On the basis of a random sample of English-language internet websites about empires, we can now formulate the first law of comparative imperialisms as follows: as an online discussion of empire grows longer, the probability of a comparison involving the Roman Empire approaches 1. (This is a variant of the general law that states that ‘as an online discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches 1’.Footnote ¹) The comparative study of empires is thriving, and the recent intensity of interest is connected, at least in part, to the international military interventions of the United States.Footnote ² But comparisons between empires are nothing new, and, in the 1960s, Peter Brunt wrote an insightful article on British and Roman imperialism.Footnote ³ That analysis was the product of the age of decolonization, an age which also acted as a spur to comparative approaches within classical scholarship: witness Nicole Loraux's suggestion that it was anti-colonial movements associated with the Algerian and Vietnam wars that led Jean-Pierre Vernant to embark on his series of comparative investigations into Greek thought and religion.Footnote ⁴ Brunt's article was written in a retrospective key at a time when it was possible to look back to the completion, or the near completion, of a major period of European colonialism and arrive at a sort of reckoning. Some two generations prior to Brunt, in the early twentieth century and at the apogee of the British Empire, Lord Cromer delivered an address to the Classical Association on ‘Ancient and Modern Imperialism’ in which he found it unimaginable to think of independence for Britain's overseas colonies.Footnote ⁵ Francis Haverfield responded sympathetically to Cromer and in his own writings associated the British and the Roman empires.Footnote ⁶ Any discussion of comparative imperialisms, therefore, will need to consider not just the recent concentration of debates over empire but also a lengthy trajectory that extends back to Cromer and Haverfield and indeed further beyond into the eighteenth century. None of the books under review reflects in detail on the intellectual history in which they may be situated, but this is a subject that at least needs to be acknowledged and that we shall have occasion to return to later.

Why compare empires? One benefit of comparison is that it helps to clarify phenomena and to sharpen the distinctive nature of the objects under scrutiny. Marcel Detienne famously urged us to ‘compare the incomparable’ and to seek out meaningful comparisons that genuinely contribute to one's understanding of the issues and challenge long-held assumptions. Comparison can lead to demystification and produce a change in outlook. For historians, as Walter Scheidel says with regard to the two empires that are discussed in Rome and China, ‘only comparisons with other civilizations make it possible to distinguish common features from culturally specific or unique characteristics and developments, help us identify variables that were critical to particular historical outcomes, and allow us to assess the nature of any given ancient state or society within the wider context of premodern world history’ (5). Another reason to undertake comparative study has to do with historical causality. Why did certain events occur in a culture, what factors contributed to their development, and could they have occurred elsewhere? ‘The main questions’, Scheidel writes, ‘are which factors were crucial rather than incidental to observed developments and how different contexts could produce similar outcomes, or vice versa’ (6). Yet another reason, which seems not to be explicitly mentioned by Scheidel, perhaps because it is so obvious, is that comparative analysis of cultures has the potential to generate inferences that have little direct bearing on either culture but nonetheless speak to the concerns of other times and places. This is a possibility allowed by Fritz-Heiner Mutschler and Achim Mittag in Conceiving the Empire, their volume on ancient Rome and China: ‘Comparing the Roman and Chinese empires contributes not only to understanding the trajectories along which the two civilizations developed, but also to heightening our awareness of possible analogies between the present and the past, be it with regard to America or China’ (xiv). Looking at the backgrounds of the contributors to the collection, the location of the conference out of which it emerged, and the funding from the VolkswagenStiftung, I am tempted to add the words ‘or Europe’ to that last phrase.

The books under review approach the subject of comparative empires from varying perspectives. The Dynamics of Ancient Empires and Rome and China are part of a new book series, Oxford Studies in Early Empires, under the general editorship of Nicola di Cosmo, Mark Edward Lewis, and Scheidel, and Scheidel played a significant editorial rôle in bringing the volumes to publication. Rome and China and Conceiving the Empire compare the Roman and Chinese empires, but are shaped by divergent concerns and editorial policies. Vaclav Smil's book, Why America Is Not a New Rome, is not an argument for comparative study but a discussion of the problems that bedevil comparisons of the Roman Empire and the USA. He is not against historical comparisons per se but finds popular assumptions about similarities between the Roman Empire and the USA to be misguided, inaccurate, and unhelpful. Collectively, these books reflect on the burgeoning field of comparative imperialisms and illustrate the benefits of comparative study; they also raise questions about method and illustrate the challenges faced by comparativists.

