CHAPTER 2
RANDALL S. HOWARTH
FROM the Midlands to Mesopotamia, from the Caucasus to the Cataracts, the armies of Rome marched and left permanent marks on the physical and cultural landscapes they traversed. Wherever a Roman army went, it left roads and depots around which the infrastructure of a cohesive empire gradually coalesced. Vast rivers of resources were mobilized across great distances, the long-term effect of which was the creation of a complex economy that linked hundreds of separate microeconomies. The army was itself a powerful social organism. It was both an agent of Romanization and a significant conduit through which flowed the potent influences of other cultures. With the demise of the Republic, the army soon became the prime arbiter of supreme power and its members a privileged class. Meanwhile, the popular culture of Rome exhibited an abiding fascination with death and blood in a reflection of the martial values of an empire sustained by a more or less permanent state of war. And, finally, many scholars point to fateful changes in the way Rome organized and deployed its army as a prime candidate for blame in the fall of the Western Empire. With these general observations in mind, one grasps not only the centrality of war to Rome but also the enormity of the topic. The chapter that follows will present a synoptic view of the evolution of Roman war and warfare; by necessity only a sketch, it will emphasize threshold moments as well as key and problematic issues pertaining to our understanding of the Romans at war.
THRESHOLD MOMENTS IN ROMAN WAR AND WARFARE
It took the Romans a long time to reach the limits of the empire they accumulated, but we can locate the threshold after which Mediterranean hegemony was certain, in the same place as did Polybius. Here is the rhetorical question that begins his History (1.1):
Who is so worthless or indolent as not to wish to know by what means and under what system of polity the Romans in less than 53 years have succeeded in subjecting nearly the whole inhabited world to their sole government—a thing unique in history?
The period to which Polybius refers (ca. 220–168 B.C.) encompassed the second war with Carthage, the defeat and humiliation of Antiochus III, and the destruction of the Macedonian monarchy. After this, two things were clear to the Romans and to everyone else: first, no external enemy of Rome could reasonably hope to win a war that Rome chose to pursue. And second, no polity bordering the Mediterranean existed except on the sufferance of Rome. We could explain Rome’s emergence to this commanding position solely in terms of the tactics, weapons, and training of the Roman army—it was often a frightfully effective machine of death—but to do so ignores the fundamentals that had already set Rome apart from other candidates for hegemony and ultimately made Roman military success more likely over the long run.
Polybius began his narrative of Rome’s rise with the first war between Carthage and Rome (264–241 B.C.). But by that time Rome had already established the basis of her long success in the preceding century. The cornerstones of that early success were constructed from a combination of good fortune and sound policy; good fortune in that the relative isolation of Italy provided a buffer that limited the appearance of outside raiders, or at least the intention and the scale of their raids—the invasion of the Gauls in 387/6 B.C. being the only important exception. Good fortune in that the manpower of Italy, a fountain of mercenaries since the late Bronze Age, was sufficiently large and acculturated to a warrior ethos that the raw material for a large military machine was ready to hand (Thuc. 7.53.1–2 and Etruscans recruited by the Athenians for Sicily). But it was sound policy, however, that from the beginning the Romans fostered an inclusive political identity that made it possible and feasible to integrate the war-making potential of conquered populations, to continuously broaden the basis of logistical support, and to create a network of allies that could and did survive disastrous setbacks. Most of all, it was sound policy to provide both allied and conquered elites genuine opportunities and real incentives to cooperate with the Romans.
Between 400 and 300 B.C., Rome developed a matrix of treaties with her neighbors that may have multiplied Rome’s manpower base by a factor of four, although numbers are always problematic. The most important of these treaties, that with the Latin League, had roots dating to the sixth century B.C. Although our primary interest in the Latin League stems from its importance to the Romans as a vehicle for mobilizing regional military assets, it was the political benefits of membership that truly differentiated the Latin League from other classical era leagues of cities and ultimately made it useful as a military organization. The most striking of these political benefits was a functional permeability of civic identity. Citizens of one member city could take residence in and exercise the rights of citizenship of other member cities. For the veteran of wars undertaken in this cooperative framework, the fruits of successful warfare were in some instances distributed in the form of colonies among whose enrollees would be counted citizens of Rome and other allied cities. The ease with which local elites could coalesce into regional elites important to Rome was an enticement to regional cooperation that did not exist in the Greek world, except perhaps in the Macedonian system of the mid-fourth century. A path to political inclusion existed at the bottom of the social scale as well. Ex-slaves could vote and the descendants of ex-slaves were full citizens. In all these we note contrasts with the more common customs exemplified in Athenian management of the Delian League. Athenians of the fifth century jealously guarded the gateways to their citizenship, to emphasize the civic and social distinctions of Athenian ancestry, and to limit colonists to Athenian enrollees. The effect was to limit both the theoretical extent of diplomatic cooperation among Delian League elites and the degree to which low-status individuals in ancillary communities felt connected to the ruling elite.1 One can hardly overestimate the benefits of these distinctions. Rome’s allies before 200 B.C., on the whole, understood the Romans as not fundamentally different from or ascendant over themselves—language and culture notwithstanding—because the diplomatic conventions promoted by the Romans gave form to and promoted inclusion in an overarching federal identity. (On the development of the conventions of the Latin League cf. Alföldi 1966, Sherwin-White 1973, Salmon 1982, and Howarth 2006, for a revisionist view).
