PRINCIPLES OF REGIONALISM

John A. Agnew

The nation-state as the fundamental geographical unit of account has been at the heart of the social sciences as a whole since the late nineteenth century. The origins of fields oriented to the "solution" of such public problems as wealth creation (economics), state management (political science), and social order (sociology) lay in providing services to the nation-state. Yet the "view from below," or that of social groups marginalized in orthodox political history and often associated with social history as a field of study, rests on the premise that the national scale typically represents the privileging of attention to the institutions associated with the interests and outlooks of modern political elites more than the reality of a homogeneous and enclosed society conforming to the political boundaries imposed by the modern system of territorial states. Moreover, not only have Europe's political boundaries been unstable over even relatively short periods of time, the geographical patterning of social life is by no means successfully captured by a singular focus on the national scale.

Of course, this is not to say that national processes of political and economic regulation are without substance in European social history. One study shows how a coherent rural region in the Pyrenees divided into separate Spanish and French national areas with the growth of effective monarchies as early as the seventeenth century. And, since the nineteenth century in particular, nation-states have played influential roles both in reinforcing and in changing various social phenomena. Rather, it is to suggest that the national is only one geographical scale among several in terms of relevance to understanding the long-term structuring of such phenomena as household and family organization, literacy, social protest, social-class formation, and political ideologies. Consequently, depending on the phenomenon in question, regions at a subnational level and regions at a supranational level are often invoked by social historians to provide more appropriate territorial units than the putative nation-state upon which to base social-historical investigation. As Otto Dann expresses it in Gli spazi del potere:

With the region, social history, liberated for some time from the weight of the national state, finally has found a more adequate concept of space. The region is the territory of the social historian, varying in its size and structure depending on the object of research. (p. 117)

The term "region" is often used without much conscious motivation other than either to group together nations that are apparently similar and thus to simplify complexity or to ground local studies within a larger geographical field of reference. The drawing of regional differences above and below the national scale also frequently involves deploying such familiar, and often theoretically unexamined, conceptual oppositions as modern-backward, commercial-feudal, and core-periphery, depending upon the theoretical orientation of the social history in question. The region, whatever its precise geographical and social parameters, seemingly cannot be avoided in social history, even when it is not rigorously defined as an inherent feature of a particular study. In the 1990s, however, there was a resurgence of studies explicitly engaging with subnational regions, not least because of the regional-ethnic revivals going on around Europe, from Spain and the British Isles to the former Yugoslavia and the Soviet Union. Regions as geographical units with which to define the contexts of study of a wide range of social structures and processes are therefore important both implicitly and explicitly in European social history.

Some "schools" of social history, particularly that associated with the Annales in interwar and immediate postwar France, have been explicitly devoted to avoiding the privileging of the state as the primary unit of geographical context. Perhaps the close link between geography and history in France led to a greater recognition by social and economic historians of the importance of assumptions about the spatial units used in research—recognition that is largely missing in the English-speaking world where an abstract sounding but usually nationally oriented sociology has tended to be more influential than geography among historians. Fernand Braudel's classic study,La Méditerranée (1949), is an excellent example of the use of a geographical frame of reference, in this case an ocean basin, as an alternative to the nation-states that had dominated historical research during the nineteenth and much of the twentieth century. For Braudel's long-term total history the relatively short histories of European states posed a significant barrier to the historical understanding that only a larger regional entity, such as the Mediterranean world, could adequately convey. Of course, even Braudel eventually succumbed to the allure of national history in his L'identité de la France (1986), though this work remains more sensitive than the typical national history to the physical geography and regional distinctions of the territory that later became France as we know it today. In addition, according to Lynn Hunt:

Despite the enormous prestige of La Méditerranée, Braudel's example did not elicit many works within the French historical community on cross-national networks of commercial exchange. Rather, French historians of the third Annales generation focused largely on France, and usually on one region of France. The best known of these great thèses were Les Paysans de Languedoc (1966) by Emmanuel Le Roy Ladurie and Beauvais et le Beauvaisis (1960) by Pierre Goubert. (p. 212)

Since the 1960s, world-systems frameworks such as that of Immanuel Wallerstein, based on distinguishing dynamic economic-geographical core macroregions, such as northwest Europe after 1700, from relatively peripheral or exploited ones, such as eastern and southern Europe; theoretical frameworks such as that of Edward W. Fox (History in Geographic Perspective, 1971), posing an opposition between "commercial" and "feudal" regions within countries such as France; and internal-colonial or mode of production arguments such as those of Michael Hechter (Internal Colonialism, 1976) and William Brustein (Social Origins of Political Regionalism, 1988), identifying different types of regions within states with respect to political and social characteristics, represent different ways of explicitly incorporating regions into social-historical analysis. Even greater emphasis on the role of regions as contexts for social invention and political affiliation can be found in the work of the economic historians Sidney Pollard (Peaceful Conquest, 1981) and Gary Herrigel (Industrial Constructions, 1996), and in that of economic sociologists such as Arnaldo Bagnasco on local economic development and the social construction of the market (Tre Italie, 1977). Demographers like Peter Laslett have found regional principles in typologies of family structure, such as East European extended families versus West European nuclear families. Much research, however, tends to operate on an implicit rather than an explicit conception of region. Even as they adopt regional frameworks in their research, social historians are not necessarily very aware of the nature of the geographical divisions that they use.

