C. Ronald Huff, Ph.D., is dean of the School of Social Ecology and professor in the Department of Criminology, Law and Society at the University of California, Irvine. He is a fellow and past-president (2000–2001) of The American Society of Criminology.

Much has been made of recent cases in which innocent defendants have been released from long jail terms thanks to DNA evidence. Yet the biological evidence upon which DNA tests are performed is not available in most cases, making proof of innocence much more challenging. Despite our justice system’s elaborate criminal procedure—the presumption of innocence, the requirement of proof "beyond a reasonable doubt"—it is likely that thousands of innocent persons are convicted of serious crimes each year, with many subsequently deprived of their liberty for years before the error is uncovered—if ever.

How extensive is this problem? What are its principal causes? Does the frequency of error vary across jurisdictions within one nation, such as the United States? Does the frequency of error vary across nations? If so, how much is that variation attributable to differences in types of justice systems (e.g., adversarial versus inquisitorial/continental)? How much variance is attributable to cultural differences, such as differing levels of tolerance for social injustice? Do societies vary in their tolerance for the conviction of the innocent as opposed to the acquittal of the guilty? Are such errors the inevitable cost of operating a system of criminal justice? What can be done to prevent or reduce wrongful conviction? How should we compensate the innocent persons who are convicted and, in many cases, incarcerated?

In societies that value the freedom of their citizens and have done so much to protect that freedom, it is arguable that being convicted of a crime that one did not commit, and being incarcerated with criminals or even put to death, represents one of the worst nightmares imaginable. And yet, when we consider the two major types of errors—false positives and false negatives—we find much more preoccupation with the question of the guilty going free, probably because of the public safety implications of freeing a criminal, who might then victimize others. The irony, of course, is that these two errors are inversely related, and every time we convict an innocent person we leave the actual offender free to continue victimizing citizens. Therefore, although the issue of wrongful conviction is generally portrayed as a so-called liberal issue focusing on the rights of the accused, it is every bit as much an issue that affects public safety and should be of equal concern to so-called law-and-order conservatives.

How and why do such miscarriages of justice occur? A colleague and I examined this question in some detail utilizing an extensive survey in the State of Ohio and a national survey of attorneys general. In carrying out our research, we utilized a very conservative criterion in defining "wrongful conviction"—that is, only those convicted of a felony and later officially cleared were included in our database. This criterion is so conservative that it excludes several important groups. From the viewpoint of injustice, the most significant of these groups consists of people who have been held before trial for lengthy periods of time before being cleared of the charges. Since they were not convicted, we do not include them, even though the sociobehavioral and system dynamics that produced their erroneous arrest and kept them confined for a long time prior to adjudication are generally parallel to the factors discussed below.

We excluded from our definition those found not guilty in a second trial or on appeal because of the exclusion of crucial evidence due to illegal searches and seizures or other violations of suspects’ rights. We do this because we believe there is an important distinction between being found "not guilty" by our legal system and actually being innocent. In many cases, one cannot infer actual innocence from subsequent acquittals or reversals.

In our database, we include cases involving official acknowledgment of error based on the following criteria: (1) a new trial was permitted and the defendant was found not guilty; (2) a pardon was granted due to new evidence; (3) innocence was established on the basis of overwhelming evidence; or (4) appellate court review proved innocence. In our earlier study, we used these criteria in examining nearly 500 cases of known error and developed an estimate of the magnitude of this phenomenon, based on our survey. Our earlier study was based on a conservative national sample of attorneys general of all 50 states and four territories of the United States (N = 54), as well as an Ohio sample of prosecutors (N = 88), presiding judges of felony courts (N = 88), public defenders (N = 28), county sheriffs (N = 88), and the chiefs of police of Ohio’s largest cities (N = 7). Our response rate from the survey was nearly 65 percent and all subsamples were well represented, with response rates ranging from 60 percent for prosecutors to 76 percent for our national sample of state attorneys general. In addition to the survey data, we identified cases through computerized searches of newspapers, magazines, and journals.

