SOURCE: Momigliano, Arnaldo. “The Lonely Historian Ammianus Marcellinus.” In Essays in Ancient and Modern Historiography, pp. 127-40. Middletown, Conn.: Wesleyan University Press, 1977.

[In the following essay, Momigliano reflects on Ammianus's motives in depicting himself as “intellectually isolated” and in keeping his emotions and opinions in check.]

I

In the second part of the fourth century the Greek East became economically stronger, militarily safer and religiously more Christian than the Latin West. The rise of Constantinople to the position of a new Rome was the most tangible expression of the new situation; but Antioch was hardly less prestigious. Latin, however, was the language of law and, to a great extent, of administration: it was also the language of the army. Many Greeks felt that they had to learn Latin just because their prospects of a career in the Roman administration had become brighter. In Egypt (and no doubt in other Greek-speaking regions as well) people read the Latin poets and historians for the first time. In Antioch Libanius came to fear the competition of teachers of Latin rhetoric (Or., 2, 44; 58, 21 Förster; Ep. 870 Wolf= 951 Förster). Competence in the Latin language was an accomplishment of Greek speakers to be recorded in prose and verse: ‘you, replete with the Laws, mixing the Italic muse with the sweet spoken honey of the Attic one’.¹ For the pagan intellectuals of the East, Latin had the additional attraction of being the language of the less Christian part of the Empire. In Rome the aristocrats had the reputation (not always deserved) of remaining faithful to the old gods.² They gladly read, translated and imitated Greek works.³ Yet any careful observer of the contemporary scene would easily have noticed that the position of the pagan Latin writers was becoming increasingly precarious. After the intellectual ascendancy of St John Chrysostom and of the two Gregorii, Latin Christianity began to regain intellectual initiative. The last years of the fourth century were dominated by St Ambrose, St Jerome and St Augustine.⁴

The historian of Latin literature therefore has to insert into a rather complex situation the stark fact that the greatest Latin poet and the greatest Latin historian of the late fourth century were both Oriental pagans whose first language was Greek. The former, Claudianus, was born in Egypt;⁵ the latter, Ammianus Marcellinus, came from Syria; both reached fame after emigrating to Italy. Ammianus had been in the élite corps of the protectores domestici (15, 5, 22) before settling in Rome about 380. The army was not only Latin in language, but fundamentally indifferent to religious conflicts.⁶

We know very little of Ammianus' life. With the exception of a letter from Libanius to Ammianus of about a.d. 392 (Ep. 983 Wolf = 1063 Förster) all we have about him comes from his historical work. Compared to, say, Livy or his own model Tacitus, Ammianus would appear to display an almost indecent readiness to speak about himself. Yet there were limits to what even an unconventional historian could say about himself in the course of his work. Furthermore, the loss of the first part of his History (Books I-XIII) includes the proemium which was the customary place for an author to give some account of the genesis of his work and of the social status of his patrons. In any case, as we shall see, the situation which Ammianus created for himself was not likely to encourage him to produce those items of information—about his education, his authentic feelings during the pagan restoration of Julian and his removal to Rome—which we should like to have.

