The University of Texas at Austin   School of Law

Main menu:

Clinical Education at UT Law

Real Cases. Real Experience.

Claims of actual innocence: Finding the needle in the haystack

by Robert O. Dawson, Bryant Smith Chair in Law, University of Texas School of Law, and co-director, UT Law School Actual Innocence Clinic

There about 150,000 inmates incarcerated in the Texas Department of Criminal Justice. If the Texas criminal justice system is 99.9 percent accurate in not sending actually innocent persons to prison, that means there are about 150 inmates currently serving time in TDCJ for offenses they did not commit. Whether that number is accurate or not is any-body's guess, but the process of identifying them truly is looking for a needle in a very big haystack.

In September 2003, the Texas Center for Actual Innocence started the University of Texas Law School's Actual Innocence Clinic as the second innocence project in Texas. (The first is at the University of Houston.) There are over 40 innocence projects in the United States, and most are associated with or are parts of law schools. Their purpose is to receive claims of actual innocence from inmates, to screen and evaluate those claims, to investigate those that survive screening, and to seek exoneration and release from prison of those inmates whose actual innocence can be proven to a court or a clemency authority. Innocence clinics in law schools are also excellent tools for teaching law students how to analyze facts and make judgments about the claims' credibility.

Barry Macha's article (see page 30) points out many of the hazards in the process of making and litigating post-conviction claims of actual innocence. The factual issues raised are often based on old evidence, involve difficult credibility determinations, and result in less than 100-percent certainty of outcome. Nevertheless, we all know that valid actual innocence claims exist and that everybody in the criminal justice system has an equal interest in exonerating a person convicted of an offense for which he or she is actually innocent.

What follow are some of the ideas that co-directors Bill Allison, Dave Sheppard, and I have developed while teaching the actual innocence clinic. These ideas and views do not necessarily represent those of any other innocence project in Texas or the United States.

Like most innocence projects, we do not charge legal fees. We pay our own investigative expenses, although we reserve the right to ask for contributions toward those expenses by family members. The Texas Center for Actual Innocence, which is the tax-exempt corporation that runs the Actual Innocence Clinic, provides in its by-laws that no director or officer of the corporation may receive any compensation of any kind for work done on actual innocence cases.

Here is how we operate and, perhaps, how we differ from some other projects.

Screening actual innocence claims

When we receive a letter from an inmate (at a rate of five to 15 per day), it is logged into our docket system and assigned to a student in the clinic. The student first determines whether the inmate is claiming actual innocence. If not, the inmate is informed we cannot help him or her and is directed to any available legal assistance that may address the problem or question presented. We don't help inmates on their post-conviction writs raising non-innocence claims.

The student does the screening under the supervision of the three co-directors of the clinic. Every letter sent and step taken is first approved by a director.

If an innocence claim is made, often it is vague or not burdened by detail. We write back asking for details, explaining that actual innocence means that the inmate had no connection to the offense (mistaken identity) or that there never was an offense (false accusation). Claims based on accident, self-defense, defense of others, and legal insanity are not innocence claims as we have defined them. We require substantial daylight between the inmate and the offense. Out of respect for the values underlying conviction finality, the law requires that proof of actual innocence must be made by new evidence, not by rehashing evidence previously considered.

We do not take cases from death row. We automatically refer all such letters to the Capital Punishment Clinic at the U.T. Law School, which has the expertise to navigate through the highly-technical body of death penalty law.

If in his or her letter, the inmate has made a detailed actual innocence claim, we then conduct an Internet search of the case. We determine whether any published or unpublished opinion was issued in an appeal of the case, and we determine whether there were any newspaper stories about the case when it was in the local courts. Any documents we find become part of our screening package.

If the actual innocence claim survives this screening process, we send the inmate a 15-page questionnaire asking about all aspects of the case, including the current location of physical evidence and legal documents.

When the inmate responds to the questionnaire, the student screening the case writes a comprehensive memorandum about the case, in which all the information gleaned (beginning with the initial inmate contact) is described in detail. The student then presents the memorandum to a clinic meeting, where it is analyzed by faculty and students. The two questions asked are: 1) is there is a real claim of actual innocence, and 2) can we prove it?

Often, as a result of the memo presentation process, students and faculty make suggestions for further investigation to the student; sometimes, the decision is that the case is hopeless and will be closed by a letter to the inmate telling him or her we are unable to help.

After the memo presentation, the investigation involves locating relevant court documents, physical evidence, and witnesses. We continually assess whether our innocence hypothesis stands the test of factual investigation. If at any point it does not, we close the case.

DNA and innocence claims

The heart and soul of actual innocence claims is DNA testing. A forensic development of the last 15 years, DNA testing has opened wide, new windows on truth in the criminal justice system and elsewhere. The most frequent use of DNA testing is made by prosecutors to provide persuasive evidence of the accused's guilt. In post-conviction testing, DNA sometimes shows that the person convicted was actually innocent and sometimes confirms that the system got it right.