I

The issue of imperial comparison is developed with a little more sophistication in Scheidel's Rome and China than in Mutschler and Mittag's Conceiving the Empire, although both volumes might have treated the conceptual and theoretical problems raised by comparison in more detail. In an extraordinary editorial decision, Mutschler and Mittag asked their contributors not to make any comparisons themselves and relegated comparative analysis to an epilogue written by the two of them. ‘In order to relieve individual contributors of the strain of constant comparison between the two cultures, we asked them to concentrate their discussion on either the Roman or the Chinese perspective according to their expertise. It seemed to us that in this way the authors’ competence could be utilized most effectively' (xvi). Readers of the book may well find themselves wishing that the contributors had made some effort at comparative analysis even if they had not burdened themselves with ‘the strain of constant comparison’. The chapters are grouped together by topic into eight pairs, and the juxtaposition of chapters affords a manner of oblique comparison. A few contributors, among them notably Mutschler himself, attempt to provide comparative reflection, but the opportunity for comparison is left to ‘a future dialogue between the two fields’ (xvi). In this sense, the collection is similar to The Dynamics of Ancient Empires, edited by Ian Morris and Scheidel, in which explicit comparison is undertaken in the first and last chapters alone, and in which the hope is expressed that the book ‘will encourage more systematic and comparative thinking about the nature and development of imperial states in early history, and serve as building blocks for cross-cultural studies’ (vi). The reader is assured that the ‘project has inspired all the participants to engage in more explicitly comparative and multidisciplinary work on early empires’ and that the editors ‘will measure this book's success by its capacity to motivate our present and future colleagues to do the same’ (vi). In defence of The Dynamics of Ancient Empires, while the book does refer to comparative study in editorial comments that describe the publication's goal (v, 103), it does not announce the comparative study of empires in its title, which rather promises a treatment of ‘state power from Assyria to Byzantium’.

It would be easy to fault the editors and the contributors for the style of their comparisons and for inconsistencies in method, rather than acknowledge their accomplishments. I might point out that the chapters in the two books about Rome and China occasionally elide the difference between trading contacts and comparison. Trade, commerce, and maritime links may shape the interpretive context, but the details of commerce and trading networks cannot serve as a substitute for comparative analysis, nor are they necessary for it. I might also observe that the scholar who interrogates his or her own location and methodological assumptions stands to accomplish more than the discovery of putative similarities or differences across empires. It is common to have one's own hypotheses confirmed, or repudiated, by the evidence when the basis of those hypotheses hides in obscurity, and at times the basis for comparison could have been subjected to greater scrutiny in these volumes. And I might have wished that the two books in the Oxford series had explained their choice of subject further. Rome and China is said to be about ‘state-society relations’ (8) but that is vague and scarcely begins to account for the particular direction taken by the chapters in the book. The editors of The Dynamics of Ancient Empires asked their authors to discuss ‘how empires were run, how they extracted resources, and what their long-term consequences were’ (vi), and, needless to say, the authors have responded to the broad mandate in very different ways.

But I do not suggest that these books are edited or written by scholars who have given little thought to the comparative method, for that is clearly not the case. The three edited collections consist of chapters written by experts who display a masterful knowledge of their subjects, and a strength of the books is the historical documentation that they offer to future comparativists. Comparative study is hard and takes the scholar out of his or her comfort zone, and I do not make light of the amount of time and effort required to work in two or more cultures. With respect to Rome and China, areas of future research are sketched out in two breathtaking sentences by Scheidel (4). The first, and much longer, of the two sets down ‘convergent trends over time’ and takes up the better part of a page as it ranges from ‘the growth of markets in land and the gradual concentration of wealth among elites’ to ‘the creation of a homogeneous elite culture and of corpora of classics’ to ‘ideologies of normative empire sustained by transcendent powers’ and many other topics. The second lays out significant differences between Rome and China and refers to ‘the Republican background of the Roman state’ and ‘the scale and functions of slave labor’, among other subjects. Exploring the areas of study described by Scheidel in these two sentences would be a worthwhile pursuit, to which he has devoted himself enthusiastically with numerous publications.

If comparison is worthwhile, does it help us answer any of the fundamental questions of human behaviour? Have we learned, for example, what drives people to conquer and establish empires? We certainly have, according to Scheidel, and the answer is ‘Sex’. In a remarkable tour de force called ‘Sex and Empire’, which appears as the last chapter of The Dynamics of Ancient Empires, Scheidel writes that, from a long-term evolutionary perspective, ‘since genetic survival (i.e., successful replication) is contingent on scarce energy resources, reproductive processes inevitably involve competition, which in turn drives evolution in response to natural selection’ (256). To be fair, Scheidel generally uses a language that is easier to understand than that quotation implies and he makes a great deal of evolutionary biology accessible to prove his point about the ultimate causes of empire. Where Michael Mann took for granted that humans were prone ‘to increase their means of subsistence’, Scheidel pushes the level of explanation further back and asks why human beings would desire to increase their means of subsistence, their control over resources, and their power.Footnote ⁷ His is an avowedly androcentric theory, grounded in Darwin's argument that men have historically sought to increase their reproductive success through competition with other men. Women are biologically different from men, and need to make a relatively high level of investment in pregnancy, breast-feeding, and the nurture of infants, but men make smaller investments than women in these areas and are free to compete with other men and deny them reproductive opportunities. When men co-operate with each other in warfare or imperialism, they are seeking to improve their reproductive capacity and to generate conditions as favourable as possible for reproductive success. The question we should thus be asking about ancient empire, Scheidel observes, is ‘whether or how it facilitated the exploitation of reproductive capacity’ (258).

