Bryn Mawr Classical Review

BMCR 2017.03.19 on the BMCR blog

Bryn Mawr Classical Review 2017.03.19

David J. Breeze, The Roman Army. Classical World Series.   London, New York:  Bloomsbury Academic, 2016.  Pp. xiv, 150.  ISBN 9781474227155.  $24.95 (pb).  

Reviewed by Duncan B. Campbell, Glasgow (


Summarizing the development of any institution over the course of a thousand years is a tall order, with only 150 pages of text available; taller still when that institution is as shadowy and poorly documented as the Roman army; but this is the task that Breeze has set himself. Consequently, the fact that he has not been entirely successful should not surprise us. However, seeing that the book in question is “designed specifically for students and teachers at late school and early university level,” we should at least expect clarity and authority.

The book shows signs of having been prepared in a hurry. Besides lapses of grammar and poorly constructed sentences (which may occasionally lead to the scratching of heads),1 key events and personalities are name-dropped without explanation.2 There is much unnecessary repetition, an unfortunate flaw in such a short text,3 while by using variations of the phrase “as we have seen”, Breeze implies that a topic has previously been discussed, which is not always the case.4 Add the blurring of distinction between fact and theory,5 along with the inclusion of spurious details that seem to be echoes of half-remembered facts,6 and the overall result is not encouraging.

The real question is: what will the book’s intended readership make of it? First of all, they may wonder why, in the “Introduction” (1-13), they are dropped straight into a summary of the army of Augustus, with its legions, its six different types of auxiliary regiments (“during the reign of Hadrian, there were 566 units”, we are told),7 its numeri and navy defending the frontiers, only to be abruptly flung back, in Chapter 1 (“The Republican Army”), to some obscure point in Rome’s past, under “one of the last kings, Servius Tullius”.8 They may then wonder why, after a dozen pages on the Republic, the next hundred pages (Chapters 2-7) seem to concentrate on themes from the Empire, with only the odd reference back to the army of Polybius, Pompey, or (mainly) Caesar (and, more confusingly, forwards to the time of Ammianus Marcellinus); and why the late Roman army of AD 235-400 merits only eight pages. Finally, they will surely be puzzled by the opening line of the “Conclusions” (131-2), that “the Roman army developed continuously throughout its long history”, as there is little evidence of this in the book.

On the contrary, the reader is reassured that “the basic structure of the [Republican] army and its procedures were maintained” (23) and, by the mid-first century AD, “continued within this framework for another hundred years and more” (37), until “a rather different army” (39) appeared in the 230s. The middle chapters of the book occasionally refer to fourth-century events (e.g. the Battle of Strasbourg and the Battle of Adrianople, pp. 67-9) without any hint of this “rather different army”, and the implication is that things continued much the same as before, despite the belated notice that “at the end of this period, a new army had been created” (123).

As an indicator of change, Breeze offers the rather trivial example of “the legions at Zama [who] shouted a war cry and banged their weapons on their shields”, contrasting this with the situation under the Empire, when “several sources remarked that the legions went into battle silently” (131). The reader must take this (along with everything else in the book) on trust, as no references are supplied (the eight pages of “Further Reading”, listing 20 ancient sources and 94 modern monographs, some out of print, can only serve to intimidate Breeze’s target readership), but it is worth noting the opinion of the writer Onasander, who recommended joining battle “with shouts and the clash of weapons”, in a work written in the first century AD (The General 29.1-2). Furthermore, observant readers may recall Breeze’s opening quote to Chapter 4 (“The Fighting Tactics of the Roman Army”), where the Hadrianic writer Arrian is made to recommend that, when the enemy comes within range, “everyone should utter a great war cry” (53)!

