Peregrine Horden and Nicholas Purcell

AHR Forum

The Mediterranean and "the New Thalassology"

Sea and ocean history is more novel than it sounds. It admirably exemplifies a new historiography of large areas. In one sense, the inquiry revisits a traditional historical geography. At the same time, both its scope and its methods are so distinctive as to make it an exciting—and quite unpredictable—area of reflection and research.

A decade or so ago, such histories were hardly encouraged. In postmodern historiography, subjects of every kind, but especially identities and power relationships, became inflections of discourse. And because it made its own world, discourse could not be anchored to a particular place, even if, for practical reasons, its historians had to use texts from a particular region and period. Preset frontiers were taboo.¹ The context, whether for an exercise in microhistory, in new historicism, or in postcolonialism, was, ideally, global.

Area studies, in many ways the creation of cultural anthropology and long suspect in that discipline, now flourish again.² But they are invigorated by what has been learned from the linguistic turn: there is no turning back. Even area studies, in their revived form, are therefore in some sense global, too. They may, for instance, be directly or indirectly influenced by debates on the nature and impact of economic and cultural globalization. The new interest in regional history derives, fundamentally, from the task of finding a different approach to world history—not through formulating generalizations about everything, but through the analysis of the whole by way of its components, and, consequently, of how those components fit together.³ The regions of the new regional history tend to have one obvious characteristic: they are big—inevitably, since they are implicitly or explicitly elements in a larger and potentially all-embracing historical project. For the same reason, their historians share a sophisticated consciousness of the problems of delimitation.⁴

The new regional history also (hardly surprisingly) retains the taste of late-twentieth-century scholarship for the erasure of established disciplinary and historical frontiers. Some existing categories seem reinvigorated; others are new. But a further common feature is that they cut across the political divisions that have shaped traditional history: the study of the lakes of the East African Rift Valley, for example, or the Silk Road.

History of seas appeals to this project because the layout of sea and land makes the oceans and their embayments a way of approaching most parts of the world. By a simple metaphorical extension, "virtual seas" can be included, too, spaces of danger and variable communications—mountain ranges, forests, or arid wildernesses such as the Sahara. Some of these will resemble oceanic expanses; others, in being more densely surrounded by populated zones and dotted with island oases, will be inland "seas" like the Mediterranean. The systematic comparison of real and metaphorical seas can suggest a new configuration of history, and one that might attain a global scale.⁵ So promising, indeed, does the notion of a sea or an ocean appear for this task that the term "the new thalassology" has seemed an appropriate coinage to denote it.⁶

The choice of sea sometimes follows traditional geographical classifications, as with the Atlantic, Pacific, and Indian oceans; but it is more enticing in many ways to propose less familiar, and sometimes smaller, maritime spaces, such as the Philippine island world.⁷ Novel species of juxtaposition and comparison become possible. They reveal unexpected coherences, such as the recurrent features of the northeast shores of the Atlantic in the premodern archaeological record.⁸ Peripheries become cores, and it is arguably one of the main attractions of the newly created or identified areas that they tend to be politically neutral. Apart from ignoring national boundaries, they subvert imperial hierarchies that privilege some powers' involvement in the areas in question. Thus, for instance, in the "new" Atlantic historiography, a "white," a "black," a "green" (Irish), and even a "red" (Marxist) Atlantic may coexist in equilibrium.⁹ Sea history also helps to expose the "myth of continents" and the precedence that historians have given to land over water as the support of social life.¹⁰ "East Asia" may be preferable to "the Far East," but the investigation of the continuum of the China Sea is more liberating still. Last, and not least, we might guess from our own experience, sea history allows sedentary landlubber historians to indulge a taste for the romance or the frisson of seafaring.¹¹

The Mediterranean has some claim to be the great original of seas as the subject of history. Sea-based political hegemony was identified as an object of historiographical inquiry in the fifth century b.c. by the first historians in the Western tradition, Herodotus and Thucydides, who invented for it the label "thalassocracy" (sea rule). From still earlier, in the Odyssey, seafaring occupied a central place in the first Greek literary imagination.¹² Even more influential in the twentieth century has naturally been the prototype sea history, Fernand Braudel's The Mediterranean and the Mediterranean World in the Age of Philip II.¹³ Braudel elaborated a unitary theory of time and causation, and applied it, on the grandest scale, in that characteristic product of the Annales school, the regional monograph in which all the social sciences are synthesized into "total history." "When I think of the individual," he wrote in the concluding paragraph of the second edition of The Mediterranean,

I am always inclined to see him imprisoned within a destiny in which he himself has little hand, fixed in a landscape in which the infinite perspectives of the long term stretch into the distance both behind him and before. In historical analysis as I see it, rightly or wrongly, the long run always wins in the end ... I am by temperament a "structuralist," little tempted by the event, or even by the short-term conjuncture which is after all merely a grouping of events in the same area.¹⁴

Yet, in that "long run" that always wins, Braudel's legacy has lain less in philosophy than in geography, through which the Mediterranean is upheld as a distinctive unity:

None of my critics has reproached me for including in this historical work the very extended geographical section which opens it, my homage to those timeless realities whose images recur throughout the whole book ... The Mediterranean as a unit, with its creative space, the amazing freedom of its sea-routes ... with its many regions, so different yet so alike, its cities born of movement, its complementary populations, its congenital enmities, is the unceasing work of human hands; but those hands have had to build with unpromising material, a natural environment far from fertile and often cruel, one that has imposed its own long lasting limitations and obstacles. All civilization can be defined as a struggle, a creative battle against the odds; the civilizations of the Mediterranean basin have wrestled with many often visible obstacles, using sometimes inadequate human resources, they have fought endlessly and blindly against the continental masses which hold the inland sea in its grip ... I have therefore sought out, within the framework of a geographical study, those local, permanent, unchanging and much repeated features which are the "constants" of Mediterranean history.¹⁵

For all its emphasis on the constraining force of a "timeless" landscape, Braudel's vividly peopled geography is maritime as well as terrestrial. It allows as much conceptual space to freedom of movement and communication across and around the sea as to the local hold of the environment. Above all, it places the Mediterranean region within a larger frame, a "greater Mediterranean" that can include Antwerp and places still more remote.¹⁶

This distinguished genealogy has helped make the Mediterranean a standard point de repère for maritime historians. The Baltic/North Sea region is "the Mediterranean of the North." The world created by later medieval Iberian navigators around the islands of the eastern Atlantic, in the period preceding that for which a "full" Atlantic history is conceivable, is the "Atlantic Mediterranean" or "Mediterranean Atlantic."¹⁷ David Abulafia has even generalized the concept of the Mediterranean, in its strict meaning of "inland sea," to a range of areas over which trade and communication has been comparatively intense. He applies the term not only to the "sub-Mediterraneans" of the Adriatic,¹⁸ Aegean, and Black seas,¹⁹ and the quasi-Mediterranean of the Caribbean, but also to the "Japanese Mediterranean."²⁰ The Mediterranean proper appears to have an unassailable position in the mainstream of historiography, and to lend itself to the comparativism of modern regional study. So it is reasonable to expect that the Mediterranean will also have much to offer the new thalassology.

But does it? In what follows, we address that question in four parts: first, by reviewing the serious criticisms that have been leveled at "Mediterranean" as a category; second, by looking at how other Mediterranean historians have responded to the challenge; third, by outlining our own suggestions for doing Mediterranean history; and finally, by debating the place of the Mediterranean thus configured in a wider scholarly world of comparative sea history and global history.

