Research on Wrongful Convictions: Half-Full or Half-Empty?

Jon Gould and Richard Leo have an interesting new paper entitled “One Hundred Years Later: Wrongful Convictions After a Century of Research.”  They are responding to a recent claim by Samuel Gross and Barbara O’Brien that researchers “do not know much about false convictions.”  Gould and Leo essentially take the “glass is half-full” position – although our knowledge may be less precise than we might like, we have nonetheless come a long way since American researchers first began to study wrongful convictions a century ago, especially as a result of studies of DNA-based exonerations over the past twenty years.  In advancing the claim, Gould and Leo provide a helpful overview of the burgeoning empirical literature on the frequency, consequences, and causes of wrongful convictions.

Here are a few tidbits I found especially intriguing:

The percentage of convictions that are wrongful remains hotly contested, but the best available studies suggest a wrongful conviction rate of somewhere in the range of two to five percent in capital cases.  Noncapital cases have not been studied as carefully, although it is at least plausible that error rates would be higher in lower-profile cases in which fewer resources are invested into the adversarial process.  Whether error rates are significantly different in plea-bargained cases also remains a mystery.

A suggestive 1996 study by the National Institute of Justice revealed that in “every year since 1989, in about 25 percent of the sexual assault cases referred to the FBI . . . the primary suspect has been excluded by DNA testing.”

A 2001 study by Talia Harmon compared a set of 76 cases in which capital inmates were released from death row because of doubts about their guilt with a random sample of capital cases in which the defendant was executed.  One variable that correlated with post-conviction success was whether the defendant was represented on appeal by private counsel or a death-penalty resource center, as opposed to a public defender.   A somewhat differently structured 2005 study by Harmon and William Lofquist made a similar finding with respect to the significance of counsel.  Of course, these findings of correlation do not say why defendants with private counsel fare better than defendants with public defenders.  Gould and Leo suggest that the explanation may lie in the greater resources available to big-firm lawyers who take capital cases pro bono and the greater specialized expertise of resource centers in capital litigation.  If they are right, then it would seem that the benefits of private counsel are specific to capital litigation.

Another intriguing finding from the Harmon and Lofquist study was that capital cases involving an African-American defendant and a white victim were significantly more likely to result in post-conviction exonerations.  This suggests that error rates are higher in cases involving allegations of inter-racial violence.