A Comprehensive study of 328 criminal cases over the last 15 years in which the convicted person was exonerated suggests that there are thousands of innocent people in prison today.The study sounds very interesting, but the news report is quite weak. In medical terms a conviction is a "positive test result". Like every other "test", conviction is imperfect; the legal system will always acquit the guilty and convict the innocent. This study gives us some insight into the prevalence of false convictions, unfortunately the journalist provided only a rough number ("thousands"). The study also suggests the causes of false conviction in rape and murder cases. The most frequent cause of false conviction in rape is the misidentification of black men by white women, the most common cause of false conviction in murder cases is deception by "witnesses" (often in return for some advantage) and false confession (often by emotionally or cognitively disabled persons).
The lessons of the study may be:
1. Identification across ethnic groups as not as strong as within ethnic groups. "They all look alike" may not be mere slur, but an accurate statement about how recognition works. This should be relatively easy to research; I suspect the work has been done. The justice system should then relatively devalue identifications where the source and target are of different ethnic groups.
2. Confessions by persons with low IQ or psychiatric problems must be highly suspect.
3. Evidence obtained by persons who have something to gain is highly tainted.
I suspect the study also suggests endemic weaknesses in our system of criminal defense -- but the failure of the journalists or his editor to provide useful data limits that conclusion. I doubt we'll see any reform coming from our current regime. We need a Dickens for the 21st century.