Europe, of course, is itself a region in the most macro-scale sense of the term. It serves to define the territorial space with respect to which European social history is practiced. Yet, analyses of Europe as a whole in social history are relatively recent, notwithstanding the tendency to make generalizations about "Europe" on the basis of studies of only small parts of it. The principles of regionalism must take this wider context into account so as to identify the various and sundry geographical divisions of the continent. Such principles, or rules, for defining the geographical basis to European social-historical variation must also pay attention to intellectual disputes about the nature of regions and to how regions have been used by social historians. The four sections of this article present, first, a discussion of Europe as a world region; second, a recounting of disputes over the character of regions as meaningful entities in social-historical research; third, a survey of some ways in which regions have been used in European social history; and, fourth, a review of the principles upon which a geographical division must rest, drawing from both the practice of social history and recent work by geographers interested in the ways in which Europe can be thought about in terms of its internal geographical divisions.

EUROPE AS A REGION

"Europe" can be thought of in geographical, historical, and institutional terms, if in practice its various meanings are often conflated. With respect to physical geography, the ancient Greeks used the term "Europe" to denote the lands to their west and north as part of a threefold division of the world that distinguished Europe from Asia to the east and Libya (Africa) to the south. Writers such as Herodotus and Strabo regarded these terms as conventional or arbitrary ones, open to systematic questioning. But, for most of the two millennia or more since they wrote, the continental scheme has been largely taken for granted as betraying some sort of essential geographical division of the world (Lewis and Wigen, 1997). Controversy has flared up over the precise delimitation of Europe from its continental neighbors, with the Ural Mountains replacing the Don River and the Sea of Azov as its eastern border by the early twentieth century, and religious, racial, and civilizational criteria increasingly substituting for physical criteria as the basis for identifying Europe in opposition to other world regions. However, Europe is still largely seen as a self-evident unit whose history has a unity too as a result of a collective destiny created by its global location and the physical attributes (physiographic range, temperate climates, location relative to oceanic wind belts, internal environmental diversity, and so forth) associated with it. One effect of this reasoning, seen in so many global histories (for example, Paul Kennedy's Rise and Fall of the Great Powers, 1987; and David Landes's Wealth and Poverty of Nations, 1998), has been to exempt Europe from the rule of absolute environmental determinism, seeing it as distinctive among the continents in offering the environmental possibilities out of which European "inventiveness," "inquisitiveness," and, finally, justifiable domination of the rest of the world, including the identification and naming of world regions, are seen as arising. Nevertheless, the logic underpinning Europe's claim to distinctiveness is still a physical-geographical one.

To most social historians, however, it is not the physical character of the continent that lies behind the appropriate use of the term. Rather, Europe's existence is understood as that of a geographical entity with a set of common or overlapping historical experiences (Wilson and van der Dussen, 1993). Thus, much of southern and western Europe was a part of the Roman Empire for at least several centuries. After the collapse of the empire, a much larger part of Europe became the global stronghold of Christianity, if with increasing sectarian divisions creating geographical ones (such as that of the tenth century a. d. between the western Catholic and eastern Orthodox traditions and the later-fifteenth-century division between Catholic and Protestant Christianity). The growth of merchant capitalism beginning in the eleventh century reintroduced city trading networks into the fabric of European society after the long retreat of trade during feudalism. With the decline of royal dynastic authority, the rise of city- and then territorial states as the premier and totalistic means of organizing political sovereignty was initially peculiar to Europe and led to political competition that then spilled out into the rest of the world and brought about the various European-based world empires.

Among other forces at work in producing a common European experience must be included the geographically differential impact of the French Revolution's (1789) call to overthrow the established aristocratic political order, the explosion of industrial urbanism from the mid-eighteenth century on, the spread of nationalist and socialist ideologies in the nineteenth century, and, above all, the slow secularization of society from the singular hold of religious authority in the nineteenth and twentieth centuries that created a Europe-wide experience of competing social allegiances and political ideologies that then distinguished the region as a whole from all others. As a result, according to the historical demographer Emmanuel Todd in L'invention d'Europe (1990), and with respect to political ideologies:

European religious and ideological passions are written in space. Each nation, each region [within Europe] adheres either to the Reformation or to the Revolution, to social democracy or to anarchism, liberalism, communism, fascism, or nazism. Each confronts its neighbors in the name of values equally absolute and undemonstrable. (p. 9)

The menu of political choices, therefore, is determined by experiences particular to the space labeled "Europe." The same goes for all manner of other phenomena that have been influenced by the common social, political, and economic experience of the region. As social history has turned more to cultural sources, a few efforts have attempted to describe how popular myths and beliefs have originated and spread across Europe.

