BMCR 2000.06.03

Rome and the Enemy. Imperial Strategy in the Principate

, Rome and the Enemy. Imperial Strategy in the Principate. Berkeley: University of California Press, 1999. xxii + 259. $35.00.

For roughly twenty-five years now, anglophone (especially) ancient historians have repeatedly tussled over Roman imperial grand strategy. The political scientist Edward Luttwak launched the current debate in 1976, and a good assessment of things as they stand is available in a recent article by C.R. Whittaker.1 Put rather crassly, the situation is this. One faction looks at the Roman empire, its size, its longevity, and can only imagine such a realm in the context of some pretty careful strategic planning. The other side is not convinced, seeing the shape of Rome’s empire as something that emerged in an almost completely ad hoc fashion. We may, in fact, never enjoy a consensus. That threat looms large especially because the Romans themselves, at least in the extant corpus of their writing, neglect the topic of imperial strategy all but utterly. Nonetheless, the mere fact of this empire fairly demands the attempt to reach some kind of agreement as to how, and thus perhaps why, it came to be and (especially) continued to be. Ramsay MacMullen three years ago made some tempting suggestions for advancing the discussion: “What is most important and most problematical are questions of intent: for example, did the Romans (and exactly which ones were in a position to make decisions, and how?) wish to move forward (in all periods? on all fronts?), or were their military forces rather meant to insure non-military objectives? According to what perceptions of the outsiders or provincial populations were Roman decisions made? Toward what ends or benefits?”2 The work here under discussion, which has been produced by one of MacMullen’s former students, raises just these sorts of question.

Susan Mattern’s fine book opens with consideration of a passage from Herodian (1.6.5-6) which details a perplexing decision faced by the young Commodus upon his father’s death. The imperial train is at the Danube frontier, tangled in a war with “the barbarians” (Quadi and Marcomanni). Should the new emperor pursue this conflict, or might he perhaps bring it to a halt and so return to the capital? His amici (Herodian has the communal sententia expressed by Ti. Claudius Pompeianus, Commodus’ brother-in-law) advise the former course of action. Why? To quit without a Roman victory would be dishonorable, and dangerous since such a culmination of hostilities would embolden the enemy; they would surely despise the Romans for their cowardice and flight. Moreover, it would be much better to return, having conquered all the way to the ocean, leading captured kings and satraps in triumph. That kind of performance has previously made Romans great ( μεγάλοι) and honorable ( ἔνδοξοι).3 Although he wavers in the face of this advice, Commodus ultimately is persuaded by others at court (the κόλακες, ap. Herod. 1.6.1), and shamefully turns his back on the Danube. Mattern’s chief concern in all this, and really throughout her book, is to portray the aristocratic mentalities that resulted in the particular kind of advice offered by Herodian’s Pompeianus. She accomplishes that task admirably and, in doing so, presents us with a forceful model for understanding generally the processes of decision-making that underlay Roman imperial foreign affairs. But, in addition to this, the model she presents offers a kind of middle ground between the two camps in the strategy debate. That is to say, there was, as Mattern tells it, a consistent logic to Roman foreign policy; it was not, however, the logic we might incline to expect and has therefore been neglected.

We find ourselves, in Mattern’s opening chapter (“The Decision-Making Elite”), among a group of men exquisitely trained in, and thoroughly shaped by, literature and rhetoric.4 These fellows simply were not in the least inclined to think like modern-day political scientists or our so-called policy wonks; hence, they also acted differently. In other words, everything points to the imposition, whole hog, of literary and rhetorical predilections, along with the Roman aristocratic code of honor, on the terrain of policy making.5 The result is, to use the example already mentioned, a striking list of topics not raised in the discussion reported by Herodian, e.g., the relative merits of the Danube as a frontier, the cost of the war, the resources potentially available for it, and the conceivable economic benefit of withdrawal. Mattern also points out the apparent utter miscalculation of the distance to the ocean. Commodus’ friends, then, did not raise what we would call practical problems; they instead worried variously about image. And, as Mattern says, while Herodian’s reportage of such situations is open to suspicion, there is nothing to suggest that the arguments here recorded are at all extraordinary.6 Rather, as she demonstrates throughout this chapter, Claudius Pompeianus is made to have spoken, within the Roman context, in absolutely conventional terms. Having adumbrated, then, the aristocratic values that ought generally to have guided Roman foreign policy, Mattern turns to some rather more concrete topics before returning in detail to this same subject in her fifth chapter (called “Values”).

