This paper is like one of those films in which the final dénouement clarifies all that went before. In his conclusion, Professor Binder announces nine theses of his own about ethnicity, nationalism, and ethnic conflict. Most of these were foreshadowed in the preceding pages, but often so subtly that their sharp statement in the conclusion comes as a bit of a revelation.
I find myself in substantial agreement with the nine general theses. I am particularly struck by the ninth, which argues that nationalism and ethnic mobilization are often driven by interstate interactions and foreign policy. Here and in few places earlier in the essay, Binder suggests that nationalism and ethnic mobilization cannot be properly understood as a matter of domestic politics. They are instead a product of interstate politics, both in the obvious example of third-party support for nationalist movements, and in the broader sense that the state system creates incentives for nationalist mobilization by and within states. I think this is correct and that the “international” sources of ethnic and nationalist politics are greatly understudied.1 The positive theses stated in the conclusion emerge indirectly out of extended critiques of five works on the subject of ethnicity, ethnic conflict, and nationalism. This makes it difficult to comment directly on the merits or promise of these claims. And while I have minor objections here and there to Binder’s several critiques, the objections are disconnected. So instead I will sketch a more general argument about the way that Binder poses “the question” about ethnicity and ethnic conflict.
What is the central question addressed by Binder’s nine theses and by the works he examines? Of course there are multiple questions, but arguably there is one that Binder poses at the outset, returns to repeatedly, and focuses on in four of the five works examined. Namely, what explains people’s attachment to and motivation to act on behalf of ethnic or national groups? At the extreme, what explains the willingness to kill and die for these groups? The question is especially puzzling if we accept, as Binder does, that ethnic groups are social constructions (i.e., conventions) rather than biological imperatives.
Binder appears to assume that an answer to this question is the sine qua non or starting point for any theory of ethnic and nationalist conflict. That is, to have a good theory of ethnic politics and conflict, we need to know what motivates people to act on behalf of an ethnic group, or how they can be so motivated. Let’s call this “the motivation question.”
It is fair to say that the works Binder discusses by Donald Horowitz, Benedict Anderson, and Russell Hardin also take this as a (or the) central question that must be answered to have a decent theory of nationalism or ethnic conflict. Horowitz (1985) argues at length for one account of motivations against another, while Anderson and Hardin frame their inquiries explicitly in terms of the motivation question.
By contrast, I think that it is not true that we need a sharp and definitive answer to the motivation question to have a theory of ethnic mobilization and conflict. Further, answers to the motivation question are not likely to be very helpful for answering questions about ethnic conflict that are of tremendous normative and social science interest. For instance, why is politics more “ethnified” in some countries (or at some times) than in others? Or why has such-and-such country or ethnic group seen much ethnic violence whereas such-and-such others have not? And why at this time but not some other time?
Let’s group questions of this latter sort under the rubric “the variation question.” My claim is that general answers to the puzzle of what motivates people to identify with and kill for ethnic groups will not be much help for explaining real-world variation in events involving nationalism and ethnic conflict. For these purposes, the grand debates about primordialism versus instrumentalism versus constructivism, or Rational Choice versus Psychology versus Discourse, are largely irrelevant.
As evidence, consider that none of Binder’s nine theses, or anywhere in the discussion of Horowitz, Anderson, or Hardin, is there a clear statement of an empirically testable proposition that follows from an argument about the motivations behind ethnicity and nationalism. The analysis and discussion do not produce factors that follow from a general theoretical answer to the motivation question, and that a researcher could identify and measure to see if their presence distinguishes between cases with and without ethnic conflict or nationalist mobilization. I am not saying that there are no testable hypotheses in any of the books at issue, but that if there are I do not think that they follow in an immediate way from an answer to the motivation question. To give an example from Binder’s own analysis, Binder comes close to a testable hypothesis when he argues that “ethnic contracts” arise as bargains to regulate competition and cooperation between ethnic groups in a society, and that “When conditions change, leading to a demand for revision of an existing ethnic contract, ethnic conflict breaks out in earnest” (p. 47). It is not clear that ethnic contracts are observable, or what conditions would lead to demands for a revision, or why these demands would necessarily produce conflict rather than peaceful renegotiation. And it is not at all clear how an answer to the motivation question – for instance, ethnicity is a “rational choice” or a deep-rooted cultural commitment -- would allow us to turn this interesting suggestion into a claim with empirical implications.
