The volume and its contents (and omissions)

Before turning to these bigger questions, let us first consider the volume itself. Sociological Studies in Roman History includes 14 articles and chapters written by Hopkins between 1965 and 2004 (two published posthumously), with a series of “Afterwords” contributed by his former students and colleagues. The volume has been very capably edited by Christopher Kelly, Hopkins's literary executor (xiii), who also provides a quite substantial Introduction (1–54). The essays republished here are characterized as “corrected reprints” (xviii). Original pagination is not continuously marked. Minor errors have been rectified, and formatting and references are standardized. Original publication details of the 14 chapters are as follows:

1. 1. “Contraception in the Roman Empire.” CSSH 8 (Reference Hopkins1965): 124–51.

2. 2. “A textual emendation in a fragment of Musonius Rufus: A note on contraception.” CQ 15 (1965): 72–74.

3. 3. “On the probable age structure of the Roman population.” Population Studies 20 (1966): 245–64.

4. 4. “Graveyards for historians.” In La mort, les morts et l'au-delà dans le monde romain, ed. F. Hinard, 113–26 (Caen 1987).

5. 5. “Economic growth and towns in classical antiquity.” In Towns in Societies: Essays in Economic History and Historical Sociology, ed. P. Abrams and E. A. Wrigley, 35–77 (Cambridge 1978).

6. 6. “Taxes and trade in the Roman empire (200 BC–AD 400).” JRS 70 (1980): 101–25.

7. 7. “Models, ships and staples.” In Trade and Famine in Classical Antiquity, ed. P. D. A. Garnsey and C. R. Whittaker, 84–109. Proceedings of the Cambridge Philological Society supplementary volume 8 (Cambridge 1983).

8. 8. “From violence to blessing: Symbols and rituals in ancient Rome.” In City States in Classical Antiquity and Medieval Italy: Athens and Rome, Florence and Venice, ed. A. Molho, K. Raaflaub, and J. Emlen, 479–98 (Stuttgart 1991).

9. 9. “Slavery in classical antiquity.” In Caste and Race: Comparative Approaches, ed. A. V. S. de Reuck and J. Knight, 166–77 (London 1967).

10. 10. “Conquest by book.” In Literacy in the Roman World, ed. J. H. Humphrey, 133–58. JRA supplement 3 (Ann Arbor 1991).

11. 11. “Novel evidence for Roman slavery,” PastPres 138 (1993): 3–27.

12. 12. “Christian number and its implications.” JECS 6 (1998): 185–226.

13. 13. “The political economy of the Roman empire.” In The Dynamics of Ancient Empires: State Power from Assyria to Byzantium, ed. I. Morris and W. Scheidel, 178–204 (Oxford 2009).

14. 14. “How to be a Roman emperor: An autobiography,” in History and Fiction: Six Essays Celebrating the Centenary of Sir Ronald Syme (1903–89), ed. R. Tomlin, 72–85 (London 2005).

As a collection of essays, this volume takes its place alongside those of other major Roman historians of the 20th c.Footnote ⁵ But it differs from those compilations in some ways. In the first place, it is not, like some of them, a comprehensive collection of Hopkins's article-length studies. The selection does illustrate the remarkable range of his interests, and it includes a number of major pieces, but important contributions from various stages of his career, on topics ranging from eunuchs and social mobility to “brother-sister” marriage in Roman Egypt and essential refinements to his “taxes and trade” model, are missing.Footnote ⁶ None of his published reviews appear, either, including the well-known and much-cited critical evaluation of Fergus Millar's magnum opus.Footnote ⁷ Nor are any of the individual chapters from his two earlier books, Conquerors and Slaves (Hopkins Reference Hopkins1978a) and Death and Renewal (Hopkins Reference Hopkins1983), included here. One would not expect chapters from books to be reprinted in a collection like this, of course, but those two volumes – billed as “Sociological studies in Roman history” volumes 1 and 2 – were not really monographs in any traditional sense. They were collections of essays grouped, sometimes loosely, around themes of recurring interest to Hopkins. And it goes without saying that there can be no full appreciation of Hopkins and his scholarship without consideration of such classics as “Conquerors and slaves: The impact of conquering an Empire on the political economy of Roman Italy”; “Divine emperors or the symbolic unity of the Roman Empire”; “Murderous games”; “Political succession in the late Republic (249-50 BC)”; or “Ambition and withdrawal: The senatorial aristocracy under the emperors.”Footnote ⁸ But the Introduction and Afterwords weave all of this additional material into the discussion quite effectively, and in any case, these 14 essays are more than enough to serve as a solid basis for an assessment of Hopkins's work.Footnote ⁹

