Document Type


Publication Date



voter fraud, election law, constitutional law


In recent months, debates over voter fraud have consumed state legislatures and blogs, courts and election commissions. The prevailing way of framing that debate has been in terms of numbers and statistics: how much voter fraud is there, and does the amount of voter fraud justify new measures to prevent it? In my essay, I argue for a shift away from statistical analysis and towards normative discourse. Only if we understand why (and whether) voter fraud is bad will we be able to decisively settle debates about what should be done about it, if anything.

The first part of my paper motivates the idea that normative debate is the proper register in which to talk about voter fraud. The second part analyzes in detail two recent cases, the Supreme Court's Purcell v. Gonzales and the 7th Circuit's Crawford v. Marion County in an effort to tease out the state interest in preventing voter fraud. In the third part, I propose a normative framework that emphasizes the harm of deterring voters from participating in elections caused by new requirements on voting. When balanced against the state interest in preventing fraud and the fear of fraud, the participatory interest will usually trump. What's more, an analysis in terms of the participatory interest is a kind of analysis courts are better suited to perform, rather than having to sort through statistics about the number of incidents of voter fraud.