The business of foreign affairs invariably confronts one with new places and odd peoples; and some degree of curiosity about these locales and folks should logically manifest itself among any group of bureaucrats making decisions in this realm. Therefore, to know what elite Romans thought, and thought they knew, about strangers and strange lands occupies the second chapter of Mattern’s book (“The Image of the World”). As she shows, both geographical and ethnographic thinking were intimately bound up with the literary genres geography and ethnography, which in themselves tended to be highly traditional and not necessarily much concerned with (again, what we would call) accuracy. Mattern argues, for example, that the average senator will have derived his geographical information chiefly from something like Pomponius Mela’s De chorographia, a work whose description of Britain (e.g.) “… seems to owe more to Eratosthenes than to the campaigns of Caesar; and the latter were already, by his (Mela’s) time, nearly a century old” (p. 65).7 But even in a text like Strabo’s Geographia, which Mattern appears to rate better (“… exhaustively researched and more up-to-date …” p. 65), we should not forget that a reasonably careful and accurate account of one region based largely on autopsy, might (without comment) stand right alongside the description of another territory utterly dependant on an earlier, and markedly inaccurate, periplous.8 In any case, Mattern reveals an entrenched pattern of faulty geographical understanding on the part of the Roman elite.9 What of course strikes us moderns as especially odd is the fact that military campaigns might time and again be undertaken with such flawed — and, to our way of thinking, needlessly flawed—geographical knowledge as foundational; but they were.10 Roughly the same kind of thinking functioned where foreign peoples were concerned. Ethnographic stereotypes were formed early on in the literature, were passed down from author to author, and came thence to a general readership. These stereotypes thus rested firmly in the Roman aristocratic mind, and as a result, a habitually creative ethnographic literature distinctly influenced real-world expeditions. An example is Aelius Gallus’ invasion of Arabia Felix: the literary tale of this region’s wealth, as opposed to (say) contemporary travel accounts, enticed Augustus and Gallus into a doomed military adventure. (pp. 78-79) In sum, the Romans often enough had the sort of information that could enable them quite effectively to mount a particular campaign or action (for example, maps and itineraries, discussed early in this chapter, facilitated that) — and they not infrequently used this information.11 However, they seem not to have apprehended their world generally in the kinds of terms that might have fostered something like grand strategy, at least not in any modern sense.

The next chapter (“Strategy”) seeks to describe, “… what determined where and how they [Roman soldiers] were deployed, and what the Romans hoped to accomplish with their military resources” (p. 82). Mattern prepares her chief argument with a series of points. The Roman military force was relatively small, this due to financial constraints and a reluctance to use the draft. As a result, any campaign would likely involve significant transfers of troops; nor could such movements be conducted with terribly great speed. All this meant, in turn, that one area would be deprived of soldiers, and thus become highly vulnerable, so as to allow fighting elsewhere. Of course, emperors nevertheless undertook large campaigns. The general shortage of troops was complicated by the fact that once conquered, a particular territory might require a fairly large concentration of soldiers to keep the peace. Having described such real constraints, Mattern proceeds to the actual defenses of the empire. She first accepts the recently emerged position regarding boundaries, attributing “… the final shape of the Roman frontiers to unconscious or accidental forces” (p. 114). The next step is to argue that the literary sources “… virtually do not recognize the idea of defensible frontiers” (p. 115). Departing from a passage in Themistius ( Or. 10.138), where the following is said explicitly, Mattern argues that the chief tool used by the Romans in defending their empire was not a physical construction of either nature or man but rather a psychological construct. Fear, she argues, as did Themistius, could stop the Scythian in a way a river or wall never could, never would, never was intended to. Mattern’s argument is that rather than to seek out or to create scientifically considered, defensible boundaries, the Romans worried most, and indeed quite consistently, about instilling an utter respect for those boundaries they did happen to establish.12 They could not, for various reasons, guarantee that these borders would resist transgression; nor did the Romans seriously try to do so. However, they did attempt to ensure that any and every breach of their territory would be met with terrifying reprisal, regardless of whether that came sooner or later. The inevitability and the terribleness of this revenge, it was thought, was the ideal protection against outsiders.

