In the post-9/11 world where the “clash of civilizations” has moved beyond classroom debates to the public realm, it is refreshing, and challenging, to see a study that does not give ethnicity an easy ride. The title of this book is slightly misleading, though, because even while it concedes that appeals for political support on the basis of ethnic categories based on “race, caste, tribe or religion” (p. 2) are frequently made, sometimes with considerable success, it asserts that such tactics do not always succeed. When they do, it is not necessarily because of their putative appeal to sentiments but, instead, because both ethnic candidates and their supporters, rather than being swayed by appeals to their nonrational selves, are actually driven by sophisticated calculations of expected gain. Their utility calculus takes the size of the ultimate prize as well as the probability of winning it into account when they choose to align themselves with one set of politicians as opposed to another. Kanchan Chandra's parsimonious and general model explains why ethic parties in India, riding on Hindu or, for that matter, Tamil nationalism, succeed in some contexts but not in others.

Chandra's core argument asserts that “ethnic parties are most likely to succeed in patronage-democracies when they have competitive rules for intra-party advancement and the ethnic group they seek to mobilize is larger than the threshold of winning or leverage imposed by the electoral system.” But since the result of ethnic appeals is contingent on “organizational, demographic and institutional variables,” the mere appeal to ethnicity is no guarantee of success (p. 15). The author derives these conjectures deductively from revealed preferences of politicians and their perception of expected gain and subjective probabilities, and tests their implications on the performance of several Indian parties that fit the definition. Her data are drawn from quantitative sources, such as the Indian census, electoral statistics, an attitudinal survey of the electorate, elite interviews, and content analysis of political discourse. The analytical capacity of this wealth of information gleaned from multiple sources is considerably enhanced through an innovative application of Gary King's ecological inference (EI) method to estimate partisan voting at the constituency level, to facilitate meticulous, empirically rigorous tests of precisely specified empirical conjectures.

The main contribution of this lucid, elegant, and rigorous study, based on meticulous fieldwork and an innovative take on new techniques, is a shot in the arm for the beleaguered practitioners of statecraft, trying to rein in ethnic politics through institutional design, legislation, allocation of office, or a touch of the old-fashion pork barrel. In an era when advocates of markets, social capital, and “unlimited access to Divinity afterwards” are doing their best to limit the scope of political analysis, Chandra achieves exactly the opposite effect by bringing the lost territory of “ethnic behavior” back into the purview of comparative politics.

The author's use of her model for cross-regional comparison in India is impressive. She explains the performance of the Bharatiya Janata Party (BJP) in the parliamentary elections of 1991 in Gujarat (its best nationwide performance, with 56% of Hindu votes, according to her estimates, using King's EI technique, p. 268), with reference to the party's successful induction of the representatives of Backward Castes into its fold. It was a masterly move by a party with a reputation for its identification with values of the Hindu upper castes. Based on her model, one finds in this case a convergence of interests between the guests, namely, the lower-status entrants to middle-level leadership status within the Hindu nationalist BJP, and the politically entrenched, upper-caste hosts, both of whom stood to gain from the electoral success of the party. To this successful demonstration of the heuristic power of her model, those familiar with the otherness of India might retort that this is so much old knowledge in a new jargon. If, after all, caste matters in Indian politics, then what is new? I shall defend Chandra's analysis here, not because her finding regarding the BJP strategy in Gujarat is new but because of the method she uses to get there. She reverses the causality: Caste matters because politics matters, not the other way round. To make this point, she contrasts Gujarat with the BJP's comparatively poor performance in the eastern state of Bihar, where the same strategy was also tried by the leadership but with considerably less success. There, the elites of the backward castes had a better choice in terms of the Janata Dal, led by their own community, much better placed to win the election, and where, in consequence, the likely pickings were much richer.

The specific empirical puzzle that Why Ethnic Parties Succeed has set out to solve, and the repertoire of tools and data that it brings to bear, complement one another. But in the course of its taking shape, the text has acquired new ambitions that set different standards against which to evaluate its significance. When one holds the findings of this book to account for its professed goal to propose “a theory of ethnic party performance in ‘patronage democracies’” (p. xvii), the empirical base appears too narrow to sustain so broad a claim. The scope of generalization from Chandra's Indian data is bounded by four restrictive conditions. First and foremost, the model appears underspecified to sustain a cross-national generalization, for it does not take into account the role of the state as an explanatory variable. India, a “responsive state” with her panoply of quotas, legislations, and judicial doctrines facilitates the functioning of ethnic parties in a manner that is not often the case (in comparison, for example, with Sri Lanka). Secondly, even within India, subnational movements and ethnic parties in Kashmir and Assam have had a different itinerary from the examples used by Chandra, which goes to show how credible generalization would require additional geopolitical variables. Thirdly, the peculiar flexibility of the inner structure of Hinduism, where categories tend to be inclusive, suits the author's model rather well, better than, for example, the Shia–Sunni schism in Pakistan or Catholic–Protestant differences might. Finally, though early in the text she attributes the appeal of ethnicity to both material and psychological rewards, the latter disappears from the analysis fairly soon (one can understand why, but that is precisely the point), and the early phases of cultural nationalist movements before they become parties are lost from the story line altogether.

As all those who analyze or merely dabble in party politics must know, many volunteers give their time and efforts to politics, not because they expect tangible rewards but because it makes them feel better about themselves. That is also true of many supporters of ethnic parties. The concept of values, in many ways reminiscent of the politics of evolutionism and idealism, is seldom present on the agenda of contemporary political analysts. This is particularly true for those of the quantitative persuasion. Not taking the intrinsic appeal of ethnicity completely on board makes political analysis vulnerable to the perils of a desiccated, disenchanting modernity and, in the process, to the loss of the empowering role of ethnicity as agency from the scope of analysis altogether. By implicitly raising this issue, perhaps unwittingly, Chandra makes a useful contribution to the scholarly agenda of the relationship of ethnicity and democracy.

Despite the minor caveats, I strongly endorse Chandra's elegant and lucid analysis of India's ethnic political parties. The effective and effortless confluence of area studies and middle-range political theory is one of the finest I have seen in recent years.