Bryn Mawr Classical Review

Bryn Mawr Classical Review 2010.01.49

Edward N. Luttwak, The Grand Strategy of the Byzantine Empire.   Cambridge, MA/London:  Belknap Press of Harvard University Press, 2009.  Pp. xi, 498.  ISBN 9780674035195.  $35.00.  

Reviewed by Anthony Kaldellis, The Ohio State University (
Word count: 2796 words


The Grand Strategy of the Byzantine Empire is not as bold in its assertions as its controversial predecessor, The Grand Strategy of the Roman Empire (1976). The basic idea, that the Byzantines preferred persuasion and co-option over decisive battles, is well established. The book's strength lies in the conceptual apparatus of strategic theory that Luttwak brings to the Byzantine military manuals in the third part, and it will provide an attractive introduction to some aspects of diplomatic and military history. Interest in Byzantium by scholars in other fields is certainly to be encouraged, but Luttwak is no Byzantinist (for example, he relies entirely on translations for the primary sources, even online translations, and too often on outdated scholarship). The following critical review is written from the standpoint of a Byzantinist who requires a closer engagement with the field by anyone who would make a "grand" argument about Byzantium.

Part I lays out the fundamental axioms of Byzantine strategy and offers a historical argument for its origin. The Byzantines avoided the risks of decisive military confrontations and sought rather to contain or co-opt their enemies, in order to preserve their own soldiers, who would be needed to contain the next enemy, and because the enemy of today was a potential ally against the enemy of tomorrow (given the waves of barbarians that the empire faced during its long history). Ideally, barbarians should be paid to go away or attack others, rather than fought. Luttwak traces the emergence of this policy to the confrontation with Attila and the Huns. Hunnic armies were fast, large, and almost impossible to destroy in the field, given their tactics and use of the composite reflex bow. It was then that Byzantine diplomacy came to the fore, and it would dominate their strategy thereafter. Moreover, it was in response to this threat that the Byzantine army became predominantly cavalry-oriented, with mounted archers replacing the heavy infantry of the Roman past.

Luttwak is generally stronger at formulating the principles of strategy than making historical arguments. Certainly the experience of Attila shaped the ongoing evolution of Byzantine strategy, but the case made here leaves too many questions unanswered. First, for a century before Attila the empire had been dealing with many Goths in a typically "Byzantine" way as with the Huns before Attila. Second, it is not clear, as Luttwak asserts (54-55, 61), that the eastern armies would have lost in battle against Attila. Luttwak has already described how the western armies (famously) defeated him in 451, though he downplays this as a temporary check (43-45). But his argument (following Iordanes) that the "proto-Byzantine" Aetius did not then utterly destroy Attila in order to use the Huns as potential leverage against the Goths indicates that it was a victory (44). He also ignores the defeat of the Huns before Toulouse in 439, where losses were so high that Attila could act against the East in 441 only in violation of a treaty and when Theodosios II had sent many units to the West, against the Vandals. The East, then, need not have regarded Attila as an "unmanageable threat" (12) or "an irresistible force" (78). It was cheaper to pay him off but he did not pose a threat to the empire's existence. This reflects a deeper problem in the assessment of Attila's significance, a point on which Luttwak dissents from his main sources, E. A. Thompson and O. Maenchen-Helfen, who regarded Attila as overrated. He became a figure of the imagination, but his handful of raids and ultimate goal (basically, extortion) did not change history. Luttwak invokes his presence in the Nibelungenlied and Icelandic sagas (18-19), but Attila's contemporary king Arthur shows clearly that late medieval literature is no guide to Roman history. Even he admits that "under Attila the Huns remained raiders rather conquerors [sic]" (36) and that they avoided combat, preferring localized attacks to "set the stage for. . . extortion" (38-39). The so-called empire of the Huns has recently been called "a protection racket on a grand scale," and "in direct encounters with the Roman army the Hun record is not particularly impressive."1

