, the “predictions” include: the reigns of Neo-Assyrian kings (column I, almostentirely reconstructed based on the content of the subsequent columns), the reigns of Neo-Babylonian kings (II 1–10), the empire’s downfall during the reign of Nabonidus (II 11–16), the institution of Achaemenid rule under Cyrus (II 17–25), Achaemenid succession

1 A. K. Grayson, Babylonian Historical-LiteraryTexts, Toronto Semitic Texts and Studies 3 (Toronto,1975); hand-copy, pp. 28–29, transcription and trans-lation pp. 30–36.

2 There is one notable exception to this interpre-tation of the text, that of M. J. Geller, “BabylonianAstronomical Diaries and Corrections of Diodorus,”BSOAS 53 (1990): 1–7. Geller’s interpretation will bepursued below.

3 A. K. Grayson and W. G. Lambert, “AkkadianProphecies,” JCS 18 (1964): 7–30; Grayson, Historical-Literary Texts.

4 William W. Hallo, “Akkadian Apocalypses,” IEJ16 (1966) : 231–42.

5 Maria deJ. Ellis, “Observations on MesopotamianOracles and Prophetic Texts: Literary and Historio-graphic Considerations,” JCS 41 (1989): 127–86.

6 The question of genre has been engaged in sev-eral studies, but most insightfully and critically by El-lis, “Oracles and Prophetic Texts.” Cf. A. K. Grayson,“Assyria and Babylonia,” Or., n.s., 49 (1980): 140–94;and the unsatisfactory treatment of Tremper LongmanIII, Fictional Akkadian Autobiography (Winona Lake,Indiana, 1991). Reserving judgment on whether thesetexts constitute a literary genre—and just what thatgenre might be—I confine myself to the nontechnicaldesignation “category,” solely determined by the pres-ence in an Akkadian text of extended use of vaticiniaex eventu.

Journal of Near Eastern Studies102

intrigues (III 1–8), the invasion of Alexander the Great (III 9–23), and, finally, what wouldappear to be predictions concerning reigns of Alexander’s successors in IV 1–6, althoughthe column is fragmentary in the extreme.7 It is the invasion of Alexander in III 9–23 thatconcerns us here.

Grayson has convincingly interpreted the opening eight lines of column III as referringto the assassination of Arses by the eunuch general Bagoas, who then placed Darius III onthe throne in 336 b.c.e. This leads to the appearance in line 9 of Alexander’s forces,called here the “army of the Hanaeans” (lúummanimes kurha-ni-i ).8 Making certain theidentification of Alexander the Great as the “Hanaean” in question is an astronomical textdated to 329 b.c.e. that specifically names Alexander as “king of the Hanaeans.”9 In thepresent text, then, the Hanaean Alexander proceeds to defeat Darius III in lines 11–13 ofcolumn III. With the scene thus set, it seems prudent at this point to reproduce the passagein question.10

Column III

9 the army of the Hanaeans [. . .]10 will attack . . . [. . .]11 his army [. . .]12 they will plund[er] his [pl]under, [his booty]13 they will carry away.11 Afterwards, [his] arm[y]14 he will refit and he will r[aise] his weapons.15 Enlil, Shamash and [Marduk(?)]16 will walk beside his army [and]17 the defeat of the army of the Hanaeans he will [effect(?)].18 He will carry away his extensive booty and19 he [will bring (it)(?)] into his palace.20 The people who had [experienced(?)] misfortune

7 According to Grayson (Historical-Literary Texts,p. 27), the curvature of the tablet is such that it is im-possible to determine just how much may have brokenoff, but he supposes it likely that the entire text origi-nally consisted of four columns, each of which is pre-served in part. This conclusion has subsequently beenchallenged on grounds of content. W. G. Lambert (TheBackground of Jewish Apocalyptic [London, 1978],p. 13) has contended that the text was originally acomposition of some six columns (see also SusanSherwin-White, “Seleucid Babylonia: A Case Studyfor the Installation and Development of Greek Rule,”in A. Kuhrt and S. Sherwin-White, eds., Hellenism inthe East: The Interaction of Greek and Non-Greek Civ-ilizations from Syria to Central Asia after Alexander[Berkeley, 1987], pp. 1–31, p. 11). According to this,the entirety of hypothetical columns *3 and *4 aremissing, leaving columns *1 and *2, *5, and *6, whichGrayson numbers I–IV. Whether there was once anaccount of the Achaemenid kings from Darius I toArtaxerxes III in Lambert’s putative columns *3 and*4 does not affect the present study. Grayson’s enu-meration will be followed here, if only to facilitate thereader’s referral to his edition.