Finally, today Europe is increasingly thought of in institutional terms, reflecting the rising importance within segments of the geographical and historical Europe of such entities as the European Union and its affiliated organizations such as the European Court of Justice and the European Parliament (Lévy, 1997). With the removal of the Iron Curtain, the ideological frontier within Europe established after World War II, the project of European unification, initiated by the Treaties of Rome in 1957 between the original six members of the post-1993 European Union, is potentially available to a large number of countries both to the east of the original core members and around the Mediterranean. The Maastricht Accord of 1992 offered a calendar for European political and monetary unification. The introduction in 1999 of the new currency, the euro, by eleven of the fifteen member countries of the European Union represents an important step in the institutional construction of a Europe with a common citizenship, political economy, and policymaking apparatus. The term Europe has become the basis for deciding which countries can be eligible for membership. Rather than singularly geographical or historical, however, the criteria are largely economic and political. Above all, conformity to a neoliberal political economy and to the practices of electoral democracy are now necessary prerequisites for joining the European Union. The project of creating a "common European home," therefore, represents a break with preexisting ways of defining Europe. Now it is a set of common values arising out of the European past but without precise geographical limits that defines who can be "inside" and who is left "outside" the European "project." Neither the physical barrier provided by the Urals nor the influence of common European experiences, such as that of Christianity, can tell who is inside and who is outside of Europe. From the institutional perspective, therefore, Europe now has a culturally virtual rather than a geographically actual existence.

WHAT ARE REGIONS?

The term "region" typically conjures up the idea of a homogeneous block of space that has a persisting distinctiveness due to its physical and/or cultural characteristics. Yet, many regions are more networks of connections between concentrations of populations and places than simply uniform spatial units. An allied claim is often that regions exist "out there" in the world, notwithstanding the prior necessity on the part of an observer of thinking that the world is in fact divided up into regions. Over the years, six disputes about regions have episodically flared up both to challenge and enliven the generally consensus view in the social sciences of regions as homogeneous, self-evident blocks of terrestrial space.

The first controversy has been about the ways in which the areas designated as regions are integrated and/or exhibit homogeneous characteristics. Typically, regions are thought of as areas exhibiting uniformity with respect to one or more characteristics. This view has been challenged by scholars who claim that such regions are often purely formal, in the sense that they are the result of aggregating smaller geographical units (census districts, municipalities, provinces, and so forth) according to statistical similarity without attending to what it is that binds the region together with respect to functional ties. Functional ties include the network or circulation linkages (transport, migration, trade, and capital flows) and central-place (settlement hierarchy) links that create distinctive regions and from which their other characteristics are derived (as described, for example, in Paul Hohenberg and Lynn H. Lees, The Making of Urban Europe, 1995). Of course, regions are often politically defined by governments (Patriarca, 1994) and political movements (such as separatist ones). They can also have affective meaning for local populations (Applegate, 1999). In such cases, the absolute formal-functional opposition fails to account for the subjective identifications that people can have with formal regions, even if it continues to serve a useful analytic purpose more generally.

Another dispute concerns the belief that regions are real in the sense of marking off truly distinctive bits of the earth's surface versus the view that they are the product solely of political and social conventions that impose regions on a much more geographically variegated world. There is a visceral tension between the idea that something is real and that is constructed. But are these ideas indeed as mutually exclusive as the dispute suggests? On the one hand, the real is like the body in philosophy's mind-body problem. It is tangible, touchable, and empirically visible. On the other hand, the constructed is like the mind making sense of itself and the body. Each of these positions rests on the same confusion between an object (a region) and an idea about that object (regional schemes). Regions reflect both differences in the world and ideas about the geography of such differences. They cannot be reduced to simply one or the other (Agnew, 1999).

A third controversy has focused on the tendency to see regions as fixed for long time periods rather than as mutable and subject to reformulation, even over relatively short periods. Leading figures in the Annales school, such as Marc Bloch and Fernand Braudel; world-systems theorists; and demographic historians have been particularly drawn to the idea of macro-regions as the settings for long-term structural history. At the same time others, particularly local historians and regional geographers, have invested heavily in the idea of fixed regional divisions and unique regional entities within countries, owing their uniqueness to "internal" characteristics. However, with the increased sense of a world subject to time-space compression, following the opening of national borders to increased trade, capital, and labor mobility and the shrinkage of global communication and transportation costs, regions are increasingly seen as contingent on the changing character of the larger contexts in which they are embedded rather than dependent on unique features of a more-or-less permanent nature (Johnston, Hauer, and Hoekveld, 1990; Gupta and Ferguson, 1997).

Less noted but perhaps more important with respect to the meaning of regions for social history, a debate has periodically erupted over regions as fundamental contexts for social life as opposed to mere accounting devices or case study settings taken as examples of national or Europe-wide norms and standards. With respect to industrialization, for example, Sidney Pollard has argued that regions are the relevant entities for considering the processes whereby different industries developed. Each region has different combinations of attributes crucial to the establishment of specific industries. In like manner, social and political processes relating to household structures, class formation, and political movements can all be thought of as embedded in regional and local contexts, "the physical arenas in which human interaction takes place" (Weitz, 1995, p. 291), rather than as abstract or national-level processes only manifesting themselves regionally, as presumed by the idea of the regional case study.