The volume is also shaped by its wide-ranging and critical Introduction, which goes far beyond the summary accounts and bland encomia usually encountered in such collections, and by the concise but trenchant Afterwords to the individual chapters, which set the essays in their scholarly and intellectual contexts and sketch subsequent developments in the many areas touched by Hopkins's research. Sociological Studies in Roman History, then, differs from most collected essays in that it offers not just a representative selection of Hopkins's work, but also a set of expert judgments of it. In this sense, this collection of Hopkins's essays comes “pre-packaged” with its own “review.” Insofar as the Introduction and Afterwords have been written by Hopkins's former students, friends, and colleagues (“overlapping categories,” xvi), however, this is very much an insiders’ account. It is true that “pleasant personal reminiscence is deliberately shouldered into the background” (xvi), but the editor and authors of the Afterwords engage with Hopkins from positions of close familiarity, and their affection for Hopkins is evident throughout.Footnote ¹⁰ So there is ample room for additional, independent assessments of the essays reprinted here.

The essays and their claims

The first four chapters address different aspects of Roman demography, with Afterwords by Caroline Vout on Chapters 1 and 2 (99–104) and by Walter Scheidel on Chapters 3 and 4 (154–59). Asking how it was that the Romans (evidently) maintained a low fertility regime, and insisting that they did not necessarily think like modern Europeans when it came to matters of social practice, Hopkins argues that, despite some confusion between contraception and abortion, the Romans did in fact practice contraception (while acknowledging that there is not much direct evidence), with varying degrees of effectiveness, and with greater frequency in the early than the late empire (Ch. 1). In the pendant piece on Musonius Rufus (Ch. 2), he proposes an emendation from ἀτοκία, “childlessness,” to ἀτόκια, “contraceptives,” drawing on his philological training to make a small but significant contribution to the evidence for the latter. From this same early phase of Hopkins's career comes a broader methodological meditation on what it is that we can and (more to the point) cannot know about the age structure of the Roman Empire (Ch. 3). Insisting that epitaphs indicating age at death tell us more about commemorative practice than about mortality, and highlighting significant regional differences in the stated age at death on tombstones, he argues that “ancient inscriptions cannot be used as the basis for calculations of mortality, absolutely or even relatively” (132–33), a conclusion reached in part through systematic comparison between the epigraphic evidence and that furnished by the United Nations model life tables.Footnote ¹¹ That the tombstones by themselves are a poor metric for mortality in the Roman Empire is confirmed in a follow-up piece (Ch. 4), published just over 20 years later. Written from a self-consciously “pessimistic, even gloomy” stance (135), it draws attention to evidentiary biases relating to the urban/rural dichotomy, class, age, and above all commemorative practice, none of which, according to Hopkins, can be overcome.

The first three of these articles, from the mid-1960s, together with a fourth not printed here, on the age of Roman girls at marriage, more or less introduced the field of demography to the study of Roman history.Footnote ¹² The historical demography of the Roman world is now part of the scholarly mainstream, of course, brought to bear on everything from the social history of the family and cultural attitudes on the life cycle to the political economy of the Roman Empire and the disease profiles of the Mediterranean basin in antiquity.Footnote ¹³ Much of this work has moved well beyond what Hopkins was able to achieve in terms of scientific complexity and precision, but the impulse to grapple with various historical problems through a demographic lens is one that Roman historians owe in no small measure to his pioneering work.Footnote ¹⁴ Nor should we lose sight of the contribution to historical method that these articles made, especially in their injunction not to focus exclusively on the surviving evidence, as if it were some sort of neutral guide to the past, but to pay equal attention to the world that produced it. Scheidel's suggestion that these essays “are as much an exercise in applied epistemology as demographic analysis” (154) captures much of that methodological spirit.