The sources for Rome’s military and political institutions begin to provide cogent detail for the period beginning with the last quarter of the fourth century B.C. Rome’s emergence then as a significant player in Greek affairs—and so in Greek historical accounts—explains this change. We see, for example, that the sources begin to associate numbers (the first through the fourth) with the legions of that period whereas they seem not to do so for any previous period. We may fairly surmise that the Romans were by then routinely deploying four legions every year. From this date through the reforms of Marius (ca. 100 B.C.), we understand each Roman legion to have a theoretical complement of 4,200, nominally divided into thirty divisions. The legion was deployed in three lines (front to back: hastati, principes, triarii) with the most experienced soldiers in the rear. Lightly armed men (velites) were deployed in the front of the formation—bringing the theoretical complement up to 5,000—and the unit was supported by three hundred cavalry. Individual units had no permanent or long-term identity and the soldiers enrolled only for specific campaigns.
By 338 B.C., the Romans were actively managing the military assets of their erstwhile partners in the Latin League as well as a varied collection of miscellaneous allies the Romans termed collectively the “Italians.” Whereas before this date the allies cooperated with Rome on a more or less voluntary basis, their cooperation was thereafter essentially compulsory. On the other hand, after 300 B.C., Roman successes were increasingly frequent and their scale increasingly impressive. Presumably the fruits of these efforts were shared; we have no particular reason to think they were not. On what proportional basis the categories of allies were deployed is a matter for conjecture until late in the third century, and even then it is inconsistent and idiosyncratic. As the Romans continued to grant their citizenship, we might suppose that the proportion of the army which was technically Roman would have increased. Against this, we note that the number of communities in the allied category continued to grow, so that the proportions may have remained more or less in balance. Without undue damage to the evidence, we may suggest that consular armies were never more than 50 percent Roman anytime after 338 B.C. and perhaps substantially less. On this question, the sources simply do not say enough. We can say that the Romans were routinely deploying at least seventy thousand soldiers in consular and proconsular armies every year by the last third of the third century B.C. This does not include any number of smaller deployments under the leadership of second-tier commanders (praetors) nor does it include naval forces.
All things considered, Rome’s matrix of treaties ensured that the costs of war in men, material, and treasure were effectively spread over a wide base. All categories of allies that provided manpower also equipped and paid their own men. Treaties—as with Naples circa 325 B.C.—provided ships, while others, such as with Syracuse circa 260 B.C., provided grain. The war with Pyrrhus and the first two wars with Carthage provide dramatic examples where Roman manpower and logistical advantages trumped both the repeated battlefield successes of enemies and the vicissitudes of natural disaster. Pyrrhus, for his part, although the apparent winner in at least two of three battles he waged with Rome between 280 and 270 B.C., ultimately gained nothing for his effort except that his name became forever synonymous with a victory won at too great a cost. In the first war with Carthage, Rome’s losses in warships and men were staggering; perhaps as many as 100,000 Roman and allied soldiers and sailors may have been lost over the twenty-three-year war and, if one counts transports and warships, well over a thousand ships. Yet in a single battle in 241 B.C., the loss of only one hundred ships by the Carthaginian side led Carthage to accept a humiliating peace. The net gain for Rome was substantial: Sicily, substantial reparations, and shortly afterwards the opportunistic seizure of Sardinia and Corsica when unpaid Carthaginian soldiers revolted. Analysis of Carthaginian silver over the course of the war shows an unmistakable pattern of repeated devaluations.Ultimately, Carthage lost that war in much the same way that the Soviet Union lost the Cold War (Lazenby 1996): it could not afford to compete.
In the second war between Rome and Carthage, Hannibal understood these advantages; one can see a consistent pattern of attempts on his part to separate Rome’s allies from Rome, a mostly unsuccessful enterprise. Although Hannibal was the superior military thinker—he won every significant battle with the Romans except his last—he could not win the war because he could not win over Rome’s allies. Meanwhile, the Romans absorbed the lessons taught by Hannibal on the battlefield and eventually stole the initiative by finally attacking Carthage itself. The result for Carthage was the same as in the previous war: humiliating terms of surrender. We should therefore understand Roman military success before 200 B.C. as a function of a structural resilience much more than we should see it as that of superior tactics, weapons, or leadership (Polyb. 3.89; Brunt 1962).