What he tells us explicitly or implicitly is this. He was born in a respectable family (Ingenuus, 19, 8, 6). In 353 he was assigned by the emperor Constantius II to the personal retinue of Ursicinus, magister equitum of the East, who was then in Nisibis (14, 9, 1). Four years later, in 357, he was still too young to be promoted to the tribunate (16, 10, 21). We may assume that he was born about 330. In 355 he escorted Ursicinus to the West (15, 5, 24). Involved in the rift between Constantius II and the Caesar Gallus, Ursicinus saved his skin and by implication that of his faithful Ammianus by helping Constantius to get rid of the usurper Silvanus in Gaul. Ursicinus organized the cold-blooded murder of Silvanus, and Ammianus must have collaborated. In Gaul he saw—but hardly spoke with—Julian, the newly appointed Caesar (16, 2, 8). Very soon Ursicinus and Ammianus were sent back to the East, where a Persian attack was expected (16, 10, 21). The unsuccessful war with the Persians provided plenty of adventures and dangers for Ammianus. He was sent by Ursicinus on a mission of espionage (18, 6, 21), for which Ammianus obviously exploited a friendship of earlier days in Syria. He managed to escape from the besieged city of Amida before it fell to the enemy. He admits that in the flight he found it difficult to keep pace with his two lower-born companions (19, 8, 6). He rejoined Ursicinus at Melitene in Lesser Armenia and went safely to Antioch with him: ‘Antiochiam revisimus insperati’ (19, 8, 12). Not surprisingly, Ursicinus lost his command. What happened to Ammianus we do not know, but three years later we find him following Julian in the new expedition against Persia which ended with Julian's death (23, 5, 7; 24, 1, 5; 25, 3, 1). Ammianus does not say that he was one of the people who listened to Julian's last words (25, 3, 15). He never tries to convey the impression that he was on personal terms with Julian. Nor is he likely to be the honoratior miles who advised the postponement of the election of a successor (25, 5, 3). The only personal detail he emphasizes is, again, the retreat to Antioch: ‘Antiochiam venimus’ (25, 10, 1). Ammianus was in Antioch eight years later in 371 (29, 1, 24), and again in 378 (31, 1, 2). The implication is perhaps that he left the army and lived there. If so, he must have interrupted his residence for travels. His journey to Greece, or more precisely to Mothone in the Peloponnesus, is dated about 365 (26, 10, 15-19: cf. Libanius, Or., 18, 392-3). He visited Thrace apparently later, after the battle of Adrianople of 378 (22, 8, 1; 27, 4, 2; 31, 7, 16 ‘nunc usque’).

Curiously, in the extant books he never says that he was a native and citizen of Antioch, though he may have said so in the lost books. We would have had to infer this from the frequency and character of his mentions of Antioch, if it were not clearly stated as a fact by his fellow Antiochene Libanius (Ep. 1063 quoted). Libanius also offers the only explicit evidence we have that Ammianus went to live in Rome—perhaps after 378. Here again we might have guessed the state of affairs from his various and, normally, hostile remarks about the present inhabitants of the old capital: especially the two digressions of 14, 6 and 28, 4 presuppose long residence in Rome. There is, however, no sufficient justification for inferring from Ammianus' indignation at the treatment of immigrants to Rome during the famine of 384 (14, 6, 19) that he was one of those thrown out of the city by order of the prefect Symmachus.⁷ What in any case Ammianus neither says nor implies is that about 392 he was giving successful readings from his own historical work to Roman audiences. But for Libanius we should never have associated Ammianus with such social events. As the conclusion of the work shows, he wanted to be known simply as ‘an ex-soldier and a Greek’ (31, 16, 9). Diligence he considered to be characteristic of a Greek historian. Timagenes, the historian of the Augustan period, was ‘et diligentia graecus et lingua’ (15, 9, 2). No wonder that in several other passages of his work Ammianus was keen to underline that Greek was his mother tongue (22, 8, 33; 23, 6, 20).

II

The lost part of his histories embraced in thirteen books the period a.d. 96-a.d. 352, whereas the following eighteen books—still extant—contain the story of 25 years, about one books for every 18 months. Arguments have been adduced to eliminate this disproportion. It has been suggested that Ammianus wrote two works which would have been conflated into one by some later editor, just as we know that Tacitus' Annales and Historiae were united in one work of thirty books (cf. St Jerome, Comm. in Zachariam, 3, 14 and the subscriptions to individual books in the Cod. Mediceus Secundus).⁸ In Ammianus, however, the thirty-one books would represent the second work only, allegedly starting with Constantine. This theory, therefore, solves nothing because it does not explain why the present text of Ammianus in thirty-one books goes, as is explicitly stated, ‘a principatu Caesaris Nervae … adusque Valentis interitum’ (31, 16, 9). There is no evidence that Ammianus wrote two different works and that the present books 14-31 are part of the second work. We must accept the fact that the lost books 1-13 told the story from the emperor Nerva onwards. It is better to confess that we do not know how Ammianus distributed his material in the first thirteen books. All we know is that even in the more summary narration of the lost part he indulged in some of his characteristic digressions, because he refers to them in the extant books (22, 15, 1; 27, 8, 4). It was normal for historians to enlarge on recent events, but there was no rule about the space they had to allocate to more remote times. Polybius distinguished sharply between introduction and main narrative of recent events. Livy gradually extended the dimensions of his narrative. Ammianus was probably nearer to Livy than to Polybius. Tacitus' shorter span was no precedent here.

