74 SOCIAL SCIENTIST how far Indian polity would go along with the Western democracies and how far it could do away with the widening influence of communism. The ^independent" Indian psephologists too emulated American models o.f voting behaviour studies. Almond, Dahl, Deu-tsch, Huntington, Key, Lane, La Palmbora, Lipset, Powel, Pye, Riesman and Weiner became their mentors. In the beginning, however, they were very few, and their accounts were sketchy, journalistic, and descriptive with the conspicuous absence of any in-depth analysis. Since the third general election in 1962, the interest in election and voting behaviour studies has become widespread, and the works have multiplied with the generous financial assistance of the Indian Council of Social Science Research (ICSSR) and the University Grants Commission (UGC). As a result, today we have more than 200 such studies. The studies, however, are too limited in scope to shed light on the crucial issues involved in the political processes in India. It is a small wonder then that there is no theory of electoral behaviour in India. Irrelevant Methodology Following the techniques and tools of American scholars Indian studies more often than not used survey method without ever questioning its adequency in the Indian context. A very important question in method is what the units are and where to begin. The survey research by its very nature has the tendency to treat the individual as the unit of analysis. This extreme individualism presumed that an individual voter was the free decision-maker in casting his vote. Thereby, it totally ignored the tremendous variations in influence on the voters in the form of superimposed consensus of groups. It also tacitly assumed that there was a high degree of correspondence between thought, expression and behavioural action. Notwithstanding its alien nature in the Indian political context the method has been eulogized. Disparate questions are put together, and correlation between the background variables of voters and the voting is made to track the determinants of the electoral behaviour without an adequate reference to societal processes and power structures. It is therefore natural that the methodology adopted in the election studies senselessly enumerated the superfluous data leading to sterile inductive inferences, and miserably failed to tap the broader context. The psephologists could get away with such shoddy work in the name of *'oases of excel--lence" amidst mediocrity and torpor.