I’ve been thinking about this error rate question a great deal since attending the Innocence Network conference in Portland, Oregon a few weeks ago (April 11th and 12th).

As a trial level public defender, one constant challenge is seeking to dispel misperceptions of jurors about the accuracy and reliability of the criminal trial process. Many jurors walk into the courtroom with beliefs that the police usually arrest the right person, the prosecutor will be able to prove the charges, police officers who testify will be truthful and accurate, and evidence introduced by “forensic science experts” is the result of rigorous, reliable scientific method. Of course, prosecutors and judges, who participate in the plea negotations that resolve most cases, share many of these sentiments (interesting recent discussion of judges’ bias toward believing cops is here and here).

But how true are these beliefs? Far too often, not very true. How well is the machinery of the criminal justice system working to produce accurate results? How often are people convicted of crimes that they did not commit?

Wrongful conviction errors come in two main categories: innocent people who plead guilty (this happens a lot, as discussed here), and innocent people who go to trial and are [falsely] convicted. Some of the people who have been falsely convicted have been subsequently exonerated. The exoneration cases are irrefutable evidence of errors in the criminal justice system. The systemic causes of error are being studied in order to advocate for reforms that will reduce this error rate.

How widespread are these errors? One speaker at the Innocence Network conference spoke of false conviction rate estimations ranging between 2% and 20% of criminal convictions, from a variety of studies using a variety of methodologies. This is a very important number to try to pin down with further study.

On April 28th, a study was published in the prestigious Proceedings of the National Academy of Sciences concluding that the error rate in death penalty cases is 4.1% (“The Rate of False Conviction of Criminal Defendants Who are Sentenced to Death” published 4/28/14). Samuel Gross, one author of the study, explained the significance of this number: “Since 1973, nearly 8,500 defendants have been sentenced to death in the United States, and 138 of them have been exonerated. Our study means that more than 200 additional innocent defendants have been sentenced to death in that period.” Many innocent people remain on death row. (See discussion here.)

One day after the death penalty error rate study was published, the botched execution of Clayton Lockett in Oklahoma on April 29th
reignited debate on the death penalty and focused attention on the fact that 3% of executions are botched, most often using lethal injection. The botched execution in Oklahoma followed a politicized process that had bullied the state supreme court and shielded secret execution processes from scrutiny. Oklahoma Governor Mary Fallin, who prevented scrutiny and delay prior to the execution, is now calling for an independent review of execution procedures. The Oklahoma execution was shocking, inhumane and cruel. (Coverage here
and here.)

It is fascinating to speak with jurors about how the presumption of innocence and the requirement of proof beyond a reasonable doubt are intended to reduce the number of wrongful convictions. I often ask jurors what they think of Blackstone’s principle: “It is better that ten guilty persons escape than that one innocent suffer.” When I ask jurors if they believe false convictions occur, some are openly skeptical. It does happen, more often than many people would like to admit.

As the events of this week have highlighted, there is an error rate in the criminal justice system. It is significant and irrefutable, and it can be reduced. Acknowledgement and study of these errors will propel reform and improve accuracy in the criminal justice system.