8 The designation requires a little explanation. Inits present context, the term is a deliberate archaism,paralleling reference to the Persians by the term “Ela-mite.” In the second millennium, the term hanû con-noted a class of soldier, for example, at Mari (CAD,vol. H, p. 82 hanû). In the Hellenistic period, thereis an apparent identification of the land of the Hanûas Thrace (A. K. Grayson, Assyrian and BabylonianChronicles, Texts from Cuneiform Sources, vol. 5 [Lo-cust Valley, New York, 1975], p. 256).

9 The left edge of the tablet in question dates thecomposition to: mu-8-kám IA-le[k-sa-a]n-dar-ri-islugal sá ta mat Ha-ni-i (A. J. Sachs and H. Hunger,Astronomical Diaries and Related Texts from Babylo-nia, vol. 1 [Vienna, 1988], text 328).

10 The following translation is heavily indebted tothe edition of Grayson; where it differs from Grayson’stranslation it does so mainly by refraining from resto-ration of certain lacunae.

11 The restoration of terms in lines 12 and 13 ismade quite certain by the use of parallel cognate accu-sative constructions: hubussu ihabbatu // sillassu isal-lalu. Cf. line 18, sil-lat-su ka-bit-tum i-sal-l[a-al-ma].

When Darius Defeated Alexander 103

21 [will experience(?)]12 well-being.22 The mood of the land [will be happy(?)].23 Tax exemption [. . .]

It seems clear that lines 11–13 refer to a victory of Alexander over the Persians; the mostfamous candidates for this allusion are the battles at the Granicus (334 b.c.e.), Issus (333),and of course Gaugamela (331). The most remarkable element of the passage follows:according to lines 13–17, the defeated Persian king retools his army and then defeats theMacedonians! The bald inaccuracy of this account is all the more striking in the face ofthe historically accurate, and occasionally quite specific, accounts contained earlier.13 Thetext is further complicated by the fact that, following this account, the first six preservedlines of the final, fourth column (after which the composition proper ends) are divided intothree sections by two horizontal lines drawn across the column width. Judging by the useof such dividing lines in this and the other exemplars of Akkadian ex eventu prophecies,this would seem to indicate three more significant reigns. Caution should, however, beurged in the interpretation of these six lines, as only fifteen cuneiform signs are at all leg-ible; further, the use of dividing lines to separate reigns is consistent neither in this textnor in the other texts of this type. Nonetheless, Grayson speculates that the three sectionsin column IV 1–6 refer to the reigns of Philip Arrhidaeus, Alexander IV, and, Seleucus I,respectively.14 Sherwin-White prefers to see in the first section the reign of Darius III (IV1–2), followed by Philip Arrhidaeus and Alexander IV in the second section (IV 3), andfinally Antigonus.15

Yet any understanding of the text that posits the inclusion of reigns by Alexander’ssuccessors simply serves to problematize further the reference to Alexander’s defeat. AsGrayson states, “It is extremely unlikely that the ‘prophet’ would deliberately falsify theoutcome and aftermath of such a famous and well-known battle” as Gaugamela.16 Heproffers no explanation, however, as to why the text does indeed contain such a reference.The lack of an explanation has in part occasioned a vastly different interpretation of thetext by Markham J. Geller.17 Geller maintains that the lines in question deal not withDarius and Alexander, but rather with the war between Antigonus and Seleucus ca. 310–307. There are two main objections to this reading, for which Geller proposes not whollysatisfying answers. First, if Geller is right, then the text would proceed from a descriptionof Darius III directly to the wars of the Diadochoi. The implementation of Greek ruleunder Alexander would receive no mention whatsoever. Despite Geller’s protestations tothe contrary, this must be regarded as passing strange. Second, again countering Geller’sprotestations, the reference to Antigonus’s army as the army of the Hanaean makes littlesense. True, it is perfectly reasonable that a late fourth- or early third-century Babylonian

12 Grayson plausibly suggests reading N-stems ofamaru in lines 20 and 21. Regardless of precisely whatverb is missing, the point of the passage is clear enough.

