Dissenting in Kansas v. Marsh, Justice Souter made the controversial claim that “among all prosecutions homicide cases suffer an unusually high incidence of false conviction.” He explained this phenomenon as due to “the combined difficulty of investigating (capital cases) without help from the victim, intense pressure to get convictions in homicide cases, and the corresponding incentive for the guilty to frame the innocent”. Doug Berman, at Sentencing Law and Policy, takes serious issue with this claim. writing:
Not only do I think that this assertion is wrong, but I think it is sad and dangerous that the four “liberal” Justices might actually believe it is true.
Unfortunately, like a couple of his commenters, I think there is a good chance that Souter is right and Doug is wrong. Why? At least six reasons.
1. The plurality, if not majority, of criminal cases involve a crime witnessed by a police officer where arrest followed immediately. In these cases – from drug cases, to DUI, to disorderly conduct – error rates are likely low. This is true for two reasons. First, notwithstanding their competitive urges, officers have relatively little incentive to lie. (This may be a bit less true in officer assault cases, where disciplinary or tort consequences for offender injuries hinge on asserting that the defender was the cause of the problem.) Second, when the arrest follows immediately after the offense, there is very low risk of misidentification. Civilian assaults – including homicides – involve seriously higher risk of jury error. In assault cases involving unknown assailants, identification errors are a significant problem. The incidents are often quick and unexpected, and witnesses have little opportunity to observe what really occurred. Cross-racial identification problems infect the accuracy of ID’s in some cases. Poor line-up and photo array techniques further undermine accuracy of these identifications. And despite all these problems, witnesses are typically very confident that their testimony is accurate. Unfortunately, juries often rely on this confidence factor to “believe” eyewitnesses; the problem is that confidence is not a proxy for accuracy. Witnesses are sure they’re telling the truth, but they are often wrong. Thus capital cases, and cases lacking officer observation more generally, involve more guilt error than average.
2. Capital juries are likely to be less sympathetic to the defense because they are death-qualified (i.e., only people who are willing to impose death are permitted to be jurors in a capital case.) This eliminates a not insignificant portion of the population that is most attractive to the defense.
3. In capital cases, defense attorneys frequently do not mount serious innocence defenses during trial for fear that, if the defendant is convicted, he or she will appear less remorseful at the punishment phase. This is basic strategy in any capital case. The sentencing tail typically wags the guilt/innocence dog.