Ammianus seems to have considered book 15 rather than book 14 the beginning of his detailed account. He prefaced it with a methodological declaration which is puzzling in its ambiguity and has not yet been adequately interpreted (15, 1): it was an answer to critics. Another preface indicates that book 26 was the beginning of the last section. Ammianus seems to imply that there had been an interval before he started to write in book 26 his account of the reigns of Valentinianus and Valens. Now 21, 10, 6, assumes that the historian Aurelius Victor had been the prefect of Rome in 388-9, but 22, 16, 12 was almost certainly written before the Christians destroyed the Serapeum of Alexandria about 391.⁹ One can date books 21-22, and probably also 23-25, about the year 390. It is more difficult to date the publication of 26-31. 27, 11, 2-8 attacks the powerful Petronius Probus in a way which clearly implies that he was dead, but all we know is that he died between 389 and 395. 29, 6, 15 has an allusion to Theodosius ‘princeps postea perspectissimus’ which does not seem to suit a living emperor: one is naturally inclined to take it as having been written after Theodosius' death in 395. On the other hand O. J. Maenchen-Helfen argued that St Jerome derived his information on the Huns from Ammianus.¹⁰ As this information begins to appear in the Adversus Iovinianum published in 393, St Jerome would have been able to read Ammianus' book 31 in that year. The verbal similarities between St Jerome and Ammianus are certainly striking (Ammianus, 31, 2, 3 ‘semicruda cuiusvis pecoris carne vescuntur’ is matched by St Jerome, 2, 7 ‘semicrudis vescuntur carnibus’) and cannot be separated from other contacts. It is prudent to leave undecided whether Ammianus ended his work before or after the death of Theodosius.

III

Ammianus had in mind a precise model when he chose to write his history in Latin. He presents himself as a continuator of Tacitus. He begins where Tacitus stops, and his thirty-one books are close enough to the thirty books of the combined edition of Tacitus' Annales and Historiae which St Jerome knew. He continues Tacitus as the Historia Augusta continues Suetonius. In the same vein, the panegyrists of the emperors of the fourth century imitated Pliny's panegyric of Trajan, and Symmachus (or his editor) distributed his correspondence in imitation of Pliny's ten books of letters. The Silver Age authors were coming back into fashion, and though Ammianus seems to disapprove of the new popularity of Juvenal (28, 4, 14) it is evident that he studied him. Claudian may owe more to Virgil than to Lucan, but he is well acquainted with the poetry of the early empire. We are therefore not surprised to hear from Libanius that Ammianus found an approving audience in Rome for his revival of the Tacitean tradition. The self-confidence which is apparent in his style seems to confirm this success.

What we should like to know is who were the audience. The rest of the evidence conveys the impression that Ammianus had few and remote, though perhaps powerful, friends and protectors. About 360 Libanius may have introduced him as a young man to friends as ‘Ammianus … a soldier in appearance, but a Socrates in his deeds’ (Ep. 234 W. = 233 F.). If he is the Ammianus of this letter (which is not certain) there is no sign that their relation between 360 and 390 went beyond mere civilities. Libanius is never mentioned in the extant part of Ammianus' work; and our Ammianus, even in the most optimistic of interpretations, fills a very small space in Libanius' large correspondence. The mere fact that Ammianus moved to Rome seems to indicate that he had severed his roots in Antioch.

