Shifts of meaning within explanations of voting and party competition

Rational choice models are increasingly used in explanations of voting and elections. Their adoption is often urged, rather uncritically, on the grounds that they provide a unified and exact explanation across political science. Far from doing so, however, their definitions of 'rationality' may differ radically even within a limited area, such as party and voting behaviour. Shifts of meaning from one model to another may be obscured by mathematical formalism, which concentrates attention on the derivation of conclusions from basic assumptions, rather than on the practical relevance of their relationship to the assumptions behind other models. A mathematical formulation may not, however, be the most relevant for purposes of electoral explanation. Rigorous verbal formulations can represent electors' and politicians' decision procedures more realistically, while remaining generally accessible and supporting detailed inter-comparisons of their working assumptions. Through these, rational choice explanations can be brought closer together and decisions between them made with a clear knowledge of what is involved, thus providing a better basis for cumulative research. The process of theoretical evaluation and assimilation is illustrated here with two original theories of voting and party competition.