Political talk, whether in elections campaigns, legislative debates, public meetings, or elsewhere, is vital to the political process. Yet very little is understood about the circumstances under which talk affects political outcomes. This research will develop game theoretic models which explore the role of debate and explanation in political decision making. Specifically, the project will examine talk in two circumstances: within legislatures and committees as members debate policy choices, and during elections as legislators attempt to explain and defend their actions to constituents. The principal analytic tool used will be the theory of games with incomplete information -- specifically games involving costless signals or communication. This is because political talk is frequently cheap. It is easy for individuals to make claims that are not true and are know to be untrue. This feature generates a fundamental credibility problem that lies at the heart of cheap talk models. When completed this project will help us better understand the impact of political communication on political outcomes.