Commentary by Edward J. Imwinkelried
Professor of Law,
University of California at Davis

The outcomes in the 28 cases documented in this report dramatize the real nature of the question of standards for determining the admissibility of scientific evidence in the United States.

Until recently, the Frye standard governed that question in most jurisdictions. In Frye v. United States,¹ the court announced that to be admissible, scientific testimony must be based on a technique that has "gained general acceptance in the particular field in which it belongs."² The court singled out novel scientific evidence and prescribed a special test for the introduction of such testimony. At one point, that test was the controlling law in both the Federal courts and 45 States.³ It is true that in 1993 the United States Supreme Court abandoned Frye and adopted a more flexible validation standard in Daubert v. Merrell Dow Pharmaceuticals, Inc.⁴ However, the Court decided Daubert on statutory rather than constitutional grounds, and, consequently, each State remains free to fashion its own standard for admitting scientific evidence. As of 1995, 22 States apparently remained committed to Frye.⁵ In short, the conservative general acceptance test is still in place in almost half the States.

Moreover, even in his lead opinion in Daubert, Mr. Justice Blackmun indicated that, at least in some respects, trial judges may continue to admit scientific evidence more cautiously and restrictively. The Justice initially pointed to Federal Rule of Evidence 403, authorizing trial judges to exclude logically relevant evidence when "its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury ... The Justice then quoted Judge Weinstein, a distinguished jurist and scholar, as declaring: "[E]xpert evidence can be both powerful and quite misleading because of the difficulty in evaluating it. Because of this risk, the judge in weighing possible prejudice against probative force under Rule 403 ... exercises more control over experts than over lay witnesses."⁶

Two points must be made. First, Justice Blackmun and Judge Weinstein are voicing conventional wisdom in suggesting that lay jurors attach greater weight to scientific evidence. The California Supreme Court has asserted that a "misleading aura of certainty ... often envelops a new scientific process."⁷ In a similar vein, the Court of Appeals for the District of Columbia, birthplace of the Frye rule, has written that jurors frequently attribute a "mystic infallibility" to scientific testimony.⁸

There have been empirical investigations into the impact that scientific evidence has on lay jurors. Although those studies are far from conclusive, they largely contradict the assertion that scientific evidence overwhelms lay jurors.⁹ After surveying the literature, two respected commentators concluded that "the image of a spellbound jury mesmerized by ... a forensic expert is more likely to reflect ... fantasies than the ... realities of courtroom testimony."¹⁰

Second, and more importantly, the advocates of special restrictions on the admissibility of scientific testimony misunderstand the fundamental nature of the question:

It is misleading to focus solely on the strengths and weaknesses of scientific evidence. In principle, the judgment must be comparative. To the extent that we discriminate against scientific evidence, subjecting it to uniquely discriminatory, restrictive rules such as Frye, we encourage the courts to rely on other types of evidence. Thus, our task is not to make an absolute judgment about the merits of scientific evidence. Rather, our task is to compare it with other types of evidence to decide whether the differential treatment of scientific evidence is justifable.¹¹

As the 28 cases collected in this report demonstrate, when we subject new scientific techniques such as DNA typing to special admissibility rules, we force the courts to rely on inferior types of evidence, such as eyewitness testimony. In all 28 cases, without the benefit of DNA evidence, the triers of fact had to rely on eyewitness testimony, which turned out to be inaccurate. In United States v. Wade,"¹² Mr. Justice Brennan noted: "The vagaries of eyewitness identification are well known; the annals of criminal law are rife with instances of mistaken identification." Those annals must now be lengthened to include the 28 wrongful convictions discussed in this report. In roughly two-thirds of the cases, the triers heard testimony based on traditional forms of expertise, such as hair analysis-testimony that passes muster under the Frye standard but that, again, turned out to be erroneous. There are numerous proficiency studies establishing that there is a significant margin of error in such traditional forensic techniques.¹³ The sobering fact is that in all 28 cases, the error was unmasked-and justice finally served-only because of the novel scientific technique of DNA typing.

