After Empire Must Come Nation?
During the last two decades a revisionist wave has gripped the historiography about imperial breakups. A venerable topic at least since World War I, generations of historians strove to explain why multinational empires crumbled to give way to nation-states. The entire scholarly field of nationalism studies was in good part a branch of this gigantic question, which the Yugoslav Wars once more thrust on historians’ minds. Yet with the memory of these wars fading, the tables have turned. The tide of revisionism has been of such vast proportions as to become the new mainstream. In the academic field of global and imperial history today, hardly anyone argues that empires were doomed to be replaced by nation-states. It was all more complex and, above all, contingent, we learn instead. As I have argued in my last book, it is time to challenge this new revisionist mainstream.
As befits a broad movement, revisionist arguments against the inevitable rise of nation-states have come in many forms and from many empires’ historiographies. In his 2006 book, Pieter Judson formulated a Habsburg version of the trend, arguing that nationality ranked low in ordinary people’s priorities at the fringes of late imperial Austria. In a Latin American variant, the Argentine historian José Carlos Chiaramonte refuted Benedict Anderson’s famous thesis that “Creole pioneers” midwifed modern nationalism in early-nineteenth-century Spanish America. These “Creoles,” Chiaramonte maintained, evoked an “American” and a local identity, but not a Mexican or an Argentine one, so that “what lacks is exactly the nationalism corresponding to the nations that would rise later.”
In recent years, historians of the French Empire have also pushed back against the inexorability of imperial decline. In his book on the case of Algeria, Todd Shepard turned against seeing decolonization as a “predetermined end point,” claiming that no one important in either France or Algeria envisaged a new independent state before 1961. Frederick Cooper’s works have become a standard reference for revisionist historians holding that imperial breakup and the rise of new nation-states in Asia and Africa were not foreordained, not foreseeable for contemporaries, and above all not wanted by key protagonists such as Léopold Sédar Senghor — until very shortly before the new states came into being. Nationalism was much weaker a political force than hitherto assumed, according to this argument. Authors such as Cooper, Shepard, and Judson thus all argue against the “teleology of the nation-state” and in favor of the “contingency” detectable in individual cases of imperial breakup.
Although all these historians excel at nuanced interpretations of their historical protagonists’ mindsets, the gist of the argument loses cogency the further we step back from individual examples. As Samuel Moyn has written, a major pitfall of the revisionist vision is that it works best when narrowing and isolating the sample, but does a very poor job at offering explanations of why the nation-state won out — not only once, but in a well-known global series of waves that David Armitage has fittingly described in the epidemic vocabulary of “contagion.” Nationalism was indeed comparatively weak and late in French West Africa and Algeria, but by the crucial years of decolonization explored by Shepard and Cooper, other components of the French Empire had long given way to independent nation-states like Vietnam or Tunisia. Since critics of colonialism from different parts of the empire spoke to each other, by 1960 Senegalese and Algerians had plenty of national models to ponder. Once we abandon the methodological nationalism of restricting our vision to eventually national units such as “Algeria,” imperial breakup and the spread of nation-states begin to look much more like a recurrent and interconnected pattern. From such a global angle, it is unconvincing to extrapolate from the “contingency” of the individual examples that revisionists so cherish.
Another problem lies in the frequent slippage between normative and empirical observations about nationalism. Revisionism has been strongest in the global academic powerhouses of North America and Western Europe, where scholars are wary of nationalism in their politics, reluctant to concede the “real” existence of nations, and keen on overcoming methodological nationalism in their professional practice. Although I sympathize with each of these three motivations, amalgamating them leads to confusing analyses. Many revisionists mix criticism of nationalist visions of homogeneity with disappointment of the track records of post-imperial states; and then season the blend with assertions that nations are “only” imagined and with warnings against the methodological nationalism of seeing the nation-state as the natural endpoint of history; or more modestly of empire. In one of his recent books, Frederick Cooper argues simultaneously that nationalism and the nation-state were inconsequential in the grand scheme of things and that the pair wrought havoc on our world. It seems to me that the first assertion weakens the second. At the heart of much of the revisionist argument is an ambivalence whether nationalism and nation-states were not or should not have been as important as earlier historians held.
Then there are conceptual issues about nationalism, which revisionists have tended to reduce to an artificially narrow category, gauged from a confusing mix of two criteria. The first benchmark someone has to pass to count as a nationalist is whether she or he publicly demanded independent statehood for their “nation.” This corresponds with Ernest Gellner’s definition of nationalism as “a political principle that holds that the political and the national unit should be congruent.” Hence, if you speak a great deal of the “soul” of the African or Wolof “nation,” yet do not propose a single African or Wolof state, you are not a nationalist.
This looks like a fair point at first, but it entails two problems. The first, perhaps minor, one is that the criterion sits uncomfortably with historical usage of the term “nationalist.” As I narrate in my book, in the 1920s “nationalism” became something like a residual category for non-communist critics of empire, many of whom did not have independence foremost on their minds. Conversely, it was the internationalist communists who clamored for national independence most straightforwardly and consistently. For instance, in 1926 the French Communist Party (PCF) nudged Algerian anticolonialists to call for the independence of their “North African nation,” while the activists themselves zigzagged over the matter, citing concerns that the term “independence” would alienate the “bourgeois nationalists” in their camp. In similarly paradoxical usage, a French intelligence report of 1921 labeled a Vietnamese spokesman a “nationalist” because, unlike his communist compatriots, he “believe[d] in the current necessity of maintaining our protectorate over Annam.”
The second, weightier, problem is that the post-Westphalian world has envisaged national (as opposed to imperial) statehood as stretching over a contiguous territory, making it a poor match for diasporic nations. The problem applies to pan-Africanists, for instance, who naturally struggled to trim their nation into a coherent territory with a state. Yet, is this reason enough to withhold the “nationalism” label from someone like Kwame Nkrumah, who also stood for the independence of the Ghanaian nation? The underlying predicament here is that the political unit never matches the national unit, which is why nationalists continue to bicker over adjustments to either or both of the two. So, if we elevate congruency between the national and the political unit into a criterion for applying the label “nationalism,” we will end up defining nationalism out of existence.
Yet this is precisely the second, more implicit, benchmark of the revisionists’ definition of nationalism: Only those who eventually got their nation-states in the form they had imagined all along can retrospectively count as “nationalists.” According to Chiaramonte, Spanish American Creoles were no nationalists because the entities they dreamed of did not achieve statehood. Recent historiography about decolonization in the French Empire has proceeded analogously with “black nationalism” or “Muslim nationalism” — both categories commonly used in the 1920s, but not crowned by nation-states. If taken to its logical consequence, this line of reasoning implies that there is no such thing as Kurdish or Basque nationalism. Although revisionists usually admonish the teleology of hindsight, their definition of nationalism is thus ultimately determined by outcomes that are visible in retrospect, artificially taking the day of writing as the cutoff date for identifying nations and their corresponding nationalisms. With this move, the revisionist wave about imperial breakup ended up misconstruing the malleable and constructed nature of nation-states that its proponents rightly and courageously set out to reveal in the first place.