More fundamental changes could have been highlighted. Undoubtedly one of the most important occurred in the mid- fourth century BC, when (in the words of Livy, 8.8.3) “what was previously a phalanx, like that of the Macedonians, afterwards began to be a battle-line composed of maniples”. From then on, the maniple remained a key subdivision of the legion right up to the time of Ammianus Marcellinus in the fourth century, who still referred to “centuries and maniples” of soldiers (e.g. 26.2.3). So it is surprising that, here, we learn that the phalanx continued to exist, “[divided] into sub- divisions known as maniples” (16), and although “phalanx” is one of the thirteen terms listed in a short glossary at the end (133),9 readers will look in vain for a definition of the maniple, which is never mentioned again, apart from the comment that “three maniples might operate together as a cohort” (20).

Instead, the change that Breeze repeatedly emphasizes is the “fossilization” of the army (3, 37, 124, 131), as if the dynamic legions of the Republic had become mired and immobile along the frontiers. This curiously pessimistic view sits awkwardly with Breeze’s final assessment of the army as “a formidable fighting force, well trained, well disciplined, well armed, and used to winning battles” (132), and it is not immediately clear to the reader how this “fossilization” could have been addressed by creating a late Roman system of “defense-in-depth”, in which “troops remained on the frontier line …, but behind them some soldiers were placed in towns and fortified supply bases” (126).

All things considered, The Roman Army does not live up to the publisher’s promise of “an invaluable book for students at school and university level, as well as a handy guide for general readers”. Regrettably, an opportunity has been missed to engage a new generation of readers and help them negotiate the diverse and difficult sources for the Roman army.10