The "junk science" controversy has made it tempting to propose special restrictions for scientific evidence, especially testimony resting on relatively new scientific techniques. One lesson to be learned from this report, however, is that before succumbing to that temptation, we should pause to pose two questions. First, have the critics of scientific evidence proven that the type of testimony in question presents a unique probative danger-or have they merely made that assertion? Further, if we impose a unique restriction on scientific testimony, on balance are the courts more likely to reach just results — or are we condemning the courts to reliance on suspect types of testimony that call into question the caliber of justice dispensed in our courts? This report should be read with those two questions foremost in mind.

Notes

1. 293 F.1013 (D.C. Cir. 1923).  [Back]
2. d. at 1014.  [Back]
3. Note, 40 Ohio St.L.J. 757, 769 (1979).  [Back]
4. 113 S.Ct. 2786 (1993).  [Back]
5. Meaney, Joseph R., "From Frye to Daubert: Is a Pattern Unfolding?" Jurimetrics, 35 (1994)  [Back]
6. 138 F.R.D. at 632.  [Back]
7. People v. Kelly. 17 Cal. 3d 24,32, 549 P.2d 1240, 1245, 130 Cal. Rptr. 144,149 (1976).  [Back]
8. United States v. Addison, 498 F.2d 741, 744 (D.C. Cir. 1974).  [Back]
9. "Standard for Admitting Scientific Evidence: A Critique from the Perspective of Juror Psychology." 28 Villanova Law Review, 554(1983) 566-570.  [Back]
10. Rogers, Richard, and Charles Patrick Ewing, "Ultimate Opinion Prescriptions: A Cosmetic Fix and a Plea for Empiricism." 13 Law and Human Behavior, 357, 363 (1989).  [Back]
11. 28 Villanova Law Review at 564.  [Back]
12. 388 U.S. 218 (1967).  [Back]
13. Giannelli, Paul C., "The Admissibility of Laboratory Reports: The Reliability of Scientific Proof," 49 Ohio St.L.J. 671 (1988).  [Back]

Commentary by Walter F. Rowe
Professor, Department of Forensic Sciences
The George Washington University

The introduction of DNA profiling has revolutionized forensic science and the criminal justice system. DNA technology has given police and the courts a means of identifying the perpetrators of rapes and murders with a very high degree of confidence.

As recently as the late 1960s, the only methods available for genetic marker analysis of blood and other body fluids were the Lattes test, the absorption-elution test, and the absorption-inhibition test. Only ABO blood group substances and ABO isoantibodies could be detected in biological stain evidence. Over the intervening years, electrophoretic methods for typing polymorphic proteins-such as phosphoglucomutase, esterase D, glyoxalase, hemoglobin, and haptoglobin-became available.

While these methods are in theory capable of greatly narrowing down the possible sources of biological stain evidence, they often fail to yield a result because of deterioration of the genetic marker. They even can yield completely erroneous results.

For a variety of reasons, DNA profiling has significantly advanced the analysis of biological stain evidence. First, these methods are intrinsically more discriminating than the methods of genetic marker analysis heretofore used. DNA profiling is more likely to exonerate a wrongly accused suspect. Second, the DNA molecule is more stable than polymorphic proteins. Third, microbial degradation does not lead to erroneous typing results.

An unforeseen consequence of the introduction of DNA profiling has been the reopening of old cases. Persons convicted of murder and rape before DNA profiling became available have sought to have the evidence in their cases reevaluated using this new technology. In some cases, DNA test results have exonerated those convicted of the offenses and resulted in their release from prison.

The National Institute of Justice commissioned a research study of such DNA exculpatory cases. Conducted by the Institute for Law and Justice and described in this report, the study has identified 28 cases in which DNA testing led to the exoneration of persons previously convicted of murder or rape.