Chapter 4 (“Income and Expenditure”) deals with three distinct, though closely related, topics. A first section looks at the aristocrats who ran the empire. These men managed significant holdings of private property. Working, then, on the assumption that the economic principles guiding them in their private affairs probably operated also in their public dealings, Mattern initially examines the way in which elite Romans approached estate management. Profit could be a motive in this arena; more important, though, was stability, which would in turn allow the aristocrat to concentrate on those things crucial to him — mainly politics.13 We move thence to the army, its cost, and its fiscal management. The terrain here is rocky indeed, and Mattern negotiates it with caution. Her conclusion is important, worth quoting (p. 142): “Purely fiscal thinking, we may conclude, was rare. And the imperial budget, at first glance a relatively simple and straightforward system, emerges as a complex entity subject to sometimes extreme pressures — an aristocratic ethos that condemned almost every means of raising money; a provincial populace sometimes willing to risk outright revolt rather than submit to increased taxation; a dependence on raw bullion, of which the supply might fluctuate wildly; incalculable expenditures on conspicuous consumption and public handouts necessary to maintain the image and status of the emperor; and so forth. Within this system emperors innovated, sometimes ingeniously, sometimes — as in the case of the late devaluations of the denarius — with cavalier disregard for what they must have perceived as potentially disastrous consequences. It is on this precarious system that the empire’s long-term military capability depended.” A second portion of this section considers the actual costs of warfare. In short, it was very expensive; yet from what we can see, purely military imperatives did not always govern expenditure in wartime. That is, large outlays on image-building projects, and these sometimes undertaken precisely while expensive wars were being fought, often made good sense to Roman emperors.14 From here we proceed in a last section of the chapter to the winnings. They too could be large: bullion, slaves, minerals (and taxes and indemnities) to be mined in the newly acquired territories, etc.15 Still, to adduce pure money-making as a cause of war was theoretically not acceptable; there were also wars waged that had little if any hope of lucrative outcomes. In short, stability seems to have triumphed over sheer money making, as did questions of image, honor, and the like over pure financial exigencies.

Rather than going into any detail, I shall quote Mattern, and thus roughly sum up the argument of her last chapter (“Values”) — and, really, of the book: “… Rome was willing to do whatever it took to retain its territory, sometimes despite severe losses and often with very harsh reprisals… The Romans, in describing this struggle, do not frame their analyses mainly in “rationalizing” economic or geopolitical terms… Instead, the Romans perceived their struggle for empire in very different terms: crucial were issues of psychology, the emotions of terror and awe that they hoped to produce in the enemy; and the moral and status issues, such as the need to repress superbia, avenge inuriae, and maintain the honor or decus of the empire. It was on these things that, as they believed, their security depended; it was for these that they fought” (p. 194).

Now, as Mattern herself intimates, the publicly alleged causes for wars are not, perhaps, always those that really bring one side to attack the other. And as she also grants, her argument against economic or geopolitical considerations having driven the Roman imperial machine is principally e silentio. Thus, in the end it might be possible to hold that Roman emperors and their friends really did discuss, though in such a fashion as to elude the historical record, such things as finances and strategically defensible boundaries before they initiated hostilities; and perhaps they said the things we do see them saying just for public consumption. Or, perhaps one could argue that at some original, pivotal point in time, or on some subliminal level, the Roman aristocrat realized that it would be very difficult (for various reasons) to defend all frontiers at all times against all marauding barbarians; and so, that aristocrat preferred to avoid unpleasant realities and to talk in more congenial, in more idealistic terms. Mattern sees the problem and confronts it directly (p. 222): “Perhaps the truth about Roman imperialism is not, after all, hidden from view, shrouded in rhetoric and myth. Perhaps the rhetoric, the myth, is the reality.” It seems to me that those, who might want to contravert her position, must bear the burden of proof. This is an excellent book, and should not be ignored by anyone interested in how the Romans ran their world.