It is important to acknowledge that there is no certain way of knowing how many innocent persons have been wrongfully convicted of crimes. Although criminologists routinely discuss the underreporting of crime and have developed alternate measures that better capture the true rate of crime, we have not developed systematic methods of reporting and measuring wrongful conviction. To address this issue, our survey respondents were asked to estimate, based on their experiences, the frequency with which wrongful conviction occurs. Based on the responses to our survey, we used 0.5 percent as our estimate. If our "panel of judges" is correct, this means that the U.S. criminal justice system might be accurate in about 99.5 percent of the cases of felony conviction. That suggests, perhaps, a level of accuracy that might inspire great confidence. However, one’s perspective on the magnitude of the problem might change when one considers the overall volume of cases processed through the U.S. criminal justice system. For example, in the year 2000 there were 2.2 million arrests in the United States for index crimes alone. We also know that about 70 percent of those arrested for felonies are ultimately convicted of either a felony or a misdemeanor. This means that if we assume that the system was 99.5 percent accurate in those cases and made errors in only one-half of 1 percent (0.5 percent) of those convictions, that rate of error would have produced about 7,500 wrongful convictions among those 2.2 million arrested for index crimes. So a small error rate in a very large system can result in thousands of miscarriages of justice and allow many of the criminals who actually committed those crimes to remain free to victimize others.

Another reason for concern regarding the accuracy of the U.S. criminal justice system is the recent, highly publicized "DNA exonerations" of persons who had been convicted of serious crimes and, in some cases, were on death row awaiting execution. In fact, the governor of Illinois declared a moratorium on the death penalty in that state following a period of time in which more death row inmates had been exonerated than had been executed. (In a highly controversial move, Governor George Ryan commuted all Illinois death sentences to life in prison just shortly before he left office in January 2003. Four inmates were pardoned and released.) Recently, in 18,000 criminal cases where biological evidence was available, DNA testing excluded more than 25 percent of prime suspects. We also know that in 25 years of experience with the death penalty, 553 persons were executed while 80 were released and had their sentences vacated, suggesting the possibility of a significant error rate. And finally, a recent study of capital cases that spanned more than 20 years raised the possibility of an even higher incidence of error.

Research has demonstrated that the following major factors contribute to wrongful
conviction:

Eyewitness error. The evidence strongly supports the conclusion that the single most important factor leading to wrongful conviction, at least in the United States and England, is eyewitness misidentification. In our own database of nearly 500 cases, eyewitness error was involved in nearly 60 percent of the cases. Such cases include those where there is a close resemblance between the actual offender and the innocent defendant, and some where the accused person bears little if any resemblance to the actual criminal—such as a foot difference in height or being of a different racial/ethnic heritage. Unfortunately, although experts and judges increasingly share the view that eyewitness identification of a suspect previously unknown to the victim/witness is highly suspect and perhaps the least reliable kind of evidence, jurors still tend to attach special importance to such identification and see the testimony of eyewitnesses as powerful. It’s as if the jurors hold the view that if a person was there when the crime occurred, then surely that person should be able to identify the criminal because, after all, they "saw it happen." But what did they really "see"? This paper cannot discuss in detail the factors that affect eyewitness identification, but these factors cluster into the following categories: (1) psychological factors, such as perception and memory; (2) systemic factors, such as the misleading and improper use of interrogation or lineups; and (3) societal and cultural factors, such as the difficulties inherent in cross-racial identification.

Unethical police/prosecutorial behavior. The police or the prosecutor, convinced of a suspect’s or a defendant’s guilt, sometimes seek to strengthen the case by engaging in overzealous and unethical behaviors such as "coaching" witnesses, fabricating or concealing evidence, and committing perjury in order to "get their man." Generally, their behavior is well intentioned, rationalized by their belief that they are taking a "scumbag" off the streets and "just helping the jury." Also, prosecutors bent on seeking convictions, rather than justice, often fail to advise defense attorneys of potentially exculpatory evidence. And research shows that in cases that are reversed due to prosecutorial misconduct, it is extremely rare for prosecutors to be disciplined at all, even when knowingly allowing perjured testimony or deliberately concealing potentially exculpatory evidence.