The "original" Mediterranean is in fact more problematic as a component in the new regional history than its illustrious forebears might suggest. First, when historians elsewhere are robustly constructing regions out of unexpected seaboards, so as to ponder recherchénetworks of communication and patterns of cultural resemblance, those studying the Mediterranean find that their subject is frequently dismissed precisely on the grounds that it is no more than a construction. Second, when the new regional vision helps to emancipate scholars from insidious political focalization, the Mediterranean stands accused of being an essentially oppressive concept, born of imperialism and deployed in the service of politically undesirable master narratives. Third, even apart from the specific ideological undesirability of links with European imperialism, the Mediterranean carries so much baggage of a more general kind that its usefulness in comparative history can seem very limited. Its problems as a historiographical subject are those of success.

Such problems tend to accentuate its uniqueness. No other sea has been the focus of a cultural history in which a unified stretch of water gives its name to a supposedly distinctive surrounding landscape and culture, and even to presumed personality traits.²¹ The relative seniority and sophistication of Mediterranean studies as a discipline is also significant, even if we date it back no further than the 1930s, when Braudel conceived his major work.²² A greater challenge for the historian is that the Mediterranean is culturally freighted in so many ways. This freighting is of course a far larger phenomenon than could be conjured up from the history of geography.²³ It hardly needs stating that the Mediterranean, as an area seldom clearly defined but often subconsciously reduced to Italy and Greece, has been the perceived center of European civilization since the Renaissance, and in many respects since classical antiquity. "The Baltic of the South": the absurdity of the inversion makes the point.

The cultural stature so often ascribed to the Mediterranean renders it unpalatable for many. It is unpalatable for anyone who sees in the conception and study of Greco-Roman antiquity simply another manifestation of imperialist ideology. (Perhaps we should avoid the Greek-derived term "thalassology" for that very reason.) It is unpalatable for those who are vigilant against all that a classical education contributed to the self-perception of the elites who invented and tried to rule the Mediterranean from the eighteenth century to World War II. The fame of the Mediterranean also lost nothing from the culture of international tourism that developed in the second half of the twentieth century, even if this culture is one from which historians may find it relatively easy to distance themselves.²⁴ These mixed reputations distinguish the Mediterranean—for now—from most other seas and oceans. Moreover, they raise serious questions about what sort of subject it might be.

Such questions underlie the most extreme attack on the whole notion of Mediterranean studies. The cultural anthropologist Michael Herzfeld has reinvigorated his critique of what he calls "Practical Mediterraneanism." His essay of that title is subtitled "Excuses for Everything, from Epistemology to Eating."²⁵ At the beginning of a new century, he finds it astonishing that people are still talking about the utility of "the" Mediterranean as a construct when almost all comparable categories (he gives no examples) have been deconstructed or reconstructed—or have self-destructed.²⁶ This conceptual durability is, however, a matter not just for astonishment or regret but for investigation. The Mediterranean is indeed "out there," and not only in the limited sense that it is a stretch of water. It exists as something "we say," as a subject of various "discourses." (The argument can be summed up, not altogether unfairly, with an analogy: theology exists, therefore God exists.²⁷) For Herzfeld, "we" are all kinds of people: inhabitants of circum-Mediterranean lands, for instance, who are fully aware of the stereotypical traits attributed to us in a massive literature, historical, literary, anthropological, and touristic. Inhabitants' excuse for misbehavior can be "the Mediterranean temperament" (hot-blooded, selfish); their excuse for political incompetence can be a culture of mañana.²⁸ "We" are also academics. As Mediterranean specialists, we insert ourselves into a global hierarchy of value, a hierarchy in which "Mediterranean" falls somewhere between "modern" and "primitive," the cozily familiar and the wholly exotic. To sustain such academic alignments, we fall back on a circular argument, from "the Mediterranean" as a distinctive and homogeneous region or culture to its particular exemplifications—and back again. Mediterraneanists always end by validating what they had presupposed.

Herzfeld's target is the strongest and most complex version of the sin of "Mediterraneanism."²⁹ It is a telling index of the methodological zeal that the Mediterranean elicits that, besides being an excuse, Mediterraneanism exists in (at least) two other forms: exclusivism and exceptionalism. The first is relatively simple—things Mediterranean are a tired subject, and their students tend to be introverted, it is suggested, when they could be interested instead, or at least also, in more stimulating and less well-trodden fields. The term "Mediterraneanism" has also been used to identify a claim that, especially in the more distant past, served to justify Mediterranean exclusivism—a claim that this area and some of its cultures have a special status among historical subjects, and that comparisons further afield are scarcely needed.³⁰ And there is a clear link between some of the forms that exceptionalism has taken and the world of the Herzfeldian excuses, which is a Mediterranean of subaltern status, explicitly modeled on the late Edward Said's "orientalism."³¹

Herzfeld reasonably requires those who use the Mediterranean category to see themselves as subjects of ethnographic investigation. They should put "the Mediterranean" within the frame rather than assume it as the frame itself. It becomes an object of research rather than an analytic tool. Historians must also put themselves within the frame: Why do they want to use this term? What (self-)interest does it serve? There can be no doubt that Braudel—for instance—fails any such reflexivity test. It is clearly no longer permissible for the answer to the Herzfeldian question "What is your excuse for doing this?" to be simply, as in the preface to Braudel's first edition, "I have loved the Mediterranean with passion."³²

To study this area without problematizing it almost out of existence is thus, for Herzfeld, to collude in a "disdainful cultural imperialism"—and not only cultural. For, as Said is taken to have shown, cultural and political imperialisms are always intimately linked.³³ Herzfeld has been criticizing cultural anthropologists of the last few decades.³⁴ But his critique clearly has a larger historical dimension. It relates to the history of politics and culture within the circum-Mediterranean area.³⁵ And it applies above all to the history of geography.

The region as it is usually defined today was delineated in the nineteenth century, not earlier. A Mediterranean world was, for example, elaborated in the early systematic anthropogeography of Germany. It emerged as a by-product of the general conceptualization of space under a heading with a terrible future, Lebensraum.³⁶ By the early 1900s, the geographer Theobald Fischer could explicitly connect geographical knowledge in Germany, as a world power and a force in world trade, with the unexploited potential of "slumbering" Mediterranean lands.³⁷ The connection between scientific knowledge and imperialism in the Mediterranean is, however, clearest in the French case.³⁸ The Mediterranean was represented as a European sea—the characterization of the Mediterranean as a region with distinctive relief or flora came more slowly, and in more specialized and not always geographical works. Striking comparisons were drawn: Kabylia as a sort of Provence, the Atlas Mountains as an analogue of the Alps.³⁹ Yet comparisons could work the other way, too—between a village in the Var and one in Morocco, or the stony plain of the fossil delta of the Durance and the Sahara. Braudel's remote intellectual ancestry comes into focus.

Although not straightforward, the nineteenth-century "invention" of the Mediterranean, as a region and not just a sea, is thus certain.⁴⁰ Only then did it become, like Bismarck's Europe, a full-fledged geographical expression. The environmental characteristics forming the "givens" of twentieth-century Mediterranean geography had not been systematically identified earlier. It is hardly surprising that no other region has inspired such hand-wringing among intellectuals over the question of whether it is an imperialist category. There is, to our knowledge, no comparable debate over the propriety of "Atlanticism" or "Pacificism." (The former is, rather, a term of art in strategic thought and refers—positively—to the Atlantic alliance between the U.S. and Western European powers; the latter has to do with peace, not water.)⁴¹ This despite the fact that "Atlantic," unlike "Mediterranean," is a name charged with classical mythology and, like all other ocean names, a European creation.⁴² Specialists in the Atlantic and the Pacific may, as elsewhere in this forum, debate the propriety of their disciplinary labels. But these are not, or not yet, seen as so politically tainted as to demand some alternative.