Most forensic scientists involved in DNA analysis have been aware that in some cases, DNA profiling has been instrumental in correcting injustices. Previously, however, almost all the information had been anecdotal. This report assembles a wealth of information on such cases, and the accounts of exculpatory DNA cases it presents will go a long way toward countering uninformed attacks on forensic DNA testing. Study results also should provide strong arguments for law enforcement officials who seek funding from State legislatures to establish forensic DNA laboratories. Furthermore, the study should completely dispel any lingering public perception of forensic DNA testing as a threat to civil liberties.

At the same time, the study also raises several important issues that need to be confronted by the legal community, law enforcement agencies, and the forensic science profession. The careful reader of this report will note the number of cases in which law enforcement agencies and prosecutors went forward with criminal prosecutions when only minimal genetic marker data were available. Critics of DNA typing, who have opposed the admission of any DNA evidence should ponder the likely consequences of such an absolute prohibition: Law enforcement agencies and forensic science laboratories would be compelled to revert to the older and less discriminating serological methods (such as ABO blood typing and polymorphic protein typing). Many innocent defendants who would be exonerated by DNA typing would instead be prosecuted because the less powerful techniques failed to exclude them.

A second important issue is the number of cases in which there was misconduct on the part of the prosecution's scientific experts. For example, the forensic serologist who testified against Gary Dotson failed to disclose that, because the alleged victim was also a type B secretor, the fraction of the male population that could have contributed the semen found on the vaginal swabs exceeded 60 percent, making the serological evidence in the case probative of very little.1 In this instance, the prosecution's expert witness failed to volunteer potentially exculpatory information but did not actually lie under oath.

Three cases discussed in this report involved expert scientific testimony by Fred Zain. Mr. Zain was a forensic serologist in the West Virginia State Police Crime Laboratory for a number of years; he then worked briefly as a forensic serologist for the Bexar County (Texas) Medical Examiner's Office. Mr. Zain's conduct as a forensic serologist was called into question when the results of a DNA test freed Glen Woodall. At Mr. Woodall's original trial, Zain testified that Woodall's ABO, phosphoglucomutase (PGM), glyoxalase (GLO), and secretor types matched those found in the semen sample. Such an event is possible but highly unlikely given that Woodall was unambiguously excluded by subsequent DNA tests. A special commission convened by order of the West Virginia Supreme Court of Appeals investigated Zain and the West Virginia State Police Crime Laboratory. As a result of this investigation, the State Supreme Court ruled that none of the testimony given by Zain in more than 130 cases was credible.² The court further ordered that Zain be indicted for perjury.³ It is sobering to reflect that but for the adventitious appearance of DNA typing, Glen Woodall would still be languishing in prison and Fred Zain might still be sending innocent persons to prison.

The advent of DNA typing will go a long way toward preventing miscarriages of justice, like the Dotson and Woodall cases, in the future. Most wrongly accused suspects will be exonerated during the initial testing of physical evidence, long before prosecution would even be considered. The quantity and quality of documentation required by laboratory quality assurance/quality control protocols preclude the wholesale falsification of test results. The minuscule quantities of DNA required for PCR-based typing procedures also allow the preservation of sufficient DNA for independent laboratory testing.

One problem that DNA testing will not remedy is inadequate legal counsel. In case after case reported here, defense counsel failed to consult competent scientific experts. Even a neophyte forensic serologist would have detected the problems with the prosecution's serological evidence in the Dotson case. It is also clear that in case after case, defense counsel failed to review the case notes of the prosecution's forensic serologists. Even a layperson would have seen that Fred Zain's written reports and sworn testimony were contradicted by his case notes. Again, one has to reflect on the likelihood that numerous innocent persons are presently incarcerated because of the inadequacy of their attorneys.

This National Institute of Justice report on DNA exculpatory cases is a unique contribution to the growing literature on forensic DNA profiling. It should be read and pondered by anyone having an interest in this burgeoning field of forensic science.

