Abstract

Many political scientists believe that partisanship is an arbitrary psychological attachment that exerts a drug-like effect on voters' decisions. An implication is that voters don't care much about policy or government performance, and instead, elections are just a roll call of intoxicated partisans. I review and reassess the evidence for this view, concluding that there is no compelling evidence to support it. For many empirical tests, partisan intoxication and policy voting are observationally equivalent. Rare opportunities to partially distinguish between these possibilities like the southern realignment suggest that policy voting is more prevalent. When I conduct new tests utilizing survey experiments about hypothetical candidates, the weight of the evidence favors policy voting.