How have Mediterranean historians responded to Mediterraneanism, now that its insidious variety has been so loudly proclaimed? Few seem to have taken up the approach advocated by Herzfeld, which would confine their studies to politically attuned discourse analysis. We have no comprehensive historical and ethnographic study of "the idea of the Mediterranean."

More frequently, the whole problem of characterizing the region has been ignored or sidestepped. Braudelian human geography—on which differences of period (ancient/medieval), and even differences of culture and religion (Christian/Muslim), press only lightly—may continue to fascinate, and to provide useful illustration for "history from below." And yet such a geography is precisely what has been missing from many of the most significant contributions to historical writing on the Mediterranean published around the turn of the millennium. Perhaps it has seemed reasonable to suppose that there is no need for further environmental study, at least on any scale, because everything that needs to be done has been done already, and excellently—by Braudel. In 2000, we argued, with deliberate provocation, that Braudel's Mediterranean book marked "the end of the Mediterranean."⁴³ We had many reasons for doing so. During the preceding decade, history and the various relevant social sciences all seemed to have moved away from grand regional studies. In any case, what more was there to write about? Braudel had done the job of studying the relations of people and environments—on plains, mountains, islands, and coastlands; along routes and in towns; and on the open sea. He had seemingly done this not just for the age of Philip II, but for all time—or certainly for premodern historical time. None of his subsequent writings on the Mediterranean added anything of substance.⁴⁴ More than twenty years after his death, who now reads Braudel on Philip II? Yet for any student of Mediterranean human geography, his book is still an invaluable resource.

Evading the problem of geographical characterization allows "the Mediterranean" to flourish as a flag of convenience (if not an excuse). To show how, we can draw a distinction between "history in" and "history of" the region.⁴⁵ It is, admittedly, not a wholly satisfactory distinction. "History of" suggests the all-encompassing, although it is meant to refer to history of the whole in a particular geographical sense, not to history of all conceivable Mediterranean subjects. Braudel's Mediterranean is of course the great exemplar of "history of." On the other hand, "history in" the region is hardly the ideal way of conveying what "history of" might sometimes leave out. "History in" the Mediterranean is only contingently or indirectly Mediterranean. That is, reference to the environment would need to come in the extreme background of analysis or explanation. It would, for example, be the point (which we shall not try to identify here) at which the historian finds the ineluctable Spanishness of the Spanish Civil War—Spanishness in a geographical sense, rather than the cultural or psychological essentialism beloved of an older national historiography. No "history in" the region is utterly detachable from its Mediterranean setting. Still, the "in/of" dichotomy is a useful one.

It is useful first of all when we confront the fact that despite—or because of—reports of the death of the Mediterranean, it is multiplying as a keyword in scholarly journal titles.⁴⁶ There are more and more of these publications, and many thousands of people must receive the electronic information service "H-Mediterranean" or visit Mediterranean history sites on the web. But much of what passes for Mediterranean history here is a recycling of old "Mediterraneanist" positions. Some of it is a cover: specialists in antiquity, for example, may have recourse to "Mediterranean" as a label that, unlike "Greek," "Roman," or "classical," avoids the stigma of intellectual collusion with the dominant culture in a socially diverse world. The majority of scholarly writing is conventional, relatively local, political, social, or economic "history in" some Mediterranean country—of no immediate wider significance, and with little attention to geography or environment.

Among books of the last decade or so, "history in" receives its consummate expression in Histoire de la Méditerranée, edited by Jean Carpentier and François Lebrun.⁴⁷ Here there is no prior definition of the subject of Mediterranean history and no obvious underlying conception to justify choices of topic beyond that of conventional geography. We find, rather, a full and helpful chronological synthesis of the political, economic, religious, and cultural history of Mediterranean lands—especially the northern Mediterranean. The siege of Toulon is included, a subject on which Braudel lectured;⁴⁸ and so is the Second Vatican Council. The latter illustrates nicely how a major event, and all the larger trends and structures it encapsulates, can be part of "history in" the Mediterranean, without in any important way requiring attention to geography. Of course, it is not geographically contingent that the council was held in a Mediterranean city—or Rome in particular. But the location of the papacy would hardly be at the forefront of any explanation of when and how the council came to be called; and the council is, very obviously, an event of much more than Mediterranean significance.⁴⁹

Even environmental histories can be read as taking it for granted that there is no need either to return to or in major ways to go beyond Braudel's human geography. With only a few exceptions, the environmental history of the Mediterranean has not attracted much attention lately.⁵⁰ And the exceptions are of a somewhat traditional kind. For example, in A. T. Grove and Oliver Rackham's excellent book The Nature of Mediterranean Europe, the Mediterranean is defined once and for all at the outset in the usual climatic terms as one of six "mediterraneoid" parts of the world. The authors can then get on with the business of showing that it is no lost Eden, subject to progressive degradation by humanity. Here is an ecological narrative that is certainly history "of" its subject. It brings the relationship between humankind and the environment back to the center of the picture, and shows the diversity of Mediterranean ecologies and the simplifications of so many earlier accounts of human intervention. But it is a history of climate, geology, and flora and fauna in which the effects of humanity—although seen as all-pervasive—are directly addressed in only one and a half chapters.

A third type of Mediterranean history sidesteps the anti-Mediterraneanist critique by confining itself primarily to the less controversial unity of the sea. This can be labeled "the maritime turn." In "Mediterranean Breviary" (to translate its original title), Predrag Matvejevi� keeps his subject manageable by restricting it to the sea and fifty kilometers inland.⁵¹ The book is unclassifiable: part historical mélange, part contemporary evocation. Here we read of the "nobility" of studying Mediterranean geology. Many Mediterranean people, we learn, "go in for horoscopes"; "theft is a Mediterranean art form"; lighthouses are an important Mediterranean institution; and "a ship that sinks does so in silence or in bedlam." Above all, there is the romance of the ports, the ships, and the seafarers: "I have listened to people living on both north and south coasts of the Mediterranean speak of sea smells. I have taken careful notes."⁵² This is Braudel as rewritten by Walt Whitman. Everything is surface, and a frictionless one at that. Generalizations meet no resistance. The book is—for the most part—a study really "of" the Mediterranean. Yet it is scarcely history.

Another fresh synopsis of Mediterranean history again virtually omits the environment altogether. It does this not because Braudel has dealt with it, but in deliberate rejection of his approach as leaving too little space for individual human initiative. In David Abulafia's edited volume The Mediterranean in History,⁵³ that subject is reduced to the sea and its immediate coastlands. It might have been entitled History in (or on) the Mediterranean. There is an opening chapter in which Oliver Rackham rehearses his arguments against progressive long-term degradation, but on this occasion he does not define the environment in question at all. After Rackham's chapter, the theme is the conflict or mutual influence of states and economies across the sea. The regional setting disappears from view. Along the same lines, Abulafia now promises what will undoubtedly be a magnificent narrative history of the Mediterranean Sea, its islands and coastlands, in which full weight is given to the decisions of named individuals. Perhaps it will be Herzfeld-proof, but the immunity may have been bought at some cost.