Notes
1. Webb, Cathleen Crowell, and Marie Chapian, Forgive Me, New York: Berkeley Books, 1986.  [Back]
2. "Court Invalidates a Decade of Blood Test Results in Criminal Cases," New York Times (November 12, 1993): A 20.  [Back]
3. Harper, Jane, "West Virginia Court Wants Forensics Expert Prosecuted," Houston Post (July 17, 1994): A 22.  [Back]

Commentary by Rockne Harmon
Senior Deputy District Attorney
Alameda County, California

The introduction of forensic DNA typing into the legal system was heralded as the most significant event in criminalistics since dermal fingerprint identification. Few developments ever live up to their advance billing-but DNA has!

Cases are now being prosecuted that never would have been possible before the advent of DNA typing. Many States have created DNA data bases on known offenders that they compare against unsolved crimes. Several States have produced matches from their data base searches, and a handful of these cases already have been successfully prosecuted.

About nine years after its introduction, forensic DNA typing is still used only selectively. This is due, in part, to several factors: the unavailability of forensic typing to local prosecutors, the time required to perform the typing, and the costs of the tests if private laboratories are utilized.

When forensic DNA typing is performed in cases under investigation or still pending in court, the results occasionally exonerate a suspect or suspects. Such cases rarely are front-page news because the tests have served their purpose. Investigators can redirect their efforts to alternative suspects. Prosecutors can dismiss charges filed against innocent suspects.

This report reviews more than two dozen cases in which forensic DNA typing ultimately exonerated suspects or defendants. Most were prosecuted at a time when forensic DNA typing was not available to police or prosecutors. Each case has a slightly different sequence and series of events. Because of these differences, each case provides additional insight into how the legal system might avoid the pitfalls of the past, whether or not the testing is performed in pending or postconviction cases.

Some already have used the cases discussed in this report to argue that hundreds more innocent defendants are in prison. They contend that the current "exclusion" rate for forensic DNA labs-close to 25 percent-suggests that a similar percentage of innocent defendants were wrongly convicted before the availability of forensic DNA typing. Unfortunately, too many variables are contained in the "exclusion" rate to draw any meaningful conclusions from it. Furthermore, nothing about the cases reviewed here necessarily supports such a conclusion.

The only clear conclusion that can be drawn is that this new technology can be used within the existing legal framework to undo past injustices. In other words, both the science and the legal system worked in these cases! This report provides additional insights into how such cases can be identified in the future.

Commentary by Ronald S. Reinstein

Presiding Judge, Criminal Department Superior Court of Arizona, Maricopa County
This report is an excellent example of the marriage between science and law and of the invaluable resource that DNA evidence has become in the forensic field. When justice can be served in such dramatic fashion by the exoneration of previously adjudged guilty individuals, science demonstrates its practical effect.
Yet the 28 cases cited in the report relate only to individuals released from prison because of DNA testing. Vastly more far-reaching in the long run is the use of DNA typing both to exclude some suspects who otherwise might be charged and to identify many other suspects who might not have been charged but for the DNA typing.

What is frustrating to many who are excited about the possibilities of the use of DNA in the forensics area is the slow pace it is traveling on the road to admissibility. Many jurisdictions do not have sufficient funds to establish their own laboratories or to send to private laboratories items of evidence for typing. Laboratories that perform testing often have backlogs measured in months. Courts, prosecutors, and defense counsel impose a great burden on laboratories' time in the usual discovery battles that occur whenever a new technique arrives on the forensic scene.

It is interesting to observe how quickly some DNA evidence opponents embrace the science when it benefits certain defendants' interests but how defensive they become when the evidence points toward other defendants. But this is not unique to DNA evidence.

It is the responsibility of the court to promote the search for truth. If that search can be assisted by science that can give reliable results, the whole system as well as society benefits. It is also the responsibility of the court to try to prevent juror confusion caused by lawyers and experts who sometimes seem unable to explain scientific evidence in language the jury understands.