Another component of Luttwak's argument is open to objection, namely that after Attila the Byzantine army came to rely primarily on cavalry (20-21, 56; cf. 26: "the core of the army," 78: "the primary force," 260: "the dominant arm"). The cavalry doctrine emerged out of nineteenth-century generalizations about the knightly culture of the Middle Ages (whereas antiquity had citizen militias), and is based largely on one text, the preface of Prokopios' Wars, which Luttwak, like many before, duly quotes (57) and takes at face value. Certainly, cavalry became more important after 500, and was more prominent in certain kinds of operations, especially against mounted enemies, but the core of the Roman army remained infantry. This emerges from scholarship that Luttwak apparently did not consult,2 and is indicated by evidence that he himself presents later from the Byzantine manuals (300-301, 312, 349, 363-364 and 369-370) that presuppose mostly infantry armies, and go beyond the concession at 273: "even at the height of the cavalry era there was a need for some infantry." Maurikios' Strategikon is about cavalry operations (267) but this is misleading: it refers to a separate work on the infantry (2.2), possibly lost. The wars with the Avars in Theophylaktos are also cited by Luttwak as proof of the cavalry thesis (60), but the narrative is not explicit and seems to me to concern infantry legions instead. As for Prokopios' preface, I have argued that it should not be taken at face value, for it was part of his ironic stance toward Justinian, to which end it concocts a fantastic warrior type.3 This text offers no sound basis on which to reconstruct military history. Prokopios was, moreover, a partisan of infantry on grounds that Luttwak inadvertently reveals (293): enemy "cavalry could be readily halted by infantry in disciplined ranks, so long as there were enough bowmen to prevent the steppe archers from simply standing in front of them to discharge their arrows." Finally, Luttwak overlooks the possibility that the Byzantines learned cavalry skills from their eastern neighbors;4 here too he possibly overrates Hunnic influence.

Part II is on the instruments and context of Byzantine diplomacy. Chapter 3 treats envoys, focusing on late antiquity, and repeats the sophism that there were "no professional diplomats. . . no minister for foreign affairs" (107, also 6). It would be refreshing for someone to challenge this. A good place to start would be Justinian's long-serving magister officiorum, Petros Patrikios, who was more of a diplomat and minister of foreign affairs than many modern professionals. Surprisingly, he is not even mentioned in this book. A more rigorous comparison would probably weaken the argument that the magister could not have been "a proper foreign minister, for sheer lack of time" (108-109), given the number of bureaus under him. The same could be said about many modern ministries, and in some countries the office of foreign minister is held by the prime minister. The account of diplomatic immunity (101-105) overlooks the crucial exchange in Prokopios between Petros by Theodahad. Chapter 4 is a brief survey of the sacred attractions of Constantinople, but shirks an analysis of how religion promoted, or was used to promote, diplomacy. The dealings of Romanos I and Symeon would have been ideal for this purpose, but they are narrated in mostly untranslated sources. Chapter 5 on court ceremonies discusses some excerpts from the Book of Ceremonies. Chapter 6 on "dynastic" (recte diplomatic) marriages lapses into a list with little analysis, and misses a major recent monograph.5 Chapter 7 on "the geography of power" is a selective commentary on the forms of address set forth in the Book of Ceremonies for addressing foreign leaders. General background information is offered here and there, but no explanation as to why this moment was chosen, why this text, or how exactly the chapter contributes to the main argument. Only the discussion of the Pechenegs (158-161) seems strictly relevant, and here Luttwak turns to the De administrando imperio. He would have found more support for his thesis in that text.

There follow two focused discussions, dealing with the Bulgarians and the Muslims (Chapters 8-9). The first is a narrative survey of warfare and diplomacy, at times anthologizing sources. Luttwak loses sight of his main argument and delights in the details of campaigns (some of which seem to contradict his main thesis: see below). The narrative is discontinuous. It offers snapshots of relations with Symeon and jumps ahead to Samuel and Basileios II. Chapter 9 begins by discussing the tax systems of late Rome and Sasanian Iran, and then turns to the treatment of religious minorities by Byzantium and the Muslims. The relevance of this to strategy is unclear (nor of the note at 453 n. 24 on Luther's furor against the Jews). Perhaps we are meant to conclude that intolerance made minorities welcome the Arabs, though this is not strictly about strategy. Luttwak seems to be unaware that some of these later narratives of "treason" may have aimed to curry favor with Muslim overlords.6 We return to strategic analysis only at the end of the chapter, with the Seljuks. Overall, part II of the book is the weakest in terms of analysis and originality.

Part III is the most successful in the book, consisting of five chapters (10-14) that survey the military manuals from antiquity to the eleventh century, and one (15) that examines the strategic dimension of Herakleios' defeat of Persia. At first sight the survey chapters might seem to paraphrase the military treatises; in fact, Luttwak uses his expertise as a strategic theorist to good effect, bringing out the logic behind the texts' recommendations. I recommend the discussion of the concept and practice of elastic defense (343-345), where comparative evidence is deployed well. At 326 Luttwak dismisses the overall value of Greek fire. At 387-392 he surprisingly omits Kekaumenos' potentially treasonous advice to foreign border lords on how they might maintain independence from Constantinople. Unfortunately, Luttwak does not discuss in chapter 13 how naval strategy was integrated with land warfare.

