BMCR 2008.06.42

Vegetius in Context: Establishing the Date of the Epitoma Rei Militaris. Historia-Einzelschrift, Band 194

, Vegetius in context : establishing the date of the Epitoma rei militaris. Historia. Einzelschriften ; Heft 194. Stuttgart: Franz Steiner Verlag, 2007. 205 pages ; 24 cm.. ISBN 9783515089890. €46.00.

Vegetius’ Epitoma rei militaris, the vade mecum of some Roman army historians and the most widely read work of Western military theory before Clausewitz’s Vom Kriege (1831), continues to fascinate and perplex: N.P. Milner’s translation is alleged to be the bestseller in the series Translated Texts for Historians, and the first ever OCT edition (with sentence numbering for the first time) has also recently appeared.1 This reform-minded tract, condemning Roman reliance on barbarian infantry and preaching a revival of the antiqua legio, has stimulated innumerable controversies on the reliability of Vegetius’ data about the Roman army and to which period(s) they apply. A reference (1.20.3) to Gratian as divus dates the text after 383, although a possible allusion (4.35.3) to a revised date for Easter would suggest c.388. In some manuscripts a subscription, dated by consuls to 450, records that an otherwise unknown Flavius Eutropius emended the text.2 The anonymous emperor, however, to whom Vegetius dedicated the initial book on recruitment and who requested a continuation of the work (hence Books 2-4), remains a mystery. Subscriptions in some manuscripts name a Theodosius (I? or II?) and others of still later date mention a Valentinianus (II? or III?). Theodosius I (379-395) and Valentinian III (425-455) have become the scholarly favorites, although advocates of Valentinian II (375-392) and Honorius (393-423) can also be found. Michael B. Charles (hereafter “C”) proposes to tilt the scale of debate in favor of Valentinian III in this revision of a 2003 University of Queensland dissertation.

For a reviewer, whose own study of this Latin text began in his undergraduate days (many moons ago), and who has tended to favor a late fourth-century date, C’s monograph aroused more than casual interest: the opportunity to be persuaded. Medievalist colleagues often are amused by Late Romanists’ fervor to debate the identity of Vegetius’ emperor: does it really matter? Indeed C (p. 149) raises this same question about debates on whether Books 26-31 of Ammianus date before or after 390. Without the subscriptions (of dubious authority) mentioning a Theodosius and a Valentinianus, would anything in the text itself pinpoint Theodosius I or Valentinian III? Explicit references, which one might expect, to neither Valens’ disaster at Adrianopole (378) nor Alaric’s capture of Rome (410) occur in the text as we have it. Vegetius the reformer clearly specifies problems to be fixed, but, except for his direct criticism that Roman infantry (allegedly) quit wearing armor under Gratian (1.20.2-5), purple-clad culprits to blame are unmentioned. Did Vegetius believe that the identity of his emperor was so obvious that it was superfluous to name him (cf. the anonymous emperors of the De rebus bellicis praef. 1), or at an early stage of the text’s transmission (Eutropius in 450?) has an editor universalized the text by omitting the emperor’s name? C (p. 37) speculates that the obscure Eutropius in 450 first united all four books as a single work. Much remains unknown and in the present state of evidence unknowable.

For C (pp. 13-16), the key to the Vegetian puzzle and a post-425 date for the treatise lies not in Vegetius the military theorist but in Vegetius the panegyrist: hence frequent comparisons of Vegetius’ prologues and epilogues to individual books with the works of Claudian, Merobaudes, and Sidonius Apollinaris, although others make similar claims about Vegetius and Pacatus’ panegyric of 389 to Theodosius I.3 Further, in C’s view Vegetius should be placed in his context as a source for the Late Empire, as another advocate of Valentinian III, C.D. Gordon, suggested 30 years ago but only summarily and militarily pursued. Yet the conceptually more varied anonymous treatise De rebus bellicis offers more fertile ground for plowing this type of furrow, and C does not fully mine the nuggets of information, conveniently summarized by the late M.R. Mezzabotta, in Vegetius’ veterinary treatise Mulomedicina. As an additional justification for the work, C’s cites (p. 13) the absence of a monograph on the date of the Epitoma (although articles abound), thus conveniently sidestepping unpublished dissertations by Silhanek (1972) and Milner (1991). Milner’s fruits were largely harvested in the “Introduction” and notes to his translation (above n. 1). Nevertheless, C frequently engages Milner’s published views and Silhanek’s arguments.4 The most that C hopes to salvage from the quagmire of conflicting views and scarce evidence on the Epitoma‘s date is a balance of probability for Valentinian III, as the case for Theodosius I is not inherently strong and unassailable.

