While in many cognate fields the study of identity has reached high levels of sophistication and has even become rarefied, in Byzantine Studies the question of Byzantine identity has barely yet been posed in terms that would meet mid twentieth-century standards. For the most part, Byzantinists struggle with the confusing and polemical notions that they have inherited from the prehistory of their field, most of which were devised in order to deny that the Byzantines were what they said that they were, namely Romans. We have been treated to various contradictory alternatives such as the "medieval Greek empire," the "multi-ethnic empire," and the "universal" or "Orthodox" empire. We have been told that, despite what they themselves believed for a thousand years, the Byzantines were not Romans but Greeks, or that Romans for them really meant Orthodox, or that their empire was so ethnically diverse that the only thing that united them was Orthodoxy and their loyalty (or mere subjection) to the emperor. And yet, in addition to being contradictory, these notions have never been grounded on the primary sources, far less a scrutiny of the sources. The elephant in the room is that the field of Byzantine Studies has been shaped by the demands of many other, often competing, parties with long-standing interests in claiming exclusive rights to the Roman legacy. A polemical tradition existed since medieval times in the West of denying that the Byzantines were Romans. Many Byzantinists have famously espoused these ideologies over the interests of their own field. As a result, we still do not know who the Byzantines were or what they meant when they said that they were Romans.
Being Byzantine is a curious book. It unselfconsciously dispenses with most of what has been said by scholars about Byzantine identity and takes its bearing directly from a selection of primary sources. Merely by doing this it represents a huge advance past the state of the field. For example, it discovers that Roman identity in Byzantium has a strong ethnic component. This is not quite correct in my view (see below), but it is far closer to the truth than all the alternatives that we have been working with so far. Page oddly does not attempt to situate her outrageous heresy among competing views, to explain why those views are wrong and why hers is so different. She just follows the sources closely, explaining what they mean by Roman here and what there. It is not clear from her scattered hints how we are supposed to get from the (allegedly) multi-ethnic empire of late antiquity to the ethnic state of Byzantine Rhomania (see, e.g., 42, 67). Page also knows (because her sources tell her clearly) that Roman and Christian did not (and could not) mean the same thing in Byzantium. But she does not confront famous declarations by leading scholars in the field that the two were equivalent, or their view that Roman-ness was so vague as to be meaningless. Page correctly realizes that being Roman was premised on a substantial degree of cultural, religious, and social homogeneity, though her book focuses on the period after 1204, when different elements of Byzantine Roman identity developed in separate directions under pressure by Frankish colonization.
Being Byzantine, then, operates largely outside the conventions of the field, which is why it manages to move forward so much and why it is so refreshing and stimulating, albeit in an innocent way. I strongly recommend it as an antidote to all the self-loathing, Roman-denying preconceptions that govern the field. But once the question is opened, there is room for theoretical reflection, refinement, and disagreement (my position is laid out in ch. 2 of Hellenism in Byzantium: The Transformations of Greek Identity and the Reception of the Classical Tradition, Cambridge, 2007, which appeared too late for Page to consider).
Firstly, while Page rejects the modern Greek nationalist interpretation of this period (9), she has not fully worked clear of the notion that the Byzantines "really were" Greeks beneath the Roman label. The subtitle gives the definite (but wrong) impression that Being Byzantine was a function of Greek Identity. She claims, moreover, that "this book examines...the identity of the Greeks--or, as the subjects of the empire tended to call themselves, the 'Romans'" (5). She too calls them Romans (Byzantine Romans) almost without exception, an admirable departure from tradition, but on one occasion indicates that she does not believe it: they were the Greeks of this period (24; cf. 67). But here we must insist that the Byzantines were not Greeks who called themselves Romans; they were Romans who were called Greeks by their enemies in the West and who are still called that today because to call them by their own name would challenge a certain view of the history of the West (and now of modern Greece too). This has nothing to do with how Greek the Byzantine's culture or their genes were; it is a question of identity.
Second, the argument of the book rests on the distinction between two different senses of being Roman: there was, first, a political meaning ("loyalty to the emperor"), and, second, an ethnic meaning (descent from other Romans, who shared the same culture, language, religion, customs, laws, etc.). The dynamic of Being Byzantine rests on the tension and divergent evolution of these two different senses after 1204. But the way in which each of them is defined separately is problematic as is the schizophrenic notion of Roman identity that results from their combination (and they were always combined in the case of ethnic Romans subject to the emperor). Let us begin with the ethnic sense. It should, more correctly, be called a national sense: the Byzantine Romans were a nation, not an ethnicity. While some believe that the two cannot be differentiated (I cite here D. Goodblatt, Elements of Ancient Jewish Nationalism, Cambridge, 2006), Page defines ethnicity narrowly as a group's belief in its common ancestry (11-13), a view shared by other historians (I cite here J. Hall, Ethnic Identity in Greek Antiquity, Cambridge, 1997). Yet the ethnic indicia that Page uses throughout the book (religion, language, customs, laws, territory and the like: 14, 17-18, 44) are, when shared by a population, indicative of a national identity and not a narrowly ethnic one; Page sometimes even forget to add belief in a common ancestry among them (93, 122, 162: "Romans by culture, religion, and so on," 276), and on one occasion refers to Greek as the "national" language of Byzantium (63). What, then, is the difference? While the sources leave no doubt that the Byzantines constituted a nation (by any standard used to identify nations anywhere), they did not necessarily define themselves ethnically. Many prominent Byzantines, whose Roman identity was never in doubt, are said to have had non-Roman ethnic ancestry (e.g., Basileios I, Ioannes Tzetzes, the Axouch family, etc.). They were Roman by virtue of their cultural profile, in terms of faith, language, dress, way of life, etc. Ever since the days of Romulus, it was possible to become Roman and this was how the Greeks had become Romans in the first place (see G. Woolf, Becoming Roman: The Origins of Provincial Civilization in Gaul, Cambridge, 1998; in general, E. Dench, Romulus' Asylum: Roman Identities from the Age of Alexander to the Age of Hadrian, Oxford, 2005). The Byzantines understood this dynamic very well.
