buckdue

Eunapius of Sardis was a pagan Greek sophist, philosopher and historian who lived from A.D. 349 until c. 414.¹ He is a prime example of the Hellenic reaction against the political, social, economic and religious changes of the fourth century, for he was virulently opposed to the Constantinian and Theodosian regimes. Although his Histories, which covered the years from A.D. 270 to 404,² have not survived, it is clear from their fragments and from the epitome by the Byzantine bureaucrat, Zosimus,³ that Eunapius wrote pagan propaganda in the guise of history. It is neither unfair nor misleading to say that today his work would be classified as historical fiction.

The emperor Aurelian has received little attention in this century, for Randall T. Saunders' 1991 Cincinnati thesis is the first comprehensive study of Aurelian since Leon Homo's book of 1904.⁴ Similarly, despite the awakening of interest in Eunapius in the past twenty-five years, the first part of his Histories remains somewhat neglected. Attention has been focused on his treatment of the fourth century, and it is now clear that he severely distorted his accounts of Julian and Theodosius for ideological reasons.⁵ The purpose of this article is to examine Eunapius’ treatment of the emperor Aurelian, for, although Aurelian ruled before emperors were missionaries and thus ready subjects for propagandistic history, it should not be assumed that Eunapius wrote an accurate and unbiased account of his reign, or even that he had the sources to do so.

L. Domitius Aurelianus was born on 9 September, probably in 214,⁶ but the literary sources agree on neither his birthplace nor his family background.⁷ Nothing is known of his career until, as the commander of the Dacian cavalry, he helped to assassinate Gallienus in 268.⁸ Aurelian became emperor himself in early October 270.⁹ He spent the first year of his reign fighting Juthungi, Vandals, Goths and the mint workers at Rome; he began building the Aurelian wall during his second visit to the city in the summer of 271.¹⁰ In the second half of that same year, he abandoned the old province of Dacia.¹¹ However, Aurelian did succeed in restoring the territorial integrity of the Roman Empire, for he conquered Palmyra in August 272¹² and the Gallic Empire fell into his hands in the spring of 274.¹³ Yet his successes were no protection against conspiracy. In the autumn of 275, one of Aurelian's officials, fearing for his own life, persuaded a group of officers to assassinate the emperor by forging a 'hit list' with their names on it.¹⁴

Zosimus reveals his ignorance of events in the western part of the empire at the very beginning of his account of Aurelian when he states that Aurelian's predecessor, Quintillus, reigned for a few months (47. 1),¹⁵ not the seventeen days found in the majority of the ancient sources.¹⁶ On the evidence of coins and papyri, Saunders concludes that Quintillus ruled just over a month.¹⁷ The difficulties worsen with regard to Aurelian's campaigns against the northern barbarians during the first year of his reign. According to Zosimus, Aurelian went from Rome to Aquileia to Pannonia where he fought an inconclusive battle against the Scyths (48). When he learned that the Alamanni were going to invade Italy, he left a garrison in Pannonia and went towards Italy, but then fought the barbarians near the Danube on the borders of the Empire (49.1). Even if there were no other sources with which to compare Zosimus, his account is clearly defective, for battles are not fought on the Danube during an invasion of Italy.

Although the evidence is not crystal clear, Aurelian does seem to have fought three wars in 270/271.¹⁸ The first was against the Juthungi on the Danube, the second against the Vandals in Pannonia, and the third against the Juthungi who invaded Italy and inflicted a defeat upon Aurelian at Placentia. The omissions in Zosimus' account are all the more unexpected because these three wars are attested in Dexippus, an historian with whom Eunapius was very familiar.¹⁹ Indeed, since both Zosimus (48.2) and Dexippus (F 6) describe the barbarians coming to parley after their defeat on the Danube, it may be assumed that Dexippus was Eunapius’ source. Zosimus' omission of the fighting in Italy may be due to his failure to epitomize Eunapius accurately, but it is suspicious that Zosimus puts the concluding battle of the first campaign at the end of the second.²⁰ Thus it is highly probable that Eunapius telescoped these two campaigns, just as he did Stilicho's two expeditions to Greece.²¹ When Eunapius took such liberties with major events which had happened only ten or twenty years before the time of writing, he can have had few qualms about doing the same to Aurelian's wars. Here, Eunapius’ motive in combining two campaigns may have been to remove the disgrace of the defeat at Placentia in order not to tarnish Aurelian's image as a warrior emperor.