The work is divided into six chapters. The first three survey past scholarship and expose inadequacies in the Theodosian views of François Paschoud, André Chastagnol, Timothy Barnes, Santo Mazzarino, Philippe Richardot, Milner, and others, besides the cases of Constantine Zuckerman for Valentinian II and Claudia Giuffrida Manmana for Honorius. Alfred R. Neumann’s RE article (Suppl. 10, 1965, 992-1020) is essentially ignored. As C’s topical approach focuses on the cases for individual emperors, the often piecemeal presentation of opponents’ views, scattered fragments lifted for refutation from the context of broader arguments, deprive a reader of the cogency of an individual scholar’s position. Despite C’s attempt to catalogue the previous scholarship (pp. 18-21), the overall view of who thought what and why is not always clear. There may be more than C allows to Zuckerman’s point (rejected as unconvincing: p. 96) about an emphasis on recruiting in the Balkans (1.28.2-4; cf. 2.11.5 on the Bessi) and silence on the manpower potential of other parts of the Empire. The second three chapters continue the assaults of the first three, but, building on the views of Otto Seeck, C.D. Gordon, Walter Goffart, and Eric Birley, attempt to consolidate the case for Valentinian III. C plays fairly and even the pro-Valentinian-III Goffart takes “his lumps” for dubious arguments (pp. 55-65).

C’s case for Valentinian III may be summarized as follows: Vegetius the panegyrist wrote in the West for a young, palace-bound emperor with no military experience, as comparison with other fifth-century panegyrical texts demonstrates; the polemic against barbarians is aimed at the increasing reliance on foederati, who provide the examples of Roman soldiers no longer wearing armor and building camps; and, finally, in C’s most original contribution, the inspiration for two useful articles on the Late Roman navy,5 the treatise’s final section on naval affairs (4.31-46) reflects the western Mediterranean situation about the time of the second treaty with the Vandals in 442.

C’s brutal honesty merits applause. Rarely does a reader encounter so many concessions about what cannot be proved or refuted. Such confessions include: Valentinian III as the addressee (pp. 14, 122-27, 180) cannot be proved: hence the emphasis on probability; not refutable are contentions that Vegetius might have written in the East (p. 49); Vegetius’ frequent combination of Goths, Alans, and Huns best fits the reign of Theodosius I (pp. 70-71); and (against Milner) the implied recruitment of coloni at 1.7.8-9 must date before 400 (pp. 139-43). Confessions and concessions, however, do not render C’s labors futile. The scanty evidence encourages debate and the case for any emperor as the addressee involves spinning the sources to suit a thesis. In particular, many of C’s bombardments of the case for Theodosius I have not missed their target. Barnes’ hypothesis, for example, of a split in the manuscript’s transmission already by 450 cannot be proved and is rendered unlikely by Reeve’s new analysis of the manuscript tradition (C pp. 32-33, 37). In contrast, the attack (pp. 75-85) on Chastagnol’s influential study, suggesting that the author of the Historia Augusta used Vegetius, not only exposes how subjective this type of argument can be, but also lowers (with some hypercriticism) historical discourse to the level of rhetorical debate. Here the result is largely inconclusive: some debater’s points scored (e.g., military terminology in the HA need not come from Vegetius), while other bullets are close to but not on the bull’s eye (e.g. Epit. 3.3 on military supply; cf. HA, Trig.Tyr. 18.4-9). Chastagnol’s comparison of Epit. 2.23-24 on military training with HA, Avid.Cass. 6.3-4 is not addressed.

Troublesome, too, is C’s resort to argumenta e silentio with implied conclusions (Vegetius does not mention X, therefore . . .): e.g., reference to a single emperor and no mention of Theodosius I’s sons (pp. 18, 120, 182), exclusive use of Latin sources in both the Epitoma and the Mulomedicina (pp. 41, 45-46), no reference to Theodosius I on the revised date of Easter (p. 17) or an emphasis on Theodosius’ victories (pp. 109-11; but cf. Epit. 2 praef. 1). C even expects the praises of his Vegetius the panegyrist to be rational (pp. 145-46). Nevertheless, the real issue is whether C’s attempt to put Vegetius in context has produced a better understanding of the man and his treatise. Approaching Vegetius as a panegyrist, although novel, downplays him as a military theorist and pays too little heed to the genre in which he wrote. Some problems merit discussion.

Who was Vegetius? C scarcely advances the issue beyond Milner’s discussion, conceding only that Publius Vegetius Renatus was an important bureaucrat (hence his Late Roman title “Flavius” in some MSS) a wealthy landowner in the West, a Christian, and also the author of the Mulomedicina (disputed by some). Manuscript subscriptions alleging the titles of vir illustris and comes (hence a conjecture of the post comes sacarum largitionum) are treated with skepticism; the attractive suggestion of comes stabuli/sacri stabuli is not mentioned,6 although the veterinary treatise reveals that Vegetius was well-traveled, a horse-keeper himself, and quite knowledgeable of equine breeds throughout the Empire (e.g. Mul. 3.6.4: properly distinguishing the Armenian breeds of Armenia [scil. Maior] from those of neighboring Sophene). Indeed Vegetius’ extensive knowledge of Hunnic horses and Roman imitation of Hunnic practices surely places the Mulomedicina in the fifth century, when Huns were better known, rather than during their initial appearance west of the Tanais River c.370. C rightly sees the veterinary treatise as a later work, although this view does not preclude the Epitoma‘s date in the late fourth century. Similarly, C doubts the prosopographical probability of Vegetius’ origo in Spain or Gallia Narbonensis, where Vegetii have frequency, whence Silhanek’s suggestion of Vegetius’ attachment to a Spanish retinue of Theodosius I. After all, this is the way that things often worked.