Where, then, does Page find Roman ethnicity? First, Byzantine sources sometimes use the language of ethnicity in part because most Romans were, after all, descended from other Romans, and also because the classical terminology of ethnos and genos was too imprecise to distinguish between these two different senses. Some authors highlighted blood-ties whereas others made it clear that being Roman was a function of one's cultural/national profile. The second reason is that Page focuses only on individuals who appear in her sources, finding that their foreign ethnic origins could preclude them from being considered fully Roman (e.g., 116-117, 158-160). But the picture changes dramatically when we look past the first generation and consider families and groups that settled in the empire and were fully integrated. A striking example are the thousands of Persians who entered in the ninth century. By converting and taking on Roman ways, they effectively disappeared as a distinctive group and became indistinguishable from other Romans. Aphanismos is how one historian called this process of assimilation-to-the-point-of-extinction (Genesios, On the Reigns 3.7). This process continued to happen throughout Byzantine history.
Let us now consider the alternative, political, version of Roman identity, which Page glosses as "loyalty to the emperor." In fact, no such thing existed. It is misleading at best, a fiction at worst, and is the one part of the argument that Page owes to the scholarship rather than to the sources. We must be clear about the way in which it has been used (and is used by her too). The claim is not that Romans (whether defined ethnically or nationally) were expected to be loyal to the emperor, because that would be obvious. No, "loyalty to the emperor" has acted in the scholarship as a definition of Roman identity in Byzantium: being loyal to the emperor in itself, taking up service under him, makes one a Roman even in the absence of ethnic/national criteria. Page frequently calls this the "dominant" meaning of Roman in Byzantium. As a definition it has, and was meant to have, dramatic implications, namely to undercut the Byzantines' view of themselves as a people defined by what we would call national criteria. But it is refuted by evidence that Page herself presents. She knows, for example, that Roman writers, before and after 1204, did not consider as Romans those barbarians from East or West who happened to take up service under their emperor (e.g., 81-83, 169). This is because those barbarians did not (yet) fulfill the national criteria, which were always the dominant ones in Byzantine ideology. Mere service under the emperor was not enough. Also, many sources state or imply that the emperor ruled over both Romans and barbarians, which means that simply being a subject of the empire was insufficient to make one a Roman. Besides, what does "loyalty to the emperor" mean anyway, when there were so many revolts against the emperor? It must mean "loyalty to the empire," but what was the "empire"? The Byzantines often called it the politeia of the Romans (Greek for respublica). In other words, "loyalty to the emperor" basically means "Roman patriotism," which is exactly what we would expect if the national version of Roman-ness was dominant. Our sources happen to be full of expressions of Roman patriotism, though many historians have labored to explain that they do not mean what they very clearly state. On one occasion, Page comes closer to the truth that lies behind this "loyalty" when she clarifies her definition as "loyalty to an institution and an ideal" (76). That ideal was a national idea. Thus, the allegedly dominant political sense collapses into the national one. Loyalty to the emperor was not a Roman identity in itself; it was only part of Roman identity before 1204, and the interest and contribution of Being Byzantine is its exploration of how this national identity survived and mutated when it was cut off from the state. But there is no question about what was primary here: "it was entirely feasible to be a Roman but live outside of the Byzantine Roman state" (123; cf. 79).
In the tradition of Byzantine Studies, Roman identity has been defined as "loyalty to the emperor" precisely in order to avoid defining it in terms of the substantive cultural (ethnic or national) criteria that Page finds written all over the sources. Being subject to the emperor is a purely formal circumstance that makes the Byzantine Romans seem like an arbitrary collection of peoples who could have any other kind of identity if left to their own devices (82: "Romans lived in the area ruled by the Roman emperor"). This might be a way to track Romans but not to define them. According to this definition, they were Roman only insofar as they were ruled by a so-called emperor of the Romans and for no reason that had anything to do with them. But the opposite was actually the case, as revealed by Page's ethnic sense. In accordance with ancient Roman tradition, in Byzantium the emperor was called "the emperor of the Romans" because the people over whom he ruled were the Romans, and not the reverse (see C. Ando, 'Was Rome a polis?' Classical Antiquity 18  5-34, here 15-17). Page gets it right elsewhere when she says that "the Romans are the object of rule, they are the essence...of the empire" (76). It is less clear what it means that the people were the "collective expression of the empire" (78). The reverse would seem to have been the case. In any case, it was not the emperor who made them Romans. What we call the empire was Rhomania, the nation-state of the Roman people, and the emperor was their ruler, answerable to them in both theory and practice.
Byzantinists clearly have much to discuss. Being Byzantine is a solid step in the right direction.
Some minor corrections. 24 their use of classical Greek does not imply that the Byzantine identified ethnically with the ancient Greeks. 58 third century AD should be second. 63 Hellen meant pagan since at least 300 AD. 92 Choniates was hardly "a man of his time." What does that mean anyway? 114-115 and 238 Romaïzontes probably does not mean "qualified Romans" but "those who sided with the Romans." 209-210 there is no reason to think that the majority of the audience for the Chronicle of the Morea were provincial ethnic Romans. 221 and 255 Perses should probably be Persai.