Columbia Law School team publishes anatomy of a wrongful South Texas execution

So we initially gave the story over to the Chicago Tribune to essentially verify, to see if they could find what we thought we had found in this case. They did in a report back in 2006, and it was basically a preliminary finding that this was a serious case, that there was a serious possibility of innocence. So we decided to keep going with the case and dig into it further. Again, the more we looked, the more we found. People started to come forward who had hidden things for years, they had been scared and they finally all came forward. We felt like it really needed to be out there and we needed to give it the full treatment a story like this really requires, which is really a book-length treatment.

What were the logistics of investigating a case like this, gathering so many witnesses years after the execution, gathering so much material, audiotape, videotape?

Eventually, we spent two years on solid investigation, using four investigators. We probably made over 100 requests under Texas' public records act to get documents. Many times we had to ask for documents two, three, even four times before the agency gave us anything. For example, the district attorney's office in Nueces County would acknowledge that they had records, then they'd say they didn't have anything, then they'd only give us 30 pages. Finally, in the third request we got the full 200 pages which really had all of the important information that we needed from that source.

We talked to over 100 witnesses, about 120. We actually traced down 300 witnesses trying to find them after 20 years, so of course it was very difficult to find some of them. We found witnesses spread all over the country, in Los Angeles, Chicago, southern Michigan, Florida, who were very important to the case. And then it actually took a very minute, meticulous examination of the materials that we had gathered. We found some incredible stuff in the photographs that we had developed off the police negatives that we didn't even think had anything. We found, for example, that there was a bloody footprint left behind by the killer at the crime scene, and that the jury never knew about it. It was never talked about at trial, even though everybody knew DeLuna's shoes had no blood on him when he was arrested very shortly after. It was a very meticulous process. Even the last couple days before we went live with this project, we were discovering new things in the records. We had about 10, 15 feet of paper we had to go through, much of it poorly photocopied, just little things that you find the second or third time you read thru it.

What's most alarming about what went wrong with this case, and why weren't the errors caught?

The most alarming thing about this case was that even though there were a lot of problems in it, it is not by any means the worst case that I've seen. Each of the kinds of things that went wrong were kind of everyday kinds of mistakes that you see in a lot of cases. For example: the eyewitness identification. That was really the single piece of evidence the prosecution had against DeLuna. They brought him to the gas station where the crime occurred about 40 minutes after it happened. The single eyewitness was there, and he was scared to death because police had essentially told him, “We found this guy under a truck, this is the guy, will you go look at him?” The witness said, “I don't want to get anywhere close to him,” but they made him go up to where DeLuna was, and they actually put a phalanx of police around him. And he could hardly even see because he was kind of short guy looking over their shoulders. The police trained their flashlights onto the backseat of this squad car where DeLuna was sitting without a shirt, his hands cuffed behind his back. And of course after 15 seconds, with a lot of people, police standing around, the witness said, “That's the guy, get me out of here.” So it was a very unreliable kind of eyewitness identification there.

> Email Michael Barajas