The distinctive blend of big questions, new approaches, sensitivity to the limitations of the ancient evidence, and bold declarations is on full display in the next trio of chapters, which explore some macroscopic aspects of the Roman economy. Afterwords are provided by Neville Morley (207–12, on Ch. 5), Willem Jongman (260–68, on Ch. 6), and Peter Fibiger Bang and Mamoru Ikeguchi (306–12, on Ch. 7). From a characteristically simple and direct question – “Was there economic growth in classical antiquity?” (160) – Hopkins launches into a wide-ranging survey of what we are now accustomed to call “economic performance” in the ancient Mediterranean world. He focuses on the Roman Empire, suggesting that the increase in urbanization and the cumulative effects of aggregation – in luxury trade, small-scale production, and urban demand – will have generated modest aggregate growth over the long term (Ch. 5).Footnote ¹⁵ “Taxes and trade” (Ch. 6), which developed some of these ideas, is organized around the following six “propositions”:

1. [1] Rome's imposition of taxes paid in money increased the volume of trade in the Roman Empire

2. [2] the tax-exporting provinces had to earn the money needed to pay their taxes by selling goods of equal value

3. [3] the Roman conquest increased the amount of rent paid, especially in the western empire

4. [4] inter-regional trade increased substantially in the period 200 BCE–200 CE

5. [5] the supply of Roman coins increased enormously in the late Republic (157–50 BCE)

6. [6] the monetary economy of the Roman Empire became integrated into a single system in the first two centuries CE

Each proposition is carefully explored not only in terms of its own logic, but also with respect to the other propositions, and each is illustrated (not proven) through different sorts of evidence. The whole elaborate edifice is meant to support the larger argument that taxation indirectly stimulated economic growth in the Roman Empire. In one of several subsequent elaborations of this model (Ch. 7), Hopkins explores the structural relationships between inter-regional, intra-regional, and inter-town trade, arguing against the grain that most transport in the Roman world was in fact short-haul and overland, and that the small-scale demand of individual urban consumers, including those living outside of the megalopoleis of Rome, Alexandria, Antioch, and Carthage, was very substantial in the aggregate and has been underrated.

These three essays, developed and refined in several subsequent publications, have given rise to a consensus picture of the Roman economy in the period from, say, 200 BCE to 200 CE, characterized in particular by long-term aggregate growth.Footnote ¹⁶ All sorts of additional quantifiable data, drawn from everything from lead pollution in the deep ice cores of Greenland to the aggregated vat capacities for the production of fish products, has been employed to confirm the basic validity of this diachronic reconstruction.Footnote ¹⁷ So influential has this “mildly developmental” view of the Roman economy become, in fact, that it underpins both the Rome chapters of the Cambridge Economic History of the Greco-Roman World and the several volumes and projects associated with the Oxford Roman Economy Project – a pretty remarkable bridge between two otherwise divergent approaches to Roman economic history.Footnote ¹⁸ It is true that we are beginning to see the first signs of a concerted reaction against this whole picture and approach, based in part on more granular investigations of rural economies than anything attempted (or perhaps even envisioned) by Hopkins, but it would require at least a generation of work to unseat the current orthodox view of the Roman economy, so deeply entrenched has it become.Footnote ¹⁹