The next stage of Rome’s military history begins with the second defeat of Carthage in 202 B.C. and ends with the demise of the Republic. This period marks the slow erosion of the presumption of equality among allies of Rome. Because after Hannibal no external entity existed that could realistically threaten Rome’s dominance, the quid pro quo implicit within treaties like that binding the Latin League was no longer necessary to sustain Roman success; the principle of equal status among allies was increasingly dishonored, even where it continued to exist by treaty. This translated into a pattern of abuse of and interference in the domestic concerns of the various Italian allies (Howarth 1999). In addition, the movement of Rome’s frontiers away from Italy changed the economy irrevocably, and not to the benefit of the lower classes of Italians (Rosenstein 2004). The combination of these conditions eventually stirred awareness of a gradual devolution to second-class status. The most destructive war in Italian history is conventionally known in English as the Social War, but it might more transparently be termed the War of the Allies. It tore Italy apart early in the first century B.C., after Rome’s allies demanded full Roman citizenship at the point of a sword. Rome survived the trauma, but the tensions that fueled the war remained fundamentally unrelieved for another hundred years.
The wars with Carthage furthered Roman advantage in the most fundamental of ways: they forced the centralization of supply and command. The logistical demands of maintaining so many men in the field for such long periods of time fostered the marshaling of enormous resources, channels of supply to distribute them, and the emergence of large-scale state contracting (Erdkamp 1998; Roth 1999). The occasional resistance of Italian municipalities to Rome’s incessant demands for manpower led inevitably toward centralization of manpower records, and, logically, to direct management of the resources to which they corresponded. These systems of supply and recruitment evolved in the third century B.C., and were taken for granted in the second century B.C.
By the end of the second century, the Romans reverted to a streamlined recruitment policy that encouraged men without property to enlist (Keppie 1998). This meant inevitably that the state was now routinely obligated to provide weapons and equipment kits to recruits. This created a practical imperative for standardization, with coincident benefits occurring in battlefield management (Smith 1958; Keppie 1998). The successive consulships of Gaius Marius (104–100 B.C.) represent the threshold by which time these reforms were functionally complete. The three ranks of the pre-Marian army were now replaced by ten cohorts of 480 men, identically equipped. This basic complement will remain the standard legionary model at least until A.D. 200. Legions hereafter bore permanent names that corresponded to enduring identities. The lightly armed velites and cavalry were replaced by separate units termed auxilia (“helpers”) and alae (“wings”), respectively. Initially these were discrete contributions from specific allies but over time it is clear that auxilia were replenished by various methods, including recruitment of citizens. Over the long run, these new categories became increasingly important to the Roman army, but gauging their varying size and tactical importance constitutes a significant problem in understanding the Imperial army.
The monarchy that was established by Caesar’s successor, Augustus, created the framework for the third significant period of the Roman military machine. Although the armies of the Republic could and did remain in the field for long periods, technically they had always been ad hoc arrangements for particular missions, at least until Marius. Armies marshaled under such circumstances were inherently insecure constituencies whose loyalties remained ripe for manipulation by ambitious dynasts. Augustus, recognizing this, after discharging and paying off a substantial portion of the combined armies of 31 B.C., created a stable professional army with a centralized command-and-control and supply structure (see further Culham, 236–60). Whereas the Republican system of supply and logistics had always left de facto organization of supply networks in the hands of commanders, Augustus modified this by creating a permanent military treasury (the aerarium militare) funded over the long term by inheritance and auction taxes (Cass. Dio 55.24.9). Eventually, perhaps as early as A.D. 200, taxation schemes began transitioning from tax in coin to tax in kind on the model of what was already in place in Sicily and Sardinia since the third century B.C.2 With the institution of a fully professional army, soldiers could count on a sixteen to twenty-five year career and the enduring patronage of a highly centralized government whose stability depended on their support. Augustus, hardly an altruist, was very careful to restrict access to leadership positions and carefully managed military deployments.3
These arrangements created the conditions necessary for cohesive and long-term military policy and, of course, left the appearance of it in the record. As a result, modern scholars have mapped patterns of deployment, plotted the construction of a hardened military infrastructure, and even proposed the existence of an enduring “grand strategy” of frontier management, although the latter notion has lost ground.4 The opposite conclusion is that the Romans had no frontier policy at all, but considered the frontiers, such as they were at any moment, as borders of their logistical network rather than the limits of their empire (Whittaker 1994). The truth may lie somewhere between.
Imperial disposition of positions within the leadership of the army was governed by at least three competing considerations: the need to dispense meaningful patronage, the necessity of keeping potential challengers from exploiting these opportunities to the detriment of Imperial protégés, and the desirability of maintaining competent commanders in key positions. The realization of these goals may coincide, but seem seldom to have done so, so that periods of what appear alternately to be either conservative or assertive frontier policy can also be construed as functions of changing Imperial confidence (cf. Lendon 1997 who argues that the army expected a code of behavior from the emperor, not just his patronage). We see that emperors gradually increased the number of legions over the third century to as many as sixty-seven but reduced their complement proportionately. If we think of the Imperial government as an essentially reactive one (Millar 1977), we might see strategic thinking as a series of weighted responses to different kinds of enemy: those from within and those from without. We see this tension early on in Domitian’s decree that no two legions could camp together (Suet. Dom. 7.3).
