In this paper, we identify a problem—the problem of which actors should serve as defenders of democracy—and propose a solution to that problem—the creation of randomly selected citizen bodies, or allotted chambers (hereafter ACs). 1 Having in place institutions that are tasked with democratic self-defense, is, we argue, a critically important pillar of democratic government, but its importance has often been neglected. This neglect is exacerbated by the evasive nature of the task that these democratic defense institutions are called to perform. Part of the problem is that the task of democratic self-defense is often mistakenly conceived as an ad hoc response to an occasional problem, rather than a routine task to which democracies should devote regular attention. Once the task of democratic self-defense is properly specified, the advantages of assigning this task to ACs, rather than courts or legislatures, become evident. Section 2 of this paper explores the task of democratic self-defense in more detail. Section 3 offers reasons for questioning the assignment of this task to existing institutions within democracies, notably legislatures or courts. Section 4 lays out the contributions that sortition—the random assignment of public responsibilities—can make to democratic self-defense. In Section 5, we sketch a proposal for an AC that could be tasked with democratic self-defense. Such a body would avoid the problems identified in Section 3 with assigning this task to either legislatures or courts. We offer three versions of this proposal—a weak, a moderate, and a strong version—and provide a tentative endorsement for the moderate model. The paper concludes by noting that the current crisis facing democracy has both generated an opening for institutional innovation and increased awareness of the necessity for democratic self-defense. It is important, we argue, that democracies make the most of this opening; doing so, we add, requires a clear understanding of the problem and a well thought-out vision of the solution.