The future should be brighter as the technology improves so that the process of DNA typing will likely become much quicker, less complex, and less expensive. The battle of the experts, it is hoped, will also subside eventually, especially in the confusing area of the statistical meaning of a match.

The conflict between various forensic experts, population geneticists, and statisticians on "the meaning of a match" is a prime example of how science and the law sometimes do not mesh, especially in jurisdictions that follow the Frye test of general acceptance in the scientific community. The numbers being bandied about by various experts are almost beyond comprehension for trial jurors.

It seems logical to allow relevant, reliable, qualitative expert opinion-for example, that the probability of a random match in DNA testing is extremely remote given a reliable multi-locus match. Likewise, experts should be able to testify from their experience about whether they are aware of random matches at four or five loci of unrelated individuals, and whether one evidence sample matches another to a reasonable degree of scientific certainty. There is a serious question about whether DNA-match testimony should be treated any differently from that of fingerprints, bite marks, hair and fiber samples, ballistics, shoe prints, and the like.

Restrictions currently imposed in some jurisdictions on the use of DNA evidence unreasonably divest such evidence of its compelling nature. If our justice system's goal is the continuing search for truth, as evidenced by the results of the study described in this report, then a similar argument can be made for the admissibility of relevant and reliable DNA-match testimony in our courts.

Commentary by George W. Clarke and Catherine Stephenson
Deputy District Attorneys
San Diego County, California

The study described in this report highlights significant aspects of the use of DNA evidence in the investigation and prosecution of criminal cases. While DNA typing is employed in various types of criminal cases (e.g., murder, robbery, kidnapping), the majority of DNA investigations entail sexual assault offenses. Indeed, in all of the cases reported in this study, sexual assault was alleged alone or in tandem with other crimes.

That the majority of DNA profiling cases concern sexual assault-usually rape-is not surprising. In few other criminal endeavors is the perpetrator as likely to deposit significant physical evidence. Occasionally, that evidence is hair, blood, or saliva; more often it is semen. Of the 28 cases reported in this study, all but two appear to have involved the analysis of the sperm component of the semen. Sexual assault cases by their very nature normally include evidence rich in DNA profile evidence.

Our enthusiasm for the use and interpretation of DNA typing, however, should be tempered inasmuch as the vast majority of sexual assault cases involving both child and adult victims do not require resolution of identity. The majority of child and adult sexual assault cases presented to us for determinations of whether to file criminal charges involve a perpetrator known to the victim. The defense normally presented is consent. In other cases, there is a denial that any sexual act occurred at all. These cases frequently do not involve physical evidence of sexual assault (injury, semen, saliva). This absence of physical evidence can be due to delay on the part of the victim in making a report to the police or to the very nature of the act, such as fondling, which is unlikely to result in the deposit or recovery of trace evidence. In such cases, the prosecutor first must resolve whether an assault even took place.

This report emphasizes that in those cases where identity is an issue, law enforcement officers must be diligent in the search for DNA evidence both at the scene and in or on the victim. Careful and timely collection and preservation of evidentiary material is critical. Collecting the bed sheets before they are washed and recovering evidence from the victim before the victim showers are important components of effective investigation. Thorough, well-documented, and honestly disseminated interviews of the victim are equally critical.

Forensic DNA typing laboratories-as numerous commentators have noted-encounter rates of exclusion of suspected attackers in close to 25 percent of cases. Careful examination of such results is commonly required whether in the pre- or postconviction setting. Typing results that exclude a suspected assailant may not demonstrate innocence. Not uncommonly, evidence collected and subjected to DNA profiling may reveal results from biological material left by other consensual sexual partners unrelated to the offense investigated or from other individuals having contact with the victim. Consideration of those results in the context of all other evidence in a specific case is essential to the determination of what took place. Law enforcement officers, prosecutors, and judges must conscientiously undertake such examinations in order to fulfill the fact-finding functions with which they are entrusted.