The only piece of information which Zosimus gives about Aurelian's defence of Pannonia against the Scyths is that he ordered all foodstuffs to be taken into the cities in order to inflict starvation on the invaders (48.1). Although Aurelian may actually have done this, it is entirely possible that Eunapius fleshed out his account by retrojecting into the third century a tactic which was common in the fourth and which would hence be accepted without question by his readers.²² Given the general dearth of information about Aurelian and the fact that none of the other, admittedly jejune, surviving sources gives any precise details about these campaigns, it is not unreasonable to assume that Eunapius transferred this stratagem from the fourth century.

After remarking that Aurelian began to circumvallate Rome and that there were a senatorial conspiracy and revolts by Septimius, Urbanus and Domitian (49.2), Zosimus proceeds directly to the campaign against Palmyra. Little is known about the conspiracy or the rebels,²³ and thus it is not surprising that Zosimus' account of these western events is sketchy. However, it is curious that Zosimus completely omits Aurelian's abandonment of Dacia and the Gothic war, both of which preceded the expedition against Palmyra, and both of which are attested in fourth century sources. The Historia Augusta has a brief account of the Gothic war (VA 22.2) and Ammianus (XXXI 5.17) and Eutropius (IX 13.1) state that Aurelian defeated the Goths.²⁴ Both Eutropius (IX 15.1) and the Historia Augusta (VA 39.7) relate that Aurelian abandoned the old province of Dacia and created a new Dacia on the right bank of the Danube.²⁵ It is thus safe to conclude that Eunapius must have known about the war and especially about the abandonment of Dacia. A possible reason for Eunapius’ failure even to mention Dacia may be found in the long digression which Zosimus retains in Book III 32. This digression cites historical examples in order to prove that Jovian was the first Roman emperor to cede territory to the enemy. Clearly a propagandist like Eunapius could not have admitted that the pagan Aurelian had anticipated the pusillanimous Christian Jovian.

On the other hand, it is probable that Eunapius had little information about Aurelian's campaign against Palmyra at least until the Roman army reached Antioch, for Zosimus only notes the bare fact of Aurelian's recovering Ancyra and Tyana (50. 2). Both the Historia Augusta (VA 22-23) and the Anonymus who is thought to be identical with Petrus Patricius (fr. 10.4)²⁶ tell how Aurelian promised his soldiers to leave not even a dog alive in Tyana if they captured the city. When the city fell, Aurelian kept this promise quite literally and thus spared the human inhabitants. The Historia Augusta attributes this act of mercy to Aurelian's seeing a vision of Apollonius of Tyana (VA 24). The dog story would surely have appealed to Eunapius’ novelistic tastes, nor would he have lost the opportunity to exploit that quasi-divine pagan sage, Apollonius of Tyana, had he known of his appearance to Aurelian.

The course of events leading to the capture of Palmyra is bedevilled by confusion, uncertainty and worse. The siege of Palmyra, which is the climax of the campaign in both Zosimus (54 ff.) and the Historia Augusta (VA 26 ff.), cannot have occurred since Palmyra lacked the necessary fortifications, and no evidence has been found of Roman siege works.²⁷ It is thus entirely possible that other episodes have also been invented by the sources. The prime candidate is the battle at Antioch which is found only in Zosimus (50.2-4). Clanville Downey's reconstruction of a battle at Immae near Antioch which reconciles Zosimus' account of the battle at Antioch with the sparse and disparate information given by the other sources is not necessarily the most persuasive interpretation of the evidence.²⁸ For example, L. Homo, taking his cue from Mommsen, suggested that the chroniclers had confused Immae and Emesa.²⁹ For his part, E. Honigmann speculated that Aurelian's victory was located at Immae because that was where Elagabalus had defeated Macrinus in 218.³⁰