A fifth controversy has involved the tendency to represent the character of regions by locating them along a temporal continuum from the backward, or traditional, at one end and the advanced, or modern, at the other. This conversion of time into space has been particularly important in historicizing certain subnational regions (such as the Italian south, the Scottish Highlands, and Andalusia) and countries as a whole (such as Italy or Ireland) into a schema representing the historical trajectory of Europe as a whole (Agnew, 1996). Thus, presumably isolated and remote regions with lower levels of economic growth than more central regions are viewed as lagging behind the more advanced ones, notwithstanding the long-term ties that bind such regions into their particular nation-states. This tendency has given rise to a contending view that poorer regions are poor because the richer ones have become rich at their expense (as in Hechter, 1976, on the British Isles)—in other words, it is not a temporal lag but rather spatial exploitation that lies behind regional differences in economic development and social change.

Finally, perhaps the dominant sense of social historians about regions, particularly regions at the subnational level, has been of entities destined to fade in significance with the creation of national markets, the emergence of national political parties with more or less uniform support across all regions, and the spread of national cultures robbing local and regional identities of their specificity. This nationalization or modernization thesis, articulated in works ranging from Eugen Weber's general study of late-nineteenth-century France, Peasants into Frenchmen (1976), to Susan Cotts Watkins's survey of demographic indicators (fertility rates, women's age at marriage, and so forth) across western Europe between 1870 and 1960, From Provinces into Nations (1991), relies on the premise that social organization in Europe has undergone a fundamental shift from local and regional levels to the national scale. This premise is a shaky one, however. Some of the data in a study such as that of Watkins can be interpreted to indicate reprovincialization after a period of nationalization, and nationalization of demographic indicators need not indicate the substitution of regional sources of social influence by national ones. Rather, demographic behavior may still be mediated through the regionally specific routines and institutions of everyday life yet yield increasing similarity of behavioral outcomes across regions. The same goes for religious affiliations, voting, consumption, and other types of social behavior (Agnew, 1987; Cartocci, 1994).

REGIONS IN EUROPEAN SOCIAL HISTORY

Four modes of usage of regions dominate social histories of Europe. The first consists of macroregions as units for the pursuit of total history. The locus classicus of this approach is Fernand Braudel's La Méditerranée (1949). The claim is that over long periods of time regions emerge based on functional linkages that then continue to distinguish one from the other. Such regions need not be ocean basins such as the Black Sea, the Indian Ocean, or the Mediterranean. They can be units determined by their relative orientations toward certain modes of production and exchange. Edward W. Fox's History in Geographic Perspective: The Other France (1971) may be used to illustrate this case briefly, as the logic of the argument need not be restricted to a single national setting.

The second and perhaps most common mode of use is that of dividing up Europe into functional regions to examine specific phenomena such as class transitions and transformations of rule, the nature of landholding and manorialism, industrialization, urbanization, and trade. Sometimes these regions are at a macro scale, as with the divisions between western and eastern Europe (or between western, central or middle, and eastern Europe) in such works as Barrington Moore Jr.'s Social Origins of Dictatorship and Democracy (1966; although this study extends in scope well beyond Europe per se), Perry Anderson's Lineages of the Absolutist State (1974), and William McNeill's The Shape of European History (1974). Sometimes the regions are more fine-grained and subnational, as in Gary Herrigel's study of German industrialization, Industrial Constructions (1996), Charles Tilly's work (for example, Coercion, Capital, and European States, a. d. 990–1992, 1992) on the logics of coercion and capital in European urbanization and state formation, and work on regional differences in artistic production as in Enrico Castelnuovo and Carlo Ginzburg, "Centre and Periphery" (1994), on Italy. Stein Rokkan's geographical template for Europe as a whole with respect to rates and degrees of state formation (for example, Rokkan and Urwin, Economy, Territory, Identity, 1983) serves as an example of work that brings together the main west-east division of the continent with the center-periphery differences that have developed within the emerging states.

The third use is to aggregate together lower-level units (counties, departments, and so forth) without much regard for national boundaries to identify persisting patterns of demographic, social, and political behavior. Regions are thus geographical areas of similarity extending across space and time. This inductive approach to regionalization is most common in studies of demography, literacy, land tenure, economic growth, and the development of political ideologies. Emmanuel Todd's L'invention d'Europe (1990) is an example of this genre of usage.

Finally, the explosion of regionalist and separatist movements in Europe has stimulated considerable interest in the emergence and roots of regional identities in relation to national ones. Charlotte Tacke's comparison of the regional bases to German and French national identities, "The Nation in the Region" (1994), serves as an example drawn from a now vast and diverse literature because of its emphasis on regionality as a source of political identities.

