Charting the history of the Med

Any resemblance between this book and Fernand Braudel's magisterial two-volume The Mediterranean and the Mediterranean World in the Age of Philip II , first published in French in 1949, is both deliberate and accidental. Any history on a Mediterranean-wide scale invites comparison.

Peregrine Horden and Nicholas Purcell - a medievalist and an ancient historian - take as their starting point the paradox that Braudel "had proclaimed the enduring unity and distinctiveness of his subject; but he had mostly confined his supporting evidence to what he thought of as the facts of geography and to 16th-century documents".

The authors' project is to explore the unity of the Mediterranean as an object of historical inquiry prior to Braudel's 16th century, broadly speaking from antiquity to the Middle Ages. They also seek to justify intellectually a history of, rather than specific economic, social, political or religious histories in the Mediterranean. This much they have in common with Braudel: the idea of the Mediterranean as a valid object of study, a well-conceived unity.

Horden and Purcell reveal that Braudel produced a monograph (as yet unpublished) on the prehistoric and ancient Mediterranean. This might seem to pre-empt their efforts, but they do not aspire to a Braudelian examination, and here is where any resemblance to The Mediterranean becomes entirely accidental. In part one of the book, "Frogs round a pond: ideas of the Mediterranean", the authors chart notions of the Mediterranean as a unity from geographical (chapter one) and historical (chapter two) points of view. They distinguish between two prominent models: an "interactionist" one, in which the sea links geographically dispersed social groups, and an "ecologising" one that emphasises the broad geographical and environmental similarities of the circum-Mediterranean lands. They identify various "Mediterraneans" - one of the imagination, particularly associated with the Roman Empire at its peak, also "Romantic" and "scholarly" Mediterraneans, chiefly products of northern European intellectuals and travellers.

More specifically, under the subtitle "Four men in a boat", they evaluate four major 20th-century figures in the historical study of the Mediterranean: Mikhail Rostovtzeff, Henri Pirenne, S. D. Goitein and Braudel himself. Braudel's study has been the most influential. He applied (to the 16th-century Mediterranean of Philip II) the three scales of analysis now familiar within the annaliste school of history. These are the longue durée , a "geohistory" dominated by the environment and enduring human practices; conjonctures , medium-term human activities with a duration of up to 50 years; and l'histoire des événements , "traditional" history of people and events.

In the authors' view, Braudel's work offered a culmination, not a starting point: "To the interactionist vision it adds an ecologising perspective. To the tradition of Romantic evocation it lends the analytical weight of a social science. And these come together in the ultimately Romantic project of an all-inclusive chef d'oeuvre , in which an entire world is subordinate to its creator-historian."

Braudel's work, despite its influence, has not been without criticism, particularly inviting the charge of environmental determinism because of the primacy of the longue durée . It is chiefly against this notion that Horden and Purcell have chosen to work, weaving a tapestry comprising many strands of evidence - historical, archaeological, geographical, ecological, anthropological - rarely combined in a single work. For them, the environment facilitates, rather than determines. In part two, "'Short distances and definite places': Mediterranean micro-ecologies", they set up the key approaches adopted. First, by examining four "definite" places - the Biqa valley, southern Etruria, Cyrenaica and the island of Melos - the authors draw out a notion of the Mediterranean as comprising an almost infinite number of tiny regions, or "micro-ecologies".

Even within the relatively constrained "definite places" selected, there are micro-ecological subdivisions, while the larger regions themselves relate to other ecologies, and so on. This notion of "micro-ecologies" is further developed in chapter four, where the phenomenon of urbanism is dismissed as almost indefinable and therefore irrelevant - one of many "givens" of Mediterranean history so treated in the book. For Horden and Purcell, "a town is an address, an arena, an architectonic agglomeration: distinctive - sometimes - for the volume and density of its buildingsI but notI for the way in which its micro-regions work". Complementary to the notion of micro-ecologies is the idea of "connectivity" (chapter five), a term for the ways in which Mediterranean regions are linked at many different scales, as they have been continuously since the second millennium BC, allowing them to coalesce or fragment in a constellation of constantly shifting groupings.

Unlike Braudel's top-down scheme involving qualitatively different scales of analysis, the authors' models of micro-ecologies and connectivity build from the bottom up and work at any scale, geographical or temporal. In part three, "Revolution and catastrophe", the longest section of the book, the authors deploy these models in a revisionist examination of the traditional big issues in Mediterranean history. Chapter six, on the "imperatives of survival" - diversification, storage and redistribution - outlines a unified view of agriculture, integrating the productivity of marginal areas with the practices of pastoralism, agriculture and tree-cropping. Micro-ecology and connectivity are vital ingredients here: variability between micro-regions is evened out by redistribution between them.