In Rome, as far as we can judge from his histories, there were few people he respected; and those he respected, he respected from afar: first of all the great aristocrat Agorius Praetextatus (27, 9, 8; 28, 1, 24), then—in less warm or less bombastic terms—Virius Nicomachus (28, 6, 28), the prefects of the City Eupraxius (27, 6, 14; 28, 1, 25), Hypatius (29, 2, 16) and Aurelius Victor (21, 10, 6). I do not see any reason to doubt that he knew these eminent men personally; this is virtually certain for ‘noster Hypatius’, an Antiochene by election, if not by birth, who was prefect of Rome in 379. Aurelius Symmachus, prefect of 384, the famine year, is conspicuously absent. Nor can we fill the lacuna by taking Symmachus' letter 9, 110 as being directed to Ammianus. As A. Cameron has definitely shown, in order to identify Ammianus with the anonymous addressee of this letter, one must believe that he was a senator (for which there is no other evidence) and that Symmachus would be so silly as to translate Greek for his benefit.¹¹ The letter is directed to a historian who was trying eloquence; why should he be Ammianus? It is a poor substitute to have Symmachus' father favourably mentioned by Ammianus (27, 3, 3). There may be an element of prudent adulation in this, as there is certainly adulation in the frequent conventional compliments paid to the father of Theodosius (27, 8, 3; 28, 3, 1; 29, 5, 15, etc., etc.) Strange to say, the only man for whom Ammianus shows real devotion and for whom his eulogy is both constant and sincere is his old patron Ursicinus, an intriguer and a bad general. In mentioning Aurelius Victor, Ammianus qualifies him as a historian. There is, however, no sign that he used or even appreciated Aurelius Victor's summary of imperial history. Victor's booklet was less jejune than other fashionable epitomes of Roman history (such as those by Eutropius and Rufius Festus), but still belonged to that type of ‘official’ and superficial history against which Ammianus was reacting.

It is even more difficult to explain why Virius Nicomachus is not given credit for his historical work which we know to have been encouraged by Theodosius (Dessau, ILS, 2947). A. Alföldi's bizarre notion that Ammianus copied Nicomachus Flavianus in his account of Valentinianus and Valens¹² was disposed of in a few lines by N. H. Baynes.¹³ But even the earlier, more sophisticated version of the same thesis by O. Seeck that Nicomachus was the ‘Thucydidean’ source of Ammianus is refuted by the simple remark that Ammianus had no Thucydidean source.¹⁴ What is evident is that Ammianus was not anxious to tell his readers what his contemporary sources were. As Zosimus did not read Ammianus, the similarities between them on Julian's reign have to be accounted for by a common source. Eunapius or Eunapius' source, Julian's doctor and friend Oribasius, are the most likely candidates for this position.¹⁵ But Ammianus is careful not to mention either, though both were renowned names, Oribasius in particular.

IV

Whatever the reasons, Ammianus himself does his best to give the impression that he is intellectually isolated. He reserves his praise for powerful political men, never for contemporary writers as such. And even the praise of political men is no sign of shared political or (what was almost the same at that time) religious convictions. Ammianus does not belong to a ‘party’ or ‘faction’.

Superficially, there would be some justification in treating him as a representative of the upper stratum of the curiales, the hard-pressed governing bodies of the cities. He is certainly very sensitive to any infringement of property rights and to any increase in taxation (16, 5, 13-15). He considers the protection of propertied classes to be one of the first duties of government. He explicitly takes sides with the curiales of Antioch against the Caesar Gallus (14, 7, 2).

Yet the mere fact that Ammianus chose to live in Rome deprives his championship of the curiales of any political weight: he is not Libanius. His sympathy for the curiales is rather one of the various elements of his conservative outlook, for which concord in the State, lack of greed and corruption, justice, readiness to fight for one's country (15, 12, 3) and simplicity in living (31, 5, 14) are even more important.

What defined a man writing in Latin in Rome was what he thought about Christianity, the Germans and the role of the Roman Senate. About these questions Ammianus was careful not to be drawn. He was quite determined not to make an issue of them. He considered Julian's persecution of the Christians a mistake and could hardly bring himself to mention the prohibition of teaching which he had imposed on Christian grammarians and rhetoricians: ‘illud … obruendum perenni silentio’ (22, 10, 7). The idea of rebuilding the temple of Jerusalem to prove the Christians wrong seemed to him odd (23, 1, 2). It is significant that in order to show that even Julian could talk nonsense he reports a mildly anti-Semitic joke attributed to Marcus Aurelius (22, 5, 5). In the same way he reproaches Julian for indulging in anti-German boasts which proved to be empty (21, 10, 8). Though he is obviously very unsympathetic to Valens' policy of transferring Goths to Thrace (31, 4, 6), he is merciless in denouncing the provocative behaviour of the Roman authorities towards them (31, 4, 9).¹⁶ He even understands the reasons which prompted the rebellion of the Frank Silvanus (15, 5, 32), though he had no compunction in helping his murderer. He never mentions the conflict about the altar of Victoria which represented a challenge to the authority of the Roman Senate and never treats the behaviour of the emperors towards the Senate as a test of the quality of their government. He does not take sides between St Ambrose and Symmachus, between the anti-German Libanius and the pro-German Themistius, and has none of the senatorial nostalgias of the Scriptores Historiae Augustae.