The third century A.D. represents the next important threshold. An essentially unstable political period, the length of average reigns by emperors decreased dramatically after 235 until Diocletian’s successful reorganization began in 284 (MacMullen 1976). The political uncertainty in Rome certainly engendered a collapse of strategic coherence. Borders abandoned by would-be emperors who marched for Rome were tested and overrun by increasingly bold invaders. The Romans initially adapted by building walls around cities and emphasizing mobility of force over static defense. Over the long run, it would appear that the Romans constructed a multilayered defensive structure that required a substantial increase in the total number of men under arms but deployed in smaller units than had been the standard before circa 200. Unfortunately, the lack of good evidence for this period makes it nearly impossible to understand these dispositions in detail.
Finally, a series of military and political shocks rocked the Roman world in the last third of the fourth century. Julian, emperor from A.D. 360 to 363, was killed in an abortive invasion of Persia, but Persian indecision and Roman territorial concessions saved his army. In 378, the impetuous Valens died along with much of his army, slaughtered by the Goths at the battle of Adrianople (Lenski 2002). His successor, Theodosius, eventually granted the Goths extraordinary powers of autonomy as well as huge subsidies. But we note an ominous development for Rome in 394 when Theodosius led an army composed substantially of Gothic confederates and defeated a Roman army led by the usurper Eugenius. With Theodosius’s death the next year the Goths became once again the enemy, stunning the Roman world by sacking the city of Rome in 410. From this point, no longer could the Romans assume dominance in military contests nor was their empire’s future a safe assumption.
THE EVOLUTION OF ROMAN MILITARY-POLITICAL INSTITUTIONS AND CUSTOMS
What we know of the political organization of Rome before 200 B.C. inspires some comparison with the Greek polis. There is an obvious and significant correlation between what we might term the political class and the primary military class upon which any organized community depended. A correlative principle is that those who were in a position of responsibility for leadership in war enjoyed the highest level of prestige. In Sparta we see this correlation clearly: the political class and the hoplite army were one and the same. Spartan kings were primarily war leaders. Likewise, we see the greatest extension of democracy to unpropertied classes in Athens, because its navy required so many sailors. Pericles, elected as strategos in Athens, enjoyed a virtual first-citizen status with the benefit of that position. The two supreme magistrates of the Roman Republic—the consuls—had no power within the city walls except to preside over Senate discussions and ritual functions, though consuls might choose to lend the prestige of their name to legislative initiatives (Sandberg 2001). Their primary portfolio was the conduct of war. There were a handful of lesser military commands available in any given election cycle so that the total number of elected and prorogued commanders in the field fluctuated from year to year.
The most exclusive elected office in Rome was that of censor, who were elected in pairs at four-to-six-year intervals. Although the censor developed wide-ranging responsibilities by the late Republic—these would include, for example, the letting of public contracts—the censors’ original and core responsibilities had to do with manpower management. The Romans also elected lower-level officials called quaestors who provided supervision of financial concerns in military districts outside of Italy. The Roman Senate existed primarily as an advisory council for war and diplomacy, and acted as a forum for the articulation of privileges associated with the conduct, rituals, and rewards of warfare. Certainly many other matters fell within the purview of senatorial comment, but the chief movers within the Senate won that status as a consequence of a family history of military endeavor. In the Comitia Centuriata, or the Assembly of Centuries, the citizens of Rome assembled to declare war and elect the consuls, the institution having evolved from an army assembled to acclaim its leaders and undergo ritual purification (see Mitchell 1990; Nicolet 1980). The clear implication is that war amounted to the first true business of the Roman state.
The internal stability of the Roman oligarchy was, like that of any other, highly dependent on internal cooperation in the management of access to the means of influence and the most effective modes of aristocratic display. The competition that is allowed in any such system must operate within consensual parameters of discretion or the dynamic balance of power within the group will become unstable. By 300 B.C., military endeavor was firmly established in Rome as the primary conduit for the accumulation of aristocratic prestige (Rosenstein 1990). Accordingly, the competition for access to leadership positions was always sharp. Even having achieved such positions, however, the ability of any member of the oligarchy to overreach was limited by the scale and duration of military endeavors. So long as short-term commands were the rule, the oligarchy had an effective tool to limit individual ambition. But, by 200 B.C., short commands had become the exception, not the rule. And from about 100 B.C., the prime movers in the Senate demonstrated an inability to restrain individual ambition on any consistent basis. We should see this as the background for Sulla’s ultimately failed attempts to enhance the power of the Senate and an excellent context within which to locate the elusive “Servian Reform” of the Comitia Centuriata (Howarth 2006).