The dynamics of agricultural technology and agrarian change come under scrutiny in chapter seven. There is a playing-down of the significance and teleology of technological change in the context of an understanding that agricultural effort can both be increased or decreased ("intensification" and "abatement" are the authors' terms).

Chapter eight explores the catastrophes that are considered so typical of the Mediterranean world: natural, such as flood and earthquake, and human induced, such as massive erosion or deforestation. Again, long-cherished views are challenged - notably through the use of geo-archaeological data gathered in the context of regional studies projects - and the picture is of a Mediterranean world substantially unchanged over the past four millennia. Pan-Mediterranean effects are rare; the micro-region is the most appropriate unit of analysis. The last chapter in this section deals with the mobility of goods and people within this connected environment. Mobility is seen as a pervasive factor, almost a natural state among Mediterranean populations.

Social aspects of Mediterranean unity form the background to parts four, "The geography of religion", and five, "Museums of man? The uses of social anthropology". The sole chapter in part four relates religious behaviour to the book's major themes: environment, production and mobility. It is these themes that fuel the religious continuity often discussed in a Mediterranean context, rather than the essentially contingent continuities from pagan to Christian to Muslim in specific locations.

The last two chapters take social anthropology as their subject. Chapter 11 examines the proposition that 20th-century ethnographies can document continuities, elements of "the past in the present". It concludes that they can, but only if they are sensitive to the history within which these communities are embedded. The book ends with a discussion of how anthropology has been used to suggest unity within the Mediterranean in the concepts of "honour" and "shame". Against those who regard such claims as examples of "Mediterraneanism" (in the sense of Edward Said's "Orientalism"), by which the region is both exoticised and homogenised in order to become an appropriate subject for ethnography, Horden and Purcell do see these as peculiarly Mediterranean categories that provide yet another example, now on the social level, of the unity of their subject.

This is not by any stretch of the imagination a "popular" work. The tone is intellectual: the reader is expected to read quotations in French, while those in Italian and Latin are given in the original and in English; Arabic sources are given only in English. But the prose is stylish and engaging and the authors' wit surfaces in some of the chapter titles, such as "'I also have a moustache': anthropology and Mediterranean unity", while their broad reading, evident on every page, is reflected in telling epigraphs at the head of each section and chapter. To avoid cluttering the text with notes, more than 100 pages of chapter-by-chapter bibliographic essays, plus a consolidated bibliography that is almost as long, round off the volume, each in itself a considerable resource. There is also a comprehensive index, essential as an aid to navigating a book whose organisation is thematic, rather than chronological or geographical.

Substantial as this volume is, the authors promise a second, containing a more detailed presentation of the bases for the ecological model in studies of climate, disease and demography. Relations between the Mediterranean and the rest of the globe will also receive attention - a view "from the outside in", an inversion of the present volume's perspective. They also undertake to respond to criticisms of this volume, reflecting their desire to initiate dialogue, rather than be definitive.

In that spirit, and recognising that there can be "no question of completeness or evenness of coverage, geographical or chronological" in their work, I wonder whether their particular conception of the relationship of humans and their environment is not more crucial than the question of Mediterranean unity. For prehistorians and classical (especially Roman) archaeologists unity is not at issue, although the authors' work offers a rationale for a unity embodying multiple local diversities. The concepts of micro-ecologies and connectivity, used to define the unity of their subject could, if we accept them, be used to undermine it by extending the analysis to links prominent in various periods with Europe (north and south) or the Black Sea area. On the other hand, these concepts perhaps over-emphasise unity in time and space, down-playing the significant differences created by human actors, whose role is perhaps under-emphasised.

The last three chapters, on social factors, sit somewhat uneasily with the rest, and perhaps ignore the fact that much of the textual evidence frequently quoted in earlier sections arises from individual responses rooted in particular periods and places. Any work of history - in the broadest sense - is a product of the historian's mind. The Corrupting Sea 's late 20th-century Mediterranean differs substantially from Braudel's created in mid-century. In recreating the Mediterranean for the new millennium, the authors offer a substantial achievement that challenges many long-held assumptions not only about the Mediterranean, but also about human relations with the environment (in its broadest sense) and even the very nature of historical writing. It certainly deserves to provoke discussion among scholars from fields as broad as its own grand scope.

John Bennet is lecturer in Aegean prehistory, University of Oxford.