Since the fall of the Berlin Wall, one of the most powerful stories recounted in the media in relation to national identities has been that of the Cold War as Deep Freeze. Under the malevolent glaciers of totalitarianism, so it is said, lay ancient ethnic identities and ethnic hatreds, perfectly preserved. As soon as the glaciers began to melt under the glare of glasnost, perestroika, and Solidarnosc, the sleepers awoke, in a scene worthy of (depending on your point of view) a gothic romance, or a horror film.

Historians and social scientists should be extremely grateful for the popularity of this story, for over the last fifteen years, a large number of them have made a living by debunking it—indeed, without it, there might not be a New York conference, or a CRN. The story, they explain, is "essentialist" and "primordialist." It substitutes Destiny for History. It disregards the radical discontinuities between older and newer forms of identification, flattens out the complex distinctions between official and vernacular forms of identification, and runs roughshod over issues of membership, geography, migration and so forth. Above all, it fails to address the fact that "identity" is not a thing, a property, a birthright, but rather a set of constantly shifting and contested narratives which we tell and retell, even as they tell and retell us. Ron Suny has been one of the most lucid, prolific and persuasive exponents of this point of view, and he gives an excellent example in his essay "Constructing Primordialism: Old Histories for New Nations," which appeared in the Journal of Modern History in 2001 and is available on the CRN website (http://www.dartmouth.edu/~crn/crn_papers/Suny3.pdf_).

As one might expect from anything connected with the Cold War, these debates and ideas have not remained obediently within the territories of NATO and the former Warsaw Pact, but have greedily gobbled up scholarship on areas throughout the world. Reinforcing, even while modulating, the influential "modernist" perspectives of Ernst Gellner and Benedict Anderson (among others), the "anti-deep freeze" constructionist school has powerfully influenced our understanding of areas that were never in danger of any sort of deep freeze, such as Martinique and Guadeloupe, and others, such as Québec, that faced it only in strictly meteorological form.

My own work and my own scholarly inclinations have long been close to the constructionists (ever since Peter Sahlins told me, in my first year in graduate school sixteen years ago, that I needed to drop everything and read Benedict Anderson). My recent book The Cult of the Nation in France took for granted the modern, constructed nature of nations—indeed, one of its points was that the existence of nation-building projects is an implicit acknolwedgment of this "constructed nature" by nationalist movements themselves. My principal purpose was not to debunk myths of French national identity, but to explore the extraordinary historical moment when having a national identity started to be seen as indispensable to a person's existence, and became the focus of unprecedented political efforts and ambitions. The essay I am currently writing for the CRN on France and Canada stresses the importance of religious boundaries, and religious forms of official and vernacular identification—as opposed to national ones—for French settlers in North America in the seventeenth and eighteenth centuries. I would argue that Québecois nationalism only took shape as a serious political force with the "Quiet Revolution" of the 1950's and 1960's, and the decline of Québecois Catholicism.

Yet despite this intellectual pedigree, I have been coming to have some doubts, deriving mostly from my work on the French Revolution, about the paths that the constructionist analysis of identities has taken. In particular, I wonder if there has not been too strong an emphasis on slow and gradual processes—on the longue durée—and not enough on the role of transformative and traumatic events. This choice has been understandable. On the one hand, it reflects a long-standing fondness of anthropologists for investigating supposedly static or slowly-evolving cultural systems; on the other, the French Annales school's privileging of slow, deep currents of geological and economic change over the random foam-spatter of "event history." Practical considerations have incidentally reinforced both influences: anthropologists conduct fieldwork more easily in stable, peaceful surroundings than in violently dislocated and dangerous ones; historians like nothing more than long rows of "serial" sources with which to chart slow change over time—marine navigators afloat on the sea of the archives. And the idea of sudden, violent shifts in outlook and identification are uncomfortably reminiscent both of the "deep freeze" school's image of the ice cracking open, and of the sort of religious conversion experiences with which very few modern Western academics have any intimate famliarity. And thus, our penchant for investigating how habits and practices slowly change, in complex interaction with the growth of states, the decline of churches, the evolution of commerce and production, longterm shifts in movement, migration and communication. A book like Peter Sahlins's shows just how fruitful this approach can be, and so does Ron Suny’s reflections on the paradoxical origins of post-Soviet nationalisms in the belly of the beast—in the "national" policies of Joseph Stalin and his successors. In both cases, the focus is largely on slow, barnicle-like processes of accretion. And so a crucial question comes to be the one that Peter Sahlins formulated for this meeting: given that these forms of identification derive in many cases from highly instrumental relations with states, from the "use and abuse of states" in the service of self-interest, how do they come to be internalized? How do the barnicles come to "stick"?

