The 28th book of Ammianus' Res Gestae adheres to the new organising principle the historian declared at 26.5.15: with the empire divided between Valentinian and Valens (a.d. 364), his narrative would no longer follow a basically fluid chronological progress, jumping from place to place, but instead advance in larger units according to geographical region. Thus, the six disproportionate chapters of Book 28 detail criminal proceedings in Rome under the nefarious official Maximus (28.1 – a hugely expansive account); Valentinian's building of fortifications in Gaul (28.2); the marvellous efforts of Theodosius the Elder in Britain (28.3); affairs at Rome again, this time merging with Ammianus' second, bitterly satirical digression on Roman life and manners (28.4); Valentinian's further campaigns against Alamanni and Saxons (28.5); and finally the complex tale of barbarian incursion and political corruption in Tripolitania, featuring the devious comes Romanus. In tandem with this compositional change is the less tangible shift in Ammianus' moral outlook after the death of Julian, who provides the imperial pattern for all other emperors in his history.
The volume is the fifteenth in the Dutch commentary series begun by P. de Jonge in 1935, and the ninth produced by de Jonge's successor team of den Boeft et al. since their 1987 commentary on Book 20 (the atelier has included J.W. Drijvers since 1995). Here, as in their previous volumes, they use de Jonge's lemmata-based procedure, supplemented with general and individual chapter introductions. There is a map and a bibliography, as well as a special chapter on chronology. For this team, ‘philological’ encompasses questions of linguistic usage, textual criticism and literary analysis. ‘Historical’ includes explication of realia, comment on legal and social matters, and speculation on chronology. They are acutely aware that Ammianus' distinctive historiography depends on a complex counterpoint of philological and historical elements. In the most recent volumes, the deftness and clarity of their scholarship has arguably exceeded even their own lofty standards (there are several notes in the present commentary which conscientiously correct mistakes in earlier ones). The speed of their production has certainly increased, with the last five volumes appearing at quite remarkable two-year intervals (the commentary on Book 29 was released at the end of 2013).
The structural and tonal difficulties of Ammianus' post-Julianic history are especially prominent in Book 28. The commentators' interpretative labour is further exacerbated by the especially poor state of the text. In addressing the various problems, den Boeft et al. engage deeply with previous scholarship, including some in their essay collection Ammianus After Julian: the Reign of Valentinian and Valens in Books 26–31 of the Res Gestae (2007). Their general opinion, echoing that of den Boeft in the collection (op. cit., pp. 293–311) is that the book does indeed suffer from disproportion and incoherence; that it presents the rule of Valentinian and Valens as falling far short of Julian's achievement, and that its lurid portrayal of the human maladministration of justice in the first and last chapters imbue it with a darker tone than any book heretofore. Nevertheless, they attribute to Ammianus a tenacity in work ethic and outlook that comes of his recognition, gained from a deep expertise in the exempla of the past, that Iustitia herself will always prevail and that Rome herself will go on (pp. xii–xiii). Space allows only a painfully selective survey of the different modes of elucidation den Boeft et al. employ.
Chronological clarity is a major casualty of the historian's revised narrative procedure, since datable events are often obscured in episodes that are telescoped or expanded. The initial chapter offers an overview of the problems dealt with specifically in the notes, and concludes with a provisional chronology beginning in 350 and running all the way to 379 (most of the book's narrative events fall between 363 and 367). The authors duly warn that some of the findings offered can be no more than educated guesses. There will be quibbles, but the reader will be grateful for a serviceable chronology.
Ammianus draws attention to the incongruity of the first chapter on Roman affairs by opening the fourth with an ‘amazing statement’ (p. ix): it has been a long time since he has dealt with Roman affairs. The authors do their best to explain the statement away as a compositional vestige of an earlier stage of writing, when the historian succumbed to his own ambivalence (expressed at 28.1.1–4) about inscribing such disasters. The statement would make sense if chapter 1 did not exist; however, Book 28 would be left disparately short by the omission (pp. ix–x). Yet in their note at 28.4.1, the commentators suggest that the statement could make sense if negotia urbana refers exclusively to the terms of urban prefects. To this problem, there indeed seems to be no ready solution.
Juridical episodes are illuminated by entries that draw heavily on the Codex Theodosianus. A particularly useful instance occurs at 28.1.17, where the Code is repeatedly invoked to explain the official terms of African corn distribution in Rome, and thus the actions of the unjustly persecuted proconsul Hymetius. This shades into a discussion of what exactly Ammianus means by horrea, the public granaries referred to in the same episode. Notes on the same dense chapter explain niceties of social rank, and also help negotiate perplexing distinctions in official nomenclature: ‘there is a difference between a vicarius urbis Romae and a vicarius praefecturae urbis, who both, rather confusingly could be called vicarius urbis’ (ad 28.1.5).
When the commentators turn their attention to Ammianus' late Latin, they help by turns to clarify and to reveal elusive shades of meaning. Very often, they do so with the various modern translations to hand, adjusting these most prevalent and persuasive, but far from infallible, filters of Ammianus' thought. A nice example falls at 28.6.29: Et ne quid coturni terribilis fabulae relinquerent intemptatum. Refusing to be carried away by the pathos of the Tripolitanian episode, den Boeft et al. posit an unspecified ‘they’ as the subject of reliquerent, rejecting the plurals coturni and fabulae as inadequate to the unity of the dramatic presentation, and pointing to Ammianus' fondness for the genitivus inversus for justification.
The above example intimates the commentators' sharp eye for literary device, which is especially keen in spotting Ammianus' allusions to earlier authors. Numerous notes convey the rich texture of the historian's literary fabric, and on multiple occasions offer ingenious solutions to textual knots. A single instance shows this acumen at work in several ways. At 28.1.33, the villainous Maximinus is described velut serpens vulnere ignoti cuiusdam attritus. Here, against Seyfarth's acceptance of Heraeus's ignoti, they defend the noti of the often nonsensical but fundamental Vaticanus manuscript by appeal to snake lore in Pliny, Nat. 8.86. They incline even more, however, to the substitution of rotae, which recalls an apposite snake image from Ammianus' beloved Virgil (Aen. 5.274). Here, as often, Seyfarth's conservative text provides ample opportunity for the authors to redeem V's text, to prefer previous emendations, or to offer their own with a compelling combination of boldness and caution.
It is difficult to offer criticisms other than picayune of a work that brings a standard and eminently illuminating series closer to its ending. It is noteworthy that these scholars, who have worked so intimately with a sometimes baffling text, tend towards an ultimately positive reading of Ammianus' pessimism: meaning is salvageable amidst gloom and corruption. In their efforts to reconstruct Ammianus' lost whole the authors give comparatively little attention to the most speculative of revisionist theories like that of C. Kelly in their After Julian collection (op. cit., pp. 271–92), who alleges that many of the distortions and inconcinnities in the book are designed to represent a less coherent world.
