Interview with Ofra Bikel. She is the producer of FRONTLINE's "The Case for Innocence". She has also produced FRONTLINE's "Snitch" and "Innocence Lost"

Over the years I have done several reports on the criminal justice system. "Innocence Lost" was a trilogy of programs about a day care child abuse investigation in Edenton, North Carolina. "Snitch," which aired in 1999, reported on the use of informers in the prosecution of drug cases. These reports all dealt with errors in the criminal justice process. And this led me to ask, 'How then, does the system correct itself when it makes mistakes?'

I began by looking at some of the really serious claims of innocence of prisoners. I found quite a few, starting with totally incompetent lawyers who were asleep or drunk during trials, to witnesses who recanted, to new evidence which emerged after the trial. ..which led me to DNA.

Since DNA tests did not exist 15 years ago, people who were convicted in cases where blood or semen played a critical role, could now ask for the evidence (if it still existed) to be re-tested. This, of course, provided a rare window through which one could see how the system rights itself when scientific evidence demonstrated that a mistake had indeed been made.

What I found was not encouraging. In spite of the fact that over 65 inmates have been released from prison because of new DNA testing, there are scores who still remain in prisons because they are denied the new test or because the results, once they favored the inmate - were dismissed, or hidden.

So in the end, you see, this report is not really about DNA. It is about how the criminal justice system responds to new unimpeachable evidence which demonstrates that mistakes have been made. When people are released because of new DNA evidence, it makes for headline stories which everyone applauds. But what is rarely discussed is the randomness of their release.

What do you mean by 'randomness?'

In our program "The Case for Innocence," the Clyde Charles case is an example. Charles was one of over a thousand files at the Innocence Project at Cardozo School of Law, where the students and their professors are known for pursuing these cases and helping to free innocent people through new DNA evidence.

The Innocence Project has a very long waiting list. So I called Barry Scheck one of the co-founders of the Project and said "I'd like to follow this case." I pointed out to him there was a 10-minute telephone hearing coming up with a judge within 4 days - and Charles's sister was desperate, afraid that this would be the very last hearing he would ever get. I said to Scheck "Let's do it. Will you personally get involved in this one right away?" He agreed. We talked on a Thursday night. By the following Tuesday he was representing Clyde Charles in front of the federal judge.

Five months later, after both Scheck and FRONTLINE pushed and pushed and pushed, Clyde Charles was exonerated and was released from prison having served there for 19 years.

Now people asked me - aren't you happy? Aren't you proud? Yes, I am happy, and I am proud that I had some part in releasing an innocent man from prison, but more than anything, I tremble with fear.

Because, what if we had decided on another case instead? Clyde Charles could have easily been in prison for another five or ten years. By the time he got the tests done in November-December 1999, the evidence kit was already badly degraded - what would have happened in another few years? And who did we not help? What unhappy family did we just toss aside because the profile was not what we were looking for?

The Innocence Project is great and deserves all the accolades it gets, but should we rely on professors and their volunteering students to free innocent people from prison? Should we rely on a journalist like myself on stumbling upon one Clyde Charles whose sister sounded good on the phone?

There is also the last story in my report, Earl Washington. He was on death row charged with rape and murder, before his sentence was commuted to life in prison. Before he got clemency he had been given a DNA test, the results of which his lawyers were not allowed to see - not at the time, 1994, and not even when we met them at the end of 1999.

When we began to research the case, we "the media" went to the lab and asked to see the results of the DNA tests. They were simply handed to us. And that is how we, stumbling into the case, told his shocked hard-working lawyers that their client had been exonerated by the test five years before, and should have probably gone home a free man, not spend the rest of his life behind bars. Can something still be done about this poor young man Earl Washington? I certainly hope so. But is this the way the system is supposed to work? I doubt it.

What about the Roy Criner case?

The Criner story, which is a rape-murder case, is a true nightmare story. Everyone agrees that new DNA tests show for certain that it is not Criner's DNA.. It is obvious to everyone that had the trial taken place today, with the DNA evidence, he would not have been found guilty. The prosecutor says so, as does the foreman of the jury. And yet, he was refused a new trial and must remain in prison for 99 years, or until the people of Texas demand his release, if this ever comes to pass.

So, again, in the end this report is not about DNA. DNA only shows how rigid the system can be, how unwilling to look at itself, and how determined it is not to admit mistakes even when faced by something as scientific and specific as DNA.

When I am interviewed for this program I am often asked: do you think that there will be a change in the law regarding DNA? My answer is yes, I think there will be a change in the law. And this is certainly good. But again if you look at DNA as a window into to how the system reacts to its own mistakes, the view is grim and foggy.