Proofiness - Charles Seife [86]
Scalia launched a violent counterattack on what he termed the sanctimonious finger-wagging of the anti-death-penalty forces. The foundation of his argument was an outrageous bit of proofiness. Scalia asserted that the justice system was almost completely foolproof: that its error rate for felonies was a mere 0.027 percent. It’s a mind-blowing figure—if true, it would mean that juries and judges have a nearly superhuman ability to prevent the innocent from winding up behind bars.
The figure came from an op-ed piece that had appeared in the New York Times six months earlier. Penned by Joshua Marquis, a prosecutor from Oregon, it attacked the 340-exonerations study, and generated the 0.027 percent figure by a back-of-the-envelope calculation:
So, let’s give the professor the benefit of the doubt: let’s assume that he understated the number of innocents by roughly a factor of 10, that instead of 340 there were 4,000 people in prison who weren’t involved in the crime in any way. During that same 15 years, there were more than 15 million felony convictions across the country. That would make the error rate .027 percent—or, to put it another way, a success rate of 99.973 percent.
Most industries would like to claim such a record of efficiency.
This is a stunningly incorrect calculation. The main problem is that it’s fallacious to look at all 15 million felony convictions across the country in coming up with an error rate. It’s comparing apples to oranges, because most of those felonies never had any chance of being overturned, even if they were wrongful. To calculate the real error rate, you have to look very carefully at which of those 15 million felony convictions you should include in the calculation. It’s not an easy task.
For one thing, not all felonies are created equal. People who are convicted of check-kiting, blackmail, and other nonviolent felonies tend to get out of jail relatively quickly; it’s the violent offenses like rape and murder that get the long sentences. Convicts with long sentences are the ones who are going to get aid from pro bono lawyers and from groups like the Innocence Project, and perhaps get their convictions overturned. Rapes and murders are just a tiny percentage of the felonies committed annually (in 2002, 0.6 percent of felony convictions were for murder and 3.3 percent were for rape or sexual assault). However, murder, rape, and sexual assault accounted for 326 out of the 340 exonerations listed by the Journal of Criminal Law and Criminology paper. That is, the vast majority of exonerations were drawn from less than 4 percent of the pool of felonies. These 600,000 people convicted of murders, rapes, and sexual assaults are the ones whose cases are really eligible for being overturned—by virtue of a high level of scrutiny and long sentences, these are the types of felony convictions that should go into Marquis’s calculation, not the 15 million felony convictions of all kinds.
However, even this 600,000 felonies figure is deceptive. It includes lots of people who never went to trial in the first place. If you want to find out the proportion of people who are wrongly convicted, you should only look at those convicts who claim that they’re innocent, rather than the ones who admit their guilt. In 2002, more than 90 percent of people who went to jail for rape and sexual assault entered a guilty plea; nearly 70 percent of murderers did the same thing and admitted their guilt. An error rate should exclude these felons, leaving only about 80,000 felons eligible to be exonerated.
This is obviously quite