At a time when our political theory was becoming puerile, two processes came together to fill the gap in our understanding. One was the old fashioned, empirical study of voting behaviour combined with political sociology. The second was the celebration of our democracy through the media, especially television. They combined to give a new lease of life and respectability to psephology. As election studies multiplied, we created one of the largest databanks of political behaviour. While that is a lasting contribution, we should not mistake statistical mastery for understanding of political processes. This process also elevated elections as the most important signifier of politics. In times of fractured political mandates, our answer was a deeply statistical one: anti-incumbency. It is a very arresting phrase; it captured our restlessness with ruling parties and individuals, while allowing us to graphically explain the outcome through ‘swing factors.’

The events of the last two weeks have brought the limitations of this framework sharply in focus. One was the response of the people to the terror strikes in Mumbai and the second was the recently concluded assembly elections. When people came out in the streets to express their grief, disbelief, anger and stoic resolve, neither they nor the media had a framework within which to understand it. Social movements and protest movements have become marginal, and we could only express our coming together in negative terms — as disenchantment with party politics and politicians. We did not construe our act as a political one, but that of turning away from politics. And yet, we went out and voted in large numbers. Clearly, we had a notion of the economic might of urban, middle-class India, understood the symbolic economy of its desires; but very little inkling of the language and form that its politics could assume.

... contd.