This is theory on the grand scale and cuts across cultures and historical periods. Scheidel's chapter is not strictly comparative in the sense that Rome and China is comparative, and he does not compare phenomena across two or more cultures in writing about sex and empire. He draws on evidence from a wide variety of cultures and on studies from a range of disciplines, though not with the stated purpose of comparison. But his chapter is concerned with ‘ultimate causation’ (260) and disregards the stated motives and ideologies of human beings and empires since these could mask the true evolutionary imperatives that are at the centre of imperialism. Not all narratives are so nakedly about power and about reproductive success as the myth of the Sabine women or the stories recounted in Herodotus' proem. However, Scheidel writes:

Scheidel offers these statistics partly in jest, and it would be wise not to take calculations that are based on the sexual obsessiveness of American men too seriously. Nevertheless, the point about observed or attested behaviour is made in earnest, and, in his view, the evidence shows that ‘imperial success increased the reproductive success and inclusive fitness of primary beneficiaries’ and also that ‘imperial success enabled those beneficiaries to engage in behavior that would ordinarily promote reproductive success’ (261). Thus, while explicit justifications are hard to find of the claim that ancient empires were ‘established in order to facilitate sexual exploitation’, an abundance of materials exists to support the argument in terms of ‘the most fundamental underlying behavioral mechanisms’ (306). Scheidel's learning is manifestly impressive and it is put to judicious use as he quotes from studies about not just the Roman Empire, Athens, Sparta, and Persia but also the Inca, the Dahomey empire, medieval Khmer rulers, and North Korea.

An important point about causality is raised by Scheidel, and the point is worth emphasizing since it is also raised, if indirectly, by some of the other chapters in the books under review. Scheidel writes in Rome and China that comparative study does not search for laws of causation but does ‘offer causal explanations of particular outcomes by identifying critical differences between similar situations and/or by identifying robust processes that occur in different settings’ (6). Causation is a complex phenomenon, and a long tradition of philosophical and scientific enquiry that goes back at least to Aristotle has been devoted to the topic. Historians are familiar with the post hoc ergo propter hoc fallacy and are trained to avoid it. They know that correlation alone does not imply causation and they do not suppose, because Germany always wins the Eurovision song contest when it is a good year for grapes in Chile, that the Chilean grape harvest is the cause of the German triumph. For the most part, the authors under consideration are sensitive to causal complexity, avoid monocausal determinism, and maintain a stance of intellectual openness in describing the relationship between human agency, events, and contexts. ‘Sex and Empire’ is exceptional in these analyses in so far as Scheidel offers a causal theory on one of the largest time-frames imaginable. Informed by social Darwinism and inflected by a mixture of ancient history, anthropology, political theory, and gender studies, his provocative argument deliberately sets out to be reductive and to divine a unifying cause. The other authors are less controversial, but the reader is occasionally stopped short by some claims. In a fine chapter in Rome and China, Nathan Rosenstein writes: ‘In the absence of a military threat that would have compelled Rome to develop in the ways that China did and because coalition building and, after 338 b.c.e., the incorporation of many non-Romans into the Republic's citizen body enabled it to meet successfully the threats it did face, the aristocracy's self-interests could determine how the Roman state evolved’ (50). I think I understand the general argument that Rosenstein is making in his chapter, but he appears to be saying here that a military threat or series of military threats could have driven the Republic ‘to develop in the ways that China did’. It is one thing to say that ‘war profoundly affected the trajectories of state-formation in China and Rome’ (49) or that differences between Rome and China can be attributed ‘in part’ to military conflicts, as Rosenstein does earlier (29); it is quite another to say that wars, even wars of a particular type, would have made the Republic develop as China did and result in the creation of ‘highly bureaucratized central administrations … capable of extending the government's reach down to the level of five household units’ (49). If the writer were rigorous in thinking through cause and effect, he would have had to eliminate every other possible factor and find that wars inexorably led to highly bureaucratized central administrations in China and would do so as well in Rome. I accept that the evidence from antiquity is limited and that some level of correlation is inevitable in judgements about the ancient world, but a level of caution is desirable in transposing causal explanations from one culture to another. It should be said that Rosenstein makes a pretty interesting case overall for the relationship between external threats and internal administrative change in Rome and China.

While Scheidel and Rosenstein draw our attention to coercion and the body in their writings, I think the body and gender could be treated more capaciously in these books and in the comparative study of ancient empires.Footnote ⁸ A few possibilities are considered by Scheidel in ‘Sex and Empire’ when he sketches out the demographic concerns that emerge more clearly from his evolutionary model of imperialism. By his count, Roman Britain held about 40,000 soldiers in their twenties and thirties and about 200,000 to 300,000 women ranging in age from fifteen to thirty-five. Of the soldiers, a small fraction would have been native but most would have been of foreign origin, and many local women would have had their choice of foreign mates or spouses, at least in theory. Scheidel believes that the gender imbalance ‘would have reached a comparable scale in other frontier zones, above all the valleys of the Rhine and the Danube’ and he observes that ‘these sexual repercussions of Roman imperialism have yet to be acknowledged in modern studies of provincial “Romanization”’ (303). Even if Scheidel is over-confident in the figures that he provides and overly reliant on a difficult sociobiology, he is right to call for more study (comparative and non-comparative) than we have at present of the kinds of sexual relationships (consenting, coercive, non-coercive) that were facilitated by the Roman Empire in frontier regions and the provinces.Footnote ⁹