Refusal to choose is not equivalent to emotional indifference. Ammianus is openly pagan and admires Julian. He may have no very clear ideas about the numen to which he so frequently alludes (14, 11, 24; 15, 8, 9; 19, 1, 4 etc.) or about Adrastia, the goddess of Justice, to which he pays homage in a famous passage (14, 11, 25-26). But he relies on the good old divination of augures and haruspices (23, 5, 10; 31, 1, 2 etc.); holds the Libri Sibyllae, the Libri Tagetici and other classics of the disciplina etrusca in high respect (23, 1, 7; 17, 10, 2; 23, 5, 13; 25, 2, 7); and finds the cradle of true prophecy in Chaldaea (23, 6, 25). He is disgusted by the disputes of Christian sects (22, 5, 3; 27, 3, 12). Similarly, he has no love for the Germans. If he has to introduce them, animal images come naturally to his mind (26, 5, 7; 31, 8, 9; 31, 15, 2). The fact that he applies the same imagery to Christians (22, 5, 4) and to rebel Roman troops or provincial tribes (15, 5, 23; 28, 6, 4) only makes the point more evident.

There are at least three reasons why Ammianus refuses to make his emotions the basis for his practical choices. The first reason is quite simply prudence. He is only too aware of the ‘pericula … veritati saepe contigua’ (26, 1, 1). Secondly, he has a genuine respect both for Christianity and for the Germans. Christianity is to him ‘religio absoluta, et simplex’ (21, 16, 18) which ‘nihil nisi iustum suadet et lene’ (22, 11, 5). He admires Christian martyrs (22, 11, 10; 27, 7, 5). The Germans are brave (25, 6, 14) and can be devoted to a good general (18, 2, 6). Thirdly, Ammianus does not like unjustified generalizations. He makes a difference between the good (27, 3, 15) and the bad bishop (22, 11, 4; 27, 3, 11), and he defends a bishop accused of having betrayed the Romans (20, 7, 7). In the same way Ammianus denounces bad German officers, but feels free to report that the barbarian Dagalaifus gave the right advice to Valentinian about choosing a colleague: ‘if you love your relatives, have your brother; if you love the State, look round for the man to invest with office’ (26, 4, 1). The separation between emotions and perception of realities was indispensable to Ammianus, if his conservatism were to have any substance at all. Choice of religion or the admittance of Germans had not been political issues in the old Rome. To yield to anti-Christian or anti-German customs amounted to dereliction of Roma aeterna. Contemporary disasters notwithstanding, Ammianus still clung to the notion of the eternity of Rome: ‘victura dum erunt homines Roma’ (14, 6, 3). Rome was the place where Julian would have deserved to be buried (25, 10, 5).¹⁷ Ammianus never really considered the empire to be in mortal danger (31, 5, 11-17). The continuity of Rome required that the successful defence of the empire, internal justice and peace, and low taxation—not religious conflicts and the recruitment of Germans—should be the true concern of the state.

A position of this kind might easily lead to Utopia or to antiquarianism or both. Indeed there is an undeniable trace of both in Ammianus. He idealizes the past without really knowing it. What he says about ancient Rome is too often derived from literary exempla. His bookish attitude is particularly noticeable in his numerous excursuses where too much second-hand knowledge is mixed with personal observations. Ammianus had an altogether exceptional direct knowledge of the empire, the greater part of which he visited: Britain, Spain and Africa west of Egypt were apparently the only three important regions he had never seen. He spoke of course both languages. It is therefore all the more remarkable that so much of his geographical information should be derivative. He gives disproportionate attention to the events of the city of Rome and is deliberately reticent about the new Rome of the East. He is, again deliberately, vague about things he must have known well, such as the Roman provincial administration and the army. He is the opposite of Polybius: he neither claims nor displays any special competence in military matters, though he had been a soldier for many years. His battle scenes are not for the expert; his knowledge of war machines seems to be elementary (23, 4). It was clearly a mistake to attribute to him the contemporary anonymous tract De rebus bellicis which is an attempt, however misguided, to use technology for the defence of the empire.

