As the study of political networks becomes more common in political science, greater attention to questions of causality is warranted. This essay explores competing visions of causality in political networks. Independent essays address issues of statistical model specification, identification of multi-step personal influence, measurement error, causality in historical perspective, and the insights of field experiments. These essays do not agree entirely on the nature of causality in political networks, though they commonly take seriously concerns regarding homophily, time- consistency, and the uniqueness of political network data. Serious consideration of these methodological issues promises to enhance the value-added of network analysis in the study of politics.