Abstract

Every ten years United States congressional districts are drawn, physically constructing political representation based on domicile. Why do we do it this way? Is territorial representation consistent with the broader normative ends of political representation). ;In section one I argue that territorial constituencies were never intended to represent local "communities of interest." Instead, physical proximity between voters was necessary to achieve the normative aims of representative government in a large nation. I begin in 13 th century England, and proceed through the political theory of Hobbes, Locke, Montesquieu, Burke, the Federalists and their critics. I also provide a novel interpretation of Federalist 10. ;In sum, territory was a habit of mind for the founders of the United States, and thus there were no explicit reasons given for using territory. The silence of territory as a definer of electoral constituencies has continued to the present day. I argue that the continued silence about territorial constituencies constitutes a substantial breech of legitimacy. ;In the second section , I argue that "constituency definition" is a normatively rich institution and how constituencies are defined ought not to be determined by interested third parties. I argue that Hanna Pitkin incorrectly conflated "political representation" and "legitimacy." I argue that citizen consent to constituency definition is a central part of legitimating any representative government. Finally, I argue that there are no persuasive normative arguments for territorial constituencies today. Justifying territorial constituencies on the normative claim that "communities of interest" ought to define constituencies for political representation is simply unpersuasive. If one believes such particularity ought to be represented in one branch of the national legislature, then other kinds of particularity---race, ideology, class, et al---simply dominate. ;In section three I argue against "particularity" for electoral constituencies. I conclude with a thought experiment: randomly assign voters into 435 national, electoral constituencies for life. Having established that the concept of "constituency" is itself normatively robust and separable from "electoral systems," I argue that important considerations of "voice," can be introduced through novel means of reform