Macroregions. Struck by a France that seemed to repeatedly divide itself since the Revolution of 1789 into two sociopolitical divisions around "order" and "movement," Edward W. Fox writes, "For an American, it was natural to begin by seeking to identify these societies in sectional terms" (p. 13). Unlike the United States, however, France has had nothing like a regional-sectional civil war since at least the medieval Albigensian Crusade. Fox finds the regional division in the different communications orbits that have emerged down the years between a Paris-oriented interior France and an externally oriented commercial France along the coasts. He gives the argument a transcendental appeal by claiming that the opposition between an agricultural-military society, on the one hand, and a commercial-seagoing society, on the other, can be found in ancient Greece and in medieval Europe as much as in the modern world. Fox is distinguishing between a subsistence society dependent on control of territory and a waterborne commercial society dependent on access to flows of goods and capital. The two "types" of society achieved their most characteristic forms during the "long" century between the revolutions of the sixteenth century and the French Revolution. The social commentators of the time, such as Montesquieu, clearly recognized them. Fox uses the dichotomous model as a framework for exploring the course of French social history since 1789, but accepts that by the Fifth Republic the opposition between two societies had largely run its material course, even if the legacy of the two Frances still "left its imprint upon the political preferences of their members" (Fox-Genovese and Genovese, 1989, p. 237).

Fox's regionalization rests on what can be called a fixed spatial division of labor between two different modes of production which though present within the boundaries of the same state nevertheless have both fractured that state and led to distinctive social orders (class struggles, inheritance systems, religious and political affiliations, and so on) within it. Thus, the history of France (and, Fox suggests, many other states) cannot be understood satisfactorily as a singular whole but only in terms of the opposition and interaction between "two Frances" based upon competing principles of social and economic organization. Though articulated in the setting of a specific (perhaps the quintessential) nation-state, Fox's argument is similar to other macroregional ones in pointing to the persistence of regional patterns of social and political behavior as the foundation for interpreting other social phenomena. Whether such phenomena can be invariably reduced to the opposition is, of course, another thing entirely.

Functional regions. The late Stein Rokkan's research enterprise was oriented to understanding the varying character (unitary versus federal, democratic versus authoritarian, and so forth) of Europe's modern states (see Rokkan and Urwin, Economy, Territory, Identity, 1983). Among other things, he noted that adjacent states tended to develop similar forms of government and that there was a fairly systematic north-south and east-west dimensionality to this variation. He represented spatial variation between states in a series of schematic diagrams transforming Europe into an abstract space by drawing on crucial periods and processes in European socio-political history. Three periods or processes are seen as crucial. The first is the pattern of the peopling and vernacularization of language in the aftermath of the Roman Empire. This produces a geoethnic map of Europe based on the south-north influence of the Romans and a west-east physical geography–ethnic geography of the settlement of new groups and their differentiation from one another. The second is the pattern of economic development and urbanization in medieval to early modern Europe, distinguishing a south-north axis drawn largely with reference to the impact of the Protestant Reformation and the Catholic Counter-Reformation and an eastwest axis with strong seaward states to the west, a belt of city-states in the center, and a set of weak landward states to the east. The third is the way in which democratization has produced different responses in different regions with smaller unitary states in the extreme west, larger unitary states flanking them to the east, a belt of federal and consociational states in the center, and a set of "retrenched empires" and successor authoritarian states yet further to the east.

This geographical template draws attention to systematic geographic variation in the forms of European states and how they arose out of different combinations of social and economic processes. It is particularly original in pointing out the distinctiveness of a long-established urbanized region running from Italy in the south to Flanders in the north. But this use of regionalization neglects the ways in which the social divisions to which Rokkan refers (ethnic identities, city-states versus territorial states and empires, religious affiliations) are translated into political power and how this in turn affects the character of state formation. An entire stage in the process of creating the political map of Europe is missing. As Charles Tilly puts it, perhaps a little too forcefully: "It is hard to see how Rokkan could have gotten much farther without laying aside his maps and concentrating on the analysis of the mechanisms of state formation" (Tilly, 1992, p. 13).

Supranational regions. A very different approach to the use of regions is to use local government areas in different countries as the basis for identifying clusters of units that can cross national boundaries and that define formal regions sharing particular attributes to one degree or another. Maps can be made of such phenomena as family types, fertility and mortality rates, rates of suicide, types of landholding, modes of agrarian organization (sharecropping, peasant proprietorship, capitalist agriculture, and so forth), literacy, religious practice (for example, attendance at Catholic mass), levels of industrial employment, civic culture, and levels of support for ideological parties of the right and the left (see, for example, Goody et al., 1976; Le Bras, 1979; Graff, 1981; and Putnam, 1993). These maps can also be correlated to see to what extent the various phenomena covary spatially with one another. For example, high suicide rates do correlate highly in some places with high rates of illegitimate births and high female autonomy (for instance, much of Sweden and Finland), but elsewhere, as in southern Portugal, they seem to correlate more with something absent in the rest of Europe, perhaps going back to the recovery of the region from Islamic conquest, matrilineal inheritance of names, equal relations in families between parents, and a nuclear ideal of family (Todd, 1990, pp. 56–61).

Various hypotheses about secularization of European society, the impact of industrialization, and the persisting effects on politics and social life of historic forms of household and family organization have been investigated by Emmanuel Todd and others taking this approach. Todd is perhaps the most forceful in his claim for basing the incidence of a wide range of social phenomena on the prior spatial distribution of family types. He shows quite convincingly that family types (communal, nuclear, stem, and so on), inheritance customs, parent-child relations, and certain features of fertility in Europe do not conform to national-level patterns. Rather, there are both localized clusters within countries and regional groupings that crisscross national boundaries. What is less convincing is the degree to which other social phenomena are truly the outcome of the "underlying" demographic and familial characteristics rather than mediated regionally by a range of economic and social pressures that have extraregional rather than historically accrued local sources. The tendency is to rigidly interpret regional patterns of "higher-level" phenomena (such as political ideologies or civic cultures) as arising from long-term regional patterns of familial and demographic features (see Sabetti, 1996).