The essays reprinted in the second half of the volume (Chs. 8–14) spiral outward from these foundational studies of demography and economy to touch on diverse facets of the social, cultural, and political history of the Roman world. Two concern the institution and experience of slavery, with Afterwords by Keith Bradley (357–62, on Ch. 9) and Catharine Edwards (425–31, on Ch. 11). The approaches in the two pieces could hardly be more different. In the first (Ch. 9), which originally appeared in a comparative volume on caste and race in world history, Hopkins contextualizes Classical slavery with reference to other forms of dependent labor and, relating its incidence to trade and empire, identifies its prevalence in Athens and Rome as “politically determined” (356). The second is rather more daring (Ch. 11). Asserting that social history can be written both from what he calls “real histories” and from fiction, and that when it comes to cultural interpretation “there is little justification for privileging one above the other” (399), he offers a close reading of the Life of Aesop as a way to reconstruct the lived, affective experience of both enslaved persons and slavers, exploring the paradox of the former as persons who were “socially despised but symbolically central” (418). This essay opened a rich vein of historical interpretation that has inspired many high-profile studies of the experience and, above all, imaginary of ancient Roman slavery.Footnote ²⁰

An essay on Roman ritual (Ch. 8), with Afterword by Jaś Elsner (340–45), stands out from the rest of the volume as an exploration in cultural anthropology. It is an attempt to understand Roman rituals not as Romans themselves experienced them, but rather as parts of a complex and composite system: “I want to envisage how the whole gamut of Roman rituals worked” (337). The defining features of Roman rituals, according to Hopkins, were (i) their subdivision of the populace into “competitive strata”; (ii) the complexity of their procedure; (iii) the differentiated and cross-cutting affiliations which they encouraged; and (iv) their obscurity and mystery (324–26). All of this is illustrated through brisk discussion of the Lupercalia, the levy, the census, and the popular assemblies (the emphasis on political ritual is marked and self-conscious). The final section considers whether or not the gradual disappearance of Republican ritual under the Principate left the empire “vulnerable to attack across its own symbolic and ritual frontiers” (339), a fascinating question to which Hopkins is reluctant to provide a straightforward answer, merely gesturing towards the eventual triumph of Christianity.

Two essays from the 1990s tackle complicated problems of social history through a characteristic mix of quantification, guesstimates, and arguments that are “deductive, hypothetical and impressionistic” (366, n. 5). In the first (Ch. 10), which is in close dialogue with the work of William Harris, who provides the Afterword (391–97), Hopkins more or less accepts the latter's figures for very low rates of literacy throughout the Greek and Roman world. He nevertheless argues, with particular reference to the documentary evidence from Roman Egypt, that the total number of adult males in the Roman Empire who could read and write was significant in the aggregate (ca. 2,000,000) and that Roman state power brought about a major increase in overall literacy, and with it a degree of cultural integration, that was without parallel in the ancient Mediterranean. The essay is notable, among other things, for its system-level approach, evident in the inferences drawn from the effects of aggregation (the total number of literate persons in the empire) and in the insights drawn from the mutually constitutive effect of two empire-wide integrating phenomena, literacy and monetization (389), not normally studied in tandem.Footnote ²¹

The article on “Christian number” is perhaps the boldest piece in the whole volume (Ch. 12, with Afterword by Kate Cooper, 481–87). It can be read as a brazen romp through the social and religious history of the high and late empire, covering everything from the size and growth of the early Christian community and the asymmetric relationship between Judaism and Christianity to the nature of parametric modeling and deductive reasoning, the concept of “community,” and the vertical social structure of the Roman world. Along the way, it makes a number of startling pronouncements – including the (in)famous calculation that there were 42 fully literate adult male Christians in the Roman Empire in 100 CE (463).Footnote ²² But it is no mere congeries of provocative observations. Indeed, this article offers nothing less than a comprehensive explanation for the particular (and peculiar) development of the early Christian church, with emphasis on the relatively small size and rapid growth of early communities of worshippers; the limited literacy that prevailed within them; and the importance of women for small-group cohesion. All of these variables together led, on Hopkins's account, to “the exclusivist and dogmatic character of [Christian] self-representation” (471) and to the magnetic idea, “which so differentiated Christians from pagans and Jews,” as he emphasizes, “that Christians were made, not born” (472), a distinctively Christian package that “gave to the Roman state a degree of symbolic unity and exploitable loyalty which it had previously lacked” (477).