13 Note especially the detailed references to Nabo-nidus in II 11–16.

14 Grayson, Historical-Literary Texts, p. 27;Sherwin-White, “Seleucid Babylonia,” p. 14.

15 Sherwin-White, “Seleucid Babylonia,” p. 14. Onthe rule of Antigonus in native Babylonian perspec-

tive, see the so-called Chronicle concerning the Dia-dochi in Grayson, Assyrian and Babylonian Chronicles,chronicle 10.

16 Grayson, Historical-Literary Texts, p. 26.17 Geller, “Astronomical Diaries,” pp. 5–7. Geller’s

view has been subsequently endorsed by MatthewStolper; see his article “Mesopotamia, 482–330 B.C.,” inD. M. Lewis et al., eds., Cambridge Ancient History, 2ded., vol. 4 (Cambridge, 1994), pp. 234–60; p. 241, n. 24.

Journal of Near Eastern Studies104

text would refer to Alexander’s successors, including Antigonus, as Hanû. The designa-tion is employed, however, in the context of the Dynastic Prophecy as a term of distinctionby which a new character is introduced. Seleucus would be no less Hanû than Antigo-nus—and the use of this epithet only makes sense if the other protagonist cannot be iden-tified as Hanû. This leads us back to Darius III. Finally, it must be noted that the mainreason for positing the exclusion of Alexander from this text is that, as Geller quite rightlynotes, no one has offered an adequate explanation of how or why this text would relate thedefeat of Alexander at the hands of Darius.

The obvious solution is that the author of the Dynastic Prophecy is here making anattempt at an authentic prediction—one which did not come true.18 This possibility has infact been considered and summarily dismissed by several commentators based on theirvarious theories of the text’s purpose.19 Grayson, for one, believes it is possible to detecta good-bad-good-bad pattern for the founder of each dynasty.20 From this he extrapolatesthat the last reign of column IV, which he identifies as that of Seleucus I, is categorized asbad. Therefore, he posits, the text functions as anti-Seleucid propaganda; the problematicaccount of Darius’s victory over Alexander is bizarre but inconsequential for the purposefor which the Dynastic Prophecy was composed. Helmer Ringgren has offered a very simi-lar conclusion, suggesting that the text was composed as a critique of the Seleucids.21 Aslightly different conclusion is reached by Sherwin-White, who seems to think that thetext culminates with a negative assessment of the “reign” of Antigonus and in fact amountsto pro-Seleucus I propaganda.22 She further claims that the inclusion of historical inaccu-racies should not occasion much surprise and invokes a passage from 2 Maccabees.23

18 Thus Lambert: “If the text ended at this pointone would say that it was composed after Alexander’sfirst defeat of the Persians at Granicus to encouragethe Babylonians to help Darius against the invader.However, the text continues with more reigns, thoughtoo broken to specify” (Background, p. 13).

19 For example, Lambert, Background, p. 13;Sherwin-White, “Seleucid Babylonia,” p. 11.

20 Grayson, Historical-Literary Texts, p. 17. Thisidea is highly speculative (as Grayson himself admits),and highly problematic. For one thing, the scribal useof dividing lines suggests against dividing the “dynas-ties” in accordance with modern historiographic con-vention. Most notable is the fact that the reign ofNabonidus (II 11–16) is isolated from what precedesand follows by dividing lines and is identified as an in-dependent dynasty, “the dynasty of Harran” (palê har-ranki, line 12). Additionally, if the text was originallysix columns in length, it would further upset Grayson’sdynasty-per-column scheme. For a general critique ofthe categorization of reigns as good and bad in theAkkadian ex eventu prophecies, see Robert D. Biggs,“Babylonian Prophecies, Astrology, and a New Sourcefor ‘Prophecy Text B’,” in F. Rochberg-Halton, ed.,Language, Literature, and History: Philological andHistorical Studies Presented to Erica Reiner, AmericanOriental Studies, vol. 67 (New Haven, Conn., 1987),pp. 1–14; see especially pp. 2–3.

21 Helmer Ringgren, “Akkadian Apocalypses,” inD. Hellholm, ed., Apocalypticism in the MediterraneanWorld and the Near East (Tübingen, 1983), pp. 379–86;p. 383.

22 Sherwin-White, “Seleucid Babylonia,” pp. 11 and14. Cf. Pierre Briant, From Cyrus to Alexander: AHistory of the Persian Empire, trans. Peter T. Daniels(Winona Lake, Indiana, 2002), p. 864.