Subnational regions. Finally, subnational regional identities have become the focus for social historians and others concerned with the history and restructuring of European political identities (for example, Applegate, 1999). Nations and regions are typically understood as categories of practice that are reified or given separate existence by people struggling to define themselves as members of this or that group. Much work seeks to identify the diversity of group identities in contemporary Europe and how they have arisen. A distinctive current, however, tries to relate regional to national identities as they have arisen over the past several hundred years. The basic premise is that regional and national identities are often intertwined rather than necessarily oppositional. In comparing the historical construction of French and German national identities, Charlotte Tacke claims that "the individual's identification with the nation . . . rests on a large variety of social ties, which simultaneously forge the links between the individual and the nation" (Tacke, 1994, pp. 691–692). The most important ties are those constituted in regions, which serve as "cultural and social space" for "civic communication" (p. 694). Local bourgeoisies in both countries created renewed regional identities at precisely the same time that the symbols they selected (honoring ancient heroes in statues, for example) were made available for appropriation by nation-building elites. In these cases, therefore, regional identities fed into the national ones and were thus lost from sight.

Elsewhere in Europe, however, regional identities appear more as acts of opposition than of accommodation to national ones. This is the message not only of the internal-colonial and mode of production approaches but also of constructivist approaches that emphasize the tendency of region and nation to become synonymous in some social-cultural contexts. Resistant regional identities, such as the Irish and Basque ones, have taken shape around claims to nationhood. Unlike the French and German cases, they have tried to develop spatial mythologies alternative to the dominant nations within their respective states (the English and the Castilian, respectively) but are often forced into terms of debate and the use of institutional forms that signify the inevitability of at least a degree of accommodation to the territorial status quo. Of course, the resistant regional identities suggest that the word "region" in political usage is itself dependent on the prior existence of nation-states of which the regions are presently part but from which they could possibly separate to become their own nations in the future. One lesson is clear. If all of the other meanings of the term discussed previously are neglected in pursuit of the currently fashionable interest in political regionalism, then we are left with thin intellectual gruel indeed: regions are only potential nations-in-the-making. The attempt to find an alternative regional accounting system to that of the dominant national one would then have come full circle.

PRINCIPLES OF REGIONALISM

The division or partition of Europe into regions cannot be reduced to one best way or a single overarching parameter. Usage is so diverse and disputes over the substance and philosophy of regions are too contentious to allow for application of a single principle of division. This being the case, it makes more sense to tailor usage to specific needs. In this spirit, I want to explore four principles of regionalism that can be applied to the analysis of different research problems based on current practice among social historians and geographers.

The first principle is that of distinctive regional communities that can share identities as well as other sociopolitical characteristics. This principle is most useful for those focusing on the vagaries of subnational political regionalism as well as the persistence of sociopolitical traits from the past. Europe has long been divided in complex ways with respect to language, religion, urbanization, the persistence or reinstatement of feudalism, agrarian systems, and the experience of industrialization. These are all symptomatic of the patchwork of social and place identities and interests that define Europe's varied communities. Nomadic and immigrant groups, most importantly, Roma (or Gypsies), Jews, and non-European immigrants, have had to fit themselves into this kaleidoscope of local and regional communities. With nation-state formation from the eighteenth century on, such groups have had to cope with the tension, and sometimes the conflict, arising between regional identities (known as Heimat in German) and national ones (represented in German by the word Vaterland). In different countries the tension has resolved itself, at least temporarily, in different ways. If in Germany identification with a Heimat has not proved inimical to the growth of a Vaterland identity, elsewhere the "resolution" has been to the advantage of one or the other.

The second principle is that of geopolitical territories under construction and challenge, often on the peripheries of states. Apparently less relevant to the interests of many social historians, this one is useful for those concerned with the tensions and conflicts associated with state formation and disintegration. As authors such as Stein Rokkan and Charles Tilly have suggested, historically based lines of geographical fracture both between and within states have emerged due to differences in state organization and the divergent histories of capitalism in different parts of Europe. Such fractures, typically involving center-periphery cleavages across the political map of Europe, have been reinforced by the popular memory of wars and the territorial claims these have entailed (such as Alsace-Lorraine in the Germany-France conflicts from 1870 to 1945). Within-state regional divisions were dampened by the growth, uneven and partial, of redistributive mechanisms associated with the growth of the European welfare state. With the advent of potentially Europe-wide organizations, such as the European Union, the fractures between states have receded somewhat as the ones within states, largely because of the perception that power now flows increasingly from Brussels as the site of the governing European Commission, have become increasingly important.