Writing in Rome and China, Karen Turner and Maria H. Dettenhofer provide different perspectives on the body from those offered by Scheidel or Rosenstein. Dettenhofer writes that court eunuchs were established in China as early as the Qin Dynasty (221–206 b.c.e.), whereas in Rome eunuchs gained in political importance only after the reigns of Diocletian and Constantine; in the Roman case, they were mainly freed slaves from Armenia and Persia and many were to occupy the rôle of imperial chamberlain, praepositus sacri cubiculi. Dettenhofer discusses the association between the status of women and eunuchs in courtly culture but notes that, with notable exceptions, Roman eunuchs were not assigned to women's quarters. Her chapter points to a dense cluster of issues connected with women, the phallus, the body, intimacy, religion, ethnicity, and power and shows that there is more to be understood about eunuchs (already the subject of a great deal of research across cultures) through cross-cultural study. Turner looks at ‘law and punishment in the formation of empire’ and weighs the difference between Chinese and Roman attitudes to corporal punishment. She argues ‘that much earlier than in the West, China developed a Foucauldian notion of biopower: the categorizing, standardizing, control, and mobilization of individual bodies to serve the state. The Chinese case challenges Foucault's contention that a passion for organizing bodies for their productive capacities marked a shift in political life from classical to modern systems’ (77). For her, ‘the criminal body served more as a site of symbolic retribution than an economic asset’ in the Roman Empire, and criminals were part of a spectacle of chastisement in which penalized bodies were put on display (78). Turner and Dettenhofer might have refined slightly the dichotomies they present, but their chapters illustrate the divergent attitudes taken toward the human body, toward its integrity and mutilation, its exploitation and deployment for labour, and its capacity for symbolic display.

Turner and Dettenhofer also rely less on quantitative data than co-authors such as Scheidel, whose agility with statistical and economic evidence is nothing less than remarkable. This raises a broader issue in so far as many of the chapters in Rome and China and The Dynamics of Ancient Empires are laden with data, statistical analyses, and quantitative studies, and in so far as many contributors make qualitative generalizations on the basis of quantitative study. The use of quantitative materials can be attractive to historians since it allows them to work toward an imagined precision. Why historians of empire indulge in this kind of analysis is not difficult to understand, given that many empires have been supported by the maintenance of archives, records, and highly evolved systems of classification and often have left behind evidence that lends itself to statistical analysis. But some vigilance in these matters is warranted in dealing with the ancient world, given the many lacunae in the evidence, the differences in conceptual systems, and the intractable foreignness of antiquity. While it would be tempting to learn how other societies compare with the Roman Empire in terms of its supposed economic growth, for example, we should register Keith Hopkins' warning that the desire to see economic growth in ancient Rome may be ‘an attempt to find the roots of modern experience in classical antiquity’ (Dynamics, 195). Perhaps inevitably, these books illustrate both the strengths and pitfalls of scholarship based on quantitative analyses; they open up new vistas onto the ancient world by coaxing fresh meanings out of the old sources and also appear to flatten out the complexity and otherness of their objects of study.

II

The universalist conceptualization of empire is broached in Conceiving the Empire by several authors, including Mittag and Mutschler, who come to it through a consideration of the terms tianxia (‘All-under-Heaven’) and imperium and, to a lesser degree, oikoumene and orbis terrarum.Footnote ¹⁰ Mittag and Mutschler argue, ‘Whereas in China the idea of tianxia was intensively discussed long before the unification of the empire, in Rome ideas of empire and imperial rule developed only with the emergence of the historical phenomenon itself’ (430–1). Given the fragmentary nature of early Roman sources, this is a hypothetical statement at best, but the issue is also discussed in three pairs of perceptive papers on geography (Helwig Schmidt-Glintzer and Katherine Clarke), large inscriptions (Martin Kern and Christian Witschel), and art and architecture (Rolf Michael Schneider and Michèle Pirazzoli-t'Serstevens). Of these papers, Kern's reading of the inscriptions of the Qin First Emperor shows that the function of the inscriptions was public, symbolic, and performative. Too remote and too elevated to be read by human beings, the inscriptions demonstrated that ‘sovereignty even extended from the human sphere into the cosmic realm, transforming a natural site into a site of civilization and human history’ (225). These inscriptions sought to efface earlier histories and claimed that ‘the normative version of history as promulgated from the sacred mountains of the former independent states was a single and unified one’ (226). The inscriptions also refer to the details of their own ritual creation and recitation, thereby ‘integrating the ritual act of reciting merit and inscribing the stele into the purview of the foregoing historical narrative and at the same time bringing the narrative to its logical conclusion’ (237). Kern explores the performative force of inscriptions elegantly, and his chapter should be read alongside not just Witschel's treatment of the Res Gestae but also Clarke's evocative analysis of the production of space in Rome, Pirazolli's chapter on the image of the other in Han art, Schneider's on the shaping of Augustan Rome, and Michael Nylan's on ‘the rhetoric of “empire”’ in China. The authors do not offer extensive comparisons, in conformity with the volume's editorial policy, but Mittag and Mutschler discuss the communicative framework and the ideological content of imperial representations as they are presented in these chapters. This is one of the most helpful sections of the editors' epilogue, and yet I cannot help thinking that an opportunity has been lost for a more detailed comparison across chapters.

