Bryn Mawr Classical Review 04.05.02

Alan Cameron and Jacqueline Long (with a contribution by Lee Sherry), Barbarians and Politics at the Court of Arcadius. The Transformation of the Classical Heritage, 19. Berkeley: 1993. Pp. xiii + 441. $55.00. ISBN 0-520-06550-6.

Reviewed by R.W. Burgess, University of Ottawa.

This is micro-history with a vengeance. Four hundred and fifty-four pages covering two years, 399-400, about which we know very little indeed. And when Cameron and Long1 are done, it turns out that most of what we thought we knew was wrong. Now to many it may seem a little extravagant to waste over 450 pages on two obscure years in the life of Late Roman Constantinople, but these years are far from unimportant, to judge from what has been made of them by past scholars, and C. has by no means wasted 450 pages. On the contrary, his approach to the topic is one I would recommend to every historian of the Later Roman empire. In fact, I would go so far as to say that this book should stand as a paradigm for the way Late Roman history must be written: C. begins by junking everything that anyone has ever said on the topic, goes back to first principles, and begins to reconstruct his account de novo, testing everything that has been said against the primary evidence, which he first re-evaluates in minute detail. In the end, as I stated above, everything we thought we knew turns out to be flawed or just plain wrong. And, most interestingly, it all began as a response to a paper by T. D. Barnes that eventually appeared in GRBS for 1986 (p. ix).

The topic that C. has analysed revolves around the famous massacre of the Goths in Constantinople on 12 July 400 and the fundamental role that the de regno and de prouidentia of Synesius of Cyrene have played in our understanding of that event, its antecedents, and its outcome. On face value it may sound trivial, but this nexus of actions in 399 and 400 has often been taken to explain why the Empire of the East survived and that of the West fell. It has been claimed that in Constantinople there were two parties, one in favour of using barbarian federates and allowing barbarians to hold high military and civil office, the other opposed to allowing the barbarians any role at all in imperial government. These parties clashed in 399-400, the pro-barbarian forces led by either Caesarius or Eutychianus (on whom, see below), the anti-barbarian forces led by Aurelian, whose spokesman and mouthpiece Synesius was. The anti-barbarian forces won the day and they purged the barbarians from the army and the administration. The East was returned to the hands of the Romans, while the West, especially its army, fell more deeply into the clutches of the barbarians, who would in the end lead to its downfall. This is stirring stuff, and it has been handed down as one of the turning points in Late Roman and early Byzantine history, with only a few serious modifications along the way. C.'s greatest contribution is to show beyond any serious doubt that there were no pro- or anti-barbarian factions, that there was no purge, that barbarians continued to occupy the same sorts of offices they had held before 400, and that Synesius and his two works were not at all what we thought them to be.

