This book is less scholarly, original, incisive, and careful than the same author's much shorter The Grand Strategy of the Roman Empire From the First Century A.D. to the Third (Baltimore, 1976). The new book's overall thesis, that the Byzantines tried to avoid war whenever they could, especially through diplomacy, has long been known to Byzantinists but is certainly worth explaining to a wider audience. Apart from this main point, however, Luttwak's treatment of strategy is undermined by a poor grasp of the army as an institution.
Luttwak treats the Byzantine Empire as if it began in 395 and ended in 1204. While such beginning and ending dates are defensible, they need explanations that are not given here. One problem is that 395 comes about a century after the end of Luttwak's earlier book, in which he assumes more unfavorable defensive conditions for Rome in the third century than he now assumes for Byzantium in 395, without explaining how matters improved in the interim. He implicitly defends his starting date by claiming that the fifth-century attacks of the Huns were what set Byzantium in a permanently defensive posture: "[Attila's] greater-than-expected threat had evoked a series of improvised reactions that soon combined into something much broader, and very long lasting as it turned out" (48). Although the attacks of the Huns may have been one cause of Byzantine defensiveness, Luttwak takes advantage of his starting date to overlook other plausible causes from the period before 395: the third-century invasions, the failure of Julian's Persian expedition in 363, the empire's defeat at Adrianople in 378, and above all the Byzantines' acceptance of Orthodox Christianity, with its extremely negative view of warfare. Later Byzantines often invoked the advice of St. Basil of Caesarea (died 379) that soldiers who killed an enemy in battle, even in a fully justified war, should be barred from the eucharist for three years. Surely this was a major cause of the Byzantines' avoidance of warfare.
Worse still, throughout this book Luttwak gives inadequate attention to the numbers, organization, and pay of the army, all of which receive more attention in his book on Rome. While in that book he quoted rough estimates for the size of the Roman army (300,000 men in AD 23, 350,000 in AD 70), in this book he mentions none of the five totals our sources give for the later army (389,704 probably in 284, 581,000 probably in 312, and after the fall of the Western Empire 150,000 in 559, 80,000 in 773, and 120,000 in 840). We also have figures for military pay and two overall payroll figures (for 641 and 867). These statistics, though they need some interpretation to be properly understood, are supported by many numbers for parts of the army, including some from official government documents of the tenth century. On p. 362 Luttwak correctly quotes a tenth-century military manual that says all Byzantine soldiers were registered in comprehensive lists. Yet he shows only vague and often mistaken conceptions of how many soldiers the Byzantines had, how they were organized, and how much they cost.
Disregarding ample evidence to the contrary, he writes on p. 285, "An acute shortage of combat-ready troops was indeed the perpetual condition of Byzantine warfare." About the tagmata, the units stationed in and around Constantinople, he says on pp. 75-76, "[I]t would have needed more than an ordinary tagma of a thousand to fifteen hundred men to guard such an extensive system of fortifications" as Constantinople's; but on pp. 178-79 he says, "Each tagma had an establishment of four thousand men, on paper at least, divided into two mere or turmae of two thousand, each in turn divided into two drungi of one thousand, each made of five bandae [an error for banda, the plural of bandum] of two hundred, in two centuries." Though neither passage has a footnote, the first is evidently based on a mistaken guess that John Haldon made in 1984 but tacitly abandoned later. The second sentence must be derived from a reliable Arab report showing that in 840 the tagmata totaled 24,000 men, more than enough to man the walls of Constantinople.
Luttwak apparently believes that the Byzantine army before the eleventh century was much smaller than it actually was. On p. 182 he says, "Nikephoros [I] had gathered every mobile force to overwhelm the Bulghars, so there was nothing left to stop them from seizing Constantinople after his ruinous defeat" in 811. Yet Nicephorus must have left most of his mobile forces behind, because the Byzantines could fight major battles against the Bulgars in 813 (another defeat) and 816 (a victory that brought a thirty-year peace), and the Bulgars never seriously threatened Constantinople at any date. Luttwak is mistaken that Bulgaria could be "a deadly threat" to the empire (171) or that "It was standard practice [for the Byzantines] to attack the Bulgarians whenever there was no active campaigning against the Arabs" (188). The Byzantines had a much stronger army than the Bulgarians, and Bulgarian aggression provoked all of the few Byzantine attacks on Bulgaria. The evidence does not even support the idea that the emperor Basil II (963-1025) originally planned to conquer Bulgaria, as Luttwak assumes. Basil seems in fact to have retaliated against Bulgarian attacks with such overwhelming force that the Bulgarians surrendered when he was demanding only that they agree to a protectorate. In the case of Basil, who was also content with a protectorate over Arab Aleppo though he could easily have conquered it, Luttwak seems to have underestimated the Byzantines' dislike for taking the offensive.