The volume concludes with two essays published posthumously, one on political economy (Ch. 13, with Afterword by Greg Woolf, 527–33), and the other an attempt to write imperial “autobiography” (Ch. 14, with Afterword by Mary Beard, 549–55). Strikingly different from one another in subject matter and style, these two late works illustrate, right up to the end of Hopkins's life, the wide range of his interests in and approaches to the Roman world. “Political economy,” in the form we have it, is a pretty austere piece that can be read as the culmination (or, as Woolf suggests, simply the “latest”) of his lifelong thoughts about the Roman Empire as a system of power.Footnote ²³ Beginning from the central state's evolution from a republican government, and covering inter alia fiscal organization, the distribution of wealth, and the key roles of the city of Rome and the imperial army as concentrations of collective power, it provides a neat and quite compelling sketch of how the Roman Empire endured over space and time, an achievement related implicitly to long-term economic growth. In “How to be a Roman emperor,” Hopkins attempts to channel the thoughts of Septimius Severus – employing an inventive reconstruction similar to that of A World Full of Gods, perhaps his most controversial work – imagining with full artistic license what it was like to be an emperor operating within the crucible of the imperial court, omnipotent and powerless at the same time.Footnote ²⁴ Even if one objects to the approach (or to its application), it is impossible not to come away from this piece without some new questions about and insights into the nature of monarchy in the Roman world.

Evidence, method, and the road ahead

In reflecting on these essays, the earliest of which was published nearly 60 years ago (the span of two full scholarly generations), it is clear that the main significance of the volume as a whole is in revealing just how different Hopkins was from most other Roman historians – then and now. One way to understand what made Hopkins so distinctive is to recognize that he was guided, from the beginning of his career to the end – and from the opening of any one essay through to its conclusions – not by evidence but by questions. This is a simple point but an essential one. It is an approach to historical interpretation that reflects his training in sociology, of course, and a commitment to viewing the Roman world from an etic perspective.Footnote ²⁵ But this interpretive stance was not simply a matter of disciplinary orientation. It was also the product of an epistemology characterized by skepticism about the representativeness, and therefore the evidentiary value, of what has survived from antiquity. On this matter, so fundamental to how we can write ancient history at all, Hopkins was quite consistent. The “truth” of “propositions” about the ancient world, in his view, “can be illustrated but not proved.”Footnote ²⁶ At the heart of such declarations is the a priori conviction that the surviving evidence from antiquity is rarely (if ever) probative by itself. As a result, it is the wrong starting point. This outlook is stated categorically in several places, such as the article on Christian number, in which he declares that “History should not be, pace the practice or presenting style of many colleagues, an amalgam of sources” (439).Footnote ²⁷ This is an attack both on induction as a method and on the logical positivism that underpins it. We simply do not have enough reliable information from the ancient world to work in that way, so the logic goes.

How, then, did Hopkins manage to mount any arguments at all? As is well known, he arrived at his substantive conclusions through deduction, as he himself points out often enough (see below), but it is worth pausing for a moment to consider three different bridges that helped him to get from questions to conclusions: archaeology, comparison, and quantification. Let us examine each in turn.

