The theory and history of nationalism, applied principally in the context of France, was my principal field of research for much of the 1980s and 1990s. However, over the last ten or twelve years my interests have shifted to a related but nonetheless separate area of study—namely the French extreme Right between the wars and the vexed question of how far fascism took root in France in that period. So this paper has no pretensions to being a tour d’horizon of recent literature on nationalism, let alone a state-of-the-art assessment of the discipline. Instead, it offers a few retrospective thoughts on the problems and pitfalls of this field of study, hopefully illuminated by a fresh methodological perspective developed from my work on the French extreme Right, and my collaboration with the French political sociologist Michel Dobry.
Ubiquitous: A World of Nation-states
The first problem is one of terminology and definition. We live in a world of nation-states. The spread of this particular model of political and social organization worldwide over the last 200 years has ensured that nation-talk, the national idiom, permeates the political discourse of governments and political movements of every ideological shade. Even those that proclaim firmly internationalist principles (whether humanist or Marxist) have had to make concessions to the national dimension of popular experience and culture, and adapt their discourses accordingly. What is the effect on popular attitudes of this constant barrage of national symbols and references from every side? Does it simply help to foster a sense of belonging to a particular spatial and social environment, a vague and largely apolitical patriotism, or is this the “banal nationalism,” as Michael Billig calls it, on which nationalist ideologies and movements draw?1
The saturation of political discourse with nation-talk also raises a difficult conceptual problem. In an age where the language of patriotism has become a routine rhetorical device, how do we distinguish what is properly nationalist? What ideology, movement, or program does not tailor its appeal to the specific historical and cultural reference-points of the community in which it operates? The term nationalism has little value as an analytical tool if it is applied to every rhetorical flourish, every attempt to hide behind the fig-leaf of patriotism, or to dress up a policy in authentic national costume.
The quest for a more specific and useful definition has to focus on the key relationship between nationalism and the issue of state legitimacy in the modern age. When I first started writing on this subject, I was briefly persuaded that the term nationalism should be applied only to state-aspiring nationalist movements, and not used at all in the context of established nation-states. Thus, in my coauthored book with Günter Minnerup, we wrote:
Nationalism […] is not a state of mind or emotion, but a political movement and ideology dedicated to establishing, by secession, unification or revolution, a national state which will unite all individuals sharing a particular set of criteria of nationality (ethinic, cultural, historical) under a common regime based on the notion of popular sovereignty.2
However, such a narrow definition was very much at odds with both academic and everyday usage, and therefore difficult to sustain. It could, anyway, be convincingly argued that issues of state legitimacy and state sovereignty remained salient long after the achievement of statehood. Indeed, according to Immanuel Wallerstein, “in almost every case statehood preceded nationhood, and not the other way round, despite a widespread myth to the contrary.”3 Accordingly, a later coauthored essay with Spyros Sofos extended the definition as follows:
Within already established ‘nation-states’, the term may be applied to discourses that seek to secure the legitimacy of that state by creating or sustaining a sense of national identity, to political programmes that seek to protect or extend state sovereignty (defence in war, territorial expansion, economic protectionism, national autonomy in decision-making etc) or, finally, to attempts to form new political communities by creating new, or retrieving older, imagined communities.4
Implicit in these formulations is the belief that nationalism is an essentially political phenomenon. This is not to dismiss economic, cultural, and geographical factors as irrelevant to the subject of enquiry. But it does challenge the view that nationalism is somehow the product of preexisting sociocultural entities called nations. Much work in this field of study has been directed to defining the criteria of nationhood. Josef Stalin provided one of the more succinct definitions: “A nation is an historically evolved, stable community of language, territory, economic life, and psychological make-up manifested in a community of culture.”5 However, there were many different paths to nationhood, and the raw material mentioned above was rarely present in the necessary or desirable combination. Examples abound where nation-states were constructed despite considerable linguistic or religious diversity, where historical traditions were invented rather than discovered and where national boundaries had little rationale beyond the realpolitik of conquest or diplomacy.6 Even in cases where a sense of nationhood had to be promoted as a precondition of the struggle for statehood, this was always within the firmly political context of opposition to existing state structures.
