Julian's Spin Doctor: Ammianus Marcellinus 24.2.22-24.3.8 and the Persian Mutiny
The capture of Pirisabora represented the first major victory for Julian's Persian expedition in A.D. 363. Ammianus Marcellinus, Zosimus, and Libanius all discuss the siege and the subsequent setback the Romans suffered the next day, when three squadrons of scouts were routed and a standard lost. Putting all three accounts together reveals substantial omissions in Ammianus' account which suggest the historian purposefully distorted his account to minimize the damage to the reputation of his hero, Julian.
On the second day of the siege of Pirisabora, Julian himself led an attack against one of the gates of the city but was repelled. He then ordered a helepolis “city-taker” siege engine to be built, the mere sight of which convinced the defenders to surrender under lenient terms of peace (24.2.18-22). Ammianus reports that the soldiers found a large stockpile of grain and weaponry in the citadel, since the city had been evacuated and 2500 men left behind to defend it from the Romans (24.2.22). Of this, the soldiers took what they needed and burnt the remainder as well as the city.
Ammianus’ chronology at this point becomes murky: he next recounts the loss of a standard by a reconnaissance force and the punishment of the men involved postera die “on day following” (24.3.1), and then he relates Julian’s speech which occurred incensa denique urbe, ut memoratum est “after the burning of the city, as I have said” (24.3.3). The reader is left to ask whether the loss of the standard (and punishment of the soldiers) occurred before or after the speech?
Ammianus seems to say that on the same day as the capture of the city, the citadel was found full of goods, the city burned, and Julian’s speech given. The following day, then, the reconnaissance force lost their standard, Julian routed the enemy, and punished the soldiers who had lost the standard. This interpretation means that Ammianus has reported the events of 24.3.1-2 out of sequence, jumping forward to the day after the city was captured and then jumping back to the day of the capture to relate Julian’s speech. Based on just the information he gives, this certainly is a possible interpretation of the sequence of events (1), but when Zosimus’ account is considered it becomes less plausible.
On the siege itself, Ammianus and Zosimus agree, but Zosimus gives much more detail following the city’s surrender. First, he says that in addition to grain and weapons, abundance τῆς ἄλλης ἀποσκευῆς “of other household stuff” was also found (3.18.5). He states that of the large amount of grain found, most was loaded onto ships and the rest split between the men. Of the weapons, the arms useful for Roman battle tactics were distributed to the men and the rest burned or thrown into the river (3.18.5-6). Zosimus' account makes good sense, but accepting it means that Ammianus’ sequence of events become awfully crowded for the day of the capture of Pirisabora: the troops had to have tried to attack the city in the morning, built a siege engine, negotiated terms with the inhabitants, found the stockpile, carried off most (if not all) of the grain and loaded it on the supply ships, burned the city, and then heard Julian’s speech.
On the other hand, the note that Julian’s speech occurred incensa denique urbe “after the city had been burned” does not necessarily place the Julian’s speech immediately before the loss of the standard by the scouts. Two alternatives are possible: either the first notice that the city was burned looks forward to the next day (having been dislocated to round out the climax of the seige in 24.2), or the second notice acts to remind the reader of the situation (the successful capture of an important city and the reason for the donative) and not act as a temporal marker. Indeed, Williams accepts without question that the speech occurred after the punishment of the soldiers (2). Zosimus’ version does not clearly put the loss of the standard either before or after Julian’s speech: while he narrates the speech before defeat of the scouts, he does not give any words which can confirm the ordering is chronological and not just topical (3.19.1).
The account of the attack and Julian’s counterattack also present difficulties when compared to Zosimus’ version, which again adds more details.
Ammianus merely says:
Postera die, quam haec acta erant, perfertur ad imperatorem cibos per otium capientem nuntius grauis Surenam, Persicum ducem, procursatorum partis nostrae tres turmas inopinum aggressum paucissimos trucidasse, inter quos strato tribuno unum rapuisse uexillum. Statimque concitus ira immani cum armigera manu festinatione ipsa tutissimus peruolauit et grassatoribus foeda consternatione depulsis residuos duo tribunos sacramento soluit ut desides et ignauos; decem uero milites ex his, qui fugerant, exauctoratos capitali addixit supplicio secutus ueteres leges. (Ammianus Marcellinus 24.3.1-2)
On the following day, while the emperor was peacefully taking his dinner, he received the unwelcome news that the Persian commander called the Surena had unexpectedly attacked three squadrons of our scouts. Our casualties had been very slight, but a tribune had been killed and a standard captured. Furiously angry, Julian flew in person to the spot with an armed force—the speed of the operation guaranteed its safety—and completely routed the marauders. The two surviving tribunes were cashiered for cowardice and neglect of duty, and ten men out of those who had run away were discharged and put to death in conformity with the ancient Roman practice. (W. Hamilton, trans., The Later Roman Empire, Penguin 1986)
Den Boeft et al. convincingly argue that though it was not technically a decimation (10 men out of an estimated 1050), Ammianus certainly was well aware of the practice and that he considers this incident to be a form of it (3). They also, though, come to the conclusion that “it is impossible to decide whether Ammianus here approves” of the measure, since they conclude that he chooses neutral language with the words ueteres leges, unlike at 29.5.23 when he “fully agrees with the strong measures taken by Theodosius” and uses the word priscus, expressing “awe and respect” (4). While the words describing the custom of decimation may be neutral, though, Ammianus hints that he does not fully approve: he places emphasis on Julian's rage (which Zosimus 3.19.2 confirms) by using the Vergilian (Aeneid 9.694) phrase concitus ira immani “shaken by frightful rage”, as den Boeft et al. themselves point out (5). In addition, the adversative uero introducing the soldiers' punishment may indicate that, unlike the tribunes' demotion, the execution of the men was not entirely justified in Ammianus' eyes.