However, the major weakness in Downey's solution is his uncritical reliance upon Zosimus.³¹ Not only is Zosimus the only source for this battle at Antioch, but he gives no geographical context except for a mention of the Orontes with reference to the positioning of the Roman infantry. The fighting is not described in any detail apart from the ruse performed by the Roman cavalry which is identical to the tactics which Aurelian employed, unsuccessfully, at the battle of Emesa (53.1). Yet Eunapius seldom if ever invented history ex nihilo and it is unlikely that he did so in this case, for Festus, Eutropius, Syncellus, Jordanes, and Eusebius-Hieronymus all place Zenobia's defeat near Antioch.³² However, given the scale of treatment in these works and the limitations of their information, Antioch need be nothing more than a geographical reference point for the conclusion of the campaign. Thus the prudent conclusion appears to be that Eunapius was inspired by this tradition of a battle in the vicinity of Antioch and retrojected into his fictional account the cavalry tactics from the battle of Emesa. Indeed, it is these cavalry tactics which reveal a possible motive for Eunapius’ historical fiction. At Emesa, Aurelian's deployment of the cavalry almost cost him the battle (vid. inf.), but by showing how well these same tactics had earlier succeeded at Antioch, Eunapius forestalled criticism of Aurelian's generalship.

Both Zosimus (52.1-2) and the Historia Augusta (VA 25) describe a brief engagement at Daphne, a suburb of Antioch. This battle is probably historical, even if Eunapius’ imagination is the source of Zosimus' graphic narrative of the assault on the hill. In the Historia Augusta, victory at Daphne gives Aurelian control of Antioch, while in Zosimus it clears the way for his march towards Emesa. The Historia Augusta is perhaps the more trustworthy witness on this point because Eunapius would have had to find another purpose for the assault on Daphne than the capture of the city, for in his Histories that was the result of the battle of Antioch. Although Daphne does command the most direct route from Antioch to Emesa,³³ this is a fact which Eunapius could easily have known and thus have deliberately used as the basis of a plausible fiction.

The decisive battle was fought at Emesa. Although only Zosimus and the Historia Augusta describe the fighting in any detail, they agree on the basic course of the battle and there is no compelling reason to suppose that it has been fabricated. In both Zosimus (53) and the Historia Augusta (25), the Palmyrene cavalry overcome the Roman cavalry, but the Roman infantry carry the day. Zosimus, however, attributes the Roman victory to the battlefield manoeuvre executed by the infantry, while the Historia Augusta credits the inspiration of a divina forma. Indeed, in Eunapius’ Histories the battle must have assumed epic proportions, for Zosimus lists the various units of the Palmyrene and Roman armies and vividly recounts the havoc wreaked by the unconventionally armed Palestinians on the Palmyrene heavy cavalry. Eunapius may, in fact, have exaggerated the scale and difficulty of the battle of Emesa. Certainly, it does not seem to have made much of an impression on the western sources. Aurelius Victor and the Epitome de Caesaribus are unaware of Zenobia, Festus (24) says that she was defeated at Immae, and Eutropius (IX. 13) minimizes Aurelian's achievement: Zenobiam ... sine gravi proelio cepit.

The siege of Palmyra, which occupies three chapters in Zosimus (54-56), reads like an historical novel written by an omniscient narrator. At first arrogantly confident in their impregnable city, the Palmyrenes eventually split into two factions, one wanting to resist to the death, the other to surrender. Ironically, the Palmyrenes, who thought that lack of provisions would compel Aurelian to raise the siege, were themselves forced to capitulate because of hunger. Eunapius, who liked feats of archery,³⁴ included an episode in which a Roman sharpshooter killed a Palmyrene loudmouth who was insulting Aurelian. He also described how disappointed Aurelian was that he had conquered only a woman and would thus have no lasting fame. This latter reflection was not unique to Eunapius, for the Historia Augusta develops the idea at some length in a fictional letter supposedly written by Aurelian himself (VA 26). Indeed, the fact that both the Historia Augusta and Zosimus relate the siege of Palmyra, but with significant differences, indicates not only that there is no direct dependence of Eunapius on the Historia Augusta, or vice versa, but also that Eunapius did not invent the idea of the siege. Hence it seems that a common tradition, albeit a false or misleading one, must underlie the historical fiction found in Zosimus and the Historia Augusta. The siege of Palmyra is thus of the same ilk as the battle at Antioch discussed above.