23 An entirely unconvincing appeal is made to2 Maccabees 13:9–24 as evidence of “the transfor-mation of defeats into victories,” and, generally, sheclaims that “the function of texts such as these is not‘objective’ history” (Sherwin-White, “Seleucid Baby-lonia,” p. 11). The same objection appears almost ver-batim, here with a reference to 2 Maccabees 11:1–15,in Susan Sherwin-White and Amélie Kuhrt, From Sam-arkhand to Sardis: A New Approach to the SeleucidEmpire (Berkeley, 1993), pp. 8–9. It must be stressed,however, that in no way could one ever suggest that2 Maccabees is a “text such as” the Akkadian vati-cinia ex eventu texts. No, the Dynastic Prophecy doesnot function as historiography; it does, however, func-tion as a piece of political propaganda in the form of amantic text, one in which events of the past are re-cited. If the audience cannot identify as accurate theex eventu predictions made, the text fails in its func-tion as propaganda. Sherwin-White has failed to shedany light on the issue of why the Dynastic Prophecy

When Darius Defeated Alexander 105

But no one has yet invoked other texts containing extended vaticinia ex eventu to see ifthey might shed light on the composition of the present work—this despite the frequentremarks in the literature that this and other Akkadian literary prophecies are in importantways very similar to certain Jewish historical apocalypses. Actual systematic comparisonbetween the Akkadian and Jewish corpora of historical reviews as predictions remains amajor scholarly desideratum in the study of these works. In the case of our present prob-lem, two texts that consist of vaticinia ex eventu and that contain historically inaccuratepredictions may profitably be brought to bear on the problem of Darius’s victory. Theseare Daniel 10–12 and Sibylline Oracles 4:40–192.

The Book of Daniel has, of course, been widely cited by those writing on Akkadian lit-erary prophecies, invoked as a parallel case of vaticinium ex eventu.24 To my knowledge,however, no one has cited the final prediction of Dan. 12:1–12 (especially vv. 11 and 12)in connection with the Dynastic Prophecy III 13–23. Chapters 10–12 of Daniel constitutea lengthy final apocalyptic vision, with an extended historical review (in the guise of pre-diction) covering the reigns of kings from the Persian era through the career of AntiochusIV Epiphanes in 11:2b–12:4. Antiochus IV is, of course, the wicked foreign king par ex-cellence in much Jewish literature of the Hellenistic era; 11:30–35 refers to his oppressionof Judea, in particular his interruption of the cult (v. 31). The same offense of AntiochusIV is likewise recorded as a prediction in the vision of Dan. 8:9–11. In both cases, theapocalyptist is concerned with predicting—legitimately—the downfall of the foreign op-pressor.25 The apocalyptist who authored Dan. 12:11 is concerned specifically with offer-ing the date of the onset of the eschaton and resurrection of the righteous dead—that is,the end of the days (µymy ≈q)—which will once and for all end foreign domination. Verse11 tells the reader in no uncertain terms how long it will be until this happens: “From thetime when the continual offering is taken away and the desolating abomination is set up isone thousand two hundred and ninety days.” I think it safe to conclude that a widespreadresurrection of the dead did not occur 1,290 days after the desecration of the Jerusalemtemple in 168 b.c.e. This fact was certainly not lost on the tradents responsible for theBook of Daniel. Hence we read in 12:12, “Blessed is the one who waits and attains to the

might report something that would certainly be identi-fied by the audience as inaccurate, thus underminingthe text’s political propagandistic function.

24 For example, Grayson, Historical-Literary Texts,pp. 20–21; Lambert, Background, pp. 13–17; J.-G.Heintz, “Note sur les origines de l’apocalyptique ju-daïque à la lumière des ‘Prophéties akkadiennes’,” inM. Philonenko and M. Simon, eds., L’Apocalyptique(Paris, 1977), p. 83; Tremper Longman III, FictionalAkkadian Autobiography, pp. 188–89; Helge S. Kvan-vig, Roots of Apocalyptic: The Mesopotamian Back-ground of the Enoch Figure and of the Son of Man(Neukirchen-Vluyn, 1988), pp. 472–74, 488–90; G. F.Hasel, “The Four World Empires of Daniel 2 against ItsNear Eastern Environments,” JSOT 12 (1979): 17–30;and E. C. Lucas, “Daniel: Resolving the Enigma,” VT50 (2000): 66–80. Note also the comparison to a pas-sage from Sibylline Oracles 3 in Grayson, Historical-Literary Texts, p. 17–19.

