Get access

What Stops the Torture?


  • A previous version of this article was presented at the 2008 annual meeting of the Midwest Political Science Association. We are grateful to Susan Hannah Allen, Andreas Beger, Justin Conrad, Jacqueline H. R. DeMeritt, Christopher Einolf, Emilie Hafner-Burton, Joseph K. Young, and three anonymous reviewers for helpful comments and suggestions. An online appendix, data, and replication files are available on the authors' websites, and at

Courtenay Ryals Conrad is Ph.D. Candidate, Department of Political Science, 113 Collegiate Loop, Tallahassee, FL 32306-2230 ( Will H. Moore is Professor, Department of Political Science, 113 Collegiate Loop, Tallahassee, FL 32306-2230 (


States whose agents engage in torture in a given year have a 93% chance of continuing to torture in the following year. What leads governments to stop the use of torture? We focus on the principal–agent relationship between the executive and the individuals responsible for supervising and interrogating state prisoners. We argue that some liberal democratic institutions change the probability that leaders support the creation of institutions that discourage jailers and interrogators from engaging in torture, thus increasing the probability of a state terminating its use of torture. These relationships are strongly conditioned by the presence of violent dissent; states rarely terminate the use of torture when they face a threat. Once campaigns of violent dissent stop, however, states with popular suffrage and a free press are considerably more likely to terminate their use of torture. Also given the end of violent dissent, the greater the number of veto points in government, the lower the likelihood that a state terminates its use of torture.