Bryn Mawr Classical Review 97.8.10

Adrian K. Goldsworthy, The Roman Army at War 100 BC-AD 200. Oxford: Clarendon Press, 1996. Pp. xiv + 311. $72.00. ISBN 0-19-815057-1.

Reviewed by Dr. Randall S. Howarth,

Goldsworthy challenges several assumptions about the way Roman armies were organized and deployed and the bases for Roman tactical success in the field. Chief among these are that legions were trained to exercise rigid battle plans and, although generally successful in set-piece battles and sieges, were ill-suited for other kinds of conflict. G. attacks the propositions that individual legionaries were merely automata in perfectly drilled military machines and that generals commanding the legions in battle exercised little influence over the outcome of battles. G. seeks to prove that Roman legions were actually quite flexible in their ability to adapt to variant conditions in the field and that Roman success was dictated in large measure both by individual actions on the field of battle and the visible leadership of Roman commanders.

According to G., the bases for the false assumptions are two-fold. First, scholars place undue emphasis on ancient sources such as Vegetius and Hyginus who focus on how Roman armies ought to be trained, equipped and deployed, and on the idealized portraits of legionary organization provided by Polybius and Josephus. Second, anachronistic notions of universal 'Principles of War' informing modern accounts of all warfare tend to gloss over the actions of actual participants in battles. Throughout the book, G. acknowledges his conceptual debts to a number of scholars' work, particularly John Keegan' s The Face of Battle (1976), The Mask of Command (1987), and The History of Warfare (1993).

Chapter one, 'The organization of the Roman Army,' surveys what we know about the organization of the Roman army. Unfortunately G. is forced to rely largely on the very same sources mentioned above, as well as the secondary materials based on them. However, G. balances this by pointing out the many exceptions to the rules. For example, there is plenty of evidence to indicate that legions were seldom at theoretical strength, partially because of normal attrition, but more importantly because it was customary practice to detach portions of legions for special duties, often for extended periods. In addition, the proportional representation of cavalry, archers, and artillery attached to infantry fluctuated wildly in identifiable units. These useful observations imply a far more flexible approach to tactics than is usually allowed.

Chapter two, 'The Opposition,' attempts to make the same survey of Rome's enemies, principally the Gauls, the Germans, and the Parthians. G. is of course hobbled by the same problem facing everyone else on this account, namely that there is scant information available, except that passed on by Roman sources. This is probably G.'s weakest chapter. Following Tacitus' Germania, G. confidently describes German institutions (42). But Tacitus' description is at the very best naive and ill-informed. Many would argue that the work is more a commentary on Rome's good old days than a genuine attempt to say anything about the Germans. In the same vein, G. equates the Latin civitas with German and Gallic tribal units (42, 54, 74). This is confusing and unnecessary, especially since G. has made no analogous attempt to correlate Roman political and military institutions. G.'s cursory description of the Parthians is largely an apology for Crassus' defeat by them in 53 B.C.

Chapter three, 'The Campaign,' surveys Roman campaigns throughout the period and makes several conclusions. First, in actual battles, individual units were committed and withdrawn according to the progress of the battle in a relatively fluid manner. Second, the only consistent tactical formula the Romans used was to attack at the earliest convenience, even against superior force, often after apparent tactical defeat. Perhaps most importantly, G. illustrates occasions on which the Romans deployed fast-reaction contingents to seek out, surprise, and destroy opponents employing guerrilla style tactics. All of this furthers G.'s thesis that Roman legions were inherently flexible and highly adaptable to different battlefield conditions and styles of warfare.

Chapters four through six, 'The General's Battle,' 'the Unit's Battle,' 'The Soldier's Battle,' develop the remainder of G.'s thesis, that is, that battles frequently turned on the actions of individuals or small groups of men. These are the best chapters of the book. G. examines many cases where minor breaches of the enemy's front line quickly turned an indecisive confrontation into a rout. The real question would then appear to be, what was there about the Romans that made them more likely to make that heroic plunge into the ranks of the enemy? G. implies that in fact this was the case, mostly for the expected reasons: a combination of fear, shame, coercion, and the boldness inspired and encouraged by good commanders. Most importantly however, G. focuses on unit morale. This is an important part of G.'s analysis, namely that the Romans projected a psychological advantage and that the psychological battle was in fact more important than the physical one. The battle is actually won by the projection of superior confidence and ésprit de corps. The decision of a force to turn and run indicates the dissolution of unit confidence and represents the actual moment of tactical victory. There are a few problems. There is no attempt to explain the significance of the earlier date in the book's title, 100 B.C., since there is only the most cursory references to campaigns up to the Social War (91-90 B.C.). There is little indication in the book of an appreciation for the historiographical problems associated with individual sources. The bibliography, mostly in English, has some glaring omissions, notably P. Brunt's Italian Manpower (1971). I was astonished by G.'s statement that "[t]he ease with which former enemies were assimilated into the Roman army as part of the auxilia was a highly distinctive feature of this period" (68). Of which period in Roman history does G. believe this was not a feature? The map on page eighty lacks several of the locations and tribes to which the accompanying text refers. Finally, G.'s frequent use of Latin terms would make a glossary useful to the undergraduate and non-specialist.

That said, the book does what is sets out to do, that is, it emphasizes the inherent flexibility of the Roman legion, and it describes Roman battles in terms of what actually happens on the ground rather than in terms of questionable theories of grand strategy and tactics.