We can observe this same threshold from a different perspective, that of loyalties demonstrated by soldiers, and the manipulation of those loyalties by competing dynasts. Before the end of the second century, recruitment of soldiers was tied to property rating. While it seems that the Romans were not routinely drafting men without property before the end of the second century B.C., it appears that on occasion they did so, mostly in response to emergency situations. Even so, it does not necessarily follow that men without property were not ordinarily enlisted; only that men without property ordinarily did not self-enlist without some sort of interaction with a patron whose property obligated them to provide a man. Manpower obligations were in some manner indexed to wealth rating; we might surmise, therefore, that the higher the wealth rating, the greater the number of bodies necessary to meet one’s obligation. For those at the top, such requirements would logically be fulfilled from one’s clientele. It had always been the custom that soldiers provided their own kit so we should expect these recruits to have been equipped with armor and weapons by their sponsoring patrons.
It follows logically that patrons should expect some enduring bond of obligation as a consequence. Although there is no confirming evidence, one might also expect a material obligation on the part of the soldier: perhaps a share of his portion of war booty or even an interest in any land that might eventually be distributed. But whether or not a formal obligation to the sponsoring patron existed, the salient point is that the informal nature of an enduring patron-client relationship would stand as an offset to the temporary influence of the commander to whom the recruit was assigned. In addition, although patronage patterns that endured between ex-commanders and veterans were only one element determining aristocratic prestige, subsequent events indicate that it was the most dangerous category. On the other hand, so long as short commands of limited scope were the rule, the growth in any individual’s prestige ultimately depended on the approval of the oligarchy as a group.
Unfortunately, all customary and institutional checks on aristocratic prestige were obliterated by the end of the second century B.C. Short-term commands remained practical only so long as armies marched from Rome and were in a position to achieve their objectives within a single campaigning season. This was already mooted by the middle 200s during which Roman campaigns evolved to accommodate war at sea against a powerful enemy on another continent. With this evolution of Roman military activity came an inevitable modification of the term limit; consuls’ and praetors’ authority to wage war was routinely prorogued for two or three years at a time. While this was a clearly practical and logical response to logistical imperatives, it also enhanced the bonding process that always takes place between soldiers and their commanders, especially successful ones. We see the implicit potential in the illegal election of the young P. Cornelius Scipio by the troops of his deceased father in 210 B.C. The army, which had been fighting in Spain for years, had been led first by the uncle and then by the father. With this, we also note the increasing scale of the plunder won in such contests, literally hundreds of thousands of pounds of silver and tens of thousand of gold, not to mention of course the many thousands of captives sold at auction.
The breaking point came with the rise of Gaius Marius at the end of the second century. Marius was an extraordinary man with boundless ambition but with one significant shortcoming: no pedigree. That is, he was the first of his family to achieve prominence in Rome. Normally men without generations of family history in Roman politics could not expect to rise to the highest levels of leadership, however capable they might be. Yet Marius, a second-generation citizen, won the consulship more times than any other man in the Republic. How can we explain Marius’s unlikely success? First, there was an emergency. A significant migration of Germans was taking place just beyond the northern frontiers of the Roman world in the last decade of the second century B.C. Several Roman armies sent to confront the Germans had been destroyed, the political effect of which was to break down, at least temporarily, some of the unity within the ruling oligarchy. Marius, fresh off a successful three-year campaign in Africa, was a logical choice for assignment in the north except that political considerations would normally forestall a repeat consulship for any commander much less an outsider like Marius. Marius’s five additional and successive consulships point to a political paralysis of sorts within the oligarchy whose collective antipathy to the “New Man” could prevent the proroguing of his command, but not his reelection (figure 2.1).
Figure 2.1 Gaius Marius, soldier and politician, creator of “Marius’s Mules.” Room 11, Glyptothek, Munich. Photo Credit: Bibi Saint-Pol.
Marius, in preparing to confront the German menace, implemented a number of initiatives in the areas of recruitment, equipment, and training that had enormous long-term impacts across the political and military landscapes. The most immediate problem facing Marius was a severe manpower shortage. The relatively low returns of the Iberian and Celtic wars of the late second century, combined with the disasters in what are now Croatia and France, would certainly generate an increasing reticence on the part of patrons to provide men and equipment. Compounding this reticence was Marius’s pedigree. While we are clearly in the range of supposition, it is not hard to envision individual members of the oligarchy turning their backs on obligations to supply his army with men. Subsequent events certainly suggest that the shortage of manpower was, indeed, an artificial shortage, because Marius did fill his army by direct enlistment. As we have already noted, this was not so much a radical reform as it was a resort to expediency. In one sense, this course of action can be construed as a sop to the aristocracy because, rather than force members of the aristocracy to provide manpower, it obviated them from giving up client labor supplies for which the burgeoning economy of the late second century had other uses. It would also unburden that aristocracy of any obligation to equip the new soldiers. Marius’s recruits, now equipped by the state with a standard complement of equipment that each man carried himself, came to be dubbed “Marius’s Mules.”