Why isn’t an answer to the motivation question a necessary point of departure for a theory of nationalism and ethnic conflict? In the first place, having a motivation or identification does not imply that one will act on behalf of it. We have reasons or motivations to do all manner of things, but we only do a few of them. Further, many different motivations can result in the same action, such as ethnic killing. So while it may be true that individual identification with an ethnic group or nation is a necessary condition (in a conceptual sense) for what we call “ethnic conflict,” it is not remotely close to sufficient. To explain real-world variation in levels of ethnic mobilization or violence, we will have to do much more than invoke motivations and identifications.2 Second, when we look at individual motivations closely, we often find them to be so complex or obscure as to be inscrutable. They are the proper domain of talented novelists, not social scientists, who may be better off trying to trace out macro-level implications of reductive, somewhat unrealistic formulations of motivations (among other things). Paul Brass (1999) nicely makes a related point in his study of Hindu-Muslim riots in India. He carefully investigated incidents labelled “communal” after the fact by politicians and the press, and found that the true motivations of those involved were hopelessly ambiguous or indiscernible. He argues on this basis that “ethnic violence” is an ex post political attribution rather than a property of social events. I think this may go too far, but it is nonetheless a valuable argument that highlights a problem with putting the motivation question at the center of our inquiries.
If we want to understand the sources of variation in ethnic mobilization and conflict, then the main question addressed in Binder’s essay and the works it considers – what motivates people to identify with an ethnic group, or what is the basis of ethnic solidarity? – may not be very important. But this is not to say that the question itself is uninteresting or unimportant. Even if ethnic and nationalist solidarities are a background condition behind tremendous variation in specific events, it is still interesting to ask where this background condition came from, given that the whole structure of the modern international order is built on top of it. The world is divided into would-be nation states, political authority is legitimated by the claim to represent a nation, and the normative prototype for a “nation” is (arguably) an ethnic group. How did these ideas take over the modern world so successfully, in the sense that they do indeed seem natural to billions of people who identify strongly with their nation and/or ethnic group? And is it a good thing that they have? Are these motivations good or bad?
These are certainly important questions, and Binder’s analysis of their treatment in the works he reviews is fascinating. I conclude by mentioning a simple and important answer that I think is missing from Binder’s discussion, though the idea does appear in a derivative form in Anderson’s Imagined Communities. The source is Karl Deutsch (1958), and more elegantly, Ernest Gellner (1983). The argument goes like this: The reason why ethnic and national attachments have been so successful as coordinating devices in the last 200 years, but not so much earlier, is that with economic modernization our life chances come to depend on our cultural heritage in a way that was not the case in the premodern era. Economic modernization creates the possibility of social and economic mobility, but makes upward mobility conditional on sharing the culture (language, in the most straightforward case) of those at the top. Thus, when there are cultural barriers to upward mobility that are linked to ascriptive differences that people cannot easily change, there are incentives for the development of nationalist movements. The main reason that people have come to care so deeply about their culture and the culture of their rulers, in this account, is that these now affect their and their children’s life chances within the parameters of the modern state.3 Surely this is not the whole story, but it is a good starting point with interesting implications for some of the questions raised above. (For instance, the story rejects the primordialist position that nationalism is powerful because it is an ahistorical biological or psychological fact, but at the same time gives a reason to see nationalism and ethnic attachments as morally defensible, at least in many circumstances.) This general argument about a secular change that has produced popular nationalist and ethnic motivations does not, however, help us much with variation questions, such as why ethnic conflict now-but-not-then or here-but-not-there. For these we need much more than general claims and arguments about motivations.
Brass, Paul. 1997. Theft of an Idol. Princeton: Princeton University Press.
Breully, John. 1994. Nationalism and the State. 2nd ed. Chicago: University of Chicago Press.
Deutsch, Karl. 1966. Nationalism and Social Communication. Cambridge: MIT Press.
Fearon, James D. and David D. Laitin. 2002. Ethnicity, Insurgency, and Civil War. Forthcoming in American Political Science Review.
Gellner, Ernest. 1983. Nations and Nationalism. Ithaca: Cornell University Press.
1 A major exception, not noted by Binder, is John Breuilly (1994).
2 Fearon and Laitin (2002) find that for the period 1945 to 1999, cross-national variation in the incidence of civil war is much better explained by variation in the conditions that favor rural guerrilla warfare than in measures of the presence or absence of ethnic or religious antagonisms or discrimination. We argue that individual motivations to rebel (often for good reason) are too common to explain the relatively rare outbreaks of major civil violence.
3 In Imagined Communities, the Deutsch/Gellner argument appears in what Anderson terms “blocked pilgrimages,” barriers to upward mobility in an administrative system due to an ascriptive difference. Anderson’s “creole pioneers” in the New World develop nationalism when the accident of their birth outside of Spain prevents them from rising within the imperial bureaucracy. Like Deutsch and Gellner, Anderson argues that strong separatist nationalist movements developed in the Habsburg empire but not Scotland because the Scottish bureaucratic pilgrims could rise, while the greater language and cultural differences prevented this for non-Germans under the Habsburgs.