Vegetius the Christian receives only cursory examination in efforts to refute a Theodosian date. His cognomen Renatus, possibly with Christian connotations (“reborn”), is not discussed, although John Lydus ( Mag. 1.47) identifies him by this cognomen. A Theodosian context of the revised date for Easter ( Epit. 4.35.3) is discounted because Vegetius does not explicitly refer to Theodosius I (pp. 16-17). But why add this somewhat superfluous reference to a discussion on the proper time to cut timber for ships, if it did not appeal to a specific emperor’s ego? Similarly, C weakly hints (pp. 27-28) that the christianized version of the sacramentum militare ( Epit. 2.5.3-5) belongs to the fifth century, but erroneously claims that Constantine I made Christianity the official religion of the Empire. Throughout the fourth century the army remained remarkably neutral in Christian-pagan conflicts and even Constantine’s compulsory prayer for soldiers (Euseb. ( VC 4.19-20) was vaguely monotheistic and not overtly Christian. A turn to christianization in all things military really begins in the Theodosian era. So why not the sacramentum as well? Vegetius’ Christianity, however, is rather curious. Apart from a passing reference to Deus (1 praef. 1) and the sacramentum, Vegetius is not very interested in religious aspects of the army, although he certainly expects that Roman soldiers will be Christians. At 4.35.2 Easter is alluded to, not explicitly mentioned, and at 3.5.4 Deus nobiscum is one battlecry added to a pagan list. Vegetius’ reforms retain the legionary eagle and the other signa militaria, which had definite pagan connections, and about which he states (3.8.15): nihil est venerabilius eorum maiestate militibus — not the statement of a zealous believer, from whom one might expect more Christian references than found in the Epitoma.7

Majority opinion puts Vegetius in the West. If Rome the city is mentioned five times and Constantinople not at all (p. 39), it is fair to point out that Ravenna (the seat of the western emperor from Honorius on) also does not appear as an imperial residence in the Epitoma. Nevertheless, except for alleged western use of Vegetius in the HA and by Claudian (rejected by C: pp. 118-19), the earliest testimonies to the text occur at Constantinople: the subscription of Eutropius (450) and about a century later references by Priscian ( Inst. 3.21) and John Lydus ( Mag. 1.47). Only Silhanek has argued extensively for Vegetius’ location in the East.

C’s western obsession, however, leads to some rather curious arguments. As C realizes, Vegetius’ composition in Latin is not evidence that he wrote in the West: non-native speakers, as he argues, could be trained to write clear, learned Latin. Lactantius under Diocletian and Priscian in the time of Anastasius and Justinian — both with jobs in the East for this very purpose — would offer examples. Ammianus’ could also have been cited as a Greek-speaker’s learned Latin despite his tortuous, rhetorical and not always clear style. Latin remained the language of government at the highest level in both East and West through the fifth century. A Gaul like the epitomator Eutropius could spend the bulk of his career in the East and still compose his Breviarium in Latin. Should the situation be different with Vegetius? In neither the Epitoma nor the Mulomedicina did Vegetius use (so far as we know) Greek sources directly, although he was well aware of Greek sources for both these subjects. Greek remained the dominant language of military treatises, but Latin seems to have become the chief tongue for work in veterinary science. Yet C (pp. 37, 46, 49 with n. 171) too eagerly follows Goffart’s “western” arguments, especially in claiming the Mulomedicina‘s composition in Latin made it useless in the East; thus an excuse to eliminate Theodosius II as Vegetius’ emperor. C misses the opportunity to have a Theodosius as the addressee and to eat the cake of a fifth-century date, too. Nor is this the only occasion when C’s western and Valentinian III obsessions miss a possibly valid argument.