Conflict and influence across the sea are indeed central to any historical writing about the Mediterranean. Yet the coherence and distinctiveness of the region as a subject is not, we think, to be sought in the relatively clear-cut facts of Norman states, linked by sea, or the trade that enriched Lusignan kings. Norman seafaring and Cypriot sugar plantations need to be set within a much wider panorama. They need to be seen within a time frame that is not only long, but also unconstrained by the dictates of narrative. If the new regional history we have been describing is to avoid being a patchwork of ill-assorted discursive constructs, the arresting qualities of its objects should be described in ways that facilitate comparison and contrast. Common denominators need to be found. For all the accumulated conceptual and cultural freight of the Mediterranean, a different type of criticism of Mediterranean history has been that in practice it lacks cohesion across time and space. There is no interesting basis, it might be said, for comparing Corsica and the Nile delta, or the age of the Sea Peoples with that of the Corsairs. Braudel, of course, thought differently. And his assertion of comparability across very wide gaps of period and place within the Mediterranean impels us to try, against the skeptics, to make a case for the value and coherence of a Mediterranean regional approach to ancient, medieval, and early modern history.⁵⁴ The following account of one possible view of the Mediterranean past offers some hope of meeting the provocations of "Mediterraneanism." It does so, moreover, by exploring characteristics of the Mediterranean that include the ways in which it related to other regions around it. It therefore supports the larger "thalassological experiment" on two fronts. First, the Mediterranean need not be a political embarrassment, a dangerously flawed foundation stone in the edifice of global comparativism. Second, the study of coherent structures in Mediterranean history throws up questions and models that may help the global historian to understand how the new constituent regions of world history have actually interacted.

At the most basic level, we have attempted in earlier work to develop a framework for interpreting certain aspects of Mediterranean history. These are aspects that shed light on the big questions of unity, distinctiveness, and continuity in the region before "modernity," however defined. The framework rests on a fourfold description of primary production, derived to an important extent from the social anthropology and archaeology of the last third of the twentieth century. The first stage is to identify a distinctive regime of risk, in which bad years of many different kinds have been common enough to make it essential to find remedies for them, but not so common as to outnumber good ones. The second stage is to identify a distinctive "logic" of production organized around coping with this risk regime. It rests on the triple strategy of (in ascending order of historical importance) diversification of productive enterprise to cope with the different kinds of bad years; storage of the surplus normally accumulated as insurance against future bad years; and redistribution of the stored as well as the fresh, using different rhythms for the different products of diversification.

These two main aspects, risk and production, are further patterned by two other crucial elements. One of them is an extreme topographical fragmentation, deriving ultimately from the tectonics of the Mediterranean region. This fragmentation can be expressed through the vocabulary of the "microecology," a deliberately loose concept. Microecologies are as much in the perceptions of primary producers as in any topographical particularity. They are interactive, both locally between people and environment, and more broadly between different microecologies, which very great distances sometimes separate. Microecologies resist mapping. They are fluid, mutable creations. This mutability bears on both of the first two stages of the argument. It vastly increases the capriciousness and unpredictability of the normal risk regime. Therefore it also promotes and rewards a certain approach to production: the three imperatives of diversification, storage, and redistribution.

Finally, there is the distinctive regime of communications made possible by the geography of the sea, with its complex coastlines and numerous islands, interlocking coastal lowlands, and frequently navigable lagoons and rivers. The effect of such a regime is primarily on the character and significance of redistribution, and through that it feeds back into the whole system. It could thus be claimed that the key variable in assessing the social and economic character of any Mediterranean microecology at a given historical moment is its "connectivity," a term borrowed, ultimately, from graph theory.

This rapid summary is inevitably abstract. But the fourfold model—of risk regime, logic of production, topographical fragmentation, and internal connectivity—is deliberately so: it is intended to embrace the characteristic variability of Mediterranean human ecology. And one advantage of it may be that it can withstand accusations of "Mediterraneanism." First, against the charge of exclusivism, it raises a number of issues that are germane to many other historical contexts. It gets away from the environmental typology of Braudel's Mediterranean, in which a residual determinism is detectable (mountain societies, for example), and brings out the two-way interaction of humanity and the environment. It allows conceptual space for the perceptions and decisions of both small producers and those who control them. It takes in the whole environment, both land and sea, and although it stresses the rural, it is meant to be applicable to the full range of human settlement, from the hermitage to the metropolis. Because it is "micro" in scale, it avoids the near-insuperable problems set by the pursuit of a single historical ecology of the Mediterranean region as a whole. For this would seem to require estimates of "global" population size and density, and carrying capacity, arrived at in the light of a not less than complete inventory of human and animal food sources.

A second "Mediterraneanist" accusation is exceptionalism. But because the ecological metaphor is structural or systemic in character—purposefully abstracted from specific environmental features—it also removes the need to define the Mediterranean environment in the standard terms that have underpinned claims about its superiority. The Mediterranean of conventional (nineteenth- and twentieth-century) human geography has been variously conceived. But three interrelated types have achieved notable prominence. First, there are the climatically based definitions, in which, for instance, certain isohyets (rainfall contours) are selected to demarcate the regions characterized by low summer precipitation. Second, there are the environmentally determined versions of these, in which it is the distribution of certain flora, and especially economically important crops, that maps the Mediterranean. (The zone of cultivation of olive and vine has been the most popular.) Finally come the cultural derivatives of the first two types, such as the supposed distinction between the lands of the oil-eaters and those of the butter-eaters.⁵⁵ The principal cavil with analyses of this kind is that none of the botanical, climatological, or environmental definitions can fully and satisfactorily map the Mediterranean region with which the historian wants to operate.⁵⁶ To mention only one serious and underestimated difficulty, climatically, as Braudel himself acknowledged, every mountaintop is outside the Mediterranean.⁵⁷

The unity and distinctiveness of the Mediterranean should be differently conceived. The unity is not that of ecological or cultural types so much as of connectivity between structurally similar (similarly mutable) microecologies. The distinctiveness, on the other hand, becomes harder to pin down. None of the four elements in the schema outlined above is in any obvious way essentially or distinctively Mediterranean. The regime of bad years is configured by the specifics of the type of climate we conventionally know as Mediterranean; but there is no reason why local climates in quite different climatic zones should not display equally complex regimes of risk. The logic of risk buffering, the intensity of fragmentation, and the possibility of connectivity can each be readily paralleled elsewhere (for example, in Southeast Asia).⁵⁸ The only way in which the Mediterranean is differentiated both from its neighbors and from comparable areas much farther away is by the sheer intensity and complexity of the ingredients. That can, for instance, be seen in the notable differences in historical character between islands in the Mediterranean and those in the oceans. The former are (sometimes) rich and populous, tightly engaged in networks of which they are often the principal nodes, and sometimes even the centers of large hegemonies, the most high-pitched individual cases of the characteristic productive and redistributive regimes. The latter, the oceanic islands, are, by contrast, known for poverty, dependence, marginality, and a subaltern status.⁵⁹

Unexpectedly, the Mediterranean therefore turns out to be identified by the unique concentration of factors that are not themselves peculiar to the region. That concentration should, at least in principle, be susceptible of calibration and mapping. Here Braudel's metaphor of the Mediterranean as an electromagnetic field, or a source of light, is helpful;⁶⁰ and still more so is his image of the Mediterranean in this sense as "pulsing," expanding and contracting to embrace adjacent zones. The shape and extent of the "paroxysm" of Mediterranean diversity has sometimes been determined only loosely by the sea's physical geography.