25 In 8:14 an angelic intermediary is overheard pro-claiming that the disruption of the cult will last “fortwo thousand three hundred evenings and mornings.”The reference is actually to the number of tamîd offer-ings (twice daily) that will be missed. This yields a pre-diction that the cult will be restored after 1,150 days.See John J. Collins, Daniel: A Commentary on the Bookof Daniel, Hermeneia, vol. 27 (Minneapolis, 1993),p. 336. Given that this differs from the length of timebefore the restoration of the cult reported in 1 Macca-bees 4:52–54, it is quite likely that this constitutes alegitimate attempt at predicting restoration at the endof a series of ex eventu predictions. In contrast to thepredictions in Daniel 12, the prediction here seems toinvolve the amount of time that will elapse before theproper functioning of the cult resumes, not to the periodof the end time and the resurrection of the righteous.All of these likely are interpretations of the “three anda half years” of Dan. 7:25 (repeated in 12:7, and in 9:27as a “half week” of years).

Journal of Near Eastern Studies106

thousand three hundred and thirty-five days.” Clearly what we have here is a real predic-tion of 1,290 days that, upon its failure, was emended by adding 45 days, yielding a pre-diction of 1,335 days.26 The original attempt at prediction has been left in the text, despitenot coming to fruition; a further prediction has been tacked on as a corrective.

This phenomenon appears again in Book 4 of the Sibylline Oracles. It is widely recog-nized that lines 49–101 of this work constitute an older anti-Macedonian oracle, not nec-essarily Jewish in origin; it was subsequently reworked by a Jewish redactor of the latefirst century c.e., who has transformed the work into an eschatological judgment againstRome.27 The author of the anti-Macedonian passage has employed two distinct frameworksfor ordering this ex eventu prediction of history. On the one hand, the text schematizeshistory according to four world-dominating kingdoms;28 on the other hand, this has beenwedded to a schema of 10 generations of world history.29 According to this text, world his-tory culminates in the fourth kingdom, which arises in the tenth generation, after whichthe eschaton ensues (4:171–92). Our passage identifies the Persians as reigning in the ninth(penultimate) generation.30 This leads the reader to conclude that the fourth kingdom,Macedonia, must be that of the tenth generation. To a first-century Jew living after theRoman destruction of Jerusalem in 70 c.e., however, the Macedonians could hardly be con-sidered the last oppressors prior to Jerusalem’s glorious restoration. Therefore, the redac-tor clumsily ignores the ten-generation scheme of the original oracle and simply appendsa lengthy section on Rome, a fifth kingdom.31 According to the book as we have it, theMacedonians herald the onset of the eschaton, and yet, paradoxically, it is not until afterthe Roman destruction of Jerusalem that the end-time comes. The situation is preciselyanalogous to what we have seen in the Dynastic Prophecy: an oracle of liberation fromforeign oppression has been incorporated into a final work that knows that the predictedliberation never took place. Hence, in Sibylline Oracles 4 just as in the Dynastic Proph-ecy, a series of vaticinia ex eventu culminating in an actual but failed prediction has beenlater expanded so that the text now names successive foreign adversaries.

Some scholars have defended ignoring the problematic passage in Dynastic Prophecy IIIby claiming that this text, and other Akkadian works like it, are not intended to be histories.And this is quite true: there is nothing behind the authoring of the Dynastic Prophecy—nor for that matter any other text, Mesopotamian, Levantine, or Egyptian, that employsextended vaticinia ex eventu—which could be mistaken for the ideology of the moderncritical historiographer. In so dismissing the bizarre episode of Alexander’s defeat at the

26 This was already recognized by Hermann Gunkelin the nineteenth century (Schöpfung und Chaos in Ur-zeit und Endzeit: Eine religionsgeschichtliche Unter-suchung über Gen 1 und Ap Joh 12 [Göttingen, 1895],p. 269). Cf. James A. Montgomery, A Critical and Ex-egetical Commentary on the Book of Daniel, ICC (Edin-burgh, 1927), p. 477; Collins, Daniel, pp. 400–401.