But Ammianus is far too aware of what is at stake to become an opinionated pedant. He has an acute, almost demonic, perception of contemporary realities: one regrets not being able to compare his account of twenty-five years of contemporary history with his narrative of the previous period. It would be illuminating to see whether he changed his method of narration and style when he left the past for the present. What we have is a picture of men and events which in its power of characterization through deformation inevitably reminds us of El Greco. Ammianus exaggerates everything. He may have found it difficult to accommodate his Greek syntactic habits to the Latin language, but his convoluted style is not to be understood as a sum of Graecisms.¹⁸

He is in the current leading from Tertullian and Apuleius to Sidonius Apollinaris and Ennodius, but with a twist of his own. He gives maximum emphasis to absurdities and follies—to violence and cunning. Roma aeterna is the ultimate judge and the ultimate hope. But the extraordinary contortions and complications of his style—which are not incompatible with its self-confidence—indicate how difficult it is for Ammianus to keep control of his world.

Ammianus does not feel safe in his own surroundings. Black magic is not to him an object of intellectual curiosity. He speaks of it as a menace (26, 3, 2; 29, 2, 17 and elsewhere). The fact that emperors encouraged charges of black magic for their own purposes does not make it any less frightful.¹⁹Fortuna and fatum are intimation of death (21, 14, 1; 30, 5, 15 etc.).²⁰ Ammianus, though intellectually in favour of clemency (29, 2, 19), accepts ruthlessness and cruelty as inevitable. We are blandly told that no torture has yet been thought up capable of compelling an Egyptian bandit to reveal his name (22, 16, 23). Brutality and horrors are on display. There is no need here to illustrate in detail Ammianus' love for the grotesque and the horrific to which E. Auerbach dedicated a famous chapter of his Mimesis.

Ammianus follows Tacitus in giving pride of place to court intrigues and to wars. But no systematic comparison between the subject matter and the narrative techniques of the two writers is yet available.²¹ It would probably show that Ammianus is more sensitive to education and tradition and less so to social status. His factual accuracy must not be taken for granted.²² He lacks technical competence and aims at literary effects. But few important errors of fact have so far been discovered in his work. Where he reflects contemporary categories of thought or expresses contemporary standards of behaviour, moods and passions, he is, of course, to be trusted unhesitatingly. The authenticity of his accounts of social attitudes is guaranteed by the stylistic elaboration to which they are submitted. This type of distortion is measurable and therefore no source of equivocation. Nobody will take literally the satirical description of the lawyers in the Eastern part of the empire (30, 4, 4): it was probably written for a public lecture in Rome. Even the famous page on the hieratic attitudes of Constantius II during his visit to Rome (16, 10 ff.) has to be appreciated as an imaginative and convincing piece of reconstruction of something Ammianus never witnessed. What is reassuring is that Ammianus never attempts to enter into the minds of dedicated Christians. The neo-Platonic speculations he summarizes (21, 1; 21, 14) still belong to Roma aeterna. Beyond Roma aeterna Ammianus becomes circumspect.

Notes

1. I. Ševčenko, ‘A Late Antique Epigram’, Synthronon, Paris, 1968, pp. 29-41 with rich bibliography. Cf. A.-J. Festugière, Antioche païenne et chrétienne, Paris, 1959, p. 411.

2. H. Bloch in A. Momigliano, The Conflict between Paganism and Christianity in the Fourth Century, Oxford, 1963, pp. 193-218, but cf. P. Brown, JRS [Journal of Roman Studies], LI, 1961, pp. 1-11 (= Religion and Society in the Age of St Augustine, London, 1972, pp. 161-82).

3. See, for instance, W. Speyer, Naucellius und sein Kreis, München, 1959, with the review by W. Schmid, Gnomon, XXXII, 1960, pp. 340-60; cf. F. Vittinghoff, HZ [Historische Zeitschrift], CXCVIII, 1964, pp. 529-74.

4. P. Brown, The World of Late Antiquity, London, 1971, pp. 115-23; J. Daniélou-H. Marrou, Des Origines à Saint Grégoire le Grand, Paris, 1963, pp. 341-8.