Plea bargaining. Plea agreements between the prosecution and the defense account for the great majority of convictions in the United States. Thus, it is not surprising that a number of these convictions are erroneous. Many innocent defendants are wrongfully convicted after "willingly" pleading guilty. From the standpoint of the typical citizen, this may be one of the most puzzling aspects to understand. Why would someone plead guilty to a crime he or she did not commit? An earlier social psychological experiment offered important insight into this question. In that experiment, it was shown that innocent "defendants" were more likely to accept plea bargains when they faced a number of charges or when the probable severity of punishment was great. Since the outcome of a legal case is never certain, many defendants can be enticed to plead guilty, even though they are innocent, in order to avoid even more severe consequences of systemic error. This also raises serious concerns regarding capital punishment. Faced with the death penalty, who would not plead guilty in order to save his or her life and hope that the truth will emerge some day?

Ineffective assistance of counsel. The assertion that a defendant’s legal counsel was inadequate has been a basis for appeal since Powell v. Alabama (287 U.S. 45 (1932). The basic rationale of such an appeal is that the original defense counsel, for whatever reasons, did not adequately represent the client’s interests in the case. Such appeals are not easily won, despite the fact that many attorneys in the U.S. are inadequately prepared for trial work. Moreover, the investigative resources available to most defense attorneys are far less extensive than those available to prosecutors, thus calling into serious question some of the basic assumptions of our adversarial system of justice in the United States. Funding for public defenders is critically deficient in a number of locations in our nation, sometimes prompting litigation, as recently happened in Wayne County, Michigan. The suit, which includes the area covering Detroit, alleges that low rates of compensation (e.g., a maximum of $225 for investigation and preparation of a case that may result in a sentence of life imprisonment) make it impossible to provide meaningful representation to clients. On the other hand, in some areas, unscrupulous defense attorneys (we call them "guilty plea wholesalers") can earn a considerable income by lining up a large number of cases and pleading their clients guilty without adequate investigation.

Community pressure for convictions. One can see this factor at work especially during community "crime waves" or whenever a serious crime occurs and receives extensive publicity. The atmosphere is often one of fear and paranoia, resulting in tremendous pressure on the police, the prosecutor, and the courts to get the offender off the streets. Nonetheless, if a case can be built on circumstantial evidence there is a tendency among the public, the police, and the prosecution to relax and to presume guilt, rather than innocence. Then, if the investigation results in the arrest of the wrong person, the actual criminal remains at large, free to commit more crimes. As police focus their investigation on the wrong suspect, other leads are ignored and the trail to the real offender may disappear. A related issue is "free press versus fair trial" and the question of the extent to which the media may contribute to the creation of an atmosphere of fear and paranoia, rather than reason. The media can, alternatively, play a constructive role in asking hard questions and monitoring the quality of justice in the community. A good example of this was the Pulitzer Prize-winning journalist Paul Henderson of the Seattle Times, who showed that the police had arrested the wrong person in the rape case against Steve Titus. Henderson discovered that the police had altered evidence to buttress their case and that their impressive record of conviction in rape cases turned out to be an artifact of their investigative techniques and their ethical violations. Though ultimately cleared of the rape charge, Titus soon died of a stress-related heart attack.

Knowledge of criminal record. Many people believe that "where there’s smoke, there’s fire." When the accused person has a record of criminal behavior, there is a tendency for law enforcement to combine that fact with circumstantial evidence and leap to the conclusion that they have identified the perpetrator. The criminal record can also become known to the jury if the defendant takes the stand or if the testimony of other defense witnesses reveals this information. It is also sometimes common knowledge in the community.