On p. 348 Luttwak says "The empire had no expendable men" in the tenth century. In reality it had so many expendable men that it was already asking them for cash contributions instead of service, and in the eleventh century it massively debased its coinage to reduce its expenses for the many soldiers it did not need. On p. 221 we read that before the Battle of Manzikert in 1071 the empire had no "organized frontier defenses" but only "akritai, the mostly Armenian border warriors, much celebrated in song and romance," who had held the frontier "for three centuries against Arabs" by "counter- raiding across the border" that was "futile against nomads" like the Turks. This explanation confuses the Byzantine epic poem "Digenes Acritas" with history, because the empire had no special force of counter-raiding akritai, but rather a great many regular troops and strong frontier fortresses whose main purpose was defense. The real weakness of the frontier in 1071 was that around 1053 the emperor Constantine IX had stupidly removed 50,000 frontier soldiers from service in return for payment of a tax.
On the other hand, Luttwak seems unaware of the catastrophic disintegration of the Byzantine army after the Battle of Manzikert, when the themes (an Anglicization of the Greek word themata), the armies that had been supported by land grants since the seventh century, disappeared as a fighting force after years of neglect and debasement of their pay. He declares, "The Seljuks could not...exploit the victory at Mantzikert...by conquering the whole of Anatolia" (229); but conquer practically the whole of Anatolia was just what the Seljuks did, and Alexius I recovered the northwest corner of it only with the help of the First Crusade. Luttwak's reference to the "spectacular restoration of the empire under Alexios I Komnenos" (163) is greatly exaggerated. Luttwak says that before the Byzantine defeat at the Battle of Myriocephalum in 1176 "Manuel I attempted a theater-level offensive to finish off the [Seljuk] Sultanate of Rum and reestablish imperial rule over all of Anatolia" (229); but the sources give us no reason to think that any emperor after Manzikert seriously tried to reconquer the whole peninsula, not even John II (1118-43), who recovered small parts of it during arduous campaigns that took up most of his reign. Even if Manuel had prevailed at Myriocephalum, retaking all of Anatolia would have required many more years and a larger army than he had. He can scarcely have entertained plans that were much more ambitious than to occupy more border strongholds.
Luttwak has similarly vague ideas about Byzantine military organization. In particular, he consistently exaggerates the importance of cavalry in the Byzantine army, despite the clear evidence of some of the sources he cites. For example, he states that the Byzantines "favored the more mobile and more flexible cavalry over the infantry," (58) and he notes that the second-century writer Arrian described Roman armies that were "one-quarter" cavalry, "very much less than in all the Byzantine field armies for which we have numerical information" (245). Yet later he says of a description of a Byzantine army in an official tenth-century manual, "In the prescribed battle array, only three out of ten [soldiers] are [cavalry] bowmen, and in the envisaged army there are 4,800 [cavalry] archers as against 11,200 heavy-infantry menthe same proportion as in the envisaged encampment of" another official manual of the period (367). Surely Arrian's proportion of 25% cavalry cannot be considered "very much less" than this proportion of 30% cavalry, which still left a field army that was 70% infantry. Moreover, these military manuals describe armies dating from a time after Leo VI (886-912) significantly increased the army's proportion of cavalry.