The third principle is that of geographical networks that tie together regions through hierarchies of cities and their hinterlands. This is most relevant to studies of industrialization, urbanization, and trade. The European settlement hierarchy has long been one of the most important integrative factors in the continent's history. Linking cities and their hinterlands into a network of centers organized by size and specialization, the European urban system has always worked against a singular territorial organization of Europe into national-state territories. Of course, this system has waxed and waned relative to the significance of national boundaries in channeling flows of goods, capital, and people. In the late twentieth century it was once again in ascendancy after a long period of relative subservience to the regulatory activities of Europe's states. Its recognition led to an emphasis on Europe as a set of connected functional regions rather than the tendency of the other principles to highlight the role of adjacency in creating formal regions homogeneous with respect to one or more social characteristic.

The fourth and final principle is that of regional societies that share a wide range of social and cultural characteristics. This fits the needs of those interested in associating social indicators to examine hypotheses about trends in social phenomena such as classes, family types, secularization, and political activities by identifying formal regions. With industrialization and urbanization since the nineteenth century, the more or less settled dimensions of social life, associated primarily with the relative social stability of rural life, have been disrupted in major ways. Initially the growth of the industrial working class was the most significant development. How this happened differed between different subnational regions, the primary geographical scale at which industrialization took place in Europe. Important social trends, also differing regionally, include the relative decline of social class as a marker of identities, the rise of so-called postmaterialist values (environmentalism and the like), growing secularization, and the development of new social identities as women and immigrant minorities acquire distinctive social imaginations. Above all is the increasing tension between established commitments to larger groups, on the one hand, such as families, occupational groups, or religious sects, and the growth of consumer and personal values that celebrate the choices of the individual, on the other. Given their divergent histories, regions, both sub- and cross-national, can be expected to differ with respect to how they cope with such social change.

Each of the four principles is recognizably related to the existing main categories of the research agenda of European social history. The first focuses mostly on the regional social inheritance from the past whereas the second is concerned with the mutual roles of regions and states in creating social and political identities. The third principle of regionalism identifies the functional regions of European urbanization as lying at the heart of the geographical organization of European economic development, notwithstanding the historically important roles in economic policy conducted by national-state governments. The fourth and final principle is directed at understanding the regional impacts of social change by means of how regions provide the contexts of everyday lives, on the one hand, in which the effects of larger-scale changes are mediated, on the other.

CONCLUSION

The point of thinking about European social history in terms of regions is not to use them, whether supra- or subnational, as a totalizing alternative to the geographical template provided by Europe's national-state boundaries. This is missed by commentators who wrongly think that nations and regions are simply opposite ways of dividing up Europe and, typically, that the former invariably, at least since the nineteenth century, trump the latter (for example, Hobsbawm, 1989). Of course, "nations" are in fact a type of region, albeit of a highly institutionalized variety. Rather, the purpose of regions is to consider the geography of Europe in a more complex way than that usually adopted: the simple coloring in of a map of Europe on the basis of its national boundaries, as in Émile Durkheim's now infamous use of national boundaries to represent a much more variegated pattern of the incidence of suicide. These national boundaries have been both too unstable over the medium term and too unimportant for representing the incidence of a wide range of phenomena (family types and agrarian systems, for example) to justify their dominating the practice of social history.

Regions are themselves obviously contestable. Hence the need to carefully adumbrate the principles upon which a given exercise in regionalizing should rest. That said, the emplacement of social phenomena is inevitably fraught when the phenomena themselves elude placement, as is increasingly the case in a world characterized by flow more than by territorial stasis. Increasingly, "social identities, geographical locations, and national allegiances all tend to be out of sync, at least more so now than in the recent past" (Rafael, 1999, p. 1210). This does not license abandoning regionalism, only attending to the potential dislocations of existing schemes of regions in a world in which a global field of forces is increasingly disrupting the territorial status quo in Europe.

See also other articles in this section.

BIBLIOGRAPHY

Europe As a Region

Foucher, Michel. Fragments d'Europe: Atlas de l'Europe médiane et orientale. Paris, 1993.

Hay, Denys. Europe: The Emergence of an Idea. New York, 1966.

Kennedy, Paul. The Rise and Fall of the Great Powers: Economic Change and Military Conflict from 1500 to 2000. New York, 1987.

Landes, David S. The Wealth and Poverty of Nations: Why Some Are So Rich and Some So Poor. New York, 1998.

Lévy, Jacques. Europe: Une géographie. Paris, 1997.

Lewis, Martin, and Kären Wigen. The Myth of Continents: A Critique of Metageography. Berkeley, Calif., 1997.

Todd, Emmanuel. L'invention d'Europe. Paris, 1990.

Wilson, Kevin, Jan van der Dussen et al, eds. The History of the Idea of Europe. Rev. ed. London, 1995.

What Are Regions?

Agnew, John A. Place and Politics: The Geographical Mediation of State and Society. London, 1987.

Agnew, John A. "Regions on the Mind Does Not Equal Regions of the Mind." Progress in Human Geography 23 (1999): 91–96.

Agnew, John A. "Time into Space: The Myth of 'Backward' Italy in Modern Europe." Time and Society 5 (1996): 27–45.

Applegate, Celia. "A Europe of Regions: Reflections on the Historiography of Sub-National Places in Modern Times." American Historical Review 104 (1999): 1157–1182.

Cartocci, Roberto. Fra lega e chiesa: L'Italia in cerca di integrazione. Bologna, Italy, 1994.