In terms of archaeology, it is true, and important, that Hopkins made much use of material evidence in his historical reconstructions. This is perhaps most evident in the macroscopic discussions of the Roman economy, in which the evidence from shipwrecks, to take the most obvious example, bulked so large.Footnote ²⁸ More generally, it was in large part through the use of archaeological evidence that Hopkins exploded an older view of the Roman economy, most closely associated with the influential work of his own teacher, Moses Finley, which had emphasized insularity, minimalism, stagnation, and embeddedness. The regional and supraregional movement of goods throughout the Roman Empire, increasingly visible in an archaeological record to which Hopkins was alert, and which he was prepared to exploit, made the old Finleyan view of the economy untenable.Footnote ²⁹ In terms of method, though, it must be noted that Hopkins had some scathing things to say about archaeologists. With respect to skeletal data, for example, he writes that “the ratio of hard work to intellectual reward is lamentably meagre” (146).Footnote ³⁰ Archaeologists have overrated the importance of pottery simply because so much of it survives (170). The polemic against the gathering of information for its own sake, as an end in itself, is directed with particular force against archaeologists, who “seem more interested in doing another dig and writing up last year's finds,” as he puts it, “than in making sense of the last generation's advances” (199, n. 79). Despite the rhetorical posturing and “signaling effects” of such comments, it is quite clear that the archaeological remains, especially when considered in the aggregate, were essential in his efforts to break free from the narrow concerns of the textual sources. And the results, as we have seen, were transformative for our understanding of the political economy of the Roman world.

Another tactic employed to shed light on a world with such a paucity of useful evidence was historical comparison. Comparative perspectives are strewn throughout the Hopkins corpus. Some material is drawn from the ancient Mediterranean world itself, such as the place of contraception in the Talmudic tradition (80–82) or the distinction between ancient slavery and Late-Antique serfdom (351). Other evidence is brought in from later periods in the premodern history of the Mediterranean, such as the Genizeh papyri from Cairo (ca. 950–1250 CE), used to illustrate patterns in trade (177–78). The modern period makes several appearances as well, with comparative insights on antiquity derived from slavery in the British West Indies and US South (353, 400–2) and from the demographic information recorded in the UN model life tables (106, 114–34). Indeed, the comparative instinct is already visible in the 1965 article on contraception (Ch. 3), in which Hopkins remarks not only on the Talmud but also on practices of birth control in early modern and modern England, the United States, and India (83–88). Though Hopkins never pursued historical comparison in a comprehensive or systematic way, his inclination to look to worlds beyond the ancient Mediterranean put him ahead of the curve.Footnote ³¹

A third key method in his toolkit was quantification. When we think about quantitative approaches in ancient history, though, it is essential to recognize that there are two very different types of quantification: the tabulation of actual datasets on the basis of empirical evidence, on the one hand, and the parametric and speculative modeling based on hypothetical data, on the other.Footnote ³² Hopkins did engage periodically in the former. In the “Taxes and trade” article (Ch. 6), for example, we find careful tabulations not only of shipwrecks (above, n. 28), but also of coins (in circulation and in production, 224–40). The aesthetics of the exposition were important, too: tables, charts, graphs, formulas, and visualized data had not been common in most writing on the ancient Mediterranean. Such empirical quantification and display are now de rigueur in much of the scholarship on the history of the Roman economy, most prominent, perhaps, in the aims and ambitions of the Oxford Roman Economy Project (above, n. 18). In this regard Hopkins has had an obvious impact.

Far more common in his own writing, though, and more characteristic of his scholarship overall, was the “quantification” of hypothetical data. Hopkins was certainly capable of rigorous work in this domain. The data assembled for calculating the age structure of the Roman population (114–27), for example, is quite complex (at least for the non-specialist), and the discussions of the interrelationships between price, money supply, speed of monetary circulation, and quantity of goods at the heart of “Taxes and trade” were unusual for the time in their econometric sophistication (223–33; cf. 240–49 on taxation and aggregate consumption). The calculations employed to quantify the growing number of Christians in the Roman Empire were also elaborate (440–53). The purpose of such “guesstimates” (442, n. 17) was “to give rough orders of magnitude to dimensions which would otherwise remain nebulous” (272), in order to sketch “the main outlines of probability” (288) and, finally, to generate “a limiting case, against which the fragments of evidence can be tested, or around which they can be fitted” (442). This parametric modeling was designed, in Kelly's words, “to establish the field of play” and “the boundary conditions – the parameters – of any proposition” (9), and it was precisely from such modeling – and the hypothetical data upon which it was based – that he arrived at many of his most provocative, and influential, conclusions. Quantification of this sort is less common in current work on the Roman world, but there are signs of renewed interest in (and, indeed, improvements upon) this approach.Footnote ³³

