The Seduction of Innocence: The Attraction and Limitations of the Focus on Innocence in Capital Punishment Law and Advocacy

Article excerpt

INTRODUCTION

Over the past five years we have seen an unprecedented swell of debate at all levels of public life regarding the American death penalty. Much of the debate centers on the crisis of confidence engendered by the high-profile release of a significant number of wrongly convicted inmates from the nation's death rows. Advocates for reform or abolition of capital punishment have seized upon this issue to promote various public policy initiatives to address the crisis, including proposals for more complete DNA collection and testing, procedural reforms in capital cases, substantive limits on the use of capital punishment, suspension of executions, and outright abolition. Advocates for the retention and vigorous use of capital punishment have been sympathetic to some, but by no means all, of these proposals. Disagreement over the nature and scope of responses to the crisis has inevitably and quite properly led to debate about the significance of the wrongful conviction of the innocent in the administration of capital punishment.

This symposium presents two common criticisms of the current focus on innocence in the debate over the death penalty in America--one from an "agnostic" on the issue of capital punishment and one from a whole-hearted supporter. Professor Ron Allen, the self-described agnostic, along with his co-author Amy Shavell, makes the argument often made by supporters of capital punishment that there is nothing distinctive about the problem of wrongful death in the capital punishment context. Rather, the execution of some innocent people is simply the unavoidable cost of implementing capital punishment and thus is comparable to the foreseeable deaths that occur whenever the government undertakes an important social project, such as building a bridge or constructing a dam. (1) Joshua Marquis, a vocal and high-profile supporter of capital punishment, criticizes those who focus on the problem of innocence in the death penalty debate for overstating the problem by overestimating the actual number of completely innocent people convicted and sentenced to die. According to Marquis, the true number, while not zero, is low enough to constitute an acceptable cost of a valuable social policy. (2)

Although it is not our focus in this paper to refute them, we think that these common critiques of innocence are deeply flawed. Allen & Shavell's critique completely misses at least two distinctive harms that flow from executing the innocent. First, unlike the innocent victims of governmental bridge-building, those who are innocent and sentenced to death suffer the additional devastation of being blamed for a terrible crime; their names, families, and entire lives are forever tainted by such ignominy, quite apart from the death of their bodies. Moreover, when such errors are discovered, as some but by no means all of them eventually will be, they deeply undermine the legitimacy of the entire criminal justice system. This latter cost, though unquantifiable, is tremendously important. Public fear of unjust violence at the hands of the state, which has a monopoly on the legitimate use of force, is the hallmark of totalitarian regimes, one of the indices that most distinguish them from free and democratic societies. There is thus ample reason to weigh erroneous executions quite differently from unavoidable deaths in the regulatory context.

We are more sympathetic to Marquis's argument about exactly who should count as an "innocent" person, but we find the conclusion that he draws from his revised number equally flawed. Marquis claims that if we apply a more rigorous definition of innocence--such as "had no involvement in the [crime], wasn't there, didn't do it" (3)--the number of wrongly convicted and sentenced to die goes down to twenty-five or thirty, out of the 7000 murderers sentenced to death since 1976. Such a ratio, argues Marquis, represents an episodic rather than epidemic rate of error. …