Other anti-East/pro-West arguments are likewise curious. C (pp. 39-40, 46) takes reference to the naval bases at Ravenna and Misenum (4.31.4-6) and their central position for controlling the Mediterranean as proof of Vegetius’ “Italocentric” perspective, but this geo-strategic view is Augustan (cf. Strabo 6.4.1) and Vegetius’ comments on the Italian fleets need be no more than his own (or his source’s) elaboration on Suet. Aug. 49.1. All of Vegetius’ references to Rome and Italy (to this reader) seem antiquarian, not contemporary — an appeal to the traditions of “old Rome” befitting Vegetius’ advocacy of the antiqua legio. Similarly, 1.17 on the two legions of mattiobarbuli in Illyricum, whose combat success with plumbatae (weighted arrows used as mini-javelins) led Diocletian and Maximianus to rename them Ioviani and Herculiani. For C (pp. 46-47, 135; argued in detail elsewhere) these legions must be the two palatine legions of seniores at Not. Dig., Occ. 7.2-4 stationed in Italy, active in the Gildonic War, but no longer (pace C) existing in Vegetius’ day, rather than the iuniores of this name at Not.Dig., Or. 5.3-4, 43-44. Hence in C’s view Theodosius I and Honorius can be eliminated as Vegetius’ emperor. Regardless of whether C’s arguments for preferring the western to the eastern legions are correct, one wonders if C has missed Vegetius’ point at 1.17. Survival of these units of Iovani and Herculiani is irrelevant. The chapter concerns training recruits to use plumbatae : two notable legions once employed this weapon, but, Vegetius implies, legions no longer do. C is aware that use of plumbatae continued. In the Ps.-Maurice ( Strat. 12B.2, 4, 5, 12, 18, 20.10; cf. 12B.16.45 Dennis), however, plumbatae belong to light infantry and not heavy infantry — the situation Vegetius wants to correct. A western orientation and a post-Theodosian date based on 1.17 seem dubious, nor does C’s attempt (elsewhere) to blur the distinction between heavy and light infantry (i.e., legionaries vs. auxilia) in the Flavian era affect which type of Late Roman infantry used plumbatae.8

For C, Vegetius’ treatment of barbarians and Sasanid Persians also points to the West. As the partition of Armenia in 387 (not 386, as C p. 168) calmed eastern tensions, C posits Vegetius’ lack of interest in Persians (pp. 49-50, 171-72). Indeed Vegetius’ Latin version of a Greek siege manual (4.1-30), in which he speaks of destroying the enemies’ civitates rather than urbes, must indicate (for C) smaller German or Goth settlements, not big eastern cities. Yet C’s concession of Vegetius’ interchangeable use of civitates and urbes ( urbs occurs about ten times in 4.1-30) nullifies the argument and C forgets his earlier rejection (p. 58) of Goffart’s failed attempt to distinguish by size civitates from urbes in discussing Vegetius’ praise of his emperor as a founder of cities. C must also minimize the Persian war of 421-22 (“Pulcheria’s crusade”). Further, C also ignores Vegetius’ note on Persian imitation of Roman fortified camps (3.10.15) and (except for a passing reference: p. 167 n. 59) the pointers to eastern warfare at 3.23-24 on camels, cataphracts, scythed chariots, and elephants. Persians were famous for cataphracts and the Sasanids revived the use of elephants, as C discusses in detail elsewhere.9 True to his antiquarian disposition, Vegetius generally did not insert contemporary references into his discussion of these themes common to many Tactica, except for notes on the employment of camels by the obscure North African Urcilliani (3.23.1) and his novel suggestion (3.24.14-16) to use carroballistae against elephants — not a likely tactic for German/Goth conflicts.

Further, C envisions (pp. 49, 171, 184) Vegetius describing a Western Empire besieged by barbarians roaming at will and a war currently in progress, which apart from the early years of his reign would not fit Theodosius I. The anonymous De rebus bellicis (6.1), however, viewed an empire beset on all sides by howling barbarians. A more generalized Roman-barbarian conflict rather than a specific war could equally be posited. For the East C seems not to consider the Hun invasion through the Caucasus, which terrorized the East 395-398 and set defense of the Caucasus passes as a sore point of Sasanid-Roman relations for the next two centuries, nor the Hun invasion of Thrace 395-396, the Roman panic after Adrianople that occasioned the massacre of Gothic troops in Asia Minor, and the anti-barbarian tirade of Synesius’ De regno, which echoes (or anticipates?) Vegetius’ own views.

Yet Vegetius’ “East” need not be east of the Bosporos or south of the Taurus. Balkan references subtly recur: Balkan recruitment and the Bessi have already been mentioned. Pillei Pannonici occur at 1.20.18-19. A fascination with the Huns, greater in the Mulomedicina than the Epitoma, points to the middle or lower Danube, and Vegetius’ self-imposed censorship about current riverboat operations on the Danube (4.46.9) surely points to the lower Danube east of the Iron Gates, especially if C’s fifth-century date and particularly one c.440 for the Epitoma is followed. Not least, the two legions of mattiobarbuli (1.17) are in Illyricum. In fact, reference to these legions points to an unnamed Diocletianic source, which also reveals itself in Vegetius’ mentioning (2.8.5) the rank of triarius prior in a legion’s cohort I, the Diocletianic date of which is confirmed by the fascinating inscription of Aurelius Gaius, once an optio triarius, who spent part of his career with Danubian units. On present evidence, survival of this rank after the Tetrarchic period is unattested.10 The case for Vegetius possibly writing in the East is not as weak as generally supposed.