The significance of universalizing discourses and practices is also raised by Josef Wiesehöfer and Ian Morris in relation to Achaemenid Persia and Athens, respectively, in The Dynamics of Ancient Empires, another volume that eschews direct comparison as a matter of principle. In his astute account of the Achaemenid Empire, Wiesehöfer writes that ‘the idea of a universal empire propagated and symbolically promulgated by the king was present in the Achaemenid Empire’ (87), and one speculates that the royal inscriptions of Darius and Xerxes would provide a stimulating comparison with the stele inscriptions of the Qin First Emperor from this perspective. Wiesehöfer traces the tension between the universalist tendencies of Achaemenid ideology and the profuse diversity of subject peoples and he shows how the Achaemenids seldom compelled inhabitants of their empire ‘to choose between an “imperial” and a “local” identity’; in his example, a resident of Smyrna could retain a Greek civic identity and simultaneously be ‘invited to regard himself as — and to be proud of being — a member of the most successful and prosperous entity of his own time, the Persian Empire’ (89–90). The word ‘invited’ is intriguing since the dominant ideology scarcely permitted the subject to break free of Persian overlordship and it is reminiscent of Achaemenid ‘tolerance’, which, as Wiesehöfer says, could be described as ‘just a way to keep the conquered peoples quiet’ (88).

In a dense and dazzling study entitled ‘The Greater Athenian State’, Morris suggests that Athens failed to convert its universalizing rhetoric and practice into a larger state that could reasonably be defined as an empire. He declares that fifth-century Athens was not an empire and that ‘far from creating an empire — defined, as above, as a territorially extensive, hierarchical, multiethnic political organization with a strong sense of foreignness between rulers and ruled — fifth-century Athens tried to develop an Ionian Greek territorial state with Athens as its capital city’ (141).Footnote ¹¹ Morris agrees with those who argue that we should not think in terms of an ‘Athenian Empire’ and draws attention to the putative empire's many diminutive features, not least in geographical extent, total revenues (1 or 2 per cent of the early Roman Empire), and population (fewer than a million against the 50 or 60 million in the Han or Roman empires). He writes that inhabitants in cities subject to Athenian rule ‘often did feel different from the Athenians, but their differences were political rather than ethnic, linguistic, religious, or cultural’ (134). For him, fifth-century Athens should be studied as part of the phenomenon of state formation, which he views as an ongoing process rather than a static or unitary event. As Morris is aware, many have criticized state formation for being overly deterministic, of limited heuristic value, and too reductive to account for socio-economic complexity, political change, and cultural difference. Few scholars have worked through the issue with greater dexterity and learning than Morris, however, and he offers as shapely a discussion as we are likely to get of fifth-century Athens from the perspective of state formation in the Aegean. Within the terms of his argument, he signals an area that will prompt further reflection among historians of ancient Greece and Rome. He asks, ‘why did Athens, the most modern Greek state with the biggest population, most advanced technology, most meritocratic administration, and deepest state capacity, fail to break through to become first a territorial Greek state and then a multiethnic empire?’ (167). Another way of asking the question is to contrast Athens' inability to make that breakthrough with Rome's success in imperial state formation, for an implication of Morris' work ‘is that the most profitable comparison for fifth-century-b.c.e. Greece might be fifth-century-b.c.e. Rome, and the most useful question to ask about such a comparison might be why Rome succeeded in breaking out of the city-state framework to become first a territorial state and then a genuine multiethnic empire, while Athens did not’ (141). A further question raised by Morris' chapter is the ideological investment, at least of modern Anglophone scholars, in the use of the expression ‘Athenian Empire’. Why the term was used in the first place and how it has determined the course of subsequent scholarship about fifth-century Athens are questions that we have only just begun to address.Footnote ¹²

As the papers by Clarke, Kern, and Schneider remind us, élites will control ideological and symbolic elements within a cultural framework in their desire to maintain and prolong their grip on state power. The point is also made by Jack A. Goldstone and John F. Haldon in the opening chapter of The Dynamics of Ancient Empires, where they offer the Chola and Vijayanagar empires in South Asia as cases in which central authority could endure ‘simply through the manipulation of key ideological and symbolic elements in the cultural system of the social formation as a whole’ (10). When élites influence the symbolic system that underpins an empire's ideology, they need to rely even less on the military or on other instruments of coercive power in order to survive and adapt over time, and the more successful the management of ideological and symbolic elements, the greater will be the appeal and acceptance of the imperial state among élite and non-élite populations. ‘A good example of more lasting imperial power is provided by the case of Rome’, Goldstone and Haldon write, ‘in which a conquest state was able to evolve an ideological hegemony that in turn generated a consensual identity among the conquered territories’ (11). If that ideological hegemony took various forms and exerted its power on subjects who lived in a land mass that now covers some forty countries, it had to accomplish its work through a language and visual culture that would speak to as many as possible: all those temples to Jupiter, statues of emperors, triumphal arches, and even the sewers and steam baths were connections to Rome and a reaffirmation of its influence. Hopkins observes, ‘To be Roman was to be sweaty and clean. The Roman Empire was an empire of conquest but also a unitary symbolic system’ (186).