Having said that, let me offer just a bit of minor criticism. There are places where the progression of the argument is probably cogent, but where the reader must do too much of the work in making it so. For example, the logic of the third chapter, where we move from a discussion of the size of the Roman army to the policy of instilling fear in the barbarians, is there; but Mattern does not bring it out as clearly as she might have. It is also surely reasonable to suppose, as is done in chapter four, that the tenets of estate management might well pop up again at the intersection of foreign policy and finances. However, having laid out aristocratic concerns in the running of large farms, she deserts the subject, and its integration with the rest of what is argued here is left (so far as I can tell) to the reader. This kind of thing nowhere invalidates Mattern’s points; however, it sometimes makes grasping them properly difficult. There is also some repetition of argument. Most significantly, the description of what might be called elite mentality as to foreign policy is split between the first and the fifth chapters. These should really be read in conjunction in order to get the full force of Mattern’s argument in this regard. In the notes above, I have made a few random additions to the material cited by Matter, and it is possible to make more such — though I stress the fact that these are purely random. Nor do I notice any missed item that would affect the book’s main arguments.16


1. E. Luttwak, The Grand Strategy of the Roman Empire (Baltimore 1976), and C.R. Whittaker, “Where Are the Frontiers Now?” in D.L. Kennedy (ed.), The Roman Army in the East (Ann Arbor 1996) 25-41. Writing about the frontiers continues, e.g.: D.F. Graf, Rome and the Arabian Frontier from the Nabataeans to the Saracens (Aldershot 1997); D. Cherry, Frontier and Society in Roman North Africa (Oxford 1998).

2. R. MacMullen, “The Roman Empire” in Ancient History: Recent Work and New Directions (Claremont 1997) 97.

3. Mattern does not, in fact, quote (or mention, so far as I can determine) this last bit of Pompeianus’ argument; but since it seems relevant, I here include it.

4. The taste for literature at Rome is something often ascribed solely to the elite. A compelling argument, however, for a rather more sophisticated level (generally) of popular culture is now made by N. Horsfall, La cultura della plebs Romana (Barcelona 1996). I thank my colleague Larissa Bonfante for bringing this excellent monograph to my attention.

5. With regard to the business of Roman honor, and its widespread influence on governmental practices, the important book of another MacMullen student should not be missed [Mattern knows it but could probably have brought it into the discussion more than she does]: J.E. Lendon, Empire of Honour. The Art of Government in the Roman World (Oxford 1997). His book and Mattern’s may quite usefully be read together for they both reveal effectively the same particular kind of elite mentality where the functioning of the empire was concerned. One minor reservation here. To say that “Roman foreign policy was conducted by wealthy but otherwise relatively ordinary men” (p. 2) is to do, I think, some injustice to these sorts. We may agree that most Roman senators most likely had no specialized training, at least, none of the sort we might consider directly useful to their administrative duties; but they were certainly, in their own terms, as well as in terms we can fully appreciate, rather extraordinary. A former Camden Professor has described the type: “Henry Pelham…belonged by birth to that circle of great families from which England has drawn, century after century, a succession of fit persons duly qualified to serve God in Church or State. From his ancestry he drew that political tradition and intuitive grasp of the principles of government which gave him a sure understanding of the growth and working of the institutions of the ruling race of the ancient world…” Pelham chose to serve merely as professor in Oxford; but he might just as well have decided otherwise. For the quote, see H. Stuart Jones, Fresh Light on Roman Bureaucracy. An Inagural Lecture Delivered before the University of Oxford on March 11, 1920 (Oxford 1920) 4.

6. On Herodian, see now H. Sidebottom, “Herodian’s Historical Methods and Understanding of History” ANRW II 34.4 (Berlin & New York 1998) 2775-2836, with 2816-2819 on the speeches.