Marius did succeed in his campaigns against the Germans, but, with the threat represented by their presence no longer a political consideration, the oligarchy closed ranks against him. Their reluctance to sanction a reward of land to Marius’s veterans should be seen as a naked political ploy designed to undermine the auctoritas that his meteoric rise had generated. However, this stunt removed a cornerstone of the compact between the oligarchy as a whole and the citizen-soldiers of Rome. The next seventy years saw a succession of dynasts who each capitalized to some degree on the combination of disaffected soldiers and a dysfunctional Senate to challenge for individual supremacy. The allegiance of soldiers to the community was inevitably displaced by allegiance to individual commanders, a recipe for an accelerating instability (Brunt 1962). All leaders cultivated strong ties to a dedicated soldiery who were willing, to one degree or another, to crash the gates of Rome in the service of their patron. In so doing they established a truism of the following century: the army would be the most important arbiter of political ascendancy. There is no particular reason, I think, to believe that any of these men, Caesar included, sought a monarchy, but merely the first position within the oligarchy. On the other hand, that Caesar’s assassination merely delayed the creation of a monarchy for a few more years demonstrates its structural inevitability.
Meanwhile, the external empire that the Romans had accumulated continued to endure, even as the Roman political system collapsed upon itself. The only serious outside threat during the civil war years came from the Parthians, whose forays into Syria were inspired by their destruction of Crassus’s army at Carrhae in 53 B.C. A new order, a monarchy wrapped in Republican titles and customs, came to power. Nevertheless, this new system retained the essential elements of the old system, if by essential we mean that the most potent forms of aristocratic prestige were still derived from the management of war and its rewards. The real difference between Republican and Imperial realpolitik was that access to the highest levels of prestige had now been stabilized under the control of a single family. What remained of the former oligarchy was unable to mount any significant challenge to the authority of the Emperor, if only because it had no opportunity to accumulate sufficient prestige to compete realistically for the affections of the army. We see, for example, that the last non-Imperial family commander allowed to triumph was Cornelius Balbus in 19 B.C. With the most impressive forms of aristocratic display firmly under the control of one family, along with, of course, control of the fruits of war and empire, others had to settle for the accumulation of empty titles and for access to the patronage of the Emperor. In the process, Rome became, for all intents and purposes, a military autocracy. All political questions paled before the most important one: to whom would the army be loyal?
The Roman emperors of the Principate and beyond continued to engage in expansionist activities in response to predictable imperatives: the need to shore up political influence at home or with the troops, and, ultimately, to transfer their accumulated loyalty to heir-designates. Lavish public games and huge donatives to the army and the citizens reinforced these political objectives, but emperors who were perceived as uninterested or unlucky in military endeavors were much more vulnerable to challenge. The best example is Nero who, for his own part, had no stomach to lead the army anywhere, preferring instead to dominate sham poetry contests. His suicide anticipated his overthrow and murder, perhaps by minutes. Later, Domitian’s failures on the frontier, despite his personal leadership there, engendered considerable hostility in the army and left him vulnerable in Rome. Finally, Commodus’s preference for the capitol over the army camp lay at the center of his own unpopularity. The unpleasant but entirely real possibility of revolts led by ambitious frontier commanders clearly conditioned Imperial policy toward the army and frontiers for the next three hundred years.5
The key problem we have for understanding the end of the Roman Empire in the West is whether we should emphasize Rome’s growing reliance on recruitment of barbarians for the army as the essential cause for subsequent problems. Without a continuous narrative it is hard to chart the history of this policy. It is certainly true that Rome had always integrated the war-making potential of her conquered foes and we should rightfully regard this as a continuing strength of Rome, not a weakness. What seems to set this period apart is the pace of absorption and the seemingly disproportionate representation of the Goths. Furthermore, one might say that the schizophrenic policies of the Romans toward the Goths actually made a problem where one had not existed. The Goths, never culturally integrated into the Roman world, sacked Rome in 410 after protracted negotiations with the Romans failed to secure them a tenable position in Italy.6 Alongside this trend, we note clear evidence that Imperial authorities were having trouble recruiting soldiers and sustaining those they had.7 The end of Rome in the West was rooted in the increasingly obvious inability of Western Roman authorities to support a significant army (see Whittaker 1994; Southern and Dixon 1996; Elton 1996; Coello 1996 for unit sizes in the late army).
NEW SCHOLARLY TRENDS CONCERNING ROMAN WAR AND WARFARE
As the other contributions to this volume suggest, much new work taking place in the study of Roman war and warfare looks at the experience of war, as opposed to its direction and outcome. What was it like to stand on the line of battle in a Roman legion? How does one feed an army as it is traversing enemy territory? What effect does a hostile army have on a host population? Such questions bring with them their own problems, not the least of which is that our sources seldom addressed them, at least directly. But there is much indirect evidence that lends itself quite nicely to questions such as these. There has always been a great deal of archaeological, papyrological, and epigraphical evidence. However, much of it has been marginalized, practically speaking, by its publication outside of historical journals. Its collection and organization into regional and Empire-wide categories gives substance to our largely anecdotal narratives, while in turn those narratives give meaning and context to the minutiae (see further James, 91–127, Adams, 261–76).
