Bryn Mawr Classical Review 1999.02.03
David Kennedy (ed.), The Roman Army in the East. Journal of Roman Archaeology Supplementary Series no. 18. Ann Arbor, MI: 1996. Pp. 320. ISBN 1063-4304. $89.50.
Reviewed by A.D. Lee, University of Wales, Lampeter
Word count: 3117 words
D. Braund, E. Dabrowa, J. Eadie, P. Freeman, S. Gregory, B. Isaac, D. Kennedy, N. Pollard, D. Potter, A. Rushworth, E. Wheeler, C. Whittaker, R. Ziegler
[The reviewer apologises for the delay in the appearance of this review.]
In the last fifteen years or so, there has been a dramatic upsurge of interest in the study of the eastern provinces of the Roman empire, particularly with reference to the frontiers in those regions and the role of the army. This interest is reflected in the successive conferences on these themes at Swansea, Sheffield, Ankara, and Cracow1 and in the monographs of individual scholars such as Bowersock, Parker, Sartre, Millar, and Mitchell.2 However, the publication which has had the greatest impact has been Ben Isaac, The Limits of Empire: The Roman army in the East (Oxford: Clarendon Press, 1990, revised edition 1992), an impact explicitly acknowledged in the title of the volume under review here. Its editor, David Kennedy, has also been a key figure in work on the subject, as co-organiser of the Sheffield conference and co-editor of its proceedings, and as author of numerous studies in the field. As he explains in the preface, the title was chosen to reflect the stimulus to further study of the subject for which Isaac's book has been responsible, but he also makes clear that the assembled papers (which did not arise from a conference, but by invitation from the editor) are intended neither as reviews of Isaac (of which there is a helpful list at p.7 n.1 -- twenty-six in all), nor as 'a collection of eulogies' (hardly appropriate anyway when the volume includes a piece by Isaac himself!); rather, 'the objective was to provide a selection of substantial essays on the theme of [Isaac's] book or on major sub-themes, whether or not dealt with by Isaac' (pp.7-8).
Before turning to the individual contributions, a few remarks about the overall spread of papers. First, chronological parameters: unlike Isaac's study, which ranged from the 1st century AD through to the 6th century, the focus of the papers in the present volume is predominantly on the Principate, with just a few venturing into the 4th century, and only one making any reference to the 6th. As for geographical parameters, the majority of pieces deal with questions firmly rooted in the eastern context (as one would expect), but it is worth emphasising that a number range more broadly than the east, tackling issues of more general significance and interest. While this feature of the volume might appear at odds with its title and potentially jeopardises its cohesion, it is consistent with the character of Isaac's study, which used the east as a base from which to address larger debates about Roman imperialism and grand strategy (or lack thereof).
David Kennedy opens with an essay bearing the same title as the volume (pp.9-24), in which he presents a useful stocktaking of where matters now stand and offers suggestions for future research. In his view, a major benefit of Isaac's study has been the example it has set of deploying archaeological evidence to tackle a 'big question', namely imperial strategy. In terms of future directions, he particularly emphasises the need, from a methodological angle, to make greater use of surface survey work in the east, given the benefits that have flowed from the use of this technique in many parts of the Mediterranean in recent years, while he suggests military logistics and demographic trends as two substantive issues particularly deserving greater attention. The latter is, of course, a difficult subject to tackle with reference to any context in the ancient world, though Roger Bagnall and Bruce Frier have recently shown what can be done.3
While Isaac was preparing his study of eastern frontiers during the mid-1980s, Dick Whittaker was independently developing his ideas on frontiers in the west, and must have been pleasantly surprised to discover that, from their different starting-points, he and Isaac had reached broadly similar conclusions about the nature of Roman frontiers and imperial strategy. In his contribution to the present volume ('Where are the frontiers now?', pp.25-41), he assesses two recent attempts, by Everett Wheeler and Arther Ferrill, to re-state the case for the existence of a Roman 'grand strategy'. His first point in response (pp.28-33) is that 'while there was some strategic thinking, it is difficult to detect any Grand Strategy in the sense of an integrated effort towards a political end' (p.31); his second is that the empire was always looking to expand, whether by direct or indirect means (as opposed to Luttwak's notion of a shift towards defensive frontiers), but that this was a matter of ideological imperative rather than 'grand strategy'. The piece is wide-ranging and stimulating, while also managing to avoid the polemical tone which marred Wheeler's article. Some of his assertions are open to question: his remark that 'in the East ... cities ... had nothing to do with defence as such, whether in the early or the late empire' (p.33) overlooks the role of Nisibis and adjacent cities in impeding Persian advances in the first half of the 4th century, while his assertion that 'the Res Gestae does not note a single new province that Augustus added to the Roman empire' (p.35) ignores the case of Egypt (27.1) -- though this was, I suppose, at the very outset of his reign. More seriously, however, he leaves undeveloped an important point towards which he seems to be moving in the third paragraph of p.33, that even if 'grand strategy' has no explanatory mileage, one can think in terms of regional frontier planning -- a point effectively acknowledged by another critic of Luttwak, John Mann, with reference to troop dispositions in the west under Stilicho at the end of the 4th century (see, e.g., JRS 69 (1979), p.182).