The success of the Roman Empire in evolving its ideological hegemony can also be measured in its political economy, as Hopkins implies in The Dynamics of Ancient Empires.Footnote ¹³ Hopkins documents the increase in money supply in the Roman Empire from the second century b.c.e. to the second century c.e. and he notes, for instance, that the volume of silver coinage in circulation in the middle of the second century c.e. was about four times larger than in 50 b.c.e. (1.76 billion denarii against 410 million, without Egypt). But what does this growth in the money supply mean and can it simply be correlated to trade and economic ‘growth’? For Hopkins, the emperors ‘produced coins as economic objects for the facilitation of trade and taxation but above all as symbolic objects of ostentation and political authority’ (201–2). Hopkins is right in this case not to equate ancient and modern economic concepts too closely and to embed the increase in money supply within the wider Roman socio-political order. Here there is further scope for comparison with Han China and the materials assembled by Scheidel in Rome and China. In a detailed study of the monetary systems of the Han and Roman empires, Scheidel uses some of the same data as Hopkins and points out that monetization was far greater in the Roman Empire, which operated on a monetary economy of silver and gold, than in Han China, which used bronze coinage and uncoined precious metal. He believes ‘in a dramatic surge of Mediterranean trade and production for a mass market in the Roman period and the growing wealth of the Roman elite that does not seem to have been matched by that of its Han peers’ and calls for more systematic exploration of the difference (205).Footnote ¹⁴

Happily, some of the socio-political implications of economic phenomena in Rome and China are discussed by Peter Fibiger Bang and Mark Edward Lewis in Scheidel's own volume. Bang describes the Roman Empire and Han China as tributary systems and briefly refers to imperial styles of consumption in the two empires. Moralizing strictures against luxury and worldly ostentation can be found in both Rome and China and, according to Bang, ‘did nothing effectively to counter the new luxurious styles of consumption’ (118). Bang is keen to point to the similarities between the Roman Empire and Han China (he calls them ‘comparable worlds’, 120). The resemblances are well stated; and it remains interesting to see an analysis of the different ways in which empires exhibit the ‘power to consume and command the world in all its great variety’ and to understand how empires have strategically used materialist consumption in order to further political domination and vice-versa (103). Lewis looks at the political and sociological aspects of gift-giving in Rome and China and notes a contrast between prevailing ideologies of imperial rule in Rome and Han China. While euergetism in imperial Rome was focused on the city and occurred in urban spaces, gift-giving in Han China was largely directed at the countryside. ‘This difference reflects the political distinction between the Roman Empire, which was structured as a multiplicity of urban centers — both old established ones and new ones built to a standard model — and the Chinese Empire, where the political power of the ruler and his agents derived directly from the registration, mobilization, and taxation of rural households’ (131–2). I hesitate over this characterization of the Roman Empire but the distinct geo-spatial orientations of Roman and Chinese gift-giving are worth reflecting on. Through a comparison of gift-giving cultures, Lewis shows differences in attitudes to the work-force in the country and the city and he discusses how imperial authority and prestige was constructed in varying ways in Rome and Han China. The appearance of the emperor in public at the Roman games enacted a visual reciprocity between the crowd and the ruler, and the emperor's gifts to the people included not just the spectacle inside the arena but also the bodily presence of himself and his family members.Footnote ¹⁵ By contrast, the emperor did not show himself to the public in Qin and Han China, and the authority of the ruler was rather augmented by his invisibility to the general population. ‘Sealed up behind layer and layer of walls, the emperor rewarded only his closest followers with the supreme honor of letting them come into his presence. This power of the inside over the outside, and of the hidden over the visible, became a longstanding principle of the spatial construction of power in imperial China’ (133). The emperor's body was strictly not for the public gaze in China, and the sight of it was heavily regulated even within the space of the palace. Lewis' argument thus circles us back to concerns raised elsewhere in Rome and China, namely, to the authority enjoyed by court eunuchs, who mediated between the sovereign and the subject, to the distinction between symbolic and economic uses of the body, and to the question of the biopolitical.

III

The most popular imperial comparison of the moment involves Rome and the USA. As the trickle turns into a flood, it is impossible to keep up with the many articles, books, internet sites, and documentaries that pose the comparison. Even after 2009, with the end of the Bush era in Washington, D.C., the election of a new president in Barack Obama, and what seems to be the advent of a changed American foreign policy, the number of works about Rome and the USA shows no sign of abating. Many classical journals and classical scholars have stayed away from these discussions, perhaps because they have doubts about the rigour of the arguments that are bandied about by commentators, or because they despair about the historical inaccuracy of the details, or simply because they wish to avoid controversies about the political present. This journal, too, has not commissioned a review of any recent book that attempts to compare ancient Roman and American imperialism.Footnote ¹⁶ A few classical scholars, more intrepid than their silent colleagues, have joined the discussions, supplemented them with their own critiques, explored the similarities between the Roman Empire and American Empire, or rejected the comparisons as unhelpful and partial. A smaller number still have even encouraged the comparisons and exhorted America to follow in the path of Rome.Footnote ¹⁷ But now here from the pen of Vaclav Smil comes Why America Is Not a New Rome, an entire book designed to finish off the idea that America and Rome are similar in any meaningful way. A professor of the environment, Smil has written prolifically about energy, food, and environmental affairs. For this book, he has immersed himself in Roman history — he proclaims Scheidel ‘the most incisive analyst of Roman society’ (118) — and digested a huge amount of information about the ancient world as well as about the United States. Among his admirers is Bill Gates, who has written a review of Why America Is Not a New Rome in which he refers to Smil as ‘a great student of history, including Roman history and the dynamics of its Empire over time’.Footnote ¹⁸