As this report notes, Judges and juries may soon routinely expect DNA typing evidence in sexual assault cases as the use of DNA technology becomes more widely known. DNA profiling evidence can speak, but not with the passion of a victim's voice. DNA typing results can shed light on "who"; it cannot explain precisely when, or how, or even why. The victim who survives the sexual assault must always be the primary and most important source of information.

Commentary by Matt L. Rodriguez
Superintendent of Police
Chicago Police Department

Criminal justice in the United States is a system founded on skepticism. "Innocent until proven guilty" and "beyond reasonable doubt" reflect more than the systematic doubt and deferred judgment that are afforded individuals accused of crime in our society. These maxims help define the incredibly high standards that the system's practitioners must meet before someone can be judged guilty.

In recent years police and prosecutors have increasingly turned to technology as a way to achieve these standards of proof with greater efficiency and effectiveness. Throughout the Nation, law enforcement agencies have entered an era in which high technology is not only desirable but also necessary to combat crime and ensure justice. Recent advances in forensic and biometric technologies, in particular, have created enormous opportunities for law enforcement to identify offenders with greater speed and certainty.

But while new technology presents opportunities, it is not without its challenges. The rate of change in technology, already fast-paced, is accelerating rapidly. And the demands on law enforcement are increasing dramatically in terms of both case volume and complexity. This environment of change exerts tremendous pressure on today's law enforcement administrators. Not only must we figure out what new technology to acquire and when to acquire it, but, just as importantly, we must ensure that our internal policies and operational procedures are keeping pace with advances in technology.

This study of DNA analysis in exculpatory cases highlights-in a very "real world" manner-both the opportunities and the challenges that this particular technology poses for law enforcement.

As a forensic science tool for criminal justice, DNA analysis has a relatively short history, dating back to groundbreaking cases in the late 1980s. What is significant about this "start date," from a law enforcement perspective, is that it stands in stark contrast to the age and experience levels of many of our police officers, especially those in larger cities. With an average age oftentimes of 40 or more, and with many police officers having 15, 20, or more years of experience, police departments today are populated with officers who did not grow up with DNA analysis and similar technologies. The result is that many agencies are still playing "catch up" when it comes to operating in today's high-technology world.

At the same time, the O.J. Simpson case and other recent sensational trials have put law enforcement under an intensely powerful microscope, examining our most basic procedures for collecting, processing, and caring for evidence. Although such scrutiny is never comfortable, it is appropriate and welcome, for the ultimate test of what we do in policing is in the courtroom. Increased scrutiny has challenged police departments to become more knowledgeable about DNA technology and more professional in evidence collection and processing. How we respond to this challenge will be crucial to our success and to the cause of justice in an even higher tech future.

Typically, when faced with challenges of this magnitude, law enforcement's first reaction is to concentrate on the specialists within our profession-in this case, the evidence technicians and crime laboratory analysts. These people are certainly critical to the effective processing of evidence, especially in the current environment of scrutiny and technological sophistication. But it is a mistake for law enforcement to focus solely on these specialists. Extensive and up-to-date training and procedures need to be provided to all of our police officers.

As the first responders to most crime scenes, patrol officers in particular must be aware of the potential opportunities and pitfalls posed by DNA technology, just as they must be extremely sensitive to the full range of evidentiary matters involved in protecting and processing crime scenes. Up and down the chain of command as well, police personnel must become more knowledgeable about DNA technology and more aware of, and responsive to, its implications for crime-scene and evidence processing. In the post-O.J. Simpson era, the handling of evidence until it reaches the crime laboratory will be as important as the laboratory technology, conditions, or procedures themselves.

Although the challenges posed by DNA analysis are many, they are outweighed by the enormous possibilities the technology presents. DNA analysis is a powerful and often necessary tool for establishing the presence or absence of someone at a crime scene. Readers of this study must remember that this issue cuts both ways.

