At their best, databases aren’t just static repositories of information. They are tools that can help tell stories. One of the newest databases, the National Registry of Exonerations, shows just how eye-opening a story data can tell when it is collected and normalized. A joint project of the University of Michigan Law School and the Center on Wrongful Convictions at Northwestern University Law School, the database profiles approximately 900 of 2000 cases in which defendants were convicted and then exonerated in the US since 1989, the year that DNA was first used as evidence. (Another 1100 “group exonerations” that occurred in response to 13 separate police corruption scandals are not included in the registry.)
A report produced from the database by University of Michigan law professor Samuel Gross reveals some of the shocking mistakes that occur in our criminal justice system: Wrongfully-convicted prisoners captured in the database spent an average of 11 years in prison before being exonerated. Perjury and false accusations were the most common causes of a wrongful conviction, accounting for 51 percent of the cases included in the database. Over 14% of the profiled exonerations were of defendants who were convicted of crimes that never happened, and around the same number of exonerations occurred among defendants who confessed to crimes they did not commit; 7% of the exonerations were of innocent defendants who pled guilty. The data also shows that false convictions vary by crime: For murder, the biggest cause of wrongful convictions was perjury, usually by a witness who claimed to have seen the crime or even participated in it. In rape cases, false convictions are almost always based on eyewitness mistakes – more often than not, mistakes by white victims who misidentified black defendants. False convictions for robbery were also almost always caused by eyewitness misidentifications. Child sex abuse exonerations were almost always from fabricated crimes that never occurred.
The registry is probably just scratching the surface, as there are no official reporting mechanisms for exonerations. To remedy that problem, the database producers have asked for help in identifying more exonerations. Each reported case must then be researched and carefully analyzed and categorized. Peter Neufeld, director of the Innocence Project, calls the registry the “Wikipedia of Innocence” and in some ways he right. In just one month since it went public with 873 entries, the database has grown to 912 entries, thanks in part to help from contributors.
Another noteworthy part of the story is the tale of the database itself. Our friend Rob Warden, Executive Director of Northwestern’s Center on Wrongful Convictions, first told us about his vision for this database nearly 10 years ago and has been working on it since then. Bravo, Rob and colleagues!