Why is it so hard to free wrongfully convicted people from prison?

It’s common to hear stories like that of Adnan Syed, who has maintained his innocence through 23 years of imprisonment on a murder conviction and whose story was chronicled through the viral true crime podcast "Serial." A judge recently vacated his conviction because prosecutors withheld information that would have exonerated him.

That's the part the public doesn't hear about as often: Why it normally takes so long for people to prove their innocence, and why there's reason to think that documented exonerations are just the tip of the innocence iceberg. GBH News legal analyst and Northeastern law professor Daniel Medwed, whose book "Barred: Why the Innocent Can't Get Out of Prison" was released last week, joined GBH Morning Edition hosts Paris Alston and Jeremy Siegel to talk about the subject. This transcript has been lightly edited.

Jeremy Siegel: First off, congratulations. The book is fantastic. It's generated a lot of buzz. And for people who haven't checked it out but should, it contains a bunch of stories of how legal procedures have blocked innocent people from getting out of prison. So we want to go through some of that and what's in the book. But first, what made you want to write this book?

Daniel Medwed: We've all read these stories. We've seen these pictures of innocent prisoners leaving the gates with their hands aloft, and their faces triumphant. But what's less well-known is why it took so long to get there. We know what put them there in the first place.

We know that eyewitnesses make mistakes. We know that police interrogation tactics can induce juveniles and people with cognitive deficits to falsely confess to crimes they didn't commit. We know that police and prosecutors hide evidence, that defense lawyers are overworked and sometimes underperform and botch their trial strategy. We know about dubious forensic evidence, junk science that can be overvalued by jurors.

But why does it take so long after you've assembled compelling evidence of innocence to convince a judge or a prosecutor to release the person? I know this firsthand. About 20 years ago, I ran the day-to-day operations of a small innocence project in New York. It was the Second Look Program at Brooklyn Law School, where my students and I investigated and litigated claims of innocence by New York State prisoners. And time and time again, we'd find really powerful new evidence, evidence pointing to an alternative perpetrator or maybe a recantation from the main prosecution witness at trial. And judges and prosecutors wouldn't take a close look. Out of those frustrations, I think, this book emerged as I began to research this phenomenon.

Paris Alston: And, Daniel, I just wonder if we hear about wrongful convictions all the time, why are they so hard to overturn? Is it the judges and the prosecutors being unwilling to reexamine a case, or is it the evidence getting caught up in the court system, which we know can move at a snail's pace?

Medwed: I think it's a little bit of both, and here's why. First, consider the psychology involved after a plea bargain or a trial results in a conviction. The presumption of innocence disappears. It's replaced by a presumption of guilt, which solidifies and hardens over time.

Cognitive psychologists talk about this in terms of the confirmation bias, or what we all know as tunnel vision: Once you've made a decision, once you've reached an outcome, it takes lots and lots and lots of countervailing evidence to deviate from that position, to get you to move from that vested position. It's why people stay in romantic relationships far longer than they should. It's why Morning Edition hosts stay in the job despite those 2:30 a.m. alarm clocks, right? We're wedded in these positions. And I think judges and prosecutors and even defense lawyers feel this. There's a conviction on the books. And cognitively, psychologically, it's hard to look at the case anew.

Second, I think the legal procedures that govern the appellate and post-conviction process sort of reflect this ideology, this emphasis on exalting finality and efficiency over accuracy. The cases have to be final at some point, or else our system would grind to a halt. There'd be no room for new cases, no room for new trials. Crime victims would never have closure. So what that means, in a practical sense, is that the appellate and post-conviction process is replete with procedural booby traps that make it hard for both the guilty and the innocent alike.

Siegel: Let's talk about some of those procedures, because in your book, you do the thing that you do so well when you talk to us, where you take these really boring, intricate pieces of the legal system and make them understandable. And talking about these booby traps and the procedure surrounding things, you've looked at how in some cases, there can be a situation where someone comes forward saying that they did a crime, but someone else is still unable to have the conviction for that crime overturned when there's someone else saying "I did this." I mean, what are these booby traps? What prevents a case like that or other cases from, you know, being a wrongful conviction overturned?

Medwed: Contrary to popular opinion, appeals aren't endless. And technically, you actually only have one right to appeal your case. It's called the direct appeal, and there are limitations on what you can do on the direct appeal. There are other remedies, too, but let's talk a little bit about the direct appeal. So when you appeal your conviction to a panel of judges in a higher court, you're limited to what happened at trial, to the record that is developed through the trial transcript. You can't bring in newly discovered evidence, and even then you can typically only raise issues that, are adequately preserved for review, that were raised at trial where the defense attorney made an objection and there was a debate on the record.

What that means in practice is that even if you can convince a judge that there was an error in your case, they might not consider it enough to reverse your conviction, in part because of another booby trap known as the harmless error doctrine. Showing that bad things happened at your trial is insufficient in most cases to get a new trial to get a reversal, you have to show that it affected the outcome. And if judges look at the rest of the evidence and they say "no harm, no foul, you would have been convicted anyway," then your conviction will stay on the books.

I think the Supreme Court has really demonstrated the emphasis on trials. In a 2017 opinion called Davila v. Davis, it said that the trial is the main event. It's in the Constitution twice. There is nothing about the appeal in the Constitution. The trial is the main event and the appeal is basically the undercard, which wouldn't be so bad, I think, if we had a better safety record at trial. That's really the problem.