More problems lie, however, in what Ammianus fails to report, but which Zosimus gives: the Surena battacked μετὰ δυνάμεως οὐκ ὀλίγης “with not a small force” (3.19.1), that Julian recovered the standard (3.19.2), and that he not only routed the enemy but burned the town from which the attack was launched (3.19.2). For Ammianus to leave these three facts out seems strange, since they would increase the glory of Julian’s lightning-fast counterattack—i.e., he fought a larger force, won back the standard, and captured a city. Why does Ammianus diminish Julian's accomplishment by omitting these facts?
The obvious answer lies in execution of the ten soldiers who lost the standard. Given what Ammianus has told the reader, Julian would seem to have acted harshly but within his rights; he lets the reader assume that the Persian force was not very large (thus the slight losses of the scouts implies cowardice), that the standard was not recovered, and that the counterattack by Julian did not accomplish the impressive act of revenge which it did. While not excusing the three squadrons, Zosimus' information certainly mitigates the culpability of scouts and makes Julian’s punishment seem excessive. In the light of the Persian expedition's ultimate failure, Julian's exection of his own men seems wasteful if not outright deplorable. Ammianus, then, seems to have cleaned up the incident as much as he could in good conscience.
Ammianus next gives the second portion of Julian's speech to the soldiers. He does not report the entire speech, in which he says Julian thanked his troops and promised them 100 pieces of silver (24.3.3). This small sum, though, caused the troops to grow angry, and Julian rebukes them, claiming poverty but offering the spoils of the rich Persian empire (24.3.3-6). Should the troops not wish to obey him, he offers his suicide or abdication (24.3.7); at this, the troops acquiesce and promise to comply, though without the overwhelming response Julian’s speeches usually evoke in Ammianus’ history. Zosimus confirms the sum of the gift, but has no mention of the near mutiny which followed (3.18.6).
That the troops were disappointed by their reward after the capture of Pirisabora is plausible: according to Ammianus, they had received only grain and weapons in addition to the silver promised by Julian. Of the abundance τῆς ἄλλης ἀποσκευῆς “of other household stuff” Zosimus mentions, much of it may have been worthless to an army on the march. Finally, the city Julian captured during his counterattack might also not have provided much plunder, or it may not have been shared with the troops at Pirisabora. When it comes to discontent among Julian's soldiers, though, Ammianus sometimes distorts the situation. When Julian, as Caesar, led his men into the Alps, they ran short of food, prompting near mutiny among troops (17.9.3-5); Ammianus, though, plays down the shortage of food and instead emphasizes their complaint of not receiving their pay, which he blames on Constantius (6). Ammianus may be employing the same tactic in Pirisabora, hiding the source of the men's discontent about the decimation behind the pretext of a small donative.
Libanius' account of the loss of the standards and the punishment describes Julian going among the defeated scouts with only a couple of his bodyguards to exact the punishment in person, making an example τοῖς ἄλλοις ἅπασιν "for all the other soldiers" (Or. 18.229-30). The orator's emphasis on Julian's bravery for going among the troops with only a few bodyguards, combined with the statement that the emperor returned to his tent θαυμαστότερος γεγονώς "having become more admirable" (Or. 18.230), suggests that Julian successfully faced some sort of disturbance as a result of his punishment of the defeated soldiers. Ammianus may have elided not just a speech thanking the soldiers, but one in which Julian ordered the deaths of the ten men as an example against cowardice. Ammianus twice refers to Julian threatening his soldiers with being hamstrung if they straggle (7); Julian may similiarly delivered threats at this occasion, provoking the near mutiny of his soldiers. The offer of 100 pieces of silver may have been Julian's attempt to win back their affections, and his offer of suicide or abdication may, as Williams thinks, have been his trump card which only barely worked, as the weak assent of the troops shows (8).
In reporting the events following the capture of Pirisabora, Ammianus has purposefully distorted his account to minimize the harshness--if not outright abuse--of Julian's command and the subsequent mutiny he nearly provokes. By omitting the recovery of the standards and Julian's capture of a city during his counterattack, Ammianus tries to make Julian's actions seem more reasonable, while his sequencing of events--which is confusing at best--may be intentionally murky to prevent the reader from associating the discontent of the troops with execution of the ten soldiers which most likely occurred just prior. Ammianus, it appears, has done his best to disguise the rapidly deteriorating relationship between the emperor and his men, as well as trying to make the best of a sordid incident which occurred on the heels of Julian's first major victory in Persia.
(1) J. den Boeft, et al., Philological and historical commentary on Ammianus Marcellinus XXIV. Brill, 2002, 69
(2) M. Williams, "Four mutinies: Tacitus' Annales 1.16-30; 1.31-49 and Ammianus Marcellinus Res Gestae 20.4.9-20.5.7; 24.3.1-8," Phoenix 51.1, 1997, p. 68.
(3) J. den Boeft, et al., op. cit., ad loc. 24.3.2
(4) J. den Boeft, et al., ibid.
(5) J. den Boeft, et al., ibid.
(6) 17.9.6-7; T. Elliot, Ammianus Marcellinus and fourth century history. Stevens, 1983, 83
(7) Directly at 23.5.21 and elliptically at 24.1.13
(8) M. Williams, op. cit., 69