It was in part through archaeological evidence and comparison, then, but above all through a particular type of quantification, which enabled parametric modeling, that Hopkins bridged the gap between limited evidence and robust conclusions. The clearest exposition of his principal working method, which we may call “deduction,” as a convenient shorthand, comes in the 1995/6 article on taxes and rents (above, n. 16); he describes the function of the model, which is the beating heart of the deductive approach, as follows:

He then goes on to sketch the main principles of the model he constructs to understand Roman economic history, including the primary importance of logical relationships and arguments, which order facts (again, not the other way around); the inverse relationship between the scope of the model and the chances of “large proportionate error”; and the building up of a “network of estimated probabilities” (193–95). And with all of this came an important corollary that again speaks to his skepticism about the value of the surviving evidence: “Because of our imperfect knowledge, each argument by itself may be weak, but in an elegantly constructed model, the imperfect, but complementary arguments prop each other up” (193). It was out of such “wigwams” that Hopkins constructed his historical worlds.

From this same piece, written well into his tenure as Professor of Ancient History, he also makes a startling admission that I do not think he would have stated quite so baldly earlier on in his career:

It is clear that Hopkins was in fact committed to detailed control over the traditional evidence, and that in one respect his history was, in Kelly's words, “as inescapably anchored to a bedrock of traditional source-citation as the ancient history-writing to which he objected” (9).

I nevertheless believe that Elsner comes closer to what made Hopkins tick when he writes, “Perhaps, in the end, the sociological theorist in him was stronger than the empirical historian” (343). For what really made Hopkins tick, as the essays collected in this volume show, were questions and arguments rather than facts. In this regard, it is my impression that the field of ancient history as a whole has not (yet) fully responded to this imperative. So much of the energy in ancient historical work today is organized under the umbrella of digital humanities, which has brought obvious benefits in terms of computation, access, and display, but which has also tended to fetishize data as an end in itself. Perhaps it is time to take our cue from Hopkins and really mobilize all of this information – the databases, the maps, the visualized networks, and so on – in order to pursue historical questions old and new. Some of this is already happening, of course. The data-intensive work associated with the Oxford Roman Economy Project certainly addresses big-picture questions, and the granular archaeological research underpinning the Roman Peasant Project (above, n. 19) promises to revolutionize our understanding of the rural world in Roman antiquity. Nor is the movement from small (discrete pieces of evidence) to big (large-scale generalization) limited to archaeology. Discussing the historiographic potential of “the new thalassography,” for example, Nicholas Purcell writes, “There are alternatives [sc. to doing global history via regional building-blocks] – above all, bigger generalisations derived from longer series of smaller-scale observations, crossovers from the iterated small-scale structures to the larger ones which are of service to history on the biggest scale…”Footnote ³⁴ The observation concerns the historiography of oceans, but the underlying principle – i.e., the making of “bigger generalisations derived from longer series of smaller-scale observations” – is surely relevant, mutatis mutandis, to all of ancient Mediterranean history and archaeology.

Ever more granular empiricism together with ever more sophisticated modeling seems the best way forward. And we will continue to rely on experts in the one or the other approach, those who are more comfortable proceeding through induction and others who prefer to reason deductively. For our collective work to move the needle, however, we must never lose sight of the underlying questions and problems. We must always examine the evidence, such as it is, not for its own sake, but rather in the service of interpretation and analysis, and all of it, the evidence and the argumentation together, mobilized, ultimately, in the interest of improving our understanding of the ancient Mediterranean world. Hopkins's essays serve as a powerful reminder of this fundamental goal. We can still read them for their substantive contributions. But the magic is in the method. We owe our thanks to Christopher Kelly for bringing all of this work out in such an exemplary package to serve as an inspiration for a new generation of Roman historians and archaeologists.