It is an excellent question. And yet the road to Tarsus exists, and many have travelled it, and for them the question is not one of slow stickiness, but of sudden, jarring transformation. Forms of identification can take shape and change with vertiginous speed.

An example is the French Revolution. Several years ago, as I read through French texts discussing national identity and national character in the eighteenth century, I spied a curious pattern. Quite suddenly, in the years 1787 and 1788, a large number of French authors began insisting that the French nation did not exist. France amounted, in the representative words of the Comte de Mirabeau, to nothing more than an "an unconstituted aggregate of disunited peoples." These peoples lacked common laws, common sentiments, even a common language. Before 1789, no one thought that this sort of diversity posed a problem, either for states or for "nations," as the word was then understood. But suddenly, French authors decided for France really to be a nation, it needed a much greater degree of what we would call political and cultural unity and uniformity. Within a number of years, Revolutionary leaders devised project after project for bringing this unity and uniformity into being: they proposed new law codes, the destruction of local dialects, changes in costume, even cuisine. Robespierre, a good student of the Jesuits and mindful of their pedagogy, advocated removing all children from their parents at age five and subjecting them to compulsory state boarding schools so as to inculcate proper republican attitudes and iron out significant cultural differences among French citizens. Politically, the Revolutionaries now identified not with a kingdom, or a province, but with an idealized, as yet unachieved French Republic and French nation, one and indivisible.

This radical shift in identifications certainly had deep roots in eighteenth-century French political culture, and one of the aims of my book was to trace these roots as far down into the soil as possible. But the shift itself did not occur in a slow and gradual fashion. It occurred in a setting of rapid dislocation and violence, of trauma and extreme emotion. Of course, before the revolutionary Terror of 1793-4, well-off writers of pamphlets did not live in conditions of daily physical violence or privation. Yet they had much to fear nonetheless: notably, economic collapse produced by the bankruptcy of the state, and the forcible imposition of royal "despotism." Both dangers already seemed in motion in the threatening summer of 1788. And once the Revolution itself began, these fears were joined by gnawing anxieties about mob violence, about counter-revolutionary conspiracies coalescing in the darkness, and also by a sense of almost indescribable exhiliration and vertigo as one pillar of the regime after another suddenly cracked, crumbled and vanished, starting with the great fortress and prison of the Bastille on July 14, 1789. It is notable that the participants frequently described the experience of Revolution as one of a compression of time—of days being the equivalent of normal years. For Robespierre, the Revolution had catapulted the French two thousand years ahead of the rest of the human race, &so that one is tempted to see them as a different species.&

William Sewell has taken the fall of the Bastille as the subject of a remarkably useful article that theorizes &events,& and urges social scientists to consider them more carefully ("Historical Events as Transformations of Structures: Inventing Revolution at the Bastille," Theory and Society, vol. 25, 1996, pp. 841-881). Sewell draws attention to the way that certain sorts of crises can dislocate existing structures and practices, producing "a deep sense of insecurity" and raising "the emotional intensity of life." He discusses the way that participants in these crises are pushed to act collectively in unusually creative ways, so that they redefine many of their most fundamental conceptions about the way human beings act—for instance, what earlier periods would have immediately categorized as a criminal mob action was now redefined as the collective rising of "the people." He explores the way in which actors spontaneously create new rituals, and develop new historical narratives to make sense of the events they are experiencing. Sewell is interested primarily in the transformation of social and cultural structures, but his work also has implications for our understanding of processes of identification. Out of the taking of the Bastille, he argues, emerged for the first time the modern concept of "revolution" itself, and of course in France after 1789, one of the most powerful forms of identification was not with the kingdom, or even with the "nation" or "patrie," but with precisely with "revolution." As Keith Michael Baker has keenly pointed out, only after 1789 did the word "revolutionary" come to be used as a noun, and as a proud self-definition. I might add the obvious point that once a form of identification gains a certain degree of legitimacy and influence in this way, banal human opportunism ensures that many more people will rush to practice it. Royal office-holders will become fervent revolutionaries; French-born Canadian fur-trappers will discover a sudden admiration for the British Navy; hardened Communist apparatchiks will reveal their heretofore-well-hidden dreams of Ukranian independence;