In the future we must reduce the likelihood of innocent persons being wrongly convicted, just as we must increase the chances of guilty parties being identified and held responsible for the crimes they commit. This can be achieved through continued refinement of DNA technology, coupled with better training and procedures to ensure that evidence is skillfully gathered, stored, and submitted for analysis. When used properly and appropriately, DNA analysis can permit us to address the skepticism and doubt that are intrinsic to our system of justice.

Commentary by Peter Neufeld, Esq. and Barry C. Scheck
Mr. Scheck Is Professor of Law and Director of Clinical Education Benjamin N. Cardozo School of Law New York, New York

Postconviction DNA exonerations provide a remarkable opportunity to reexamine, with greater insight than ever before, the strengths and weaknesses of our criminal justice system and how they bear on the all-important question of factual innocence. The dimensions of the factual innocence problem exceed the impressive number of postconviction DNA exonerations listed in this report. Indeed, there is a strong scientific basis for believing these matters represent just the tip of a very deep and disturbing iceberg of cases. Powerful proof for this proposition lies with an extraordinary set of data collected by the Federal Bureau of Investigation (FBI) since it began forensic DNA testing in 1989.

Every year since 1989, in about 25 percent of the sexual assault cases referred to the FBI where results could be obtained (primarily by State and local law enforcement), the primary suspect has been excluded by forensic DNA testing. Specifically, FBI officials report that out of roughly 10,000 sexual assault cases since 1989, about 2,000 tests have been inconclusive (usually insufficient high molecular weight DNA to do testing), about 2,000 tests have excluded the primary suspect, and about 6,000 have "matched" or included the primary suspect.¹ The fact that these percentages have remained constant for seven years, and that the National Institute of Justice's informal survey of private laboratories reveals a strikingly similar 26 percent exclusion rate, strongly suggests that postarrest and postconviction DNA exonerations are tied to some strong, underlying systemic problems that generate erroneous accusations and convictions.

It must be stressed that the sexual assault referrals made to the FBI ordinarily involve cases where (1) identity is at issue (there is no consent defense), (2) the non-DNA evidence linking the suspect to the crime is eyewitness identification, (3) the suspects have been arrested or indicted based on non-DNA evidence, and (4) the biological evidence (sperm) has been recovered from a place (vaginal/rectal/oral swabs or underwear) that makes DNA results on the issue of identity virtually dispositive.

It is, of course, possible that some of the FBI's sexual assault exclusions have included false negatives. False negatives could occur, for example, because of (1) laboratory error; (2) situations where the victim of the assault conceals the existence of a consensual sexual partner within 48 hours of the incident and the accused suspect did not ejaculate (if the suspect ejaculated, the DNA should be identified along with the undisclosed sexual partner); or (3) multiple assailant sexual assault cases where none of the apprehended suspects ejaculated (the FBI counts the exclusion of all multiple suspects in a case as just one exclusion). Nonetheless, even with these caveats, it is still plain that forensic DNA testing is prospectively exonerating a substantial number of innocent individuals who would have otherwise stood trial, frequently facing the difficult task of refuting mistaken eyewitness identification by a truthful crime victim who would rightly deserve juror sympathy.

Without DNA testing, the prospects of wrongful convictions in these exclusion cases are evident. Even if one assumes half the normal conviction rate (State conviction rates for felony sexual assaults average about 62 percent), one would expect that hundreds of people who have been exonerated by FBI DNA testing in sexual assault cases over the last seven years would have otherwise been convicted.

