Rachel A. Harmon*
Arrests are the paradigmatic police activity. Though the practice of arrests in the United States, especially arrests involving minority suspects, is under attack, even critics widely assume the power to arrest is essential to policing. As a result, neither commentators nor scholars have asked why police need to make arrests. This Article takes up that question, and it argues that the power to arrest and the use of that power should be curtailed. The twelve million arrests police conduct each year are harmful not only to the individual arrested but also to their families and communities and to society as a whole. Given their costs, arrests should be used only when they serve an important state interest; yet, they often happen even when no such interest exists. Governments have allowed constitutional law to become the primary constraint on arrest practices, and it has proved a poor proxy for good policy analysis. The Fourth Amendment permits arrests whenever an officer has probable cause: it has no mechanism for ensuring that the state has any interest in making an arrest—as opposed to starting the criminal process in another way. More broadly, traditional arguments for arrests cannot justify existing arrest practice. Arrests are usually unnecessary to start the criminal process effectively, to maintain order, to collect evidence, or to deter crime. In most cases, reasonable, less intrusive, alternative means exist or could exist for achieving these ends. Even arrests for some serious crimes might be curbed significantly without risking substantial harm to public safety or order. If the state can achieve its law enforcement objectives without arrests, then police departments should conduct far fewer arrests than they currently do, and states should restrict the statutory authority to arrest accordingly. Though there are risks to reducing arrests, those risks are far less problematic than continuing what is presently a massive, and largely unnecessary, enterprise of state coercion.
*Professor of Law, University of Virginia School of Law. I am grateful to participants in the University of Virginia Policing Roundtable (especially Jeff Fagan) and attendants of the University of Arizona Solls Lecture for their feedback on an early version of this idea, and to Caleb Nelson, Bob Newman, Lauren Ouziel, and participants at Crimfest 2015 and the University of Virginia and UCLA law schools’ faculty workshops for their more recent comments. I have also benefitted from exceptional research assistance by Adam Stempel and especially by Andrew Manns, who helped me with every aspect of this project.