If (for C) Vegetius’ barbarians can only be western, they can also only be (following C.D. Gordon) foederati (pp. 129-30), although Scharf’s monograph on foederati escaped C’s notice. C subscribes to the views popular in some circles (e.g. Elton, Nicasie) that Germanization/barbarization of the army has been overemphasized and that the Roman army of the Late Empire was just as efficient and disciplined as its predecessor during the Principate. Hence the army’s lack of discipline, seen in no longer fortifying its marching camps (1.21.1-3) and the problematic criticism that Roman soldiers quit wearing armor in the time of Gratian (1.20.2-10) must post-date Theodosius I, whom C sees as an enforcer of discipline (Jord. Get. 139). Besides, Vegetius, if addressing Theodosius, would hardly be so impolitic as to criticize the emperor’s brother-in-law Gratian, whose half-sister Galla he married in 388 and of whom he was much enamored (Zos. 4.44.3-4; C. p. 39). Yet other scholars are more skeptical about this political marriage, and C (p. 164) seems unaware that Jordanes uses the topos of a successor correcting the lax discipline of his predecessor (cf. Amm. 31.14.2 on Valens’ maintenance of discipline). Zosimus’ view (2.15.1) that Constantine I began the barbarization of the army in 312 (emphasized by Liebeschuetz) is not mentioned.11 Although the army of the Empire had always been multi-ethnic, Constantine’s resort to enrolling Germans against Maxentius initiated a new phase in external recruitment and the need to replace quickly an army destroyed at Adrianople began another. C (pp. 155-64: largely relying on Grosse) assures us that fourth-century Romans continued to fortify their marching camps, although he concedes that not all camps in Ammianus need indicate fortified sites and the argument is padded with irrelevant references to fortification of passes and building bridges. Another scholar, however, notes occasional evidence of no or inadequate fortified encampments.12

Epit. 1.20.2-10 on Roman abandonment of body armor has always been one of Vegetius’ most enigmatic passages, not only for its direct criticism of Gratian, but also because Vegetius ties the lack of armor to Gothic victories and the destruction of many cities. The temptation to see allusions to Adrianople and Gothic depredations in the Balkans should be resisted, as Vegetius’ scenario of unarmored Romans becoming pincushions for Gothic arrows does not fit Ammianus’ account (the only detailed one extant) of Adrianople. Romans wore armor at that battle. Vegetius’ allusion to Adrianople (already recognized by Goffart and Milner) occurs at 3.11.7-9. For C (pp. 52-55) 1.20.2-5 refers to Alaric’s successes before 410.

Although many simply reject Vegetius 1.20.2-10 as incredible, C has investigated the use of body armor in the late fourth and early fifth centuries (summary at pp. 126-30 of an article elsewhere). Here and throughout his Vegetian studies, C mysteriously avoids use of the Ps.-Maurice, where Roman use of armor c.600 is clear. As with fortified camps, so with body armor, C’s solution is German foederati, who of course as infantry generally wore little or no body armor to begin with. However attractive this solution may be from one perspective, C does not provide evidence that Rome imposed use of armor or Roman tactics on the foederati. Nor is there a hint in 1.20.2-10 that the infantry are barbarians and not citizens, who at 1.20.4 ab imperatore postulant to be relieved of their breastplates and helmets — not a statement to be expected from barbarian foederati. In his article C hits upon a plausible solution to the problem of 1.20.2-10, but rejects it.13 C notes the similarity between 1.20.2-5 and Orosius 7.33.13-14: exposed infantry ( nuda) at Adrianople, when deserted by the cavalry on its flanks, fell victim to clouds of Gothic arrows. The situations described in the two authors are not identical, as C realizes. But if Vegetius used Orosius and either carelessly or intentionally misinterpreted him — panegyrists and military theorists, after all, are not held to strict standards of historical accuracy — C would have further support for his desired fifth-century date. In any case, Vegetius’ misuse or misunderstanding of his source at 1.20 would seem to solve the problem of this passage better than a resort to foederati.