The Institute for Law and Justice report does not purport to be more than a quick survey, based primarily on press clippings and summary interviews, of postconviction DNA exoneration cases, and it does not undertake any systematic analysis of them. Since we have been, through the Innocence Project at Cardozo Law School, either attorneys of record or assisting counsel in the vast majority of these cases, we have attempted to investigate, with care and in detail, some of the factors that have led to the conviction of the innocent.²

Interestingly, in many respects the reasons for the conviction of the innocent in the DNA cases do not seem strikingly different than those cited by Yale Professor Edwin Borchard in his seminal work, Convicting the Innocent (Garden City Pub., 1932), which reviewed 65 cases, and more recently by Hugo Bedau and Michael Radelet in In Spite of Innocence (Northeastern University Press, 1992), which reviewed 416 erroneous convictions in death cases from 1900 to 1991. Mistaken eyewitness identification, coerced confessions, unreliable forensic laboratory work, law enforcement misconduct, and ineffective representation of counsel, singly and often in combination, remain the leading causes of wrongful convictions.

There are, however, historically unique aspects to the DNA exoneration cases. Most significantly, both the postconviction cases described in this report and the prospective sexual assault exclusions produced by the FBI and other laboratories create an opportunity for groundbreaking criminal justice research.

Take, for instance, just the FBI's sexual assault cases. One can confirm among these cases, with greater scientific assurance than is ordinarily provided by a trial verdict, which suspects charged were truly innocent and which suspects were truly guilty. We believe it crucial to identify, prior to any DNA testing, precisely what factors in the investigatory and charging process produced incorrect results in some of these cases and correct results in others. Are there systemic weaknesses that can be identified in eyewitness identification procedures, crime scene investigations, non-DNA laboratory tests (hair, fiber, etc.), police interrogation techniques, or other investigatory methods used by police and prosecutors that are conducive to false or true arrests and convictions? Perhaps there has never been a richer or more exciting set of cases for criminal justice researchers to explore in terms of shedding light on how law enforcement methods impact the crucial problem of factual innocence.

Finally, notwithstanding, the research opportunities presented by the postarrest and postconviction DNA exoneration cases as to how wrongful accusations and convictions occur, the most significant implication of these cases is already apparent-the extent of factually incorrect convictions in our system must be much greater than anyone wants to believe. Postarrest and postconviction DNA exonerations have invariably involved analysis of sexual assault evidence (sperm), even if a murder charge was involved, that proved the existence of mistaken eyewitness identification. Since there does not seem to be anything inherent in sexual assault cases that would make eyewitnesses more prone to mistakes than in robberies or other serious crimes where the crucial proof is eyewitness identification, it naturally follows that the rate of mistaken identifications and convictions is similar to DNA exoneration cases.

The recently passed anti-terrorism bill contains a sweeping and unprecedented curtailment of the fight to obtain postconviction habeas corpus relief in the Federal courts: Strict time limits (one year in nondeath cases, three months in death cases) have been set for fling the writ; State court factual findings are "Presumed to be correct"; State court misinterpretations of the United States Constitution are not a basis for relief unless those misinterpretations are "unreasonable"; and all petitioners must show, prior to obtaining a hearing, facts sufficient to establish by clear and convincing evidence that but for the constitutional error, no reasonable factfinder would have found the petitioner guilty.

In short, just as DNA testing, the most important technological breakthrough of twentieth century forensic science, demonstrates that the problem of wrongful convictions in America is systemic and serious, Congress and the President, in our view, have eviscerated the "great writ" that for two centuries provided relief to those who were unjustly convicted. Hopefully, before this century closes, as the ramifications of the DNA exoneration cases become better understood, this triumph of political expediency over America's traditional concerns for liberty and justice will be redressed.

Notes

1. Although there is no sure way to determine what the results would have been on the inconclusive tests if results had been obtainable, it seems a fair assumption, given the strong trends over a seven-year period, that the percentages of exclusions and inclusions of the primary suspect would have run about the same as the cases where results were obtainable. Indeed, since most of the FBI's cases since 1989 involved RFLP tests, which require greater amounts of sample than PCR-based testing, it would be interesting to test this hypothesis by performing PCR tests on some of the old inconclusive cases where primary suspects were either acquitted or convicted.  [Back]