One productive category of this work focuses on logistics (Le Bohec 2000; Alston 1995; Erdkamp 1998; Roth 1999). It is absolutely clear that ancient writers were aware of the importance of logistics in the outcomes of war: Polybius notes for the third centuryB.C. that “the advantages of the Romans lie in inexhaustible resources in supplies and men” (3.89). According to Frontinus, Caesar sought to “conquer the enemy with hunger rather than with steel” (4.7.1). The Romans were nothing if not organized, and everything we know about Roman training suggests that Roman soldiers were kept constantly busy training, building, or foraging. Romans did not maintain separate support units in conjunction with combat units so that all the support activities—supply, mess, armory, engineering, foraging—would normally be carried out within the structure of the legion. Each legion on the move required thousands of mules, hundreds of carts, and perhaps thousands of muleteers, drivers, and other “helpers,” especially if the legion commander had a staff or if the unit were equipped for siege warfare. The Romans made elaborate provision for gathering water, firewood, and fodder, only a few days usage of which can reasonably be hauled at a time by any army on the move. Keeping the army in supplies entailed a wide range of discretion from pillaging to requisitioning, the choice of which was related to the desired impact on the host population (Roth 1999).
Since the third century B.C., the Romans made systematic provision for regular supply networks and the seemingly durable nature of these networks helps explain how the various players who vied for power at the end of the Republic could maintain their armies. The combined figures for Romans under arms when the civil wars finally ended in 30 B.C. are in the vicinity of one-half million: a testament to a stunning achievement of logistics and one not equaled in the West for another eighteen centuries. What were tactical depots in the time of Augustus hardened into permanent infrastructure in the following decades. The construction of the navy bases at Misenum and Ravenna were tied to the need to protect supply shipments into Italy. All told, there must have been a significant and salutary impact on local economies, especially in underdeveloped areas. The Roman armies built an incredible network of roads, of course, not so that soldiers had a hard path upon which to walk, but to keep the cart traffic supporting them from sinking into the mud (Chevalier 1976). All of this suggests a rich area for further investigation.
Another direction taken with respect to Roman warfare is that we are increasingly interested in what is going on in the minds of the men on the line. What was a Roman infantry battle actually like and how did it differ from a hoplite- or Macedonian-style contest? In a hoplite battle it makes little sense to hold men in reserve if the goal is to force the enemy backward until the formation breaks down and the army becomes a mass of individuals. Ancient sources usually do not make it easy to estimate how long these encounters lasted, but, by comparison, it seems easy to prove that typical Roman infantry battles lasted longer than typical Greek hoplite battles, in some cases hours longer (cf. Pritchett 4: 46–51). Our ancient authors note that the Romans’ regular rotation of fresh troops for tired troops was a significant Roman advantage and reason for success. One can see little opportunity for rotation if continuous pressure existed between the front lines of opposing forces, if only because there is no space between men (Sabin 2000).
The model we are given by Hollywood, where disciplined formations of soldiers devolve into full-scale melees, while presumably allowing for continuous reinforcement by fresh troops, nevertheless is also unsatisfactory because it demands casualty figures commensurate with a mutual slaughter lasting for hours. There is no consistent evidence that suggests that this was the norm. An alternative and much more plausible model is that of two armies in a “dynamic balance of mutual dread” (Sabin 2000: 15). We would see a series of individual or group clashes along the fronts of two armies which were barely separated. The space between the armies also allowed the Romans to rotate, as was their reputation, a constant supply of new men to the front. Opponents anxious for the clash and too undisciplined to wait, hasten their demise at the hands of the constantly refreshed Romans. The first to step back, usually that side which tires first, precipitates a general movement backward and heralds an ultimate defeat. Hannibal’s practiced approach was to choose topography that would allow him to attack Roman flanks, thereby disrupting or precluding an orderly rotation of fresh troops to the point of contact. The ultimate influence of Hannibal on Roman strategy is that he taught the Romans the art of pre-battle maneuver. After Hannibal, Romans always gave much more thought to battlefield choice and psychological advantage (see Sabin 1996: 59–80). Once the battle was joined, however, it was the actions of the few that mattered the most. When one tries to reconstruct the progress of any Roman battle, it seems that we should place much more emphasis on the acts of individuals than we have previously done and regard legions as flexible structures (Goldsworthy 1996).8
In conclusion, we see that there is scarcely any element of Roman history and culture that should be seen in isolation from the habits of the Romans in war. In the beginning, Roman diplomatic success was predicated on partnership in the conduct of war and the sharing of its benefits. The most obvious path to prestige in Rome was found in the management of war and its rewards. Eventually the practical problems of managing warfare at great distances from the city of Rome overwhelmed both the institutions and customs of the city and the diplomatic instruments that had facilitated early success. The system that replaced the Republic increasingly favored the army as a privileged class of citizens and it continued to expand. The demise of the political idea of Rome was tied to changes in the way the army was organized and deployed. On the other hand, we see that the Roman army and the habits of Rome at war should not be considered in isolation from the social, diplomatic, and political contexts in which they existed. Happily, all of these realizations are resulting in a fresh synthesis of narrative evidence and material culture, and a host of new questions being asked of long-familiar material.