Problematic also are aspects of C’s Vegetius the panegyrist. C claims to have demonstrated (pp. 21, 87-89, 102-105, 114, 118-19) that Vegetius’ emperor is a young, isolated ruler with no personal military experience. On C’s reading of the text (pp. 102, 145), Vegetius’ condescension to the emperor can even be detected (e.g., 4 praef. 4-5), although this reviewer does not discern this subtlety in Vegetius’ comments. The addressee’s youth, first argued by Seeck in 1876 and popular among advocates of Valentinian II and III and Honorius, is derived from Vegetius’ praise of the emperor’s foot-speed and skills in equitation, archery, and the armatura (3.26.35-38), traditional aspects of military training for young males and unlikely to be praised in an emperor like Theodosius I, who assumed the purple when already beyond age 40. Claudian, however, would seem at least partially to compromise Seeck’s argument: at Cons. Stil. 1.64-68 (C p. 105 n. 80) the Persians praise Stilicho (hardly still a teenager when sent in 384 to negotiate with the Sasanids about Armenia) for his equestrian skill and archery. Moreover, unknown to C, Ammianus’ obituary of Constantius II (21.16.7) includes praise of that emperor’s skills in equitation, javelin-throwing, archery, and the armatura without restricting such skills to his youth. As fleetness of foot was a standard Homeric epithet of Achilles, one wonders if this trait should be taken so literally in a panegyrical passage.

Further, C’s hypothesis of an isolated, militarily ignorant emperor (repeatedly asserted) is never proved. The absence in Vegetius of references to an emperor’s bonding techniques with the rank-and-file soldiers (e.g. marching on foot rather than mounted, keeping watches, etc.) indicate to C that Vegetius’ emperor has no field experience — another argument from silence: Vegetius would have given examples if any existed. Comparisons with Merobaudes and even Pliny’s panegyric to Trajan are adduced. Indeed, the absence of such bonding techniques in Vegetius and their presence in Claudian becomes a means to reject Richardot’s argument that Theodosius’ advice to Honorius at IV. Cons. Hon. 320-51 derives from Vegetius (pp. 118-19 with n. 50). C does not recognize that such bonding techniques are topoi, traceable in historical and military texts from Xenophon’s Agesilaus to the HA. Vegetius’ exclusion of these topoi, proves nothing. Nor can much faith be put in C’s arguments (pp. 111-12) about the simplicity or obviousness of much of Vegetius’ advice, if given to an experienced general. Such criticisms underlie a basic misunderstanding of the genre of military theory (ancient or modern). To paraphrase Clausewitz, everything in war may seem simple, but in war even the simplest thing is difficult. Observance of basic principles in the field under the pressures of real circumstances differ significantly from the hindsight, security, and presumed omniscience of the armchair strategist and the desktop historian. Polyaenus, as C realizes, faced much the same situation as Vegetius in offering his much more rhetorical and less practical Strategika (not Strategemata, as pp. 111, 121) to Marcus Aurelius and Lucius Verus in the early years of the Parthian war of 161-166.

C’s diligent pursuit of contemporary references yields both successes and red herrings, some of which have already been mentioned. A possible reference to a British fleet (4.37.3), much discussed among students of Roman Britain, has no real chronological value (pp. 47-48), as another scholar had already noted.14 But C’s attempt to pin down an exact date for the end of gladiatorial combats is a futile exercise: use of gladiatorial techniques (1.11.3-4) for training would seem to derive from Vegetius’ source (Cato? Frontinus?) and (to this reviewer) was never meant to be a contemporary reference. Likewise the attempt (p. 113) to wring the drop of a contemporary reference from the dry 3.25.13 (a lost battle does not mean a lost war; cf. Onas. 36.3-5; Front. Strat. 2.10).

C’s lack of attention to Vegetius’ sources and the genre of military theory perhaps constitutes the work’s greatest flaw. Although C follows (p. 26) the general opinion that Vegetius was not a vir militaris, this author’s negative reputation for lacking military expertise stems largely from scholars seeking something he was not: an historian of the Roman army. Vegetius was up-to-date on contemporary cavalry (1.20.2, 3.26.34), for which he proposed no reforms, and avoided having his loose lips sink ships on the Danube (4.46.9). His knowledge of contemporary military “slang,” some of it Celtic or German (e.g., barritus, drungus, globus,), may not prove his locus in either West or East (cf. C p. 49), but like Arrian’s use of Celtic terms (e.g. Tact. 37.4, 42.4, 43.2) it displays a knowledge of “insider” details. Nor does Vegetius lack originality (cf. Lenoir’s useful paper unknown to C): his careful application of Republican terms to contemporary infantry deployment is masterful, to give only one example.15