1.   A few examples: on p. 33, the wording implies that the imperial fleets at Ravenna and Misenum were disbanded in AD 68; on p. 58, it is misleading to represent the Parthian army’s ‘many camels laden with arrows’ (Plut., Crassus 25.1) as an “artillery train”; on p. 78, a misplaced bracket and an unnecessary dash confuse the intended link between Domitian’s restriction on soldiers’ savings and the rescheduling of the soldiers’ annual oath-taking (see also note 6, below); on p. 79, the word order creates the impression that soldiers trained with wooden helmets as well as wooden practice swords; on p. 92, readers will assume that the emperor Otho had to bribe centurions to grant soldiers leave; on p. 104, a misplaced subordinate clause implies that the legionary wore his sword on the right because it was short, whereas the intended meaning appears to be that it was its shortness that permitted it to be worn on the right; on p. 120, it is misleading to characterise a scarcity of stamped tiles as evidence that the practice “did not take off in Britain”.
2.   A few examples: “Domitius Corbulo”, mentioned on p. 43, is only introduced as one of Nero’s generals on p. 57; no context is given for “Julian’s army in Persia” (43), and readers are left to infer from p. 57 that he was attacking Ctesiphon; on p. 4, the identity of the “army” that crossed the frozen Rhine on the last day of 406 is not divulged (was it a Roman army?); it is not clear how “the state was prepared to break the rules” (54) by appointing the consular Claudius Pompeianus as a general.
3.   A few examples: “ox herds at the wood and pigs” (84) already mentioned as “the oxherds at the wood, Lucco in charge of the pigs” (81); “the ranks of the army contained many specialists …, including surveyors, architects, glaziers, and smiths” (120), already mentioned as “the army contained architects, surveyors, …, smiths, …, and glaziers” (118), and “each unit contained … surveyors, …, architects, …, glaziers, smiths, …” (96). The reign of Marcus Aurelius as “the last time a whole legion was moved” (124), already mentioned on p. 37 (see also note 6, below).
4.   For example, “we have already seen that it became more difficult to recruit soldiers” (128), referring back to the statement that “military service became enforced, with recruits provided by cities and landowners, or hereditary” (75), but no mention of any difficulty; “the [fourth-century] field armies were usually billeted in cities and this caused local friction, as we have already seen” (129), referring back to two third-century cases of soldiers offending against households with whom they had been billeted (95), but nothing from the fourth century.
5.   A few examples: the idea that the legion’s first cohort “increased in size from 480 to 800 men” (36) arranged in “five double-strength centuries” (1) is based on a theory proposed by Alfred von Domaszewski , Die Rangordnung der römischen Heeres (1908). The alleged “move of … the annual taking of the oath to the Roman state from January 1 to January 3” (78) is based on a theory proposed by Robert Fink, in R.O. Fink et al., “The Feriale Duranum”, Yale Classical Studies 7 (1940), 50-162. The idea that the Sasanians “used poisonous gases … in the mines” at Dura (72) is based on a theory proposed by Simon James, “Stratagems, Combat, and ‘Chemical Warfare’ in the Siege Mines of Dura-Europos”, American Journal of Archaeology 115 (2011), 69-101.
6.   A few examples: the treatise of ‘pseudo-Hyginus’ has not “variously been ascribed to the second or the fourth century” (6) within living memory; currently favoured dates span the century from Domitian to Marcus Aurelius. In Polybius’ description of the Roman army, it was not “the soldiers [who] chose their own centurions” (20), but the tribunes. At the Milvian Bridge, Constantine did not “place the Christian symbol on his flag” (56) but on his soldiers’ shields. Corbulo and Paetus did not invade Armenia in a “pincer movement” in AD 54 (57), and Paetus only took up his command in AD 61. The letter of Claudius Terentianus does not say that “he had failed to join a legion because of a poor reference” (76), but pours scorn on references, claiming that ‘nothing gets done here without money’ (P. Mich. 468). The supposed “move of the pay-day … from January 1 to January 3” (78) is mistaken, as several papyri show that the first stipendium of the year (there appear to have been three) continued to be paid on 1 January (e.g. ChLA 446; 473; 495). The reimbursement of 1458 denarii to the cavalryman Dionysius “sometime in the reign of Hadrian” (108) was actually 1459 denarii and is thought to date from the reign of Marcus Aurelius (P. Fay. 105). The “Lauriacensian auxiliaries” of ILS 774 (111) are actually the late Roman milites auxiliares Lauriacenses, a different class of troops from the earlier auxiliaries. The reign of Marcus Aurelius was not “the last time a whole legion was moved” (124), as the Second Italica Legion moved to a new fortress at Lauriacum around AD 200 (AE 1912, 293), and the Second Parthica Legion occupied a new base at Apamea in preparation for both Caracalla’s Parthian War and the Persian Wars of Severus Alexander and Gordian III (e.g. AE 1993, 1571-97), returning to Albano in between times. The comitatus was not the emperor’s “field army” (125), but rather his entourage (cf. AE 1949, 38).
7.   According to P. Holder, “Auxiliary deployment in the reign of Hadrian”, in J.J. Wilkes (ed.), Documenting the Roman Army (London 2003), 101-145, there are 563 units currently attested for this period.
8.   Readers are assisted by a “Select list of Roman emperors” (135), some of whom do not feature in the text (note that “Valeus” should be Valens, and Otho, mentioned on p. 92, is omitted), but there are no signposts for Roman Republican history.
9.   Curiously, three of the 13 words do not appear to have been used in the text; most of the others (apart from ‘decurion’) are explained as they arise, rendering the glossary unnecessary.
10.   A few editorial mistakes: on the map, Caparcotna appears as “Caporcotani” and Africa as “Atrica”, and Sicily is labelled “Belgrade” (xii); Figure 2 is not “the front and back” but the two inner leaves of a diptych (7); in Table 4, one of the arrows showing the chief centurion’s promotion prospects has gone adrift (99); counting from 30 BC to AD 14 makes 43 years, not 44 (26); “barrage train” (58); “secondy” (79); “Julius Apollinarius” (96) also appears as “Iulius Apollinaris” (108, 120); limitanei (125, 129) also appear incorrectly as limitanenses (132); and horses are surely “which” rather than “who” (80).

Read comments on this review or add a comment on the BMCR blog

Read Latest
About BMCR
Review for BMCR
Support BMCR

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010