Dann, Otto. "La regione: Una cornice elastica per la nuova storia sociale." In Gli spazi del potere. Edited by Franco Andreucci and Alessandra Pescarolo. Florence, Italy, 1989. Pages 115–118.

Gupta, Akhil, and James Ferguson, eds. Culture, Power, and Place: Explorations in Critical Anthropology. Durham, N.C., 1997.

Hechter, Michael. Internal Colonialism: The Celtic Fringe in British National Development. Berkeley, Calif., 1976.

Hohenberg, Paul, and Lynn Hollen Lees. The Making of Urban Europe, 1000–1994. Cambridge, Mass., 1995.

Johnston, R. J., J. Hauer, and G. A. Hoekveld, eds. Regional Geography: Current Developments and Future Prospects. London, 1990.

Patriarca, Silvana. "Statistical Nation Building and the Consolidation of Regions in Italy." Social Science History 18 (1994): 359–376.

Pollard, Sidney. Peaceful Conquest: The Industrialization of Europe, 1760–1970. New York, 1981.

Watkins, Susan Cotts. From Provinces into Nations: Demographic Integration in Western Europe, 1870–1960. Princeton, N. J., 1991.

Weber, Eugen. Peasants into Frenchmen: The Modernization of Rural France, 1870–1914. Stanford, Calif., 1976.

Weitz, Eric D. "The Realms of Identities: A Comment on Class and Politics in Milan." Social Science History 19 (1995): 289–294.

Regions in European Social History

Anderson, Perry. Lineages of the Absolutist State. London, 1974.

Bagnasco, Arnaldo. Tre Italie: La probematica territoriale dello sviluppo italiano. Bologna, Italy, 1977.

Braudel, Fernand. L'identité de la France. Paris, 1986. English translation by Siân Reynolds. The Identity of France. Vol. 1: History and Environment. New York, 1989.

Braudel, Fernand. La Méditerranée et le monde Méditerranéen à l'époque de Philippe II. Paris, 1949. English translation by Siân Reynolds. The Mediterranean and the Mediterranean World in the Age of Philip II. New York, 1972.

Brustein, William. The Social Origins of Political Regionalism: France, 1849–1981. Berkeley, Calif., 1988.

Castelnuovo, Enrico, and Carlo Ginzberg. "Centre and Periphery." In History of Italian Art. Vol. 1 Edited by Peter Burke. Cambridge, U.K. 1994. Pages 29–112.

Fox, Edward Whiting. History in Geographic Perspective: The Other France. New York, 1971.

Fox-Genovese, Elizabeth, and Eugene D. Genovese. "Social Classes and Class Struggles in Geographic Perspective." In Eugene D. Genovese and Leonard Hochberg, eds. Geographic Perspectives in History. Oxford, 1989. Pages 235–255.

Goody, Jack, Joan Thirsk, and E. P. Thompson. Family and Inheritance: Rural Society in Western Europe. Cambridge, U.K., 1976.

Graff, Harvey J., ed. Literacy and Social Development in the West. Cambridge, U.K., 1981.

Hechter, Michael. Internal Colonialism: The Celtic Fringe in British National Development. Berkeley, Calif. 1976.

Herrigel, Gary. Industrial Constructions: The Sources of German Industrial Power. New York, 1996.

Hunt, Lynn. "French History in the Last Twenty Years: The Rise and Fall of the Annales Paradigm." Journal of Contemporary History 21 (1986): 209–224.

Le Bras, Hervé. L'enfant et la famille dans les pays de l'OCDE. Paris, 1979.

McNeill, William. The Shape of European History. New York, 1974.

Moore, Barrington, Jr. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston, 1966.

Pollard, Sidney. Peaceful Conquest: The Industrialization of Europe, 1760-1970. New York, 1981.

Putnam, Robert. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, N.J., 1993.

Rokkan, Stein, and Derek W. Urwin. Economy, Territory, Identity: Politics of West European Peripheries. London, 1983.

Sabetti, Filippo. "Path Dependency and Civic Culture: Some Lessons from Italy about Interpreting Social Experiments." Politics and Society 24 (1996): 19–44.

Tacke, Charlotte. "The Nation in the Region: National Movements in Germany and France in the Nineteenth Century." In Nationalism in Europe: Past and Present. 2 vols. Edited by Justo G. Beramendi, Ramon Maiz, and Xosé M. Nunez. Vol. 2. Santiago de Compostela, Spain, 1994. Pages 658–675.

Tilly, Charles. Coercion, Capital, and European States, a. d. 990–1992.

Oxford, 1992. Wallerstein, Immanuel. The Modern World-System. Vol. 1: Capitalist Agriculture and the Origins of the European World-Economy in the Sixteenth Century. New York, 1974.

Conclusion

Hobsbawm, Eric. "La dimensione statale come fondamento delle articolazioni regionali." In Gli spazi del potere. Edited by Franco Andreucci and Alessandro Pescarolo. Florence, Italy, 1989. Pages 230–237.

Rafael, Vicente L. "Regionalism, Area Studies, and the Accidents of Agency." American Historical Review 104 (1999): 1208–1220.