The microecological approach also gives us a way of delimiting the Mediterranean—or rather, of avoiding delimiting it in any fixed, "timeless" way. The region has only shifting and "fuzzy" boundaries. As one moves away from the sea, microecologies cease or radically change character, although in differing ways and at differing rates in the many kinds of frontier zones that the region presents to its neighbors. The Mediterranean may also be conceived as a large zone of net introversion or involution. It is an area within which internal contacts are, overall—if we could ever arrive at some supreme reckoning—more numerous, dense, or durable than external ones, even though the Mediterranean's "external relations" (Braudel's greater Mediterranean) have reached right across the globe.

Arguments of the microecological kind also reassure on the question of whether the Mediterranean is a proper subject for any sort of history (except possibly the history of imperialist learning), let alone the new regional variety. The many-sidedness of the model helps the historian avoid reification. Weaving together the ecological conditions of Mediterranean coastlands with the nature of seaborne connectivity produces an approach to Mediterranean distinctiveness that may be less vulnerable to the political and heuristic difficulties associated with imperial geography than are other, simpler accounts.

Our model is also to a surprising extent compatible with the history of Mediterranean self-perception. The awareness of living in a fragmented world, nonetheless united by the connectivity that overcomes that fragmentation, has a long history in Mediterranean societies. Although the semantic extension of "the Mediterranean" from sea to region first occurred in the nineteenth century, the sea itself had first been perceived as a unity more than two thousand years earlier. For at least some Greeks, the sea we now call the Mediterranean was intuited as something more than an ensemble of linked smaller seas.⁶¹ It was the Great Sea, the "sea in our part of the world," the sea around which "we" (never explicitly identified) "live like ants and frogs around a wetland," as Socrates puts it in Plato's Phaedo (109b). This sea was one term of a dual classification of the world's surface water. The "mediterranean" (without the capital letter—a geographical term, not a name) was the only one of its kind, the sea in the middle of the land, as distinct from the lands in the middle of the sea, that is, the continents surrounded by outer ocean. It happens that the precise phrase mediterraneum mare is not attested until after the Roman Empire had ended in the West.⁶² The concept to which it refers, however, is far older.

Because it was part of a reflection on the layout of the whole globe, the ancient invention of the Mediterranean was essentially maritime. It was only very indirectly a contribution to regional thought. (Its counterpart, the invention of the continents, was quite different: continental land was indeed already in the fifth century b.c. split three ways, into Asia, Africa, and Europe, to each of which was attributed a genuine regional identity.) Something similar to the ancient conception of the Mediterranean Sea is revealed in the cosmological "Book of Curiosities" from eleventh-century Cairo, the unique manuscript of which, now in Oxford, was rediscovered in 2000.⁶³ Here, on a striking map, the Mediterranean is a unity containing and clearly characterized by 120 islands and 121 mainland anchorages. Once again it is a remarkable type of sea, rather than a larger area.

The definition of the Mediterranean as a region was not, therefore, a by-product or concomitant of ancient imperialism.⁶⁴ Nor did it appear in medieval or early modern European thought, even that of the Ottomans, whose mastery of the sea in the earlier sixteenth century was almost as extensive as Rome's. Thus it is not, despite the fears and scruples of some contemporary writers, part of the master narrative of the aggressive elites of premodern empires. It is true that in ancient diplomatic negotiations—a central vehicle of hegemonic thought—the unity of the Mediterranean lands did sometimes find expression. Most famously, in a settlement between Athens and Persia attributed to the middle of the fifth century b.c., one clause was held to have forbidden Persian messengers to come within fifty miles or a day's ride of the sea (even called the Greek sea in one version of events).⁶⁵ In a still more imperialist context, the command of Pompey the Great against piracy, and his later control of the redistribution of cereal staples, were defined as running all over the Mediterranean and inland for the equivalent of fifty miles.

In this way of thinking, however, something rather different from Mediterranean empire is to be seen. Once again, we encounter a conception of how the maritime world worked, and what it meant to the land. It was the domain of naval war, but also of redistribution, especially of staples. And it was the place where the rhythms of redistribution were threatened by endemic piracy. Such thinking is identical to Plato's opinion about the incompatibility of good government with the sea and its activities. His ideal city was to be ten miles inland, well away from mercantile vice.⁶⁶ When the Romans wanted an excuse to destroy the great entrepôt of Carthage in 149 b.c., the deliberately preposterous condition they laid down was that Carthage should be uprooted and moved inland, away from its great ports. The distance they appointed was ten miles.⁶⁷

Ancient thought about the Mediterranean thus concerned itself not with the creation of a geographical expression to denote a region that might be subdued, but with an analysis of the society (in a very loose sense) to which the sea was central. It was a characterization of the world around the sea, not a labeling of a world that the sea happened to adjoin. It reflected, although obliquely, the distinctive interaction of Mediterranean connectivity and terrestrial ecology that can be discerned in many other periods of Mediterranean history. Such a Mediterranean, far from being a relatively simple political construction, will be extremely hard to define, and will map onto the concerns of human communities, value systems, economic ambitions, and interregional allegiances in an intrinsically unstable way. This Mediterranean should indeed be an "essentially contested" category.⁶⁸ The legacy of ancient and nineteenth-century geography and other elements in the "meta-subject" of discourse about the Mediterranean, the area's long-accumulated cultural baggage, are reflections of such contestation. They form an aspect of the subject that we have to work with, and through. But they do not invalidate it. On the contrary, in a history of perceptions of the Mediterranean that goes beyond the detection of evil empires, what we catch is an intrinsic part of a microecological world.

Our aim here is emphatically not to impose a new paradigm on Mediterranean historians, forestalling discussion of alternatives. It is, rather, to discover on what basis we can treat ancient and medieval—or, more widely, premodern—Mediterranean history as a single field of inquiry, and how far an approach derived ultimately from Braudel enables us to cross the boundaries that scholarship seems to have erected. This involves "decentering" the Mediterranean,⁶⁹ as well as adopting an extremely long time frame. The goal is comparable with the ways in which some of the other new regional historians have striven to break traditional molds and escape older disciplinary constraints. It is encouraging that others have found in maritime history a satisfying route to this destination.

It is a further happy consequence of developing new models for the interpretation of regional history that there is common ground between Mediterranean history and the most outspoken anti-Mediterraneanist of all, Michael Herzfeld. To think in terms of Mediterranean history is indeed an "excuse"—an excuse for (as he generously puts it in response to our ideas) "defining new alliances and agglomerations capable of generating novel and interesting heuristic options."⁷⁰ This convergence with Herzfeld may reflect the extent to which "our" Mediterranean is no longer Mediterraneanist. But unlike the other authors mentioned above, such as David Abulafia, we may have thrown the baby of global comparability out with the bathwater of traditional Mediterranean analysis. Although the Mediterranean appears to be a conspicuous example of the type of network through which other new regions tend to be defined, its distinctive historical regimes of connectivity actually turn out to be very unusual, and notably hard to parallel in other parts of the world.