7. On Mela, see now K. Brodersen, Kreuzfahrt durch die Alte Welt (Darmstadt 1994), and now F.E. Romer, Pomponius Mela’s Description of the World (Ann Arbor 1998)with esp. 22-27 for the present matters. And on Mela’s continued influence in geographical education, id., ” Principia geographiae. Antike Texte im frühen Erdkundeunterricht” Anregung. Zeitschrift für Gymnasialpädagogik 42 (1996) 29-43.

8. See M. Knight, A Geographic, Archaeological, and Scientific Commentary on Strabon’s Egypt (Geographika, Book 17, sections 1-2) (diss. New York University 1998) 14-18 and 20-24, delineating careful, personal investigation of Egypt by Strabo, but only indirect and inaccurate knowledge of the rest of North Africa (in particular Libya), which completes his book 17. And note that while Strabo seems to have used a periplous for his description of Libya, he generally disdains that genre (Knight, op. cit. 21 n. 63). Strabo has elsewhere been criticized for using long outdated accounts of places he describes; e.g., R. Baladié, Le Péloponnèse de Strabon. Étude de géographie historique (Paris 1980) 13.

9. To add one further example to Mattern’s list, Polybius reached the opinion that the Pyrenees ran north-south, rather than east-west, since they extended from the Mediterranean to the Atlantic; and this opinion, with the odd exception of Josephus, prevailed through the rest of antiquity. See F. Beltrán Lloris & F. Pina Polo, “Die Pyrenäen als Grenze und die geographische Sichtweise der Römer” in E. Olshausen & H. Sonnabend (eds.), Stuttgarter Kolloquium zur historischen Geographie des Altertums 5 (1993): “Gebirgsland als Lebensraum” (Amsterdam 1996) 210-211. And as they point out, this was also Strabo’s (3.1.3) position on the matter (cf. Knight, op. cit. 8 on Strabo and Polybius). Mattern notes a similar mistake with regard to the Carpathians. (p. 61)

10. Mattern pp. 60-61 discusses, e.g., various campaigns ostensibly undertaken in order to extend Rome’s control to the northern ocean — campaigns doomed to almost certain failure because of the stunning miscalculations of the distances involved conjoined with the stubbornly persistent misimpressions about the intervening territory and peoples.

11. One might also think here of the creation of the eastern limes in Germany. Given the fact that it runs a perfectly straight line over more than 81 km (between Welzheim and Walldürn), we must assume a painstaking survey of the land — land which was, of course, at the moment of survey, beyond the frontier. The action was supervised by C. Popilius Carus Pedo, then governor of Germania superior, and probably completed in about A.D. 155. See G. Alföldy, “Caius Popilius Carus Pedo und die Vorverlegung des obergermanischen Limes” FBW 8 (1983) 55-67. The ethnographic and geographic traditions about Germany, as most recently portrayed, seem in harmony with Mattern’s overall picture (though a bit more optimism is displayed here): J.B. Rives, Tacitus, Germania (Oxford 1999) 35-41.

12. One of the factors involved in creating this fear must have been the demarcation of the border. That is, the barbarian had to know where he was leaving the barbaricum and entering the forbidden, civilized territory of Rome. For some interesting comments in this regard, see D. Potter, “Empty Areas and Roman Frontier Policy” AJPh 113 (1992) 269-274. Cf. also, for rivers and mountains as boundaries of the empire, the articles “Fluss” (F. Schön, pp. 149-150) and “Gebirge” (H. Sonnabend, pp. 161-162) in H. Sonnabend (ed.), Mensch und Landschaft in der Antike. Lexikon der Historischen Geographie (Stuttgart-Weimar 1999). Note also that foreign affairs was not the only place where fear was at work in running the Roman world. Dio Chrysostom, for example, says explicitly that the ideal king (emperor) ought not function by means of terror ( ἔκπλήξις) or fear ( φόβος), but rather that reverence ( αἰδώς) of him is a better and stronger tool ( Or. 1.25). Presumably, there was a reason for stressing this. Note also that the emperor, even if he did not intend it, could still be terrifying, especially in his role as judge. See Epict. Diss. 2.19.17-18 or CJ 7.67.1 (A.D. 293). Judges generally were apparently rather frightening — at least in part, I guess, by intention: Damigeron-Evax [ap. R. Halleux & J. Schamp, Les lapidaires grecs (Paris 1985)] LI.2, p. 283, Hoc lapide (i.e., the lapis arnosteatites) quicumque usus fuerit mitescit iras regum et iudicum furias, cum habueris eum strictum in manu sinistra.