C.'s method is exact and painstaking in its detail. Since the works of Synesius figure so prominently in the interpretation of events, C. goes right back to the beginning, and after a short introduction begins (in Chapter Two) with a complete re-evaluation of the life and works of Synesius (pp. 13-69), omitting his trip to Constantinople, to which he devotes Chapter Three (pp. 71-102). Here, as elsewhere, C. strips away a century of accretions that have built up in the study of Synesius and presents a strikingly new account of the man and his life, re-evaluating, for instance, his family (there is no mention of children in Hymn 7, which dates to 401/2, not 411/2), his 'conversion' (there wasn't one), the conflict between his supposed Christian 'orthodoxy' and his Neo-Platonic Hellenism (a black and white dichotomy that C. rightly maintains is false and unrealistic), his baptism (he was baptized before ordination), Hypatia and his relationship to her (a fascinating and wide-ranging section on Late Roman philosophers and scholarship), and the 'Panhellenion' in Constantinople and Synesius' crypto-pagan patrons (C. effectively demolishes both ideas). Most important for the thesis of the book (though it had already been set out in earlier articles by T. D. Barnes and C. himself), he establishes beyond a doubt the true dates for Synesius' trip to Constantinople: 397-400. In Chapter Four (pp. 103-42) he analyses the de regno, the famous anti-barbarian speech that Synesius supposedly delivered to Arcadius on the occasion of the presentation of Cyrene's crown gold for Arcadius' tertia quinquennalia. C. (following Barnes, though they differ on the date) rightly argues that the speech was not in fact delivered before Arcadius, but read to a select group of like-minded individuals in 398, after the real speech -- certainly shorter and more boring -- had been delivered. It is here that the standard idea of the pro- and anti-barbarian parties begins to come off the rails, for Synesius' anti-barbarian rhetoric concerns Alaric, not Tribigild (as Peter Heather has already shown; cf. p. 112). Chapters Five and Six analyse the de prouidentia, Synesius' Egyptian allegory, the other work he composed while in Constantinople and the key to any interpretation of the years 399-400 (pp. 143-252). In Chapter Five C. re-establishes the identity of Typhos as Caesarius and rejects the view of A. H. M. Jones (et al.) that Typhos was Eutychianus. The presentation of the dates of all laws addressed to a PPO Orientis between 395 and 405 in tabular form under the headings 'Eutychian', 'Aurelian', and 'Caesarius' (pp. 156-8) is typical of C.'s detailed and careful exposition and concern for the reader's understanding of such complicated material (as is the handy chronological table, pp. xi-xiii). He also sorts out the other holders of the Eastern prefecture through to Aurelian's second, in 414-416. In Chapter Six he examines what Synesius really says about the barbarians and the events of the years 399-400, and concludes with a reconstructed narrative of the activities of Gaïnas (including a completely new account of the massacre), Tribigild, Aurelian (returned Sept./Oct. 400), and Fravitta (killed in 405). Chapter Seven changes tack completely and suddenly shifts to a literary analysis of the background to and sources for the de prouidentia (pp. 253-300). This chapter seems completely out of place in what has been almost completely a rigorous historical analysis. There is a weak protestation about its being what Synesius would have expected (p. 253), but it fails to smooth over the fact that this chapter looks as though it was produced for the planned commentary that this volume grew out of (p. ix) and was included rather than abandon the effort that went into it. As regards its content, I should only say that I still cannot help but think that the wolf of Synesius' riddle (pp. 258-9, 268, 296-7) is Romans/the Empire in contrast to the lion, which represents the barbarians (cf. pp. 313-4). Chapter Eight discusses the few odds and ends left over from the earlier chapters (such as G. Dagron's bizarre theory that Osiris and Typhos represent the two halves of the Empire), and neatly ties up most of the loose ends (pp. 301-36). Chapter Nine is a translation of and commentary on the de prouidentia (pp. 335-98), which at first seems out of place but is in reality essential to the book since so much hinges on a thorough understanding of the entire piece (and how many readers would dig up PG and wade through it all? As C. says (p. 11): 'We hope thus to facilitate study of a complex work written in extraordinarily difficult Greek.'). Earlier he includes summaries of the de regno (pp. 103-6) and the de prouidentia (pp. 144-9). Other commentators on and analysts of similarly obscure or difficult works would do well to follow C.'s lead in this. The work finishes up with three appendices on 'Aurelian and Pulcheria', 'Chrysostom's Movements in 400-402', and 'Synesius's Visit to Athens' (pp. 399-411).

It is impossible in a review of this size even to try to detail the myriad of major and minor revisions that C. has made in our understanding of Synesius, his works, and the historical background to the period. In my break-down above I have simply tried to note what struck me as the most noteworthy. Suffice it to say that I find most of C.'s arguments convincing, presented as they are with full and clear documentation of primary and secondary material. On the negative side, C. is sometimes too ruthless in his evaluation of the work of others (even your reviewer comes in for an [undeserved] jab) and those on the receiving end may be tempted to evaluate his reconstructions on form rather than content. This would be a mistake.

Most of you who have read this far will probably want to know how C.'s opus compares with the portions of Wolf Liebeschuetz's recent work that cover exactly the same topic.2 In a nutshell, it is like Hyperion to a satyr, even discounting the gross incompetence of the actual writing and production of the latter.3 C.'s volume is the one to which future scholars and graduate students will turn, which is a shame since Liebeschuetz had access to a fairly advanced version of C.'s work and seems not to have been greatly swayed by it.

The force and accuracy of C.'s argumentation and re-evaluation carry the reader along effortlessly, except at one crucial crux where he uncharacteristically flounders. In Synesius' de prouidentia the two protagonists are hidden behind the allegorical disguises of Osiris and Typhos. Osiris is obviously Aurelian, praefectus praetorio Orientis of 399, but scholars have been split over whether Caesarius or Eutychianus should be Typhos, who in the de prouidentia is explicitly stated to have directly succeeded Osiris as King of Egypt (i.e. PPO Orientis). Notable names have cast their weight behind each of the identifications -- Seeck and Barnes for Caesarius, Jones, PLRE I, and Liebeschuetz for Eutychianus -- but no one recently has analysed the problem in as much detail as C. He rightly shows that most of the evidence supports an identification of Typhos with Caesarius. However, three laws in the Codex Theodosianus, dated to December 399, are addressed to Eutychianus as PPO Orientis and thus imply that he, not Caesarius, was the successor to Aurelian (CTh. 12. 1. 163-5; cf. C.'s p. 158). C. wanders around the problem when he comes to it, citing other valid instances where the dates of laws in the CTh are corrupt, and suggests finally that the consular date may perhaps be incorrect, picked up by mistake from some other extracts (pp. 175-7). C.'s vagueness seriously weakens his argument, which can be presented more effectively.