5. A. Cameron, Claudian, Oxford, 1970, a book of general importance.

6. A. H. M. Jones in A. Momigliano, The Conflict quoted, 23-6.

7. H. P. Kohns, Versorgungskrisen und Hungerrevolten im spätantiken Rom, Bonn, 1961, pp. 168-82. The date of Ammianus' journey to Egypt is unknown (17, 4, 6; 22, 15). For other details of his life see 18, 4, 7; 18, 6, 5; 18, 8, 11.

8. H. T. Rowell, Ammianus Marcellinus, Soldier-Historian of the Late Roman Empire, Cincinnati (Semple Lectures), 1964; Id., Mélanges Carcopino, Paris, 1966, 839-48. On Tacitus, F. Brunhölz, Abh. Marburg Gelehrt. Gesell. 1971, 3, 111-43.

9. J. Schwartz, La fin du Serapeum d'Alexandrie, in ‘Essays in Honor of C. Bradford Welles’, New Haven, 1966, pp. 97-111.

10. AJPh [American Journal of Philosophy], LXXVI, 1955, pp. 384-99. Contra R. Syme, JRS, LVIII, 1968, p. 218; Ammianus and the Historia Augusta, Oxford, 1968, pp. 12-24. But see A. Cameron, JRS, LXI, 1971, 259-61.

11. A. Cameron, The Roman Friends of Ammianus, JRS, LIV, 1964, pp. 15-28. Cf. A. Selem, A proposito degli amici romani di Ammiano, Annali Libera Università della Tuscia, III, 1972, pp. 1-50; Ammiano e la morte di Giuliano, Rend. Ist. Lombardo, 107, 1973, pp. 1119-35.

12. A Conflict of Ideas in the Late Roman Empire, Oxford, 1952.

13. JRS, XLIII, 1953, p. 169.

14. Hermes, XLI, 1906, pp. 481-539, supported by R. Laqueur in Probleme der Spätantike, Stuttgart, 1930, pp. 33-6.

15. W. R. Chalmers, CQ [Classical Quarterly], LIV, 1960, pp. 152-60; A. Cameron, Ib., LVI, 1963, pp. 232-6 (also LIX, 1965, pp. 289-98). But cf. also L. Dillemann, Syria, XXXVIII, 1961, pp. 87-158; H. Gärtner, Abhandl. Mainz. Akad., 1968, n. 10; C. J. Classen, Museum Africum, I, 1972, pp. 39-47; D. A. Pauw, Karaktertekening bij Ammianus Marcellinus, diss. Leiden, 1972.

16. A. Demandt, Zeitkritik und Geschichtsbild im Werk Ammians, Bonn, 1965, represents a new departure on Ammianus' attitude towards the Germans.

17. F. Paschoud, Roma aeterna, Paris, 1966; M. Fuhrmann, Die Romidee der Spätantike, HZ, CCXII, 1968, pp. 529-61. Cf. also H. Tränkle, Antike und Abendland, XI, 1962, pp. 21-34.

18. E. Norden, Antike Kunstprosa, II, Berlin, 1898, pp. 646-50. The best analysis of Ammianus' style is perhaps H. Hagendahl, Studia Ammianea, diss. Uppsala, 1921. For literary allusions, G. B. A. Fletcher, RPh [Revue de Philologie], LXIII, 1937, pp. 377-95. See also in general the excellent essay by J. Vogt, Ammianus Marcellinus als erzählender Geschichtsschreiber der Spätzeit, Abhandl. Mainz. Akad., 1963, No. 8 and J. Fontaine, Ammien Marcellin historien romantique, Lettres d'Humanité, XXVIII, 1969, pp. 417-35; R. MacMullen, Some Pictures of A. M., Art Bulletin, XLVI, 1964, pp. 435-55.

19. Cf. A. A. Barb, in A. Momigliano, The Conflict between Paganism and Christianity, pp. 100-25; H. Funke, Jahrb. Antike und Christ., X, 1967, pp. 145-75.

20. C. P. T. Naudé, Fortuna in Ammianus Marcellinus, Acta Classica VII, 1964, pp. 70-88; W. Seyfarth, Ammianus Marcellinus und das Fatum, Klio, XLIII-XLV, 1965, pp. 291-306; L. Bonfante Warren, Parola del Passato, 1964, pp. 401-27.

