Decades of research suggest that social interaction influences opinion formation and affects voting behavior. However, recent work concerning the nexus between deliberation and democratic practice--particularly in the American context--has re-focused attention on the normative consequences of socially-driven political behavior. Among the most common criticisms of interpersonal networks are that most people have very insular social circles, and that when they do not they are unlikely to engage in politics. In this paper we provide evidence that such pessimistic assessments are unwarranted, though for somewhat unexpected reasons. Using data from the American Component of the 1992 Cross-National Election Project and the 2000 American National Election Study, we examine whether and under what conditions social networks facilitate interest-based voting. Our findings indicate that when networks provide unambiguous signals regarding candidates, that they serve as potentially useful information shortcuts, facilitating connections between individuals' vote decisions and their underlying preferences. And, because many Americans reside in reasonably supportive social environments, networks often help citizens make ``correct'' voting decisions (Lau and Redlawsk 1997). In the end, social networks appear to help shoulder the demands of democratic theory, but not by helping people learn about politics in any traditional sense.