"We need to be more skeptical."

A great interview with Dallas DA Craig Watkins in Reason. Read the whole thing, but the best quotes are below:

We need to guard against being a rubber stamp for every case the police department sends our way. We need to be more skeptical. We also need to train prosecutors to think about their jobs in a different way. We shouldn’t be judging young prosecutors by how many convictions they win, or by how many people they put in jail.
...
But take eyewitness identification. It’s been proven time and time again in studies that eyewitness identification is extremely unreliable. Yet police, prosecutors, and juries still tend to put a lot of faith in them. And these same studies show there are some basic steps you can take make eyewitness identifications more reliable, but that also would result in fewer identifications, and fewer prosecutions. But if there are procedures available to increase the validity of a form of evidence, and police and prosecutors aren’t using it, then they’re deliberately increasing the chances of a wrongful conviction in order to get more convictions. And defendants aren’t getting a fair trial.

DNA acquittals shaking up forensic science

Mainstream media attention that brings up the unreliability of eyewitnesses:

But Neufeld and many others argue that is a tiny concern compared to what real crime labs really can and cannot do.

People are often convicted by eyewitnesses, an informant, or someone with a grudge. Even witnesses with the best of intentions can be totally wrong, psychology tests show.

Physical evidence will always be important. Unfortunately, there is no DNA to be found at the scenes of many serious crimes. Forensic techniques need to be a good as possible and courts and juries need to understand the limits.

It is not just a question of freeing those wrongfully convicted, but catching the real perpetrators of the crimes.