A starting point for the study of nationalism is not whether nations exist; it is rather how the category operates in practice, that is, how nationalist logics and frames of reference are formulated and deployed. To quote from a source cited earlier: “in a way similar to proposed conceptualizations of ‘race’, we would argue that ‘nation’ should be conceptualized as ‘an unstable and decentred’ complex of social meanings constantly being transformed by political struggle.”7 The relationship should thus be inverted. Nationalism creates nations, not the other way round. Nations are political artifacts called into being by nationalist ideologies and movements.
I would, however, take this argument a stage further, and would use as illustration an issue raised by Eric Hobsbawm in his classic study Nations and Nationalism since 1780 . Although Hobsbawm acknowledges the role that the State may play in creating national identities after independent statehood has been achieved, he is extraordinarily reluctant to accept that nations may be forged by the actual political struggle for independent statehood itself, insisting instead on the primacy of economic geography and cultural affinities like religion and especially language. He dismisses many contemporary West European separatist movements as no more than “reactions against centralization […] or various other local or sectional discontents capable of being wrapped in coloured banners,” not real nationalism in fact.8 He may be right that some of the movements do not seriously envisage total state independence as their final aim in which case the validity of the tag nationalist is certainly debatable. But his underlying assumptions about what makes a nation are deterministic, and are contradicted by the pattern of nation-state formation in the last two hundred years.
I would agree, rather, with the Welsh Marxist historian Gwyn Alf Williams in his influential essay “When was Wales?” who wrote: “Wales is now and Wales has always been now […] it is not a mystical presence ubiquitous through our history […] Wales is an artefact which the Welsh produce; the Welsh make and remake Wales day by day and year after year […] Wales will not exist unless the Welsh want it.”9 This is not to deny that collective historical myths, a common language, economic viability, etc., are valuable as mobilizing themes and coagulants of identity, but doing creates a sense of being, identities are in a constant state of flux, and the solidarities wrought by the tide of political events may be as durable as those based on ethnicity.
Elusive: the Complexities of Nationalism
Even after some preliminary clearing of the ground, the term nationalism remains extraordinarily diverse in character, function, and field of application. Nationalism may be mobilized by established states, or by movements that challenge existing state structures. It has, at different times, been harnessed to ideologies of every shade across the political spectrum, to projects of social emancipation and to those of social integration, to the cause of imperialist expansion and to that of liberation, to authoritarian populism and to liberal democracy. Nationhood has been defined by the determinist principles of descendance ( droit du sang ) and by the voluntarist principles of residence (droit du sol), has been inspired by both ethnic and civic solidarities, has been achieved by state-led assimilation, or by a more tolerant incorporation of subsidiary identities in civil society. The sheer diversity of this experience warns against attempts to generalize about the processes of nation-formation or the ideological characteristics of nationalism(s), and indeed case studies of individual countries would reveal an infinitely more nuanced picture than can be conveyed by the set of contrasts outlined above.
The classic distinction is, of course, the one made by Hans Kohn between Western and Eastern nationalisms in Europe, the former political, rationalist, voluntarist, largely democratic (and often seen as French in inspiration), the latter cultural, mystical, exclusivist, authoritarian (and often seen as German).10 This dualism has been influential in subsequent historiography, where there is sometimes a tendency to see nationalism in terms of simple and over-schematic dichotomies—“open” versus “closed,” “civic” versus “ethnic,” “French” versus “German,” “Left” versus “Right” (not to mention “good” versus “bad”). France provides an especially interesting illustration of this polarity, because here (to a far greater extent than elsewhere) these rival forms of nationalism coexisted and competed for political ascendancy. In this scenario, the civic republican nationalism bequeathed by the French Revolution was challenged in the three decades before World War I by an exclusivist and authoritarian brand of nationalism, described by Zeev Sternhell as “particularist and organicist […] taking shape as a local variety of biological and racial nationalism, very close to the völkisch tradition in Germany.”11 Though it was only under the Vichy regime that this version briefly became the ideology of the state, arguably it has retained at the very least an intermittent political appeal in France for the last century.
The reality is, of course, more nuanced and complex. For example, the term “republican nationalism” seems inadequate to cover all the different historical situations in which the French Left has deployed the idea of nation. On the one hand it evok...