13. Here, Mattern has missed a large body of important work by Dennis Kehoe, whose conclusions generally add much support to her arguments. See his: “Allocation of Risk and Investment on the Estates of Pliny the Younger” Chiron 18 (1988) 15-42; “Approaches to Economic Problems in the ‘Letters’ of Pliny the Younger. The Question of Risk in Agriculture” ANRW II,33.1 (Berlin & New York 1989) 555-590; Management and Investment on Estates in Roman Egypt during the Early Empire (Bonn 1992); rev. of Rathbone, Economic Rationalism in JRA 6 (1993) 476-484; “Investment in Estates by Upper-Class Landowners in Early Imperial Italy. The Case of Pliny the Younger” in H. Sancisi-Weerdenburg & H.C. Teitler (eds.), De Agricultura. In memoriam P.W. de Neeve (Amsterdam 1993) 214-237; “Investimento e sicurezza del possesso nell’affitto agrario romano” in E. Lo Cascio (ed.), Terre, proprietari e contadini dell’impero Romano (Naples 1997) 61-73; Investment, Profit and Tenancy. The Jurists and the Roman Agrarian Economy (Ann Arbor 1997).

14. But as Mattern’s discussion of the via nova Traiana in Arabia shows (pp. 114, 145, 148), it is not always easy to distinguish practicality from image-building, nor is it at all clear that an emperor like Trajan did so. In large part, too, this is precisely Mattern’s point, even though the sword can, here and there, cut the other way.

15. Another impressive building can now be added to the list of constructions put up ex manubiis, which in this case must have been pretty substantial: the Colosseum was paid for by the spoils from the Jewish War. See G. Alföldy, “Eine Bauinschrift aus dem Colosseum” ZPE 109 (1995) 195-226. The original dedicatory inscription seems pretty certainly to have been: Imp. Caes. Vespasianus Aug. / amphitheatrum novum / ex manubis fieri iussit.

16. At p. 18, Mattern says that Egypt was turned over to an equestrian because it was too important to be entrusted to a senator. Note, however, Brunt’s argument that Augustus’ original decision was ad hoc, the province having been given to the equestrian Cornelius Gallus because he happened to be there, and in charge, at the moment when provincialization was decided upon. P.A. Brunt, “Princeps and Equites” JRS 73 (1983) 62-63 [Mattern does have this article in her bibliography]. See further on this problem, and with reference to Brunt’s arguments, W. Eck, Die Verwaltung des Römischen Reiches in der Hohen Kaiserzeit. Ausgewählte und erweiterte Beiträge. 1. Band (Basel 1995) 40-42, 99-100. In considering the reasons for which particular men were chosen for governmental posts (cf. p. 20), two articles (in particular) by P. Leunissen should also be taken into account: “Herrscher und senatorische Elite. Regierungsstil und Beförderungspraxis im Zeitraum vom 180-235 n.Chr.” SIFC 10 (1992) 946-954; “Conventions of Patronage in Senatorial Careers under the Principate” Chiron 23 (1993) 101-120. With regard to the records kept by provincial governors (pp. 68-69), including military records, see R. Haensch, “Das Statthalterarchiv” ZRG 109 (1992) 209-317. As for the size of the imperial legion (pp. 82, 104-105), see J. Roth, “The Size and Organization of the Roman Imperial Legion” Historia 43 (1994) 346-362. His book has also now appeared (Mattern had access only to the dissertation): The Logistics of the Roman Army at War (264 BC – AD 235) (Leiden 1998). A number of the essays in a volume that has just appeared, and that was obviously not available to Mattern, are relevant to a number of the matters discussed by her: A. Goldsworthy and I. Haynes (eds.), The Roman Army as Community (London 1999). Also relevant to, and generally supporting, the arguments of Mattern is now W. Eck, “The Bar Kokhba Revolt: The Roman Point of View” JRS 89 (1999) 76-89.