27 See David Flusser, “The Four Empires in theFourth Sibyl and in the Book of Daniel,” Israel Orien-tal Studies 2 (1972): 148–75. Cf. John J. Collins, “TheSibylline Oracles,” in James Charlesworth, ed., The OldTestament Pseudepigrapha, 2 vols. (Garden City, NewYork, 1983–85), vol. 1, p. 381.

28 Cf. Daniel 2, 7; the Qumran text 4QFour King-doms (4Q547); and the first chapter of the Zoroastrian

composition Zand-i Vohuman Yasn.29 Cf. the Apocalypse of Weeks in 1 Enoch,

11QMelchizedek, and Sibylline Oracles 1. SeeFlusser, “Four Empires,” on both these schematiza-tions of history.

30 The Persians rule for one generation (line 66).Before them, the Assyrians rule for six generations (line50), then the Medes for two generations (line 55). Thistallies to nine generations. Line 47 specifies that theend-time occurs during the tenth and final generation.

31 Flusser (“Four Empires,” pp. 157–59) makes theinteresting observation that, even in later Jewish textsthat employ a historical scheme with Rome as the lastoppressive kingdom prior to Jerusalem’s restoration,the scheme is always one of four kingdoms, not five.

When Darius Defeated Alexander 107

hands of Darius III, however, these scholars fail to consider carefully the purpose for whichthese texts were in fact composed. The fact is that the Dynastic Prophecy is not history butpropaganda. Everyone who has studied the text recognizes this. Saying that this text is not“history,” however, is not enough; one must consider in detail why the composers of thetext should have structured their propaganda in such a way. Framing political propagandaas a prediction of the future seals the position of the propagandist with the approval of thegods, those who control the fate of humankind. The trick is to have the audience believethat the prediction is reliable. This is done by “predicting” historical events. If the audi-ence believes that the text is authentically old, then they are led to conclude that the “his-tory” in the text was legitimately predicted, assuring them of the reliability of all thepredictions in the text. Now, this is not to say that the content of the ex eventu predictionsis necessarily “historically reliable” by modern criteria; that is by no means to be as-sumed. What must be assumed, however, is that the author of such a propagandistic workincluded events which—whether they happened or not—are recognizable and presumedverified by the intended audience of the text. Is it reasonable to suppose that any Mesopo-tamian audience after 330 could recognize as accurate the claim that the Persian GreatKing had successfully repulsed the Macedonian onslaught?

The main objection against viewing Darius’s victory over Alexander as a legitimate at-tempt at prediction has been the fact that the composition continues beyond this point, andseems to narrate later reigns. As we have seen, however, both Daniel 12 and Sibylline Or-acles 4 contain a legitimate prediction that failed but that nonetheless was preserved. Thetradents of the original oracles simply revised their predictions, adding to the text ratherthan rewriting it. The entire purpose of employing vaticinia ex eventu is that the techniqueinstills confidence in the predictive powers of a given text. It would completely subvertthe text’s purpose to include anything that a reader might construe as historically inaccu-rate—to say nothing of grossly misrepresenting the most important political event of thelast 200 years. The only reasonable explanation for the inclusion of the account of Alex-ander’s defeat is that it was the original conclusion to a text that sought to gain support forDarius—likely composed on the heels of Alexander’s victories at the Granicus or Issus.32

With the surrender of Babylon by the satrap Mazaios and the death of Darius in 330 theoriginal prediction had obviously failed. Some Babylonian scribe or school, instead ofdisregarding the Dynastic Prophecy, obviously felt the text important enough to pre-serve—and what is more, to update. This in turn explains the illogical (probable) appear-ance of the Diadochoi (col. IV) in a text celebrating the defeat of Macedon.

32 Cf. Amélie Kuhrt, “Survey of Written SourcesAvailable for the History of Babylonia under the LaterAchaemenids,” in Heleen Sancisi-Weerdenburg, ed.,Achaemenid History I: Sources, Structures and Syn-theses (Leiden, 1987), pp. 147–57. Refraining fromsuggesting a date for the text, Kuhrt nonetheless won-ders, “Could the text have been composed before Gau-

gamela in order to rally Babylonian support for theAchaemenids and thus be shifting into the realm ofgenuine prophecy at this point?” (p. 155). A case fora post-Gaugamela date of the text is argued in Ga-briele Marasco, “La ‘Profezie Dinastica’,” Annali dellaScuola Normale Superiore di Pisa 15/2 (1985): 529–37; p. 533.