The discussion about communalism has come to the centre stage in national discourse once again. The allegedly beef-related lynching in Dadri, followed by the protest by writers, has raised fears of a saffron - induced surged in communal incidents. The fact that communal riots are driven by political considerations is well - understood. However, there is a large number of bbooks about what increases the probability of communal riots. The book 'Votes and Violence' by Wilkinson was one such book that I happened to read. It is a painstakingly put together econometric analysis of communal incidents in India since independence. The results, while reinforcing certain widely-held beliefs, also discredits other notions that we have.
While researching for my article on Islam for The Indian Opinion, my focus was very much on the question of law and order. Wilkinson's book helped me think through many such issues. Some facts were validated - for example, that communal incidents saw a major spike in the 1990s (coincident with the rapid rise of the BJP). Wilkinson set up several different specifications of his basic model with the probability of riots as the dependent variable. In each such specification, whether there were elections in the next 6 months was a statistically significant independent variable. Using instrumental variables, Wilkinson also argued that economic competition between Hindus and Muslims does not impact communal incidents at all. Evidently, the premiss for a communal conflagration has to be conjured up from nothingness. Hence the freak incidents of the kind we saw in Dadri are either engineered or exploited to create communal tension.
What is interesting, however, is the role of caste - based politics in containing communal riots. When Lalu Prasad talks about 'mandal' (caste) versus 'kamandal' (communal) politics, he is (intentionally or unintentionally) making a very valid point. Communal riots are seen to be more likely in bi-polar states, where political consolidation between two parties reduces the importance of the muslim vote. Caste politics, by fragmenting the Hindu vote, makes the muslim voters more important at the margin, and therefore reduces the likelihood of communal incidents. Law and order is a state subject, but communal riots are no less likely to happen in Congress-ruled states than in other states. There are many other such findings that Wilkinson presents; I would therefore strongly recommend the book to anyone interested in this topic.
2 comments:
What exactly would you classify as a bi-polar state? Isn't caste prevalent among muslims in India as well? While currently they may vote as a single block in states such as Bihar what would stop a contest between castes among muslims when the saffron wave subsides?
The authors of the book use a particular definition which extends the Herfindahl index to a political setting, but basically a state like Gujarat or Madhya Pradesh where the top 2 parties (BJP, Congress) take up almost the entire vote. The second questions you pose is very interesting. My gut feeling is that given their lower socio - economic status, the amount of dispersion among the different castes in muslims will be lower, and unlike Hindus, security is a major concern that unites them. Hindu castes probably have fewer such unifying things.
Post a Comment