Zenobia fled from Palmyra while it was still under siege, but was captured by the Romans before she crossed the Euphrates (55.3). She and her supporters were put on trial at Emesa, and the rhetor and philosopher Longinus was condemned to death (56.2-3). Up to this point, Zosimus and the Historia Augusta are in essential agreement on the subject of Zenobia. For example, both name Longinus and note that Zenobia and her entourage rode camels on their flight (VA 28). Whether or not these details are historically accurate, they derive from a common tradition which both the western author of the Historia Augusta and the eastern Eunapius knew and accepted. However, Zosimus' account of what happened after the trial is quite heterodox, for he states that Zenobia died either from sickness or from not eating, and that all the other Palmyrenes whom Aurelian was taking to Europe drowned in the strait between Chalcedon and Byzantium (59), with the sole exception of Zenobia's son. However, both the Historia Augusta (VA 34) and Eutropius (IX. 13) describe how Zenobia was displayed in Aurelian's triumph, after which she seems to have lived comfortably in Rome.³⁵ The weight of the opposing evidence is definitely against Zosimus, not only because of its quantity and basic consistency, but because it is western.³⁶ How then is Zosimus' version to be explained? There seems to be no political or ideological reason for Eunapius to have invented her death and the mass drowning. The explanation may be that, as far as the eastern sources were concerned, Zenobia and her companions disappeared from the page of history after she left Emesa, and imagination filled the vacuum. There is an indication that Eunapius was uncertain about the cause of Zenobia's death, for Zosimus introduces the two alternatives with the cautionary "they say" (59).

While Aurelian was returning to Europe, Palmyra revolted. Zosimus gives a detailed account of how a certain Apsaeus tried to persuade Marcellinus, the governor of Mesopotamia, to revolt. When he demurred, the Palmyrenes proclaimed Antiochus emperor (60). This version appears to be grounded in fact, for these three characters are attested in epigraphic and other evidence.³⁷ The Historia Augusta also records the revolt of Palmyra, although the details are different and the usurper is called Achilleus (VA 31). Zosimus tells a curious story about how, when Aurelian returned to Antioch, he astonished the people by unexpectedly appearing at a horse race (61.1). This episode must be a variant of a story in the Historia Augusta: when Aurelian displayed himself in the purple on a hill near Tyana the city surrendered (VA 24). In Zosimus, of course, there is no siege of Tyana to which this story could be attached and, indeed, it may have existed independent of a particular context. Both Zosimus (61.1) and the Historia Augusta (VA 31) state that Aurelian destroyed Palmyra after suppressing the revolt, but, once again, there is no archaeological evidence to support the historians' claims.³⁸

Zosimus deals with the rest of Aurelian's reign in one chapter (61). He briefly mentions the suppression of the Alexandrians and then notes that Aurelian built the Temple of the Sun, reformed the currency, and organised a bread dole for the people of Rome. Zosimus differs from the Historia Augusta and other western sources in placing Aurelian's triumph before the deposition of Tetricus, the last Gallic emperor, and thus once again reveals Eunapius’ ignorance of western events.³⁹ Zosimus devotes all of the final chapter on Aurelian (62) to the intrigue which led to his assassination. Since very much the same story is told by the Historia Augusta (VA 36) and Eutropius (IX. 15), it must have been well known in both the east and west. Clearly, Eunapius was not alone in having a livelier interest in tales of conspiracy and betrayal⁴⁰ than in the finer points of currency reform.

