BMCR 2003.09.44

Late Roman Warlords

, Late Roman warlords. Oxford classical monographs. Oxford: Oxford University Press, 2002. xiv, 347 pages : illustrations, map ; 23 cm.. ISBN 0199252440. $85.00.

Penny MacGeorge’s new volume on late Roman history after 455 is a revised version of a 1996 Oxford DPhil thesis, which offers a biographical treatment of the history of the last twenty-five or so years of the West, as told through the careers of Marcellinus, Aegidius, Syagrius, Ricimer, Gundobad, Orestes, and Odoacar. In large part M. has undertaken a reworking of J. M. O’Flynn’s Generalissimos of the Western Roman Empire of 1983, and although the focus is in many ways different the approach is the same. The problem is, biography is an inherently dangerous genre when one doesn’t, and indeed can’t, know much if anything about one’s subjects, a situation that M. herself surprisingly admits (p. 305 n. 24). But if this is the case, how can M., or indeed how can anyone, write a history of the end of the Roman Empire?

The sad fact is that most of what we think of as the history of the fifth-century Western empire, especially after 455, is conjecture. The ‘facts’, or more accurately the raw material that we rely on for reconstructing the past — literary sources, archaeology, coins, inscriptions, and the other physical remains of late Roman civilization — are so scarce that we historians are forced to hypothesize just to connect one isolated and often tiny fragment of information to another. By the time we start asking important questions like why, how, when, and what for, we have erected huge edifices of unverifiable conjecture. The necessity of having to provide so much sheer invention just to form a basic narrative encourages us to rely on the work of our predecessors rather than the sources themselves. After all, which of the two provides the greater detail? The sources end up a very distant third behind previous scholarly interpretation and our own interpretations, and they are often shunted into the notes, incomplete, misunderstood, cited from earlier scholars, or just plain ignored. And because our sources are so jejune, contradictory, and poorly understood, modern historiography can become little more than a huge Rorschach test: one can see pretty much what one wants to see.

M. is forced to play this game, but she refuses to be carried away by the tendencies of some of her predecessors (who are duly noted). However, she does tend to see her role more as a guide to the existing than as an advocate of the new, so the tralaticious tendency of the genre is much in evidence. But she knows her way around the period and remains in command of the material even when the going gets tortuous, as it frequently does. As a result, the book overall is solid and sensible, and it provides a good background for anyone who is trying to come to grips with the fifth-century, both as a historical period and a historiographical subject.

However, it does still read very much like a thesis and it has a thesis’s tendency to over-explanation and excessive detail: M. seems to feel that it is her primary duty to rehearse just about every theory for every action, motivation, or development of her protagonists. This is good and useful stuff in a thesis, but one needs to be more incisive in a book and M. often fails to make or clearly enunciate choices among the (many) competing theories.

In other ways, too, this book’s obvious thesis origin remains. M. can’t read Greek (!) and has had John Matthews, Michael Whitby, and Michael Sharp translate and interpret the Greek texts for her. As a result, in her notes she constantly thanks them (with full name and title each time) and refers to their advice. She quotes Oxford seminars as sources (e.g. pp. 31 n. 86, 42 n. 40, 44 n. 48, 162 n. 9, 163 nn. 15-17), even including one in the bibliography (p. 322). When listing sources she often just arbitrarily stops with ‘etc’ (the worst run is pp. 5-14, nn. 1, 17, 19, 21, 22, 25, 28, 34). There are many problems of detail with the bibliography and citations that aren’t worth reciting here, but they are the sort of errors that should have been caught and fixed before the viva.

The maps have been attacked by gremlins as well. Map one, ‘Dioceses and Provinces of the Late Roman Empire’, is a map of the prefectures and provinces, and a number of names are either incorrect or strangely Anglicized. On map two, many of the names of cities, places, and Germanic tribes are in French (the French of the source map?). I can only assume that it was an editor at the press who ensured that nearly every single date throughout the entire book, even in such expressions as ‘the 460s’, ‘the fourth century’, or ‘post-408’, comes with an attached ‘AD’. On p. 34, for instance, it appears nine times! Could any reader be confused enough to think that M. was talking about the fifth century BC?

Because of the nature of the work, a discussion of the content and M.’s interpretations would by necessity devolve into a detailed and boring catalogue of nitpicking, so instead I shall simply note the basic structure of the work and then discuss a few general points and how M. has handled them. The book begins with a too swift (seven-and-a-bit-page) run through earlier military leaders in the West, men like Merobaudes, Stilicho, and Aëtius. The shift from emperors like Constantius, Julian, and Valentinian I, who themselves took to the field, to those like Honorius and Valentinian III, who left their military affairs in the hands of these increasingly powerful Germanic military leaders, is of fundamental importance for understanding the militarily chaotic situation in the fifth century that is the subject of the book, but M.’s single analytical paragraph (p, 6) does not do justice to this development.

Each of the next three sections — covering Marcellinus; Aegidius and Syagrius; and Ricimer, Gundobad, Orestes, and Odoacar — starts with a chapter called ‘The Background and the Sources’, a title that clearly signals the split in each chapter, the first part offering some geographical background to what is to come (Dalmatia, northern Gaul, and Italy), the second half discussing the sources. It hardly needs to be said that the two don’t belong together. The sections on the sources are quite cursory (about four pages each); they do not say anything new or solve problems that arise in later chapters, This is a fundamental problem since this is where any historian of the fifth-century West should spend most of his or her time. Nevertheless, M.’s approach is still an advance over many other late Roman historians, who do not discuss the sources in this way at all.

The remaining chapters in each section follow the careers of the named subjects from emergence to death. The second section is different from the other two in that it deviates from the biographical approach and goes into great detail about the ‘Kingdom of Soissons’. As a result, it includes chapters on material evidence from northern Gaul, later medieval sources for the ‘kingdom’, the army of late Roman Gaul, and a summary chapter that presents alternative models, offering a middle course between the received tradition and the recent demolition of that view by Edward James. These are the best chapters in the book and clearly demonstrate that the biographical model is not the best one for analysing this period. There is a short concluding chapter and an appendix on naval power in the fifth century.

The odd thing about M.’s subjects is that they are such a disparate bunch. Aegidius was a lesser military commander who revolted against the emperor and briefly set up shop for himself in northern Gaul; Syagrius clearly was a warlord; Ricimer, Gundobad, and Orestes were the legitimate supreme commanders of the imperial army (such as it was); Marcellinus was an independent general who became a legitimate supreme commander; and Odoacar was a freebooter who joined the imperial army and was eventually proclaimed king. M. rightly expresses dissatisfaction with her title (p. v), but the problem is not so much with the title as it is with the association of such different individuals without an overarching purpose or theme.

When there is as little information on offer as there is for the end of the Roman empire in the West, it is absolutely vital to understand where our sources got their information, how reliable they are, and how they are related. The late Roman historian must work like an editor, treating his sources as manuscripts that need to be collated and analysed to determine their relationships and filiation. No one would attempt to edit a text without first constructing a stemma and understanding the interrelationships and character of his manuscripts. No late Roman historian should write a history without knowing the same information about his sources. Just as the editor runs the risk of producing a corrupt text, so the historian may produce corrupt history. And just as an editor should describe his manuscripts and explain his editorial methods, so a fifth-century historian should spend a considerable amount of time discussing and analysing his sources. M. tries, but she rarely engages directly with the sources and the various accounts they represent. This is the greatest failing of the book.

A good example of a difficult problem presented by the sources appears in the biography of Marcellinus, the least effective of M.’s chapters. The ‘textus receptus’, as it were, of Marcellinus’ career is implausible and convoluted. M. is sharp enough to realize the problems, but, instead of starting from the sources and building upon them, she tries to rework the modern interpretations and so is still held in thrall by them. As a result an alternative account of Marcellinus, starting from the sources, which appeared in the same year as this book, will instead be the starting point for his career from now on.1

A major problem for modern scholarship is whether certain Western emperors were recognized by Eastern emperors. Like many others, M. accepts that Majorian and Nepos were so recognized (pp. 199 and 273). The problem is that no one, M. included, pays sufficient attention to the relevant sources, which are unambiguous and unproblematic: superscriptions to Eastern laws, Eastern coinage, Western consulates proclaimed and disseminated in the East, and official Eastern documents, like the De caerimoniis. These show that after the death of Valentinian III only Anthemius was ever accepted as a legitimate emperor, that is as a full member of the imperial college, in the eyes of the Eastern emperor.

M. correctly rejects the modern ‘politically correct’ view that eschews any negative comments about the fifth-century West and talks about change and transition rather than decline and fall. For much of the fourth and fifth century this approach has been salutary and opened up important new ways of interpreting and understanding so many important aspects of the thought and material culture of Late Antiquity. But for the fifth to seventh centuries it gives us a completely distorted and overly positive view, and as M. notes in a number of places (esp. pp. 136, 167-72), the archaeological record shows that decline and fall is the only sensible and honest way to describe what happened to the West, and that is how she presents it. This supports the general thrust of her overall narrative that the military situation after 455 is another vivid example of the sorry state of an empire on the way to its final dissolution.

M. wisely rejects the view of Walter Goffart that barbarians were allotted revenues rather than actual land when settled in the West (e.g. pp. 161-2, 282 n. 58). This interpretation was once in vogue but is now on its last legs, and M. does not allow it to influence her depiction of the break-up of the empire and the loss of territory to Germanic settlement.

M. also raises the question of the lack of usurpers (p. 296). From 31 BC until 455 the emperor’s greatest worry was usurpation. Yet between 455 and 476, four full years out of twenty-two had no emperor, more than the entire rest of the history of the empire combined, and not only did no one step in to fill the gaps, but during these interregna the empire continued to function as usual. And within this period there were only three usurpers on record, Romanus in 470 (unsuccessful), Olybrius in 472 (successful), and Julius Nepos in 474 (successful), all against reigning emperors. Why? M. mentions the ‘radical change in political realities’ of the period and in particular explains that ‘the title of emperor was by now hardly worth usurping outside Italy and that the position of emperor was, to an intelligent man, no longer worth aiming for’ (p. 296). This shift in ‘political realities’ had occurred within an almost unbelievably short time, within thirty or forty years, and is another sign of the enormous and swift changes that affected the empire of the West in the fifth century. One wishes that M. had again set aside her biographical approach to try to explain why and how this happened.

Also worth mentioning is M.’s marvelous piece of detective work that shows how a square exagium (a weight for solidi) was transmogrified into a bronze dagger by modern tralaticious scholarship (p. 181 n. 20).

M.’s work is sensible and thorough, but because of its analysis of so much previous scholarship and its narrow biographical structure it really brought home to me the limits of traditional approaches to this period, articulated above. I think the history of the end of the Western empire calls out for a new historiographical paradigm, something probably structured in the first place around a sort of translation of and commentary on the sources, something that establishes a more solid base than what currently exists upon which we can build with more confidence.


1. Michael Kulikowski, ‘Marcellinus “of Dalmatia” and the Dissolution of the Fifth-Century Empire’, Byzantion 72 (2002), 177-91.