First of all, I think it highly probable that all three laws in the CTh (which deal with curial obligations) are in fact extracts from the same law. In this I follow Seeck, Barnes, and C. This is supported by the dates of the three extracts ('III id. Dec.', 'V kal. Ian.', and 'III kal. Ian., Theodoro vc', i.e. 11 Dec., 28 Dec., and 30 Dec., 399), which all appear to be corruptions of 'III kal. Ian., Honorio IIII et Eutychiano vc', i.e. 30 December, 398 (Eutychianus is known to have been PPO Orientis between 4 Sept. 397 and 25 July 399, and CTh 1. 2. 11 and 15. 1. 4 also date to December 398). The consular date of Theodorus (i.e. 399) was probably picked up from an accepta or proposita date, given that the original law was issued so close to the end of the year (cf. CLRE, p. 77). Then at some point early in the compilation process, the first 'kal.' became an 'id.' In semi-uncial and cursive the standard abbreviations 'Id' and 'kl'4 are easily confused; I have seen a number of examples of this in manuscripts (and the CTh) and there is a case in the chronicle of Hydatius where Hydatius had two dates for the invasion of Spain in 409, both identical except that one had 'id' and the other 'kl' (§ 34 [Mommsen 42]). As a consequence of this copying error, the first law was hypercorrected from 'Ian.' to 'Dec.' since the laws were originally in chronological order, the previous law was dated 'kal. Dec.' (i.e. 1 Dec.), and 'id. Ian.' was eleven months earlier and thus out of sequence. Emending 'Ian.' to 'Dec.' 'corrected' the sequence (to 11 Dec.). The 'III' became a 'V' at some point between the original copying of the excerpt and the date of MS V ('III' to 'II' to 'V', a standard error as well).

Second, C. shows beyond a doubt (pace Liebeschuetz, PLRE, and others) that Aurelian survived in office until April 400 (when he was exiled). Contemporary practice and the surviving evidence prove beyond any doubt that Aurelian actually entered the consulship on 1 January 400, and so he obviously cannot have been exiled before that date (pp. 161-75, esp. 164-8; this is absolutely fundamental and overrides any other evidence or argument). Therefore, even if one were to accept that Eutychianus was Typhos, the dates on the three extracts must be emended, since they all date to December 399 and Aurelian's successor cannot have taken office until April of 400. The next surviving law addressed to a PPO Orientis dates to 8 December 400 and is addressed to Caesarius (CTh. 1. 34. 1).

Every scholar and graduate student interested in the history of the Late Empire should buy or at least read this book, if not for its content, then for its method. No one will agree with everything C. has to say (witness Liebeschuetz) -- that is the nature of the discipline -- but I cannot help but think that C. has got closer to whatever recoverable truth there may be concerning these events than anyone else before him.


  • [1] The book is essentially Cameron's, with Long contributing small sections of Chapters Three and Four (16 pages), and most of Chapter Seven (on which, see below), and being responsible for the final version of the translation of Synesius' de prouidentia, which makes up Chapter Nine. Sherry was involved with Long and Cameron in producing the translation and its commentary (see p. ix).
  • [2] J. H. W. G. Liebeschuetz, Barbarians and Bishops: Army, Church, and State in the Age of Arcadius and Chrysostom (Oxford, 1990).
  • [3] While the text has its fair share of errors, the notes seem to have been compiled and checked over successive Bank Holiday weekends: they are chock-a-block with compositional and proof-reading errors of every kind imaginable (I gave up trying to count the appearances of 'see pp. 000-000'), and some notes do not even appear to have been completed. The only serious typographical error I found in C. appears on the back cover, where a politically correct gremlin has turned the poet 'Claudian' into 'Claudina'! Surprisingly, C. has slipped into that bureaucratic vulgarism of using non-reflexive reflexives ('The bulk of the narrative was written by myself...'; p. ix), something also present in the pages of Liebeschuetz '... the soldiers turned against the prefect [Rufinus], killing first his personal guard and then himself'; p. 92). Can this be stopped?
  • [4] 'kl' is just as frequent as, if not more so than, 'kal' in manuscripts; it just gets regularized out in printed editions.