The book's strengths reflect its author's expertise in strategic theory, but from the standpoint of Byzantine studies here are too many annoying errors throughout.7 More troubling are problems in the book's methodology. Other than the manuals, the literary sources are taken at face value (as we saw with Prokopios). For example, there is no analysis of the ethnographic conventions behind Ammianus' account of the Huns.8 Sidonius, we are told, was "not led astray by poetic needs -- he is describing [riding skills] quite accurately" (28), but elsewhere "panic" or "poetic needs" are allowed (43). This is about the extent of literary sophistication brought to bear on the sources. Contra 63, the reasons why Priskos wrote his account of the embassy to Attila had little to do with the context (personal, literary, or ideological) of Tacitus' Germania. Luttwak's dismissal of Prokopios' account of the plague on the grounds that it imitates classical models (87-88) is about sixty years out-of-date;9 he then paradoxically endorses that account because other sources confirm it. He viciously dismisses Said and all classicists who study the representation of "the Other" in literature (448 n. 1) -- "an evil fashion." At 252 he laments the takeover of philosophy by linguistics, preferring that it instill "tranquility." There is no Linguistic Turn here, in more senses than one.

There is also no analysis of the interplay between command structure and strategy nor of the empire's military organization and the nature of its units (they are briefly mentioned at 178, with a reference to the entry in the Oxford Dictionary of Byzantium). We are left with a vague impression of the actual army with which the grand strategy was implemented. There are also no discussions of the criteria for appointing officers; of the relation between the army and civilians; of how the Byzantines perceived war in the first place; of their idea of Holy War (mentioned in a parenthesis at 412 and a note at 470), which would surely influence strategy; or of changes, gradual or dramatic, in strategy between 400 and 1100 (a glimpse only at 369-370, and an acknowledgment in the last paragraph: 417-418). One passage implies a more evolutionary view (13: "it was only under Herakleios...that the distinctive grand strategy...was fully formed"), but this methodological challenge is not taken up. At 112 the shift from force to diplomacy is upheld as the turning point between (later) Rome and Byzantium, but Luttwak states at the top of the same page that emperors from Augustus to Marcus Aurelius "preferred gold to iron whenever enemies were more cheaply bought off than fought." So despite the showcasing of Attila, the transition from Rome to Byzantium remains hazy.

Strategic doctrine prevails over history, and is mostly static. Except for Nikephoros I, who "decided to rely entirely on his own military strength" (177) and so came to grief (see 183 for his error), there is no analysis of the generals who ignored what Luttwak postulates as Byzantine strategy but succeeded. He thinks these were only Justinian and Basileios II (284), but the long reconquest produced many who preferred war over diplomacy (Kourkouas, Nikephoros Phokas, Ioannes Tzimiskes, etc.). The most important methodological problem, however, is that Luttwak seems not to know that the Byzantines fought civil wars about as often as they fought foreign ones,10 and that, accordingly, their command structure and strategy were designed to cope with internal threats, both real and imagined. Even the very emergence of their distinctive strategic mode may have had more to do with playing against each other the Gothic warlords absorbed into the system in the late fourth century than it did with Attila.

The exposition is punctuated by weird statements and outdated notions. There is no validity to the claim that Byzantium after 1259 was a Greek kingdom rather than an empire (6, 56, 70, 234). Iconoclasm was not a struggle between "the Hellenic proclivity for imagery" and "abstract Jewish monotheism" (118). There was no thing such as "European civilization" in the early Middle Ages (124). No regional "zone" rejected Hellenism in late antiquity and non-Greek-speakers did not overlap with non-Chalcedonians (410). Why are Penelope's suitors called "the Ithaca provincials"? (25) What does it mean exactly that Belisarios "is still remembered today by unlettered Romans" (80), that the modern names of rivers are more "accurate" than their medieval ones (34, 42), or that for the Byzantines only an ancient Greek text could be classical but not a Roman one (249)? And is the following a joke? "Christianity certainly helped to combat prejudice--not only because of its universal embrace but also because it dissuaded its followers from bathing, and therefore removed the barrier of smell that greatly inhibited Roman intimacy with barbarians" (145).

Such statements are made parenthetically and do not affect the argument. But there are times where substantive disagreement is possible, for example that "war cries and arms waving served exactly the same function as displays of nuclear weapons during the Cold War" (111), or "that almost all the Byzantines we know of were intensely devout Christians is beyond question" (113). Contra 108, late Roman arms factories were not a "military-industrial complex" as they were owned by the state. Contra 137, the reluctance to marry princesses to foreigners had little to do with "the claimed position of the emperor as God's viceroy on earth. . . who must exist on a higher plane than all other rulers." This is a function of the modern "theologization" of Byzantium. Konstantinos VII gives a different excuse (quoted at 139), premised on national differences. If Greeks and Byzantines did not know the word "strategy" (412), then what did the title Strategikon and its like mean? At 129 we are told that imperial power was "unlimited by laws" but at 280 that "it was regulated by laws." Luttwak is more confident that the sixth-century plague "wrecked the entire state and its army" than are the historians on whom he relies (13, 89-92). The jury is still out on that one. At 125 Luttwak mocks the court hydraulic devices as "little more than childish foolery," but a few pages later he has observed that the "immanent presence of power" at political capitals is "scorned only by those with no access to it" (129).