David Braund's piece ('River frontiers in the environmental psychology of the Roman world', pp.43-7), the shortest in the volume, aims essentially to make one main point -- but it is a thought-provoking one, and perhaps all the more effective because of its relative brevity. As he observes, it has increasingly become accepted in modern scholarship on Roman frontiers that rivers did not make good frontiers in strategic terms. Why then did so many Roman frontiers nevertheless run along rivers? He suggests we need a better appreciation of Roman attitudes to rivers -- that they were seen as religiously-charged entities which were dangerous and needed appeasing. 'From a Roman perspective, rivers were indeed natural boundaries in a sense that includes their religiosity, their natural power and their tendency to divide and to bound' (p.47). (N. Purcell cites further epigraphic evidence on the taming of rivers in G. Shipley & J. Salmon (eds.), Human Landscapes in Classical Antiquity (London: Routledge, 1996), p.199.)
David Potter's paper ('Emperors, their borders and their neighbours: the scope of imperial mandata', pp.49-66) presents a detailed and carefully constructed discussion of imperial instructions to governors during the Principate with a view to showing that it is possible to observe mechanisms for making foreign policy in operation and that Isaac's claim that the empire was consistently aggressive oversimplifies matters. This is particularly welcome here because it is one of the few papers in the volume which takes serious issue with some of Isaac's conclusions -- and does so with considerable effectiveness. It also makes interesting reading alongside Whittaker's contribution, since at a number of points the two offer almost diametrically opposed interpretations of the same material (e.g., on Stephen Dyson's 1985 study (pp.31, 50), on the Tabula Siarensis (pp.35, 51), on imperial policy in the mid-2nd century (pp.30, 53), on the imagery of the empire surrounded by walls (pp.36-7, 54)). This all seems to be fortuitous -- there is nothing to suggest the contributors had the opportunity to read one another's papers -- but whatever the case, it is a positive feature of the volume.
From this point onwards, the papers focus more explicitly on the east (with the exception of the concluding piece, which is deliberately comparative). David Kennedy's essay, 'Parthia and Rome: eastern perspectives' (pp.67-90), attempts to escape from 'Rome-centric' approaches to Roman-Parthian relations and explores the possibilities for achieving a Parthian perspective on relations with the Roman empire. As he acknowledges, the limited character of indigenous Parthian sources presents serious problems here, but he suggests some more indirect approaches. He offers the thought-provoking observation that it was unusual, historically, for Greater Syria (defined as the land between the Mediterranean and the Zagros Mountains) to be divided between two major states, as was the case with Rome and Parthia; it was more common for one state to control the whole area (p.73). He then examines Roman-Parthian dealings, especially in the 1st century BC, and argues that both states sought at various points to control all of Greater Syria. Roman policy in the 2nd century AD offers ready support for this; more novel is his interpretation of Parthian policy in the mid-1st century BC along the same lines (pp.77ff.). Implicit in all this is some qualification of Isaac's emphasis on Roman aggression and Parthian non-aggression. The discussion includes comparison of Roman and Parthian military resources and organisation with a view to understanding Rome's increasing success in the 2nd century AD. In the end, I was not sure how successful Kennedy had been in escaping a Rome-centric perspective, but this paper is certainly a valuable overview of Roman-Parthian relations with some novel ingredients. (I noted one typographical error in this piece: the reference in the third paragraph on p.72 to fig.2 should be to the inset map on fig.1.)
Philip Freeman's contribution ('The annexation of Arabia and imperial grand strategy', pp.91-118) presents a very detailed analysis of the Roman occupation of the Nabataean kingdom in 106 and the ensuing creation of the province of Arabia. Throughout his discussion, he is concerned to emphasise the role of contingent factors, both in the 'decision' to occupy and in the establishment of provincial structures. With regard to the former, he suggests a greater role than has usually been allowed for initiative on the part of the governor of Syria -- a suggestion which sets up an interesting point of debate with Potter's emphasis earlier in the volume (pp.56-60) on the parameters within which their mandata allowed them to act -- while on the latter aspect, he argues that the real organisation of the new province may not have got underway until a decade after the initial military intervention. Freeman prefaces all this by outlining a number of ways in which he regards Isaac's approach as unsatisfactory, arguing that the latter's emphasis on Roman expansionism does not make sufficient allowance for the role of contingent circumstances. This criticism seems to me misdirected, not taking sufficient account of the distinction between ideology and policy (a point stressed in Whittaker's essay): Isaac emphasises (perhaps over-emphasises) Roman aggression -- the ideological aspect -- but also stresses the lack of coherent planning -- the absence of grand strategy or the like. (And indeed at the end of his piece, Freeman ends up acknowledging that his conclusions are in substantial agreement with Isaac's views.)