What conclusions can be drawn from this examination of Eunapius’ treatment of Aurelian? Although western sources were better informed about western events and eastern sources about eastern, it is abundantly clear that the fourth century knew very little about Aurelian. The epitomators and chroniclers stated the few facts which they had, while Eunapius and the author of the Historia Augusta composed historical novels. Yet Eunapius’ purpose was not to entertain his readers, but rather to indoctrinate them with his Hellenic ideology, and so the first emperor of his Histories was made into a sanitized and successful pagan who restored an Empire which still enjoyed the favour of the gods.⁴¹

Footnotes

1 R. Goulet, 'Sur la chronologie de la vie et des œuvres d'Eunape de Sardes', JHS 100 (1980), 64.

2 Photius, Bibliotheca, cod. 77.

3 Photius, Bibliotheca, cod. 98. R. C. Blockley, The Fragmentary Classicising Historians of the Later Roman Empire (Liverpool, 1981), vol. I, 2.

4 R.T. Saunders, A Biography of the Emperor Aurelian (A.D. 270- 275) (Diss. Cincinnati, 1991). Leon Homo, Essai sur le règne de l'Empereur Aurélian (270-275) (Paris, 1904).

5 D.F. Buck, 'Some Distortions in Eunapius’ Account of Julian the Apostate', The Ancient History Bulletin 4 (1990), 113-115. D. F. Buck, 'Eunapius of Sardis and Theodosius the Great', Byzantion 58 (1988), 36-53.

6 Saunders, op. cit., 104.

7 Ibid., 106-110.

8 Ibid., 121-130.

9 Ibid., 145.

10 Ibid., 163-197.

11 Ibid., 199.

12 Ibid., 237.

13 Ibid., 258.

14 Ibid., 273 ff.

15 References to Book I of Zosimus will be in this form.

16 François Paschoud, ed., Zosime, Histoire Nouvelle, tome 1, livres I et 11 (Paris, 1971), n. 75, 162. Subsequent references will take the form: Paschoud, n. 00, p. 000.

17 Saunders, op. cit., 143.

18 Saunders, op. cit., 163 ff.

19 For Dexippus, FGrHist 100 F 6 & 7. For Eunapius’ knowledge of Dexippus, see FHG IV, fr. 1.

20 As noticed by Saunders, op. cit., 164.

21 Alan Cameron, Claudian (Oxford, 1970), 474-7.

22 For its use against Gainas, see Zosimus V 19. 7.

23 Paschoud, n. 77, 163-4.

24 Cf. Saunders, op. cit., 197 f.

25 Ibid., 199 f.; other sources reporting the abandonment are Festus (8), Jordanes (Romana, 217), Malalas (12.301) and Syncellus (721-722).

26 Müller, FHG IV, 191 ff.

27 M. Gawlikowski, 'Les défenses de Palmyre', Syria 51 (1974), 231-42. Gawlikowski, Palmyre (Paris, 1985), 65. Saunders, op. cit., 224 ff. and 228-9.

28 G. Downey, 'Aurelian's Victory over Zenobia at Immae, A.D. 272', TAPA 81 (1950), 57-68.

29 Homo, op. cit., 93, n. 1 citing Th. Mommsen in A. von Sallet, Die Fürsten von Palmyra, 47 n. 73.

30 E. Honigmann, 'Syria', P-W, 4A (1932), col. 1692.

31 Downey, op. cit.: 'Scholars have agreed that the narrative of Zosimus (who used excellent sources, notably Dexippus and Eunapius) is our best account of the campaign.'

32 Ibid., 58-9: Festus (Brev. 24), Eutropius (9.13.2), Syncellus (vol. I, 721.10-12 Bonn ed.), Jordanes (Rom. 291), Eusebius-Hieronymus (an. Abr. 2289, Olymp. 263).

33 Downey, op. cit., 61.

34 Paschoud (n. 82, 49) compares the Lydios episode in 170.

35 For the other sources, see Paschoud, n. 87, 168.

36 Zosimus' version is explicitly rejected by H. Mattingly, CAH XII (1957), 305, n. 1.

37 Paschoud, n. 88, 169.

38 Saunders, op. cit., 246.

39 Cf. Paschoud, n. 70, 170.

40 Cf. the story of the general Timasius, the sausage-seller Bargos, and the latter's concubine (V 8.3-10, 3 and frr. 70 and 71).

41 I wish to record my gratitude to the Department of Classical Studies at the University of Ottawa for their hospitality and support during the sabbatical in which this article was prepared.