Science Vs. the Death Penalty

This post by Philip Yam illustrates an common tendency to inflate the proportion of an argument that is proved by science. Yam claims that "Science has shown that our death penalty system is deeply flawed."

Yam makes two "scientific" arguments against the death penalty. First, he points out that science -- specifically DNA testing -- has exonerated at least one person and possibly more. Second, science has demonstrated the unreliability of eyewitness testimony, which is a crucial element of the evidence against most death row inmates.

All that is scientifically shown, then, is that our death penalty system cannot be relied on to be perfect in executing only the guilty. Or rather, that our criminal justice system is imperfect -- after all, eyewitness testimony is used against defendants sentenced to prison, too. It requires additional steps -- steps I happen to agree with, mind you -- to go from "imperfect" to "deeply flawed." One must evaluate the system's mistake rate as being too high, and believe that there's a meaningful difference between sentencing this mix of criminals and innocents to death or to life without parole. Yam treats this step as rather obvious, implying that if Americans only knew how many innocents were sentenced, they would reject the death penalty. But that fact is not so obvious -- after all, the guilty folks are pretty dangerous characters who richly deserve death (according to some ethical systems). Either way, though, this step is not a scientific one.