Smil's major aim is to refute the claim that Rome and America are similar, his secondary aim is to show that America is not an empire, and he seeks to meet these in a prose that is clear, direct, and forceful. He says his book ‘is not a book of comparative history — not histoire comparative but histoire comparée using Marc Bloch's (1928) classic distinction — and not preoccupied by any specific methodological concerns or written to meet the approbation of professional reductionist historians’ (2).Footnote ¹⁹ He has a number of targets to choose from, including Charles Krauthammer, Cullen Murphy and Joseph Nye, in the USA, and Peter Bender, Jonathan Freedland and Giovanni Viansino, in Europe.Footnote ²⁰ These writers have been influential at a variety of levels, and I can attest to the popularity among university students of Murphy's book, Are We Rome?, and Freedland's article ‘Rome, AD … Rome, DC?’ (published in the Guardian, in 2002, it accompanied a television programme ‘Rome: The Model Empire’, presented by Freedland). Smil's attacks are rarely ad hominem, and he writes that his ‘only goals are to inquire, illuminate, and explain and thus, I hope, to understand’ (2). The book contains many pages of useful information, and it will be read eagerly by students looking for help with their assignments and papers.

The relentless argument of the book is to insist that Rome and America are not the same: when he writes that he is repudiating ‘comparisons’, what Smil means is that he is fighting against claims about the similarity of Rome and America. Smil's analysis might be more effective were he to work with a deeper sense of ‘comparison’ and to recognize that the best comparisons are products of the creative imagination and not merely exercises designed to prove a statistical outcome. He tends to focus on similarity and difference, when a subtle grasp of the relationship between past and present is called for. But the study of the classical tradition is a difficult business. Tracing influences across time and place, working out the uses of Rome over generations and centuries, sifting through the legacies of empire and colonialism, analysing how individuals or institutions receive and refashion the past in responding to the pressures of the present, all of these processes require a layered sense of history and defeat the best intentions of pamphleteers, polemicists, and political theorists. Often carried away by the enthusiasm of his own arguments, Smil offers refutations that are presentist and unduly extended and sometimes also unnecessary. He asks, ‘Is not the fact of dramatically different lives in the Roman world and in the modern United States a truism not worth belaboring?’ (33). And so we are given page after page of data on ‘average income, affordable consumption, wealth distribution, and housing space’ and ‘energy use, reliance on inanimate prime movers, infant mortality, or common morbidity’ (146). The many differences between ancient Rome and the USA are weighed up, Rome is found wanting, and we learn that ‘by any comparative measure, an average Roman lived in misery, and an average American enjoys a level of affluence unprecedented in human history’ (146). I find that an unsurprising conclusion but also think it unhistorical and inadequately contextualized. A little more useful is the conclusion Smil arrives at later in the book where he states that ‘the most notable commonality between ancient Rome and modern America is the (vastly exaggerated) perception of their respective powers’ (158).

One of the interesting features of the comparisons between ancient Rome and the United States is their sheer volume and intensity. No one thinks that Rome and the USA are the same, and we may differ with each other on whether the USA can legitimately be described as an empire. But a pressing question is, why do so many feel the need to make the comparison? And why do so many feel the need to call the USA an ‘empire’? The answers to those questions are different but also connected, and the truth is that many critics and intellectuals have come to believe that the USA is an empire, that it resembles ancient Rome in this respect, and that its own decline lies in the near future. The subject of empire remains an emotive issue globally, and critics of American foreign policy will continue to use terms such as ‘empire’ and ‘imperialism’ in expressing their opposition.Footnote ²¹ It was precisely because empire roused strong passions that George W. Bush said, in 2004, in his State of the Union speech, ‘We have no desire to dominate, no ambitions of empire’, and it was to take the wind out of the sails of their critics on the left that neo-conservative writers sought for a while to redefine imperialism as a desirable strategy for America. Smil's remark that he conceived his book without ‘any ideological message in mind’ is, therefore, puzzling and somewhat belated (2). Today, only the most disinterested intellectual, or the most naive, or the most disingenuous could write a book that denied America was an empire and also claim that the book had no ideological import. The more urgent task would be to understand why so many around the world feel sufficiently oppressed and disheartened by the USA and its allies to want to call it an ‘empire’. For all the good that it has accomplished in the world, the US government has followed a policy of military intervention in Vietnam, Granada, Afghanistan, and Iraq, to name just a few places, that has led many internationally to think of it as imperialist. Frankly, America has given its attackers plenty of reasons for criticism, and whether or not they are right to call it an ‘empire’, the heavy weight of the charges still sticks. Take away the name ‘empire’, but the militarism, unilateralism, regime changes, attempts at nation building, and hypocrisy in foreign policy remain inextricably associated with the history of the USA.