Take the example of the movement of people and goods. In traditional Mediterranean history, what was visible was high commerce, the glamorous movement of high-value cargoes. This is also what was most prominent in the evidence from antiquity, and it was regularly marginalized as "only" luxury trade, small-scale and of interest primarily to the elite. Even for the early modern period, Braudel had to insist that high-value cargoes mattered economically, too. In oceanic spaces, such networks were traced in antiquity, as in the extraordinary first-century survey of the coasts of the Indian Ocean, the Periplous of the Erythraean Sea. With Braudel, we acknowledge the importance of high-value, low-bulk cargoes. But the interdependence of the Mediterranean is based on a different kind of process entirely: a redistribution that is an extension of primary production induced by the need to buffer risks.

This basic engagement of the primary producer with maritime connectivity, already visible in the landscape of Hesiod's Works and Days at the end of the eighth century b.c., underlay the familiar ancient notion of "our sea." Such possessiveness was not, to repeat, a cultural assent to the aggressive proprietorship of ancient imperial states. On the contrary, it represented, however self-consciously, a rural, "peasant," microregional view. Similarly, in the late fifth to early fourth century b.c., the author of the Hippocratic Regimen claims to be writing for the majority of men. He is addressing those who use ordinary, accessible food and drink; who exert themselves as much as is essential—and who undertake both overland journeys and sea voyages to collect their livelihood.⁷¹ This again nicely reflects the way in which the outlet to the medium of redistribution is part of every microregion's ideal portfolio of resources, and is labeled accordingly: "our mountain pasture," "our wetland," "our irrigable meadows," "our scrublands"—"our sea." The Mediterranean of the ancient Greeks was a microregion writ large.

So when we come to scrutinize what this Mediterranean has to offer the systematic comparison of sea histories, there is, as far as we can tell at this stage, little alternative to "agreeing to differ"—both with other Mediterranean scholars and with those at work on other seas. The terrestrial environment is of relatively little importance in sea historiography elsewhere, just as it has often been marginal to Mediterranean historical writing since Braudel. "Oceans Connect" is the great theme, as reflected in the title of an ongoing Ford Foundation project at Duke University. It is not "Environments Unite."⁷² To study the Baltic Sea may require a good deal of environmental knowledge, and invites a broad climatic and geological backdrop, but on nothing like the scale that has been evinced, rightly or wrongly, either by Braudel or in the microecological model offered above.⁷³ The Atlantic and Pacific oceans are too large and their coastlands too diverse for any ecological, cultural, or psychological (stereo)types to have been widely associated with them.⁷⁴ On the contrary, what matters—what makes these seas new and exciting historiographical categories—is the density and variety of human connections across them. Thus David Armitage's typology of Atlantic history writing is all about contacts, and the polities, economies, and cultures that are making them; it is not about the ecology of the Atlantic's various European, African, and American littorals. "Circum-," "trans-," and "cis-Atlantic" studies do not even presuppose common environmental or climatic features, let alone an Atlantic personality.⁷⁵ The Atlantic, the Indian Ocean, Pacific archipelagoes, and even the Sahara are spaces negotiated by high-profile trade and through the movements, migrations, and diasporas connected with it or with the concerns of states. Similar movements have of course had a long history in the Mediterranean, but they are not the secret of its identity. We wait to be told that there are maritime spaces—perhaps in the Japanese archipelago, or in the early modern Caribbean—in which it is an ecologically driven interdependence that weaves the fabric of the region. Until that point, the Mediterranean remains striking for its otherness.

The agenda for folding the Mediterranean into long-range comparative regional history cannot, then, rely wholly on typologies of harbors or routes, on densities of port cities or maritime confederations, on the penetration of economic networks by valuable seaborne goods, on patterns in the recruitment of seafarers, or on technological innovations in transport. What of the other theme implicit in the new thalassology—how the seas that are its objects join up to constitute a changing global history?

Plato was too sanguine in thinking that at ten miles inland, even in mountainous Crete, his ideal city would escape the "corruption" engendered by the complex ecology of the sea. The question is how to model the ways in which the networks of movement that characterize Mediterranean space reach out into the surrounding zones, zones in which the conditions that promoted the movements fall off or are altogether absent. Ian Morris has coined the notion of "Mediterraneanization" to capture the degree to which connectivity may intensify or abate within the traditional Mediterranean—and in the zones, sometimes even quite distant ones, with which it has contact. Many transitional zones around the Mediterranean exhibit what Brent Shaw has called recursiveness, reversing polarity so that the land exerts a more potent influence on the coastlands than does the world of the sea.⁷⁶ There is the additional possibility that the integrated, involuted Mediterranean milieu has at certain times had a repellent effect on the economies and social systems of nearby regions. These turn their backs on the world of the sea, and establish lines of communication bypassing the Mediterranean interior. Two examples would be the corridor from the Black Sea to the Rhine, and the routes that cross the highlands of eastern Anatolia from the Caucasus and reach the Red Sea without impinging on the inner Mediterranean zone. They respond to Plato's corrupting sea, but not through being directly engaged with it—rather the opposite. Equally, there are the "hemorrhage" points at which—again at particular moments in millennial history—different gateways open. Through these gateways, the distinctive Mediterranean world interacts, sometimes very intensively, with, for instance, the zones to which the rivers of the Ukrainian steppe or the coastal trade routes of the Red Sea or Persian Gulf give access.

The part played by our Mediterranean in the wider assemblage of regional pieces that makes up world history seems likely to prove an unusual one. Of course the Mediterranean fits in. But it is probably the only piece of its kind, and not, despite the reach of Braudel's greater Mediterranean, actually very like the other sea-centered components. Although it obviously belongs in this gallery of maritime portraits, its interest will be in how different it is from the rest of the show.

We are not, on the other hand, attracted by "Mediterranean exceptionalism." And that is why we think that it is in the study of long-distance interaction even more than in comparison that the greatest advantage of the new regional history lies. Our hope is that specialists in one of the oceans, the Central Asian steppe, or the Sahara will join with students of the Mediterranean in dialectic. Its ultimate objective will be to refine the ways in which we frame a global history—a history of which the Mediterranean has been such a peculiar and important segment.

Peregrine Horden is Reader in Medieval History at Royal Holloway, University of London. He co-authored The Corrupting Sea: A Study of Mediterranean History (Oxford 2000) with Nicholas Purcell, and is also currently working with him on the sequel, Liquid Continents, which will address the methodological problems of fitting the distinctive Mediterranean of Corrupting Sea into a wider context of large-scale and long-term history. He has also written extensively on the history of medicine and charity in medieval Europe and Byzantium, and on the history of music therapy. A monograph on the global history of the earliest hospitals is in progress for Yale University Press.

Nicholas Purcell is Lecturer in Ancient History at Oxford University, and a Fellow of St John's College. His work has addressed a number of problems of ancient social, economic, and cultural history, and much of it has connections with geography and the study of landscape. He is co-author with Peregrine Horden of The Corrupting Sea (Oxford, 2000) and its forthcoming sequel, Liquid Continents.

Notes

¹ J. J. Cohen, "Introduction," in Cohen, ed., The Postcolonial Middle Ages (Basingstoke, 2000), 7: "a postcolonial Middle Ages has no frontiers, only heterogeneous borderlands with multiple centers." See also W. D. Mignolo, Local Histories/Global Designs: Coloniality, Subaltern Knowledges, and Border Thinking (Princeton, N.J., 2000).

² Rena Lederman, "Globalization and the Future of Culture Areas," Annual Review of Anthropology 27 (1998): 427–449.

³ To be distinguished from the older world-systems theory: e.g., P. N. Kardulias, World-Systems Theory in Practice: Leadership, Production, and Exchange (Lanham, Md., 1999); T. D. Hall, A World-Systems Reader (Lanham, Md., 2000).