21. Cf., however, Chr. Samberger, Die Kaiserbiographie in den Res Gestae des Ammianus Marcellinus, Klio, LI, 1969, pp. 349-483; and above all D. Flach, Von Tacitus zu Ammian, Historia, XXI, 1972, pp. 333-50.

22. Cf. K. Rosen, Studien zur Darstellungskunst und Glaubwürdigkeit des Ammianus Marcellinus, diss. Heidelberg, 1968; but contra N. J. E. Austin, In Support of Ammianus' Veracity, Historia, XXII, 1973, pp. 331-5. See in the same instalment of Historia the two papers by R. T. Ridley and T. S. Burns respectively on Julian's Persian Expedition and the Battle of Adrianople.

A Select Bibliography

L. Angliviel de la Beaumelle, Remarques sur l'attitude d'Ammien Marcellin à l'égard du Christianisme, ‘Mélanges W. Seston’, Paris, 1974, pp. 15-24.

J. Bidez, La vie de l'Empereur Julien, Paris, 1930.

K. Bringmann, Ammianus Marcellinus als spätantiker römischer Historiker, Antike und Abendland, 19, 1973, pp. 44-60.

M. Büdinger, Ammianus Marcellinus und die Eigenart seines Geschichtswerkes. Eine universalhistorische Studie. In ‘Denkschriften der phil.-hist. Klasse der Wiener Akademie der Wiss.’, XLIV, Abh. 5, 1896.

P. M. Camus, Ammien Marcellin, Paris, 1967.

L. Dautremer, Ammien Marcellin. Étude d'histoire littéraire, Lille, 1899.

A. Demandt, Zeitkritik und Geschichtsbild im Werk Ammians, Bonn, 1965.

S. Dill, Roman Society in the Last Century of the Western Empire, London, 1898.

W. Ensslin, Zur Geschichtsschreibung und Weltanschauung des Ammianus Marcellinus, Klio Beih. 16, 1923.

H. Gärtner, Einige Überlegungen … zu Ammians Charakteristik des Kaisers Julian, Wiesbaden, 1969.

J. Gimazane, Ammien Marcellin. Sa vie et son œuvre, diss. Bordeaux-Toulouse, 1889.

A. Von Gutschmid, Ammianus Marcellinus, in ‘Kleine Schriften’, V, Leipzig, 1894, pp. 567-84.

W. Hartke, Römische Kinderkaiser. Eine Strukturanalyse römischen Denkens und Daseins, Berlin, 1951.

M. Hertz, Aulus Gellius und Ammianus Marcellinus, Hermes, 8, 1874, pp. 257-302.

W. Klein, Studien zu Ammianus Marcellinus, Klio Beih. 13, 1914.

M. L. W. Laistner, The Greater Roman Historians, Berkeley, 1947.

S. Mazzarino, Il pensiero storico classico, III, Bari, 1967.

A. Momigliano, Pagan and Christian Historiography in the Fourth Century A.D., in ‘The Conflict between Paganism and Christianity in the Fourth Century’, Essays edited by A. Momigliano, Oxford, 1963 (Cf. above, pp. 107-26).

Th. Mommsen, Ammians Geographica, in ‘Gesamm. Schriften’, VII, Berlin, 1909, pp. 393-429.

Ch. P. Th. Naudé, Ammianus Marcellinus in die lig van die antieke geskiedskrywing, diss. Leiden, 1956.

R. Pack, The Roman Digressions of Ammianus Marcellinus, TAPhA [Transactions of the American Philological Association], 1953, pp. 181-9.

G. B. Pighi, Studia Ammianea, Milano, 1935.

Idem, I discorsi nelle storie di Ammiano Marcellino, Milano, 1936.

Idem, Nuovi studi Ammianei, Milano, 1936.

W. Richter, Die Darstellung der Hunnen bei Ammianus Marcellinus, Historia, 23, 1974, pp. 343-77.

R. Syme, Ammianus and the Historia Augusta, Oxford, 1968.

E. A. Thompson, The Historical Work of Ammianus Marcellinus, Cambridge, 1947.

Z. V. Udalcova, Mirovozzrenie Ammiana Marcellina, Vizant. Vremennik, 28, 1968, pp. 38-59.