A better understanding of the genre of military theory, however, would have saved C from several missteps. C (p. 43) posits Vegetius’ familiarity with Aeschylus’ (or another’s) Seven against Thebes to account for Capenus’ invention of siege-ladders (4.21.2). More probably, Vegetius reproduced an exemplum of a primus inventor found in his source. The anthropological interests of some fifth-century B.C.E. sophists initiated catalogues of “first inventors.” Ephorus composed one ( FGrH 70 T 1) and the genre enjoyed some Hellenistic and hence Roman popularity (cf. Plin. HN 7.191-215). Military inventors feature prominently in such catalogues. Further, apart from the barbarian polemic and advocacy of reviving an antiqua legio, C misses Vegetius’ major military “lessons”: a small, well-trained, well-disciplined army can defeat barbarian hordes, but, more importantly, the good general avoids battle, unless it is to his advantage, and relies on intelligence and indirect means to defeat an enemy. This is the “Odysseus ethos” and the doctrine of stratagems. By no means was Vegetius an advocate of “decisive battle,” as he has sometimes been erroneously portrayed. In all probability, Vegetius reproduced and endorsed this doctrine from Frontinus, one of the explicitly named sources. Thus C’s whole discussion of Vegetius’ relationship to Greek military thought (pp. 43-45) is misguided. Vegetius was well aware of the Greek military treatises — whether through Frontinus, an attested philhellene with definite interests in Greek military theorists, or another source(s) must remain unknown. Rhetorically, Vegetius (1.8.9) distances himself from Greek tactici in favor of the disciplina militaris populi Romani, but the claim is specious. Cato, a named source, romanized parts of Polybius’ lost Tactica and Frontinus’ Strategemata was inspired by Greek theory, as was in all likelihood his lost comprehensive work of military theory. Greek Tactica were by no means lightweight literary works of no practical value, as C alleges (pp. 44-45). C ignores Roman deployment of the legion in phalangical formations, a phenomenon for which Vegetius offers a bridge in the evidence between Arrian’s Tactica and Acies contra Alanos in the Hadrianic era and Byzantine practices.16

C’s lack of attention to genre also has implications for his naval thesis (pp. 174-80), the “crown jewel” of his case for a dating under Valentinian III. I leave for others to assess critically C’s specific arguments. C, however, does not realize that Vegetius’ treatment of naval affairs is the only such theoretical discussion to have survived from Antiquity. Other theorists — from Aeneas Tacticus on — had promised to discuss naval affairs, but, if they did, those portions of their works have not survived. Placement of naval affairs in a treatise on war, given the genre’s stress on terrestrial over naval affairs, would seem to have come at the end of the work (cf. Aen. Tact. 40.8) — the very place in which Vegetius’ naval discourse occurs. Nothing can be argued from one manuscript’s detachment of the naval portion (4.31-46) as a separate Liber belli navalis (Scorialensis L.III.33, IX/X s., not L.III.3, as C p. 34). Thus, if the end of a treatise was the proper place for discussion of naval affairs, C’s view (p. 180) that Vegetius’ anxiety about Vandals occasioned his treatment of naval warfare as a last-minute addition is probably not seaworthy.

C has chosen to enter the treacherous minefield of Vegetian studies, a research topic notable for its booby traps and IEDs. Few scholars escape unscathed from this terrain and C has made a valiant effort.17 If, as C knows, the case for Valentinian III has not been proved, advocates for other emperors as the addressee must, nevertheless, now tread more lightly themselves. Vegetius remains an enigma. C has further muddied the waters.

Notes

1. N.P. Milner, tr., Vegetius: Epitome of Military Science, 2nd edition (Liverpool 1996); cf. my review of the first edition (1993): Journal of Military History 58 (1994) 136-38; M.D. Reeve, ed., Vegetius, Epitoma rei militaris (Oxford 2004).

2. A Eutropius not to be confused (e.g., J.B. Campbell, “Vegetius Renatus, Flavius,” OCD, 3rd edition, 1996, 1584) with the consul of 387 and author of a Breviarium dedicated to Valens: noted by M.D. Reeve, “Notes on Vegetius,” PCPS 44 (1998) 182 n. 3. Vegetius used Eutropius the epitomator as a source, as this reviewer hopes to demonstrate elsewhere.

3. G. Viré, “L’ Epitome rei militaris de Végèce: entre traité technique et oeuvre littéraire,” in P. Defosse, ed., Hommages à Carl Deroux, II: Prose et linguistique, Médicine (Brussels 2002) 427-28; cf. Nixon, in C.E.V. Nixon/B.S. Rodgers, In Praise of Later Roman Emperors (Berkeley/Los Angeles 1994) 502 n. 127. Viré’s paper escaped C’s notice: e.g. p. 122.

4. C.D. Gordon, “Vegetius and His Proposed Reforms of the Army,” in J.A.S. Evans, ed., Polis et Imperium: Studies in Honour of Edward Togo Salmon (Toronto 1974) 50; M.R. Mezzabotta, “Aspects of Multiculturalism in the Mulomedicina of Vegetius,” Akroterion 45 (2000) 52-64; D.K. Silhanek, Vegetius’ Epitoma, Books 1 and 2: A Translation and Commentary (diss. New York University 1972); N.P. Milner, Vegetius and the Anonymous De Rebus Bellicis (D.Phil. thesis. Oxford 1991).

5. “Transporting the Troops in Late Antiquity: Naves onerariae, Claudian and the Gildonic War,” CJ 100.3 (2005) 275-99; “Vegetius on Liburnae : Naval Terminology in the Late Roman Period,” SCI 24 (2005) 181-93.