In his essay on 'Civic coins and imperial campaigns' (pp.119-34), Ruprecht Ziegler analyses the minting activity of communities in Asia Minor during the first two centuries AD as a guide to troop movements. I am a little sceptical about his claim that numismatics has been unduly neglected in considerations of this subject,4 but his paper illustrates well what can be gained from giving attention to this category of evidence, with reference to issues such as whether Pliny's governorship of Bithynia-Pontus had preparations for Trajan's Parthian campaign as a major objective, and Domitian's plans in the early 90s.
John Eadie's piece ('One hundred years of rebellion: the eastern army in politics, AD 175-272', pp.135-51) aims to see what larger patterns can be detected in the involvement of eastern forces in the political instabilities of the period indicated. He is particularly interested in seeing how significant indigenous political activity was in the various upheavals of the period, and argues that this was more the case after 235 than before. This is not a particularly startling conclusion at the end of a paper a little top-heavy with narrative, but it does direct one's attention to a different perspective from which to think about the volume's theme.
Ben Isaac's contribution ('Eusebius and the geography of Roman provinces', pp.153-67) pursues the important issue of geographical knowledge through an analysis of Eusebius' Onomasticon. Because this work contains extensive reference to public roads, garrisons and city territories, Isaac argues that Eusebius must have derived his information from official archives in Caesarea, in which case the geographical character of the Onomasticon can be taken to reflect the state of regional geographical knowledge at the official level. If one accepts all this, then the prognosis is gloomy, since Eusebius shows no sign of being able to conceptualise geographical space in the way that moderns can. Much of this makes good sense of the evidence, but the chain of argument is not without its vulnerablities. Reliance on government archives at Caesarea is seen as accounting for the fact that Eusebius only has information about city territories in the province of Palestina; yet the fact that Eusebius includes details about public roads lying outside Palestina, in Syria and Arabia, is not seen as problematic in this respect. The question of Eusebius' purpose in writing the work, and his intended audience, is also barely addressed -- specifically, was it intended for practical use in locating the sites mentioned? If not, then, its weaknesses in geographical conceptualisation becomes less significant.
In her paper, 'Was there an eastern origin for the design of late Roman fortifications? Some problems for research on forts of Rome's eastern frontier' (pp.169-209), Shelagh Gregory provides a good overview of the state of knowledge of Roman military architecture in the east (with numerous maps and photographs). She begins by endorsing Isaac's argument that the traditional use of the term limes, 'was somewhat astray, especially for the later periods' (p.169) -- a view that has now been strongly challenged, specifically with reference to the late Roman period, by Constantin Zuckerman5 -- but her paper also provides a valuable critique of the work of early investigators such as Musil, Poidebard and Stein. Gregory adopts what she describes as a 'minimalist' approach to the evidence: 'sites are assumed not to be Roman unless there is fairly conclusive evidence for their being so' (p.176 n.3). This means that her conclusions are also limited -- the question in the title, e.g., is only answered in the most tentative manner -- but her paper is nevertheless useful for the salutary warnings it provides against over-interpreting archaeological evidence.