Luttwak's impressions about military finance (and state finance in general) are also mostly baseless and mistaken. He shows little understanding of the Byzantine system of taxation (admittedly an obscure and difficult subject) and no understanding of the system of military land grants that largely supported both the themes and the tagmata. He explains the Byzantines' "perpetual" shortage of combat-ready soldiers by saying, "The taxes that could be collected from the modest surplus of a primitive agriculture could not pay for the permanent upkeep of a sufficient number of trained troops" (285- 6), without explaining how the situation differed for the Romans, who had just the same sort of agricultural economy. He implies that matters were different for the Arabs after they conquered Syria and Egypt: "Muslim taxes could be low because the cost of Muslim rule was very low at first" (201). We however have no evidence whatever that the Arabs lowered taxes. Our best statistic for the revenue of Egypt (2 million Byzantine nomismata) dates from after the Arab conquest and seems to have been the same as the revenue under the Byzantines. Around 800 the Arab Caliphate recorded total revenues equivalent to some 35 million nomismata, about seventeen times as much as the greatly diminished empire at the same date.
The explanation of Luttwak's vagueness about the Byzantine army as an institution seems to be the influence of John Haldon, whom he thanks in the preface for "a detailed critique of an early draft" of the book (x). Haldon makes it clear that he writes "within a historical materialist framework--that is to say...from a 'Marxist' perspective" on p. 6 of his Byzantium in the Seventh Century (Cambridge, 1990). He dismisses most Byzantine statistics as unreliable (usually without explaining why or making any estimates to put in their place) because they show the empire must have maintained a large army through the seventh century by distributing state lands to its soldiers as military land grants. That the empire survived by this form of privatization, however unacceptable it may be to a doctrinaire Marxist, was decisively demonstrated by Michael Hendy in his Studies on the Byzantine Monetary Economy (Cambridge, 1985), which consequently refutes most of Haldon's earlier (and later) work on the army. Yet Haldon has continued to ignore all the evidence that is incompatible with his ideology. Far from being a Marxist himself, Luttwak pointedly ridicules an earlier Marxist, E. A. Thompson, calling his "class hatred...absurd and hilarious" (439 n. 70). I am therefore puzzled as to why Luttwak prefers Haldon's work to dozens of Byzantine and Arab primary sources, to Hendy's work, which Luttwak occasionally cites but seems not fully to have understood, or to my own, which he never cites though he once favorably reviewed my History of the Byzantine State and Society (Stanford, 1997). That book, with its companion Byzantium and Its Army (Stanford, 1995), cites and discusses all the statistics I mention here.
It would be a waste of space to try to catalogue the many misprints, misspellings of medieval and modern names, and miscellaneous mistakes that, even if they have no effect on the argument, show that both Luttwak and his press did an unusually sloppy job on this book. The index is too short and not analytic enough. Two errors seem such serious slanders of Byzantine Christianity that they need correcting here. To say that in Byzantium "Pagans who refused to convert were to be killed" (p. 205; cf. p. 209) is a gross distortion of the fourth- century prohibition of pagan sacrifices, which never extended to pagan beliefs, or even (before the sixth century) to teaching paganism. No Christian emperor executed pagans for refusing to practice Christianity, as several pagan emperors had executed Christians for refusing to practice paganism. Another serious mistake is to say that the Monophysite Christians of Syria and Egypt "had been harshly persecuted by the Byzantine authorities to persuade them to accept the christology of the Council of Chalcedon of 351 [a slip for "451"]" (206). In reality, from 451 to the late seventh century most emperors went out of their way to conciliate Monophysites. The "persecution" of Justinian consisted of detaining Monophysite bishops in the imperial palace to keep them from ordaining more priests, which they nonetheless managed to do with the help of Justinian's Monophysite wife Theodora.
Besides popularizing the valid thesis that the Byzantines systematically avoided warfare, the book has some other merits. The description of the devastating impact of the plague under Justinian is excellent, and that the plague was one of the reasons for later Byzantine defensiveness is implicitly acknowledged. The significance of the natural defensibility of Constantinople is well understood (67); so is the importance of the tightly organized army camps that the Byzantines had continued to use since Roman times (248). The treatment of Byzantine diplomacy is extensive and accurate. The Byzantine manuals of tactics are often (though not always) explained perceptively, even if some general readers may find the detail excessive. Many comments are intelligent and felicitous, like the remark that in diplomacy the Byzantines' "apparent proclivity for pointless formalities was usually rather purposeful" (169). The style is readable but not simplistic, resembling that of good journalism. The book is nevertheless seriously flawed, and presents little that is truly new.