<< Cone, Again | Main | News Scan >>

More Hysteria on Deterrence


We have more reaction to the Associated Press article noted here, which brought to the attention of the general public the studies on death penalty deterrence that people involved in the debate have known about for years.

An article by Cassy Stubbs in the Huffington Post begins:

Among the many factors in the debate about the death penalty is whether capital punishment deters violent crime. Although solid research indicates that there is no valid evidence of such deterrence, recent attention has been given to a few flawed studies concluding that the death penalty does deter murder.

How does Stubbs know what is "solid" and what is "flawed"? Her bio indicates that she is a staff attorney with the ACLU Capital Punishment Project, and it does not indicate any expertise in the sciences in question.

She says the Emory and Colorado studies have been rejected by "top social scientists," citing Jeffrey Fagan. What makes Fagan "top" in comparison to the authors of the studies in question, other than the fact he reaches a result Stubbs likes? She doesn't say.  She quotes Fagan saying that the studies "fail to reach the demanding standards of social science." Oh, really?

I'm not an expert in the field either, so where do we nonexperts turn to know what is up to "standards" and what isn't? Probably the best criterion available is publication in a reputable, peer-reviewed journal in the field, one in which articles are selected, reviewed, and edited by people who are experts. Such publication is no guarantee of validity, but at least it serves as evidence that the author is willing to submit the work to scrutiny of other experts, and it has survived the scrutiny.

The Emory study and the University of Colorado study are both published in respected journals. Here are the cites:  Hashem Dezhbakhsh, Paul H. Rubin, & Joanna M. Shepherd, Does Capital Punishment Have a Deterrent Effect? New Evidence from Postmoratorium Panel Data, American Law & Economics Review, vol. 5, no. 2, pp. 344-376 (Fall 2003); H. Naci Mocan & R. Kaj Gittings, Getting Off Death Row: Commuted Sentences and the Deterrent Effect of Capital Punishment, Journal of Law and Economics, vol. 46, no. 2, pp. 453-478 (October 2003).

And where is Fagan's critique published? Stubbs cites congressional testimony, which means very little. Witnesses in congressional hearings are generally chosen by staffers, not experts in the field, and their testimony is not reviewed by anyone. There is also a lecture published in a law review, 4 Ohio St. J. Crim. L. 255, which also does not imply any selection or review by an expert in the field. Fagan's own publication page lists his "Refereed Journal Articles and Chapters," and the death penalty deterrence critique is not there.

Stubbs also relied on the Donahue and Wolfers article. As I noted earlier, they chose to bypass the peer-review process and publish in a law review. If you see a person avoid the kitchen, you have to wonder if he knows he can't stand the heat.

Unlike Stubbs, I am not going to condemn work in a field that is not mine and where experts disagree. Perhaps it is valid. If so, though, it is more than a little strange that it hasn't gone through the usual peer-review process.

Stubbs notes that "researchers Lawrence Katz, Steven Levitte[sic] and Ellen Shustorovich analyzed state data between 1950 and 1990 and did not find a correlation between the death penalty and crime rates." This study is published in a respected journal, and the researchers are well known. [Levitt is a member of CJLF's Academic Review Board.]  However, a big problem obvious to every capital punishment attorney is immediately apparent. They mixed the completely different death penalty regimes before and after Furman v. Georgia. As we all know, the Supreme Court held in Furman that the systems then in effect made far too many crimes potentially capital and then actually imposed that penalty on a few with no systematic way of separating them out. Such a system may very well have no deterrent effect, but it no longer exists. Post-Furman, the study period ends just as executions in significant numbers were getting started. The study is academically interesting and may well be methodologically valid, but a lack of deterrent effect in the pre-Furman regime tells us nothing of interest to current policy.

In the AP article discussed in the previous post, some of the researchers noted that the attacks on their work were particularly determined to refute their results, a more intensely result-oriented assault than is the norm in academic debates. Why does the anti side go into such hysteria over deterrence? Because they know that this is an argument that is persuasive to "swing voters." There are a great many people who are not convinced either way by the moral debate. Given that ambivalence, the practical consequences are what matter to them. Thoughtful people such as Stuart Taylor and Cass Sunstein, who apparently aren't bothered by the injustice of Charles Manson and Richard Speck getting life in prison instead of death, are bothered by the possibility that lack of an effective death penalty kills innocent people through loss of deterrence.

Deterrence matters, and the debate over it matters. There are studies, there are criticisms of studies, and there are responses to the critics. Such is the normal path of academic discourse. Sophomoric articles that anoint as a top expert anyone who reaches the writer's preferred result and condemn as a hack anyone who reaches the opposite result do not help.


I think your general point in this specific post is fair, but I don't think it's justified to call Stubbs' blog post "sophomoric" merely because you don't find the studies she cited to be as persuasive as some other studies. The final graph of her post, in fact, makes the same point you do: "when it comes to analyzing data and reaching statistical conclusions that are used to affect our nation's policy and legislation on a matter as dire as capital punishment, it is critical that the research use statistically valid methodology." If there *is* ambiguity in the data, I would argue that such ambiguity argues against the use of an inherently irreparable sanction. But that's another can of worms.

On a broader note, though I often disagree with this blog's stance on many issues, I frequently appreciate the attention you give to discussing new research and empirical data.

"sophomoric," I shoulda wrote... :)

Yes, it certainly would be unjustified "to call Stubbs' blog post 'sophomoric' merely because [I] don't find the studies she cited to be as persuasive as some other studies." But that wasn't the reason at all. The reason was the way she trashes the studies that point toward a result she doesn't like, claiming they are methodologically flawed and the contrary studies are solid without a sufficient basis for the distinction. I specifically refrained from doing the same, noting that Donahue and Wolfers might be right for all I know, but they should put their work into the crucible of peer review.

Leave a comment

Monthly Archives