Nigel Pollard ('The Roman army as "total institution" in the Near East? Dura-Europos as a case study', pp.211-27) sets out to address the question of Romanisation and military-civilian relations in the east by studying Dura-Europos with reference to the sociologist Ervin Goffman's concept of a 'total institution' -- that is, 'place of a residence or work where a large number of like-minded individuals, cut off from the wider society for an appreciable period of time, together lead an enclosed, formally administered life' (p.211). If the Dura situation could be shown to conform closely to Goffman's model, then it would provide an example where the army's role as an agent of Romanisation was minimised. Pollard's discussion is clear and well-written, and shows the utility of the model he applies, but he is constantly hampered by the limited evidence on which he must base any conclusions. It is also interesting to compare his discussion with that of Richard Alston concerning the army in Roman Egypt,6 who concludes (on the basis of more plentiful evidence) that soldiers and veterans in the Fayum were more integrated with local society than has traditionally been assumed. Nevertheless, the paper deserves to provoke the discussion of wider issues of the Roman army and Romanisation which he hopes it will (p.226). (I was a little surprised that there was no reference in this paper to Ramsay MacMullen's seminal paper on 'the legion as a society'.7)
In the longest paper of the volume ('The laxity of Syrian legions', pp.229-76), Everett Wheeler provides an exhaustive discussion of the topos of the laxity of legions stationed in Syria, which he argues 'has much to do with traditional Roman morality, if not also ethnocentrism, and very little to do with the reality of Syrian legions' (p.237). After explaining the relevance of the issue to the wider question of frontiers, he investigates the Republican origins of the topos, and then methodically discusses the ancient evidence for each of the major periods -- Julio-Claudians, Flavians and Antonines, and the 3rd and 4th centuries. After such a comprehensive demolition of the notion, it will be interesting to see whether or not it has been laid to rest for good in future discussions of the eastern army.8
The penultimate paper, by Edward Dabrowa ('The commanders of Syrian legions, 1st-3rd c. AD', pp.277-96), mainly comprises a prosopography of the category indicated in the title, designed to update the last such listing, by Ritterling in 1925 (IV Scythica and X Fretensis are not included here because they have been dealt with recently elsewhere). Only brief discussion accompanies the prosopographical entries, but the latter should provide a useful basis for further interpretive work.
In the final contribution ('North African deserts and mountains: comparisons and insights', pp.297-316), Alan Rushworth aims to introduce a comparative perspective into the equation by comparing 'the activities and deployment of the Roman army on the empire's two desert frontiers -- North Africa and the East' (p.297). He begins by acknowledging the very significant differences in the political-strategic landscape of the two regions, and therefore limits his comparison to the Arabian-South Syrian desert and mountains and the African frontier west of Cyrenaica. Discussion focuses on three subjects: military roads, nomads and linear barriers, and policing of mountainous zones. Close similarities between Africa and the east are noted with respect to the first and third of these subjects, whereas the second highlights significant differences. The discussion of this second area is not as clearly set out as it might be, but overall, the inclusion of this comparative paper is a valuable addition to the volume and a good way to round it off.
In conclusion, the papers in this volume do avoid eulogising Isaac's study, but Potter's is the only one which seriously challenges some of his conclusions. There are many interesting points of disagreement between some of the papers, particularly in the earlier part of the volume, and most of the papers make for informative and stimulating reading. Above all, they demonstrate that there remains plenty of mileage for research in this area. The volume has been well produced and carefully proof-read.
1. S. Mitchell (ed.), Armies and Frontiers in Roman and Byzantine Anatolia (Oxford: British Archaeological Reports, 1983); P. Freeman & D. Kennedy (eds.), The Defence of the Roman and Byzantine East (Oxford: British Archaeological Reports, 1986); D. French & C.S. Lightfoot (eds.), The Eastern Frontier of the Roman Empire (Oxford: British Archaeological Reports, 1989); E. Dabrowa (ed.), The Roman and Byzantine Army in the East (Cracow, 1994).
2. G. Bowersock, Roman Arabia (Cambridge, MA: Harvard UP, 1983); S.T. Parker, Romans and Saracens: A History of the Arabian Frontier (Winona Lake, IN: American School of Oriental Research, 1986); M. Sartre, L'Orient romaine: Provinces et sociétés provinciales en Méditerranée orientale d'Auguste aux Severes (31 avant J.-C.-235 après J.-C.) (Paris: Seuil, 1991); F. Millar, The Roman Near East, 31 BC-AD 337 (Cambridge, MA: Harvard UP, 1993); S. Mitchell, Anatolia: Land, Men and Gods in Asia Minor (Oxford: Clarendon Press, 1993).
3. R. Bagnall and B. Frier, The Demography of Roman Egypt (Cambridge: Cambridge University Press, 1995).
4. See K. Harl, Civic Coinage and Civic Politics in the Roman East, AD 180-270 (Berkeley: University of California Press, 1987), p.19 for references to previous discussions.
5. 'Sur le dispositif frontalier en Arménie, le limes et son évolution, sous le Bas-Empire', Historia 47/1 (1998), pp.108-28 at 112-22.
6. Soldier and Society in Roman Egypt: A social history (London: Routledge, 1995), chap.7.
7. 'The legion as a society', Historia 33 (1984), pp.440-56, reprinted in his Changes in the Roman Empire (Princeton, NJ: Princeton University Press, 1990), pp.225-35.
8. Wheeler's paper is accompanied by a massive bibliography, but I did note the absence of one relevant item: S. Braund, 'Juvenal and the east: satire as an historical source' in French and Lightfoot, The Eastern Frontier of the Roman Empire, pp.45-52.