‘To those who have long analyzed global politics through the lens of empire’, Jennifer Pitts observes, ‘whether from postcolonial or Marxist or other perspectives, the spate of literature on American empire produced since 2001 often seems to misperceive the phenomenon, to see novelty and exception in the American case where students of empire recognize reiterations of older patterns and tropes, or conversely to seek too simple lessons from past empires.’Footnote ²² Debates about whether a particular nation was an ‘empire’ go back, in English alone, to the later years of the eighteenth century if not earlier. The word ‘empire’ itself has a fraught and complicated history over the nineteenth century, a history that many twentieth-century and twenty-first-century intellectuals appear to forget when they hold forth on the subject.Footnote ²³ While it remains an urgent task to understand the status of comparative imperialisms at this particular conjunction of political, economic, and social forces, American imperialism should not be examined in isolation but seen within a large frame of global and historical concerns.

The comparative study of empires has enjoyed a resurgence because of the rôle played by the United States of America in world affairs since 1989, and, especially, because of US-led interventions in Iraq and Afghanistan. Yet the comparative study of empires is explicable in so many ways that it can also be thought of as overdetermined — the dominance of the USA, the rise of China on the international stage, the debates over globalization and multiculturalism, and the second and third waves of analyses prompted by decolonization are all factors that could be mentioned in this regard. The contributors to these volumes are aware of broader geo-political contexts, in which they formulate their analyses, and they seek to account for these wider spheres of activity in their discussions. They are also attuned to developments in a variety of fields such as legal history, political theory, and post-colonial studies. But one difficulty raised by their work is the apparent lack of attention given to older reflections about comparative empire and the history of scholarship on the issue. Shmuel Eisenstadt, Moses Finley and Michael Mann are frequently acknowledged in these chapters, but there seems to be little sense of previous treatments of empire, whether from the early twentieth century or the nineteenth century, let alone the eighteenth century. I did not expect to see obsolete titles by obscure authorities but rather a feeling for how questions and problems developed over time, and for an intellectual genealogy that engaged with the history of political thought. The comparative study of empires will benefit from an awareness that questions about empire have been posed in the USA, Europe, and elsewhere for centuries.

Consider the tradition of American exceptionalism, which I just mentioned, and which not only has a rich history but also developed in relation to ancient Greece and Rome. In colonial America, intellectuals and statesmen already felt compelled to articulate a relationship with Rome, to take sides with the Republic or the Empire, and to search for moral and political lessons in the writings of the ancients. But some also emphatically rejected parallels with ancient Greece and Rome and insisted on the dynamic singularity of America. At the Constitutional Convention, Charles Pinckney of South Carolina asked, ‘Can we copy from Greece or Rome? Can this apply to the free yeomanry of America? We surely differ from the whole. Our situation is unexampled.’Footnote ²⁴ America's difference and uniqueness was emphasized from an early date, before it was an empire, and was often formulated in opposition to the civilizations of the past. This sense of uniqueness illustrates how antiquity was deployed to nurture a particular vision of nationhood and how it could be used, by later generations, to mark out the nation's special status so as to cast America outside the field of comparative imperial history.

Or consider the issue of decline, which has been frequently invoked in discussions of empire. In drawing attention to imperial decline as an interpretive problem, Edward Gibbon was not immune to the historical forces swirling around him — European colonialism, the post-humanist fascination with Rome, and the growth of international maritime trade. His letters reveal a curiosity about developments in the American colonies and the East India Company and implicitly compare the British and the Roman empires.Footnote ²⁵ It would be important to understand how and why Gibbon developed a conception of imperial decline and how it was adopted or refuted by writers after him. Of course, Gibbon has been heavily studied and modern authors often refer to him and the Decline and Fall by name, but few who talk in the present about American or Roman decline seem to appreciate when the question of ‘decadence’ was framed in the Enlightenment or why it took on a particular urgency then.Footnote ²⁶ The answers are complex, but part of the explanation lies in the fact that, for many historians in the eighteenth and nineteenth centuries, the idea of decline was predicated, in the words of David Mattingly, on ‘a sense of wonder that something so emblematic of civilization and order should have been brought down by “barbarians”’ and on a strong identification between ancient Rome and modern Europe.Footnote ²⁷

With the comparisons of China and Rome, Scheidel, Mutschler and Mittag recognize that their publications appear at a time when many are writing about China's economic growth and a putative geopolitical rivalry with America. How this environment shapes their volumes is less clear than the fact that it looms large in the background, although it can be said, at least, that an avowed emphasis on ‘empire’ places these books in a separate category from comparative works on ancient scientific and philosophical thought.Footnote ²⁸ Further, the relatively small number of China-based scholars in these collections suggests that their approaches are shaped largely by contemporary Euro-American agendas and concerns. But these, in turn, have developed out of a long history of Western and non-Western encounters with China, and what is missing in these fine volumes is a treatment of how evolving political contexts have altered the shape and nature of imperial comparisons over time. We require a denser archaeology of knowledge in order to comprehend the connections between the analyses of the present and the cultural, intellectual, and political actualities of the past. In a word, we need to understand what makes it possible for us to grasp, or wish to grasp, two or more empires together. So dominant a force is imperialism in Euro-American history that we must speak of its historical influence across the centuries and examine the extent to which our own questions remain shaped by it.