⁴ R. Bin Wong, "Entre monde et nation: Les régions braudéliennes en Asie," Annales 56, no. 1 (2001): 5–41; Maurice Aymard, "De la Méditerranée àga l'Asie: Une comparaison nécessaire (commentaire)," Annales 56, no. 1 (2001): 43–50.

⁵ David Abulafia, "Mediterraneans," in William V. Harris, ed., Rethinking the Mediterranean (Oxford, 2005), 64–93. On the Sahara, see also Méditerranée 99, no. 3–4 (2002), "Le Sahara, cette 'autre Méditerranée.'t"

⁶ From the ancient Greek thalassa, "sea." Edward Peters, "Quid nobis cum pelago? The New Thalassology and the Economic History of Europe," Journal of Interdisciplinary History 34 (2003): 49–61; Brent Shaw, "A Peculiar Island: Maghrib and Mediterranean," Mediterranean Historical Review 18 (2003): 93–125.

⁷ Frederick L. Wernstedt and J. E. Spencer, The Philippine Island World: A Physical, Cultural, and Regional Geography (Berkeley, Calif., 1967), is still the starting point.

⁸ Barry Cunliffe, Facing the Ocean: The Atlantic and Its Peoples (Oxford, 2001). See also Rupert A. Housley and Geraint Coles, eds., Atlantic Connections and Adaptations: Economies, Environments and Subsistence in Lands Bordering the North Atlantic (Oxford, 2001).

⁹ David Armitage, "Three Concepts of Atlantic History," in David Armitage and Michael J. Braddick, eds., The British Atlantic World, 1500–1800 (Basingstoke, 2002), 14–15.

¹⁰ Martin W. Lewis and Kären E. Wigen, The Myth of Continents: A Critique of Metageography (Berkeley, Calif., 1997); Daniel Finamore, ed., Maritime History as World History (Gainesville, Fla., 2004).

¹¹ Peregrine Horden and Nicholas Purcell, The Corrupting Sea: A Study of Mediterranean History (Oxford, 2000), 38 on Braudel. Contrast Felipe Fernández-Armesto, "Maritime History and World History," in Finamore, Maritime History, 8.

¹² Horden and Purcell, The Corrupting Sea, 24–25.

¹³ Fernand Braudel, The Mediterranean and the Mediterranean World in the Age of Philip II, trans. Siân Reynolds, 2nd ed., 2 vols. (New York, 1976). For Braudel's intellectual biography and his influence on sea and ocean history, see references in Horden and Purcell, The Corrupting Sea, 542–543. Add now John A. Marino, ed., Early Modern History and the Social Sciences: Testing the Limits of Braudel's Mediterranean (Kirksville, Mich., 2002).

¹⁴ Braudel, Mediterranean, 1244.

¹⁵ Ibid., 1239.

¹⁶ Ibid., 170, 394–396.

¹⁷ Robert S. Lopez, The Commercial Revolution of the Middle Ages, 950–1350 (Englewood Cliffs, N.J., 1971), 20, 23; Abulafia, "Mediterraneans," 66, 81.

¹⁸ See also Pierre Cabens, ed., Histoire de l'Adriatique (Paris, 2001).

¹⁹ Charles King, The Black Sea: A History (Oxford, 2004).

²⁰ Abulafia, "Mediterraneans," 85–90.

²¹ Paul Sant Cassia, "Authors in Search of a Character: Personhood, Agency and Identity in the Mediterranean," Journal of Mediterranean Studies 1 (1991): 1–17.

²² Braudel, Mediterranean, 15.

²³ For some bibliography, see Horden and Purcell, The Corrupting Sea, 26–29, 538–540.

²⁴ Bertram C. Gordon, "The Mediterranean as a Tourist Destination from Classical Antiquity to Club Med," Mediterranean Studies 12 (2003): 203–226.

²⁵ In Harris, Rethinking the Mediterranean, 45–63. See also Herzfeld, "Performing Comparisons: Ethnography, Globetrotting, and the Spaces of Social Knowledge," Journal of Anthropological Research 57 (2001): 265–267.

²⁶ For Victoria A. Goddard, Josep R. Llobera, and Cris Shore, the Mediterranean was invented in the 1960s and passed its sell-by date in the 1980s. "Introduction: The Anthropology of Europe," in Goddard, Llobera, and Shore, eds., The Anthropology of Europe: Identity and Boundaries in Conflict (Oxford, 1994), 4, 20–23.

²⁷ By Roger Bagnall, at the conference held at the Center for the Ancient Mediterranean, Columbia University, September 21–22, 2001, at which Herzfeld's paper originated.

²⁸ For the political dimension of "the Mediterranean" in Israeli "public discourse," see Yaacov Shavit, "Mediterranean History and the History of the Mediterranean: Further Reflections," Journal of Mediterranean Studies 4 (1994): 313–329.

²⁹ Horden and Purcell, The Corrupting Sea, 486–487, 522–523; Herzfeld, Anthropology through the Looking-Glass: Critical Ethnography on the Margins of Europe (Cambridge, 1987); J. de Pina-Cabral, "The Mediterranean as a Category of Regional Comparison: A Critical View," Current Anthropology 30 (1989): 399–406. See also Yaacov Shavit, "The Mediterranean World and 'Mediterraneanism': The Origins, Meaning, and Application of a Geo-Cultural Notion in Israel," Mediterranean Historical Review 3 (1988): 96–117. It is worth attempting to plead for consistent use of labels. In both Ian Morris, "Mediterraneanization," Mediterranean Historical Review 18 (2003): 30–55, and Greg Woolf, "A Sea of Faith," ibid., 126–143, "Mediterraneanism" is used to mean Mediterranean exceptionalism, rather than the subtler object of Herzfeld's critique; or even more baldly, "great interest in the Mediterranean as a historical object" (Morris, 35–37).

³⁰ Cf. Woolf, "Sea of Faith," 140: "the most useful units of study ... are either larger or smaller than the Mediterranean world."

³¹ Edward Said, Orientalism (New York, 1978); also James G. Carrier, ed., Occidentalism: Images of the West (Oxford, 1995). For further bibliography, see U. Freitag, "The Critique of Orientalism," in Michael Bentley, ed., Companion to Historiography (London, 1997), 620–638.

³² Braudel, Mediterranean, 17. Braudel's experience in Algeria may be more relevant than his passion. See Pierre Daix, Braudel (Paris, 1995).

³³ See also his Culture and Imperialism (New York, 1993).

³⁴ We attempt an anthropological rejoinder in Horden and Purcell, The Corrupting Sea, 487–523.

³⁵ See, e.g., Jane Schneider, ed., Italy's "Southern Question": Orientalism in One Country (Oxford, 1998); Nelson Moe, The View from Vesuvius: Italian Culture and the Southern Question (Berkeley, Calif., 2002); Thomas W. Gallant, Experiencing Dominion: Culture, Identity, and Power in the British Mediterranean (Notre Dame, Ind., 2002), esp. 35–36, on the "Mediterranean Irish" of the Ionian Islands; Ussama Makdisi, "Ottoman Orientalism," AHR 107, no. 3 (June 2002): 768–796.

³⁶ R. Peet, "The Social Origins of Environmental Determinism," Annals of the Association of American Geographers 75 (1985): 309–333.