6. R. Scharf, “Der comes sacri stabuli in der Spätantike, Tyche 5 (1990) 135-47: a pro-Valentinian-III view; Milner (above n. 1) xxv-xxvi.

7. A more detailed discussion of these issues in E.L. Wheeler, “Shock and Awe: Battles of the Gods in Roman Imperial Warfare, Part I,” in Y. Le Bohec and C. Wolff, edd., L’armée romaine et la religion sous le Haut-Empire. In press.

8. ” Mattiobarbuli in Vegetius’ Epitoma Rei Militaris : The Ioviani and the Herculiani, AHB 18.3-4 (2004) 109-21; “Mons Graupius Revisited: Tacitus, Agricola and Auxiliary Infantry,” Athenaeum 92 (2004) 129-40.

9. “The Rise of the Sassanid [sic] Elephant Corps: Elephants and the Later Roman Empire,” IA 42 (2007) 301-46; cf. P. Rance, “Elephants in Warfare in Late Antiquity,” AAntHung 43 (2003) 355-84.

10. AE 1981.777 with T. Drew-Bear, “Les voyages d’Auréius Gaius, soldat de Dioclétien,” in La géographie administrative et politique d’Alexandre à Mohamet (Strasbourg 1979) 93-141; cf. E.L. Wheeler, “The Legion as Phalanx in the Late Empire, Part II,” RÉMA 1 (2004) 170-73; P. Richardot relates 4.46.9 to the naval victory over the Greuthungi in 386: “La datation de De Re Militari de Végèce,” Latomus, 57 (1998) 146.

11. K.G. Holum, Theodosian Empresses (Berkeley/Los Angeles 1982) 45; S.I. Oost, Gallia Placidia Augusta (Chicago 1983) 47; S. Williams/G. Friell, Theodosius: The Empire at Bay (London) 61-62; R. Scharf, Foederati. Von der völkerrechtlichen Kategorie zur byzantinischen Truppengattung (Vienna 2001); cf. now T. Stickler, “The Foederati,” in P. Erdkamp, ed., A Companion to the Roman Army (Oxford 2007) 495-514; H. Elton, Warfare in Roman Europe, AD 350-425 (Oxford 1996): not cited by C; M.J. Nicasie, Twilight of Empire: The Roman Army from the Reign of Diocletian until the Battle of Andrianople (Amsterdam 1998); J.H.W.G. Liebeschuetz, Barbarians and Bishops (Oxford 1990): apparently unknown to C; correcting lax discipline as a topos : E.L. Wheeler, “The Laxity of Syrian Legions,” in D. Kennedy, ed., The Roman Army in the East (= JRA Suppl.18 [Ann Arbor 1996]) 229-76.

12. R. Grosse, “Das römische-byzantinische Marschlager von 4.-10. Jahrhundert,” BZ 22 (1913) 90-121; P. Richardot, La fin de l’armée romaine 284-476, 3rd edition (Paris 2005) 244 = 1st edition (Paris 1998) 183.

13. “Vegetius on Armour: The Pedites Nudati of the Epitoma Rei Militaris, AntSoc 33 (2003) 127-67, esp. 160-65.

14. B. Rankov, “Now You See It, Now You Don’t: The British Fleet in Vegetius IV.37,” in P. Freeman et al., eds., Limes XVIII: Proceedings of the XVIIIth International Congress of Roman Frontier Studies Held in Amman, Jordan (September 2000) (Oxford 2002) II, 921-24.

15. M. Lenoir, “Le littérature De re militari,” in Les littératures techniques dans l’Antiquité romaine: statut, public et destination, tradition (Vaneouvres/Geneva 1996) 77-108, esp. 86-90; on barbarian terms see P. Rance, ” Drungus, δρουγγος, and δρουγγιστι : A Gallicism and Continuity in Late Roman Cavalry Tactics,” Phoenix 58 (2004) 96-130; C. Zuckerman, “Le δεύτερον βάνδον Κωνσταντινιακῶν dans une épitaphe de Pylai,” Tyche 10 (1995) 233-35; cf. on barritus M.P. Speidel, “The Four Earliest Auxilia Palatina,” RÉMA 1 (2004) 132-46.

16. See E.L. Wheeler, “The Legion as Phalanx,” Chiron 9 (1979) 303-18, “The Late Roman Legion as Phalanx, Part I” in Y. Le Bohec and C. Wolff, edd., L’armée romaine de Dioclétien à Valentinien I (Lyon/Paris 2004]) 309-58, and (above no.10) 147-75.

17. Most of C’s factual inaccuracies have been corrected above. Typographical and editorial errors are minimal except for the list of arguments at p. 83 (initially presented as a, b, c, but then discussed as 1, 2, 3). At p. 81 n. 171, C’s work cited but omitted in his bibliography is: “Imperial Cuirasses in Latin Verse: From Augustus to the Fall of the West,” AC 73 (2004) 127-48.