DNA testing in criminal cases, unlike DNA testing to determine paternity, is almost never the entire picture of guilt or innocence. The results of a competently-conducted DNA test are impossible to ignore, but their significance is almost always subject to argument when combined with the other evidence in the case. That is where traditional lawyer factual analysis becomes critical and one must always remember that we deal only with probabilities, not certainties.

Innocence cases as a group are here to stay because of DNA testing. However, DNA testing is possible in only a small percentage of criminal cases – primarily in murder and sexual assault cases. Many innocence claims do not depend on DNA testing, but on re-examination of scientific evidence, recantation by witnesses, or other new evidence tending to show actual innocence.

Probability vs. certainty

We all would like 100-percent certainty in the criminal justice system, but we all know that is not possible. Probably every prosecutor in the history of the state of Texas has reminded the jury on voir dire and in closing argument that proof beyond a reasonable doubt does not require proof beyond all doubt, proof beyond a shadow of a doubt, or proof to a moral certainty. In other words, jurors are told that they can have a doubt of guilt and still convict, so long as the doubt is not too large.

The same phenomenon occurs at the exoneration end of the process. There is no such thing as absolute certainty. Even with a favorable DNA result, one can almost always dream up a version of the facts that makes that result somewhat consistent with guilt. Texas law requires that proof of guilt must be beyond a reasonable doubt to convict and that proof of actual innocence must be by clear and convincing evidence to exonerate. Neither standard requires absolute certainty because absolute certainty is beyond the power of mere mortals to achieve.

That means, unfortunately, at the end of the day that some people will remain convinced that a convicted person is actually innocent. It also means that some people will remain convinced that an exonerated inmate is actually guilty. We cannot change that; we can only seek to minimize the numbers of such persons.

Role alteration in exoneration cases

Texas prosecutors are told by the legislature that their primary duty is not to convict but to see that justice is done. I have always understood that to mean, among other things, that a prosecutor must subjectively and substantively believe in the guilt of the person he or she is prosecuting to ethically pursue a charge. If that is not what the prosecutor believes, then he or she should move for dismissal or give the case to a prosecutor who views it differently.

Defense attorneys are not required to believe in the factual innocence of their clients, nor are they permitted not to present a defense just because they do not "believe in it." While they cannot suborn perjury, defense attorneys must present all plausible defenses within sound trial strategy. Their burden is only to raise a reasonable doubt.

When a defense attorney is presenting an actual innocence claim, however, I believe he is required ethically to believe in his client's actual innocence. In that circumstance, the burden of persuasion rests squarely on the defense lawyer's shoulders. A claim of actual innocence should and does carry such morally persuasive weight with prosecutors and judges that it cannot ethically be made by a defense attorney who does not subjectively and substantively believe in it.

How exoneration cases (ideally) should be handled

The reaction of a prosecutor to a post-conviction claim of actual innocence should initially be one of skepticism. The prosecutor who prosecuted the case believed in it and thought the result was correct. Now, along comes a defense attorney to tell him he has convicted an innocent person. And worse yet, the defense attorney is now asking the prosecutor to assist him in proving that a miscarriage of justice occurred.

Nobody wants to see a person in prison for an offense he did not commit. That is everybody's worst nightmare. At this stage, the prosecutor's job is to challenge the evidence of actual innocence presented by the defense attorney to determine in her own mind whether it is solid or not. If by that process, the prosecutor concludes that the inmate is very probably actually innocent of the offense for which he is serving time, then the prosecutor should join the defense attorney in further investigation with a view to obtaining an exoneration and release if that proves factually appropriate. At this stage, it should be a team effort. This is no more a time for prosecutorial game-playing than it is for defense game-playing.

If at the end of the day the inmate is exonerated and released, there will be credit enough to go around. The public will see that former adversaries are capable of working together to secure the correct result, and the public's confidence in the integrity of the criminal justice system will be strengthened.

If the prosecutor remains unpersuaded by the evidence of actual innocence, the case can still be litigated to conclusion in an adversarial context – one hopes in a civilized manner.


Each time a convicted person is exonerated by post-conviction proof of actual innocence, two consequences ensue: First, the event shows that the criminal justice system, while far from broken, is fallible because it is run by humans. Second, it shows that humans within the system, operating in good faith, can jointly correct some of these mistakes. The process of exoneration through post-conviction proof of innocence should be thought of not as an external challenge to the criminal justice system but rather as a necessary and integral part of it.

Reprinted from The Texas Prosecutor (Volume 35, Number 1, January/February 2005) with permission of the Texas District & County Attorneys Association.