What do we talk about when we talk about nationalism? Too much, it would seem. One sociologist complains that ‘the scholarship on ethnicity, race, and nationalism has become unsurveyably vast’; a leading intellectual historian deems it ‘intolerably protean’.footnote1 In his recent primer, the British sociologist Anthony Smith—who is, amongst other distinctions, the chief bibliographer of nationalism studies—describes an intellectual sprawl rather like Los Angeles: ‘These debates are diffuse and wide-ranging. They concern not only competing ideologies of nationalism nor even just the clash of particular theories. They involve radical disagreements over definitions of key terms, widely divergent histories of the nation and rival accounts of the “shape of things to come”.’ Amongst the currently warring camps, Smith distinguishes ‘primordialists’, ‘perennialists’, ‘neo-perennialists’, ‘instrumentalists’ and ‘modernists’. (He might have added ‘constructivists’, ‘neo-Weberians’ and ‘neo-Beardians’ as well.) He describes himself meanwhile as an ‘ethno-symbolist’, investigating nationalism as the modernization of pre-existing cultural identities.footnote2 In the face of so many categorical elisions, conflicting typologies and incongruent disciplinary perspectives, nationalism studies is seemingly embalmed in what Clifford Geertz called a ‘stultifying aura of conceptual ambiguity’.footnote3

Yet until recently most of the voices in this cacophony have shared three core and rarely challenged assumptions. The first is a ‘methodological nationalism’ that equates modern societies with nation-states and the state with the political nation. The second is the autonomy, or even primordiality, of nationalism as a historical force. (Political philosopher Erica Benner has lampooned this belief in a ‘unique set of national values, cherished by most of a nation’s members, which easily takes precedence over other values and interests whenever nationalists say they should.’)footnote4 The third presupposition is that liberal and reactionary nationalisms can be fundamentally distinguished. This dichotomy was given its most influential form by the Czech exile Hans Kohn in his monumental The Idea of Nationalism (1944), where he opposed ‘Western civic’ (political) versus ‘Eastern ethnic’ (cultural) nationalisms.footnote5

These cornerstone assumptions, along with the warring paradigms itemized by Smith, came under radical scrutiny from younger sociologists of Bourdieusian and neo-Weberian persuasions during the 1990s: a paradoxical decade defined both by the integration of formerly state-planned economies into the global market and an unexpected wave of extreme nationalism and civil war in what was once called the ‘Second World’. Whereas the previous generation of scholarship in the 1970s and 80s (Gellner, Anderson, Smith, Hobsbawm) had been primarily interested in the conditions and transformations that created modern nation-states, taking for granted their subsequent existences as ‘static, bounded, homogenous entities’, the new generation, confronted with the sudden emergence of catastrophic post-communist nationalisms in a supposedly ‘globalized’ world, has had a greater interest in ‘the dynamics of relatively rapid changes in degrees of ethnic, racial or national groupness’. The Kohnian dichotomy, in particular, began to seem irrelevant. Nationalisms, wrote Rogers Brubaker, are too ‘normatively and empirically unruly’ to be parsed ‘into types with clearly contrasting empirical and moral profiles’, especially when modifiers like ‘ethnic’ are equally abstract. Elsewhere he proposed that ‘groupness’, whether as ethnicity or nation, ‘is a variable, not a constant; it cannot be presupposed’. Therefore it was necessary to ‘decouple the study of nationhood and nationness from the study of nations as substantial entities, collectivities or communities.’footnote6

Brubaker has been one of the prime movers of this revolt against ‘substantialism’. In his seminal 1996 book Nationalism Reframed, which surveyed the resurgence of nationalism in the collapsing state systems of the ussr and Yugoslavia, he asked how complex, layered identities could suddenly be nullified by ‘the terrible categorical simplicity of ascribed nationality’. He rejected the ‘Sleeping Beauty’ thesis that the nations federated by Communism were simply waiting for a wake-up kiss from Western democracy. He proposed instead that theorists needed to abandon the search for the Holy Grail of the essential ‘nation’ and concentrate instead on the ‘processual dynamics of nationalism’:

Nationalism, moreover, ‘is not engendered by nations. [In Bourdieu’s terminology] it is produced—or better, it is induced—by political fields of particular kinds.’ ‘Reification’, he added, ‘is a social process, not only an intellectual practice. As such it is central to the phenomenon of nationalism.’footnote7