BIBLIOGRAPHY
Alföldi, A. 1966. Early Rome and the Latins. Ann Arbor.
Alston, R. 1995. Soldier and society in Roman Egypt: A social history. London.
Birley, A., and R. Birley. 1994. “Four new writing tablets from Vindolanda.” ZPE 100: 431–45.
Brock, R., and S. Hodkinson (eds.). 2000. Alternatives to Athens: Varieties of political organization and community in ancient Greece. Oxford.
Brunt, P. 1962. “The army and the land in the Roman revolution.” JRS 52: 69–86.
Cameron, A. 1993. The later Roman empire. London.
Campbell, B. 1975. “Who were the viri militares?” JRS 65: 11–31.
Chevalier, R. 1976. Roman roads. Berkeley.
Coello, T. 1996. Unit sizes in the late Roman army. Oxford.
Cornell, T. 1993. “The end of Roman imperial expansion,” in Rich-Shipley, 139–70.
———. 1995. The beginnings of Rome: Italy from the Bronze Age to the Punic Wars (ca. 1000–264 B.C.). London.
Duncan-Jones, R. 1974. The economy of the Roman empire: Quantitative studies. Cambridge.
———. 1990. Structure and scale in the Roman economy. Cambridge.
———. 1994. Money and government in the Roman empire. Cambridge.
Elton, H. 1996. Warfare in Roman Europe, A.D. 350–425. Oxford.
Erdkamp, P. 1998. Hunger and the sword: Warfare and food supply in Roman republican wars, 264–30B.C. Amsterdam.
Goldsworthy, A. 1996. The Roman army at war, 100B.C. to A.D. 200. Oxford.
Goodburn, R., and P. Bartholomew. 1976. Aspects of the Notitia Dignitatum. Oxford.
Howarth, R. 1999. “Rome, the Italians, and the land.” Historia 48: 282–300.
———. 2006. The origins of Roman citizenship. Lewiston.
Isaac, B. 1992. The limits of empire: The Roman army in the east. Rev. ed. Oxford.
Kennedy, D. 1996. The Roman army in the east. JRA Supplemental vol. 18. Ann Arbor.
Keppie, L. 1998. The making of the Roman army: From republic to empire. Norman.
Lazenby, J. F. 1996. The First Punic War. A military history. Palo Alto.
Le Bohec, Y. 2000. The imperial Roman army. Trans. by R. Bate. London.
Lendon, J. E. 1997. Empire of honour. The art of government in the Roman world. Oxford.
———. 2005. Soldiers and ghosts: A history of battle in classical antiquity. New Haven.
Lenski, N. 2002. Failure of empire: Valens and the Roman state in the fourth century A.D. Berkeley.
Lomas, K. 2000. “Cities, states, and ethnic identity in southeast Italy,” in E. Herring and K. Lomas (eds.). 2000. The emergence of state identities in Italy in the first millennium BC. London.
Luttwak, E. 1976. The grand strategy of the Roman empire from the first centuryA.D. to the third. Baltimore.
MacMullen, R. 1976. The Roman government’s response to crisis:A.D. 235–337. New Haven.
———. 1982. “The epigraphic habit in the Roman empire.” AJP 103: 233–46.
Mattern, S. 1999. Rome and the enemy: Imperial strategy in the Principate. Berkeley.
Millar, F. 1977. The emperor in the Roman world (31B.C.–A.D. 337). Ithaca.
Mitchell, R. 1990. Patricians and plebeians: The origin of the Roman state. Ithaca.
Nicolet, C. 1980. The world of the citizen in republican Rome. Trans. by P. Falla. Berkeley.
Rees, R. 2004. Diocletian and the Tetrarchy. Edinburgh.
Rosenstein, N.. 2004. Rome at war: Farms, family and death in the middle republic. Chapel Hill.
Roth, J. 1999. The logistics of the Roman army at war, 264B.C.–A.D. 235. Leiden.
Sabin, P. 1996. “The mechanics of battle in the Second Punic War,” in T. J. Cornell, N. B. Rankov, and P. Sabin (eds.). 1996. The Second Punic War: A reappraisal. BICS, Supplemental vol. 67. London. Pp. 59–80.
———. 2000. “The face of Roman battle.” JRS 90: 1–17.
Salmon, E. T. 1982. The making of Roman Italy. Ithaca.
Sandberg, K. 2001. Magistrates and assemblies: A study of legislative practice in republican Rome. Rome.
Sherwin-White, A. N. 1973. The Roman citizenship. 2nd ed. Oxford.
Smith, R. E. 1958. Service in the post-marian army. Manchester.
Southern, P., and Dixon, K. 1996. The late Roman army. New Haven.
Thomsen, R. 1980. King Servius Tullius: A historical synthesis. Humanitas 5. Copenhagen.
Whittaker, C. R. 1994. Frontiers of the Roman empire: A social and economic study. Baltimore.