How Many More Are Innocent?
America's 250th DNA exoneration raises questions about how often we send the wrong person to prison.

Radley Balko | February 8, 2010

Listen to Audio Version (MP3)

Freddie Peacock of Rochester, New York, was convicted of rape in 1976. Last week he became the 250th person to be exonerated by DNA testing since 1989. According to a new report by the Innocence Project, those 250 prisoners served 3,160 years between them; 17 spent time on death row. Remarkably, 67 percent of them were convicted after 2000—a decade after the onset of modern DNA testing. The glaring question here is, How many more are there?

Calculating the percentage of innocents now in prison is a tricky and controversial process. The numerator itself is difficult enough to figure out. The certainty of DNA testing means we can be positive the 250 cases listed in the Innocence Project report didn't commit the crimes for which they were convicted, and that number also continues to rise. But there are hundreds of other cases in which convictions have been overturned due to a lack of evidence, recantation of eyewitness testimony, or police or prosecutorial misconduct, but for which there was no DNA evidence to establish definitive guilt or innocence. Those were wrongful convictions in that there wasn't sufficient evidence to establish reasonable doubt, but we can't be sure all the accused were factually innocent.

Most prosecutors fight requests for post-conviction DNA testing. That means the discovery of wrongful convictions is limited by the time and resources available to the Innocence Project and similar legal aid organizations to fight for a test in court. It's notable that in one of the few jurisdictions where the district attorney is actively seeking out wrongful convictions—Dallas County, Texas—the county by itself has seen more exonerations than all but a handful of individual states. If prosecutors in other jurisdictions were to follow Dallas D.A. Craig Watkins' lead, that 250 figure would be significantly higher.

If the numerator is tough to figure, the denominator is even more controversial. One of the more farcical attempts at writing off the growing number of DNA exonerations came in a concurring opinion that Supreme Court Justice Antonin Scalia wrote in the 2005 case Kansas v. Marsh. Scalia began by dismissing the idea that an innocent person may have been executed in America, explaining that if such a tragedy had occurred, "we would not have to hunt for it; the innocent’s name would be shouted from the rooftops by the abolition lobby."

Scalia has probably since become acquainted with the name Cameron Todd Willingham, the Texas man executed in 2004 who was likely innocent. But the justice's pique also betrays an unfamiliarity with how death penalty opposition organizations work. While Scalia is right that proof of an executed innocent would be good rhetorical fodder for death penalty abolitionists, legal aid groups aren't about to waste their limited resources hunting down mistaken executions when there are living, breathing innocents still to be discovered. Moreover, in many jurisdictions, prosecutors destroy the case files after an execution, making any post-execution investigation rather difficult. That we don't know for certain about more executed innocents doesn't mean they haven't happened.

Scalia then cited some absurd math from Josh Marquis, an Oregon prosecutor who has held various executive positions for the National District Attorneys Association. According to the Marquis formula Scalia endorsed, at the time there had been about 200 DNA exonerations. For posterity, Marquis then arbitrarily multiplied that number by 10, to come up with 2,000 wrongful convictions. Marquis then took every single felony conviction over the previous 15 years as his denominator, to come up with a meager .027 wrongful conviction rate. Move along, America. Nothing to see here. Your criminal justice system's performing just fine.

The figure is absurd. First, the subset of cases for which DNA testing can prove guilt is exceedingly small. It's generally limited to most rape and some murder cases. You can throw out the entire body of drug charges and nearly all burglary, robbery, assault, and other classes of felonies. As University of Michigan Law Professor Samuel L. Gross wrote of Marquis in a 2008 article (PDF) in the Annual Review of Law and Science, "By this logic, we could estimate the proportion of baseball players who’ve used steroids by dividing the number of major league players who’ve been caught by the total of all baseball players at all levels: major league, minor league, semipro, college and Little League—and maybe throwing in football and basketball players as well."

If the aim is to calculate the percentage of people who claim they're innocent and who actually are, you might throw out all cases decided by a guilty plea, too. But this can also get tricky. According to the Innocence Project, more than a quarter of DNA exonerations included a false confession or guilty plea. The plea bargaining process can also induce innocent people to plead guilty to lesser crimes to avoid charges with more serious prison time, particularly in drug cases.

The Innocence Project cites a study by Seton Hall's D. Michael Risinger that puts the percentage of innocents in prison at 3 to 5 percent. But that study looked only at capital crimes, and there's yet more debate over whether data gleaned from those accused of crimes that are eligible for the death penalty would translate into higher or lower wrongful conviction rates for those accused of lesser crimes. (Those who argue that it would be higher note that there's more pressure on prosecutors and jurors to hold someone accountable in murder cases. On the other hand, defendants tend to have better representation in capital cases.) But even dropping below the study's floor, using the 2008 prison population, a 2 percent wrongful conviction rate would mean about 46,000 people incarcerated for crimes they didn't commit.

Whatever the percentage, DNA testing has exposed some gaping flaws in the system, calling into question traditional assumptions on the value of eyewitness testimony, forensic evidence, confessions, and the appeals process. (In several cases in which a defendant was later exonerated by DNA testing, appeals courts not only upheld convictions, but noted the "overwhelming evidence" of the defendants' guilt.) Scalia stated in Marsh that an exoneration "demonstrates not the failure of the system but its success," but it would be naive to believe the same systemic flaws exposed by these exonerations in the small subset of cases for which DNA testing is available don't also exist in the much larger pool of non-DNA cases. Put another way, if we now know because of DNA testing that misconduct by police and prosecutors produced a wrongful conviction in a high-profile murder case, it's probably safe to assume that the same problems led to the wrongful conviction of a number of routine drug suspects over the years, too. The difference is that there's no test to clear those people's names.

So these 250 DNA exonerations aren't proof that the system is working. They're a wake-up call that it isn't. Instead of falling back on groups like the Innocence Project to serve as unofficial checks against wrongful convictions, lawmakers, judges, and law enforcement officials should be looking at why there's so much work for these organizations in the first place.