similarly normative claims that the sanctity of human life precludes state-sanctioned killings, regardless of any possible social benefits of capital punishment. Separate from normative considerations, deterrence is not the only empirical issue relevant to the debate over capital punishment. Other considerations include whether capital punishment can be administered in a nondiscriminatory and consistent fashion, whether the risk of a mistaken execution of an innocent person is acceptably small, and the cost of administering the death penalty in comparison with other sanction alternatives.

Although there is empirical evidence on the issues of discrimination, mistakes, and cost, the charge to the committee does not include these questions. Nor have we been charged with rendering an overall judgment on whether capital punishment is good public policy. We have been tasked only with assessing the scientific quality of the post-Gregg evidence on the deterrent effect of capital punishment and making recommendations for improving the scientific quality and policy relevance of future research.

In including recommendations for future research, the study’s statement of task recognized that potential remedies to shortcomings in the evidence on the deterrent effect of capital punishment on homicide might also be used in the study of the crime prevention effects of noncapital sanctions. Thus, this report also offers recommendations for improving the scientific quality and policy relevance of that research.

The post-Gregg studies can be divided into two types on the basis of the type of data analyzed. Panel data studies analyze sets of states or counties measured over time, usually from about 1970 to 2000. These studies relate homicide rates over time and the jurisdictions covered to the legal status of capital punishment or the frequency of executions or both. Time-series studies generally cover only a single geographic unit, which may be as large as a nation or as small as a city. These studies usually examine whether there are short-term changes in homicide rates in that geographic unit in the aftermath of an execution. We review and critique these two types of studies separately because their design and statistical methods are quite different.

Assessing the deterrent effect of the death penalty is much more than a question of interest to social science research. It is a matter of importance to U.S. society at large, and we expect that a potentially broad audience will want to understand how the committee reached its conclusions. Yet the research that the committee has had to appraise is a body of formal empirical work that makes use of highly technical concepts and techniques. The committee has been mindful of the importance of reaching as broad an audience as possible while meeting the fundamental requirement that the report be scientifically grounded. With this in mind, Chapters 1, 2, and 3 (as well as the summary) have been written for a broad, largely policy audience, largely avoiding technical language. In contrast, Chapters 4 and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement