Massive city walls, military factories, horrea, forts, watchtowers and other fortified structures dominated the late Roman landscape. Much had changed since the first centuries of Empire. Accordingly, for a long time the scholarly world, influenced by the notion of decay, stressed that the late antique army was only a shadow of the glorious, powerful force that had once been capable of conquering new peoples. Poor discipline, difficulties in recruitment, barbarization, weakening in weaponry and armour characterized, it was believed, the late antique imperial army. However, in recent decades, as part of a general revaluation of Late Antiquity, a different set of interpretations has been proposed for the late Roman military.

The book under review here, edited by Sarantis and Christie, represents an ambitious effort in that direction. In fact, several of the contributors are those who have been the most active proponents of the new approach. Consequently, it is of interest to see their combined forces in a single volume — the first specifically focused on war and warfare in Late Antiquity. The aim is stated concisely by the editors: ‘This work examines the actual practice of war, rather than the military as an institution. It sets out to consider what determined the course and outcome of military conflicts in Late Antiquity, how these were witnessed and experienced by soldiers and civilians, and how warfare interacted with societies, economies, settlement patterns and wider political changes' (xvii).

The volumes are organized along lines that will be familiar to the readers of the series Late Antique Archaeology, with preliminary chapters which offer bibliographies and a status quaestionis. The first volume is made up of papers dealing with basic issues: literary sources and material evidence, military equipment and weaponry, tactics, strategy, diplomacy and frontiers, not to mention, of course, fortifications (101–370). Generally, the essays offered by Alexander Sarantis, Neil Christie and Conor Whately are clear and up-to-date, though some readers may feel that the complex organization of the bibliographical material means repetition. It would be pedantic to point out all the omissions, nonetheless it is unfortunate, for example, to find no mention of the important work of M. Gschwind, some in collaboration with H. Hasan, on the frontier segment covering the left bank of the Euphrates from Sura to Circesium. The bibliographical essays are preceded by a long paper on ‘Waging war in Late Antiquity’ (1–98), written by Sarantis, the author of more than 250 pages (approximately a quarter of the entire text of the two volumes), including the aforementioned paper, bibliographical essays — one of them jointly with Neil Christie — and another paper on the Balkans. S. is convinced that the Roman Empire was capable of adopting a grand strategy with ‘the prioritisation of different theatres of war and the distribution of military forces and diplomatic efforts within those areas according to their opponents, political goals and available resources' (8). Moreover, he endorses the view that ‘regional strategies were more positive and effective than is suggested by much of the secondary literature’ (8). S. notes that the widespread presence of fortifications, towns and forts represented the fundamental, positive element of a system capable of giving major strategic and tactical strength, and of providing the Empire with the intelligence, resources and manpower for aggressive campaigns (6).

A central, innovative point of S.'s papers is that he does not assume that fortified sites located along major axes of communication can only be explained as part of a defence-in-depth system. He argues that in several cases the fortified sites were also used as a point of support for the army in offensive campaigns, conducted into enemy territory. S. draws such general conclusions relying chiefly on his excellent paper discussing imperial strategy in the fifth- and sixth-century Balkans (759–808). In fact, the sources reveal some notable examples of imperial campaigns in enemy territory. This runs counter to the accepted view that the Roman army in the Balkans had been greatly weakened. Moreover, S. stresses the importance of diplomacy in dealing with neighbouring peoples. Finally, he observes that the fortifications scattered through the territory were not islands of Roman administration disconnected from rural areas. John Wilkes's paper (735–57) investigates the character of the south-western Balkans in Late Antiquity and the building up of fortifications, while Florin Curta's (809–50) is focused on the character of sixth- and seventh-century settlement in the Balkans.

James Crow (397–402) examines different kinds of fortifications in the East: the massive city walls of Antioch; the long walls of Thrace (a structure with small forts and towers, 58 km long, running from the Black Sea coast to the Sea of Marmara) built against the incursions of the Bulgars in the later fifth century; and the linear defences, with a total known length of 41 km, in eastern Bulgaria, possibly built in the fifth century. Christie (927–68) demonstrates how fortifications were the result of internal warfare rather than barbarian attacks. Moreover, he investigates what public monuments can tell us about civil wars.

Michael Whitby (433–59), relying on literary sources, provides a good overview of the tactics employed by the Romans and by their enemies in besieging cities. Hugh Elton (655–81) is particularly helpful in his overview of granaries, warehouses, army factories and stables, the elements that made the difference between a centrally-organized state and the barbarians. One point of disagreement might be noted here: Elton does not think that Diocletian carried out an important programme of fortifications in the East, concluding that ‘it is probably better to see construction of new fortifications and repair and upgrading of existing defences as ongoing tasks carried out by all administrations' (669). However, Diocletian had a particular impact in the East: finding the military apparatus weakened to an unacceptable degree, he introduced significant changes in Arabia and Palaestina, deploying three legions in new bases and strengthening a series of minor structures.

Michael Kulikowski (683–701) argues that the fifth-century Roman West was not submerged by wave upon wave of barbarian invasions. In fact, barbarian leaders commanded armies of first-, second- and third-generation provincials. Consequently, these events were not invasions, but civil wars fought in the interior of Roman provinces. Moreover, focusing on Spain, Kulikowski argues that it is often impossible to reconstruct a specific event by bringing together archaeological and literary documentation. Oriol Olesti et al. (703–31) relate the intriguing story of a macaque (monkey) buried with military objects, at some point in the fifth or sixth century, in the town of Iulia Libica. Moreover, they show that the Pyrenees were a nodal point, with many fortified segments designed to block the road to potential invaders.

James Howard-Johnston (853–91) offers a fascinating study of the territory north and south of the Armenian Taurus, which in the sixth century emerged as a zone of confrontation between Rome and Persia. The Romans deployed armies and built forts in the area. This again demonstrates that the Empire was still capable of non-defensive initiatives. John Haldon's paper (373–93) examines the ways in which Byzantines reacted to the Arab threat in the Middle Byzantine period (c. 660–1025), constructing forts and watchtowers in Anatolia, and organizing an effective logistical system. The importance of logistics is also stressed by Whately (893–924), who, in his discussion of the legionary fort of El-Lejjun in the sixth century, shows how imperial logistics varied from region to region, and also from generation to generation.

J. C. N. Coulston (463–92), Michel Kazanski (493–521) and John Conyard (523–67) rely on archaeology to show how the Roman and barbarian armies evolved, each influencing the other, with barbarian equipment ending up in the Roman army and vice versa. By contrast, Susannah Belcher (631–52), Ian Colvin (571–97) and Maria Kouroumali (969–99) discuss literary sources. Belcher shows how Christian writers offered a view on the loss of Nisibis to the Persians in 363 that is completely different to the one provided by Ammianus Marcellinus. Colvin argues that Procopius and Agathias drew much of their information on the events in Lazica from documents they had found in the imperial archives. Kouroumali, looking at Procopius, elucidates the attitude of Italians towards the Greeks and Goths who were fighting over their territory. The paper by Christopher Lillington-Martin (599–630) is particularly interesting because, thanks to a careful use of literary and topographical evidence, he reconstructs the features of two important sixth-century battles in Procopius (Dara in 530 and Rome in 537–538).

Regrettably, Sarantis' papers in the first volume have more misprints than are really acceptable. Italian language publications have been particularly victimized. In one of the bibliographies no title in Italian is without misspellings (188)! This is a great pity because the two volumes are of the greatest interest. They have managed to integrate various sources and disciplines — most importantly, revealing a late antique Empire still more than capable of looking after itself.