Luttwak has much that is useful to say about the Byzantines' strategy, especially when discussing their military manuals. His book, however, is not so strong when it comes to Byzantine history. It is out of touch with the state of the field. More precision and better information would have smoothed the union of history and strategic theory that is attempted here.


1.   C. Kelly, Attila the Hun: Barbarian Terror and the Fall of the Roman Empire (London 2009) 47 and 227; cf. 49: "their relatively brief domination of the middle Danube."
2.   W. Treadgold, Byzantium and Its Army, 284-1081 (Stanford 1995) 50-53, 108. The absence of Treadgold's works is a striking omission, given that he is one of the leading Byzantine military historians. J. Haldon, Warfare, State, and Society in the Byzantine World, 565-1204 (London 1999) 193-196 also paints a mixed picture. E. McGeer, 'Infantry versus Cavalry: The Byzantine Response,' Revue des études byzantines 46 (1988) 135-145, here 136 refers to "the underestimated role and importance of infantry in Byzantine armies."
3.   A. Kaldellis, 'Classicism, Barbarism, and Warfare: Prokopios and the Conservative Reaction to Later Roman Military Policy,' American Journal of Ancient History n.s. 3-4 (2004-2005 [2007]) 189-218. This is one of the few points on which Luttwak (at 274) disagrees with J. Haldon, Warfare, State, and Society in the Byzantine World, 565-1204 (London 1999) 216, who likewise called Prokopios' mounted archer/lancers "something of a myth."
4.   A. D. H. Bivar, 'Cavalry Equipment and Tactics on the Euphrates Frontier,' Dumbarton Oaks Papers 25 (1972) 273-291.
5.   A. G. Panagopoulou, Oi diplomatikoi gamoi sto Byzantio (6os-12os aionas) (Athens 2006), 500 pages.
6.   Most recently in H. Kennedy, The Great Arab Conquests: How the Spread of Islam Changed the World We Live in (Da Capo Press 2007).
7.   E.g., 34-35: the Hunnic raid through the Caucasus was in 395 not 399, and Luttwak does not know G. and M. Greatrex in Byzantion 69 (1999) 65-75. 36: Rome and Persia did not agree to garrison the Caucasus in 562, but over a century earlier. 78: Theodosius I for Theodosius II. 83: costitutiones (twice). 84: Justinian's Novellae were not "collected in a separate compilation" -- that is an illusion created by the modern edition. 86: Justinian was not named Flavius as a child -- that was an imperial name. 98: Sodgians. 111: apkrisarion. 128: the first known Byzantine-Arab exchange of prisoners was not in 805 but 769: Theophanes, Chronicle a.m. 6261 (p. 444). 131: Stamford Bridge is not in Greater London. 186: 1913 for 913. 237: dates the Strategikon to "the late part of the reign of Justinian" (elsewhere, e.g., 267, is it attributed to the reign of Maurikios). 266: by the end of the fifteenth century "Greek was still an unknown language even for the most learned scholars of western Europe." 338: the Byzantine counteroffensive did not begin in the middle of the tenth century but two centuries earlier. 391: archenos. 394: Khusrau II never went to Constantinople. 436 n. 23: Averil Cameron Claudian, Poetry and Propaganda should be Alan Cameron, Claudian: Poetry and Propaganda. Harald's epithet Hardrada is translated as "stern counsel" at 155 and "hard ruler" at 131 (and his story is told twice in different chapters). At 40 the Slavic "monoxyla (= one tree = dugouts)," but at 463 n. 20 we get "= single trunk. But they were not dugout canoes." At 226 we are told that Mantzikert was "a catastrophic strategic defeat" but at 162 that "it was not a catastrophic military defeat" because "the catastrophe came in the aftermath."
8.   See C. King, 'The Veracity of Ammianus Marcellinus' Description of the Huns,' American Journal of Ancient History 12 (1987) 77-95; C. Kelly, Attila the Hun: Barbarian Terror and the Fall of the Roman Empire (London 2009) 17-28.
9.   See A. Kaldellis, Procopius of Caesarea: Tyranny, History, and Philosophy at the End of Antiquity (Philadelphia 2004) 26-27, citing the relevant studies.
10.   For a tabulation, see W. Treadgold, 'Byzantium, the Reluctant Warrior,' in N. Christie and M. Yazigi, eds., Noble Ideals and Bloody Realities: Warfare in the Middle Ages (Leiden 2006) 209-233.

Comment on this review in the BMCR blog
Read Latest
Index for 2010
Change Greek Display
Books Available for Review
Bryn Mawr Classical Commentaries

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

HTML generated at 09:33:26, Wednesday, 27 January 2010