³⁷ Theobald Fischer, Mittelmeerbilder: Gesammelte Abhandlungen zur Kunde der Mittelmeerländer (1906; repr., Leipzig, 1913), v. See also Lucien Febvre, A Geographical Introduction to History (London, 1932), 41; W. Stroch and G. Meiring, La méditerranée allemande (Paris, 2000).

³⁸ Daniel Nordman, "La Méditerranée dans la pensée geographique française (vers 1800–vers 1950)," in Claude Guillot, Denys Lombard, and Roderich Ptak, eds., From the Mediterranean to the China Sea: Miscellaneous Notes (Wiesbaden, 1998), 1–20. For the background, see Christopher Drew Armstrong, "Travel and Experience in the Mediterranean of Louis XV," in Harris, Rethinking the Mediterranean, 235–237.

³⁹ Nordman, "Méditerranée," 18.

⁴⁰ See also Shavit, "The Mediterranean World and 'Mediterraneanism.'"

⁴¹ An impression derived from "googling."

⁴² Armitage, "Three Concepts," 12.

⁴³ Horden and Purcell, The Corrupting Sea, 39–43.

⁴⁴ Braudel, The Mediterranean in the Ancient World, trans. Siân Reynolds (1998; repr., London, 2001). See also Braudel, "Fernand Braudel, l'antiquité et l'histoire ancienne" [interview with J. Andreau, M. Aymard, and R. Etienne, April 29, 1985], Quaderni di Storia 12, no. 24 (1986): 5–21.

⁴⁵ Horden and Purcell, The Corrupting Sea, 2, 9. See also Shavit, "Mediterranean History and the History of the Mediterranean," for a third category, "Mediterranean history," which presupposes cultural uniformity as well as unity.

⁴⁶ Susan E. Alcock, "Alphabet Soup in the Mediterranean Basin: The Emergence of the Mediterranean Serial," in Harris, Rethinking the Mediterranean, 314–338; Morris, "Mediterraneanization," 34–35.

⁴⁷ Jean Carpentier and François Lebrun, eds., Histoire de la Méditerranée (Paris, 1998).

⁴⁸ Une leçon d'histoire de Fernand Braudel: Châteauvallon, Journées Fernand Braudel, 18, 19 et 20 Octobre, 1985 (Paris, 1986), pls. 19–20, facing 161.

⁴⁹ See Carpentier and Lebrun, Histoire de la Méditerranée, 231, 502–503, for Vatican II (also, e.g., 167 on Gregorian Reform).

⁵⁰ Ted Steinberg, "Down to Earth: Nature, Agency, and Power in History," AHR 107, no. 3 (June 2002): 798–820. The Mediterranean gets little attention in John F. Richards, The Unending Frontier: An Environmental History of the Early Modern World (Berkeley, Calif., 2003).

⁵¹ Predrag Matvejevi�, Mediterranean: A Cultural Landscape, trans. Michael Henry Heim (1987; repr., Berkeley, Calif., 1999).

⁵² Ibid., 30, 43, 32, 63, 38.

⁵³ David Abulafia, ed., The Mediterranean in History (London, 2003).

⁵⁴ Vol. 1 of our Study of Mediterranean History was The Corrupting Sea, its title explained in the opening epigraphs. Vol. 2, Liquid Continents, which concentrates on the boundaries of the Mediterranean, is in progress. What follows also draws on Purcell, "The Boundless Sea of Unlikeness? On Defining the Mediterranean," Mediterranean Historical Review 18 (December 2003): 9–29, and on Horden and Purcell, "Four Years of Corruption: A Response to Critics," in Harris, Rethinking the Mediterranean, 348–376.

⁵⁵ Braudel, Mediterranean, 168–170, 234–238.

⁵⁶ Braudel, ibid., 234–235, recognized that the Mediterranean was actually the overlap of other climatic zones, and that it was climatically diverse itself; but he maintained its impressionistic coherence in the face of this realization. See also Henk Driessen, "Pre- and Post-Braudelian Conceptions of the Mediterranean Area: The Puzzle of Boundaries," Narodna umjetnost: Croatian Journal of Ethnology and Folklore Research 36, no. 1 (1999): 53–63.

⁵⁷ Braudel, Mediterranean, 26–27.

⁵⁸ To cite only one instance: C. Higham, The Archaeology of Mainland Southeast Asia from 10,000 B.C. to the Fall of Angkor (Cambridge, 1989), 1–14, for ecological diversity in this region. See also D. Ludden, An Agrarian History of South Asia (London, 1999).

⁵⁹ Stephen A. Royle, A Geography of Islands: Small Island Insularity (London, 2001).

⁶⁰ Braudel, Mediterranean, 168.

⁶¹ For what follows, see Horden and Purcell, The Corrupting Sea, 10–13, 530–533; Purcell, "Boundless Sea of Unlikeness?" 13–16.

⁶² Isidore of Seville, Etymologiae, 13.16.1, in Isidore, Isidori Hispalensis Episcopi Etymologiarum sive Originum Libri XX, ed. Wallace Martin Lindsay, 2 vols. (1911; repr., Oxford, 1985), 2: 98. There are likely to have been earlier such works, now lost.

⁶³ Bodley, MS Arab. c. 90, fols. 30b–31a. The MS is due to be published in summer 2006 at www.bodley.ox.ac.uk/bookofcuriosities.

⁶⁴ Purcell, "Boundless Sea," 15–16. The Romans, for instance, did not incorporate any "Mediterraneanness" into their self-definition as empire builders, but they gave full symbolic play to the contrast between communicable littoral and the impenetrable forests, mountains, and deserts of the continental interiors.

⁶⁵ Plutarch, Life of Kimon, 13, 19.

⁶⁶ Laws, 704–705.

⁶⁷ Appian, Libyca, 73.

⁶⁸ Horden and Purcell, The Corrupting Sea, 523.

⁶⁹ Morris, "Mediterraneanization," 37.

⁷⁰ Herzfeld, "Practical Mediterraneanism," 50.

⁷¹ Regimen [Peri diaites], 68.

⁷² http://www.duke.edu/web/oceans/project.html.

⁷³ Nils Blomkvist, The Discovery of the Baltic: The Reception of a Catholic World-System in the European North (AD 1075–1225) (Leiden, 2005).

⁷⁴ For the possible counterexample of the Indian Ocean, see Kurti N. Chaudhuri, Trade and Civilisation in the Indian Ocean from the Rise of Islam to 1750 (Cambridge, 1985), and Chaudhuri, Asia before Europe: Economy and Civilisation of the Indian Ocean from the Rise of Islam to 1750 (Cambridge, 1990), 70. But the "unities of material life" are identified in only very abstract terms. Cf. Alain Bresson, "Ecology and Beyond: The Mediterranean Paradigm," in Harris, Rethinking, 111–112, citing André Wink, "From the Mediterranean to the Indian Ocean: Medieval History in Geographic Perspective," Comparative Studies in Society and History 44 (2002): 416–445, on the very un-Mediterranean ecological milieus surrounding the Indian Ocean. Note also the avowedly maritime perspective of Michael Pearson, The Indian Ocean (London, 2003), 5.

⁷⁵ Armitage, "Three Concepts," 15–25.

⁷⁶ Morris, "Mediterraneanization," 43–46; Shaw, "Peculiar Island."

Content in the History Cooperative database is intended for personal, noncommercial use only. You may not reproduce, publish, distribute, transmit, participate in the transfer or sale of, modify, create derivative works from, display, or in any way exploit the History Cooperative database in whole or in part without the written permission of the copyright holder.