False and coerced confessions and suggestive interrogations. Another important factor in wrongful convictions is false and coerced confessions, often related to suggestive interrogations. In a recent analysis of 62 postconviction DNA exonerations, for example, about one in four involved false confessions

Inappropriate use of informants or "snitches." Another important contributing factor is the widespread and often unprincipled use of informants, or "snitches," by police, prosecutors, and jail officers. For example, the perjury and other unethical and illegal conduct in the highly-publicized "snitch crisis" in the Los Angeles County Jail in the 1980s was ultimately linked to 225 felony convictions. Also, five of the first 13 Illinois death row inmates found to have been wrongfully convicted were prosecuted using jailhouse informants. Many DNA exoneration cases have revealed that the initial convictions involved the use of "jailhouse snitches."

Other factors. Other factors include judicial error, bias or neglect of duty; voluntary and deliberate false confessions (such as those from compulsive, guilt-ridden "false confessors" who come forth frequently in highly publicized cases); and mental incompetence of the accused.

Research indicates that in most cases of wrongful conviction, more than one of these factors is involved. For example, eyewitness misidentification may be exacerbated or even caused by unethical police practices, such as "coaching" witnesses. Or an inattentive or biased judge may not "perceive" unethical conduct by the police or prosecution or inadequate representation by counsel at trial. Generally speaking, wrongful conviction is more likely to occur when more than one part of the system breaks down.

Sometimes, the best way to learn about our own culture, or system of justice, is to view it in the context of other cultures and justice systems. With respect to the problem of wrongful conviction, some issues that seem worthy of cross-national study include: (1) the role played by eyewitness error in different cultures—including those that are more homogeneous with respect to racial and ethnic composition than is the United States; (2) the degree to which police misconduct occurs and is or is not tolerated in various societies and in various systems of justice, such as those in England, Japan, Germany, Spain, and France; (3) the extent to which the type of criminal justice system (e.g., inquisitorial/continental versus adversarial) affects the pressure placed on the prosecutor to get convictions, which, in turn, may affect the rate of error; and (4) the extent to which various societies are willing to tolerate wrongful conviction as "the price that must be paid" to control crime.

The following public policy recommendations focus on (1) the prevention/reduction of wrongful convictions, and (2) the compensation and reintegration of those who are the victims of the system—those who are convicted but innocent. Recommendations include the following:

• There is a need for public education on the subject of wrongful conviction and its dysfunctional implications for society, including both violations of individual rights and increased threats to public safety.

• In cases where eyewitness identification is the key evidence, a jury or a judge should hear, in a special pretrial session, all information related to the issue and should decide the adequacy, validity, and reliability of the eyewitness identification in each case.

• In cases where eyewitness identification is involved, the court should always permit the use of qualified expert witnesses and should issue precise cautionary instructions to juries.

• When a crime occurs, law enforcement authorities should make a special effort to conduct investigations of all eyewitnesses and victims as soon as possible, in order to minimize the consequences of memory distortion.

• No identification procedure (pre- or post-indictment) should be conducted in the absence of the defendant’s attorney once the attorney has been appointed or retained.

• Police interrogations in the United States should be electronically recorded, in full, as is now commonplace or mandatory in England, Canada, and Australia.

• The basic and advanced training of law enforcement personnel, prosecutors and their staff, and judges should include consideration of wrongful conviction, including its causes, its implications for our society and for the criminal justice system, and methods of prevention. This training should include the presentation of detailed case studies of wrongful conviction, focusing on witness errors, police and prosecutorial errors, official misconduct, and judicial failure to detect errors at trial and on appeal.

• Innocence commissions should be established at the state and federal levels in the United States. These commissions could review appeals, refer appropriate cases to trial-level courts that could then entertain collateral attacks on the conviction. They could hold hearings and decide to dismiss the appeal/referral, order a new trial, or vacate the conviction. Great Britain’s Criminal Cases Review Commission (CCRC), an independent body responsible for investigating suspected miscarriages of justice in England, Wales, and Northern Ireland, is probably the best current example. According to the most recent data available, the commission had received 5,051 applications and had reviewed 4,353 from 1997–2001. Of 101 cases that have subsequently been heard by the courts following CCRC referrals, 67 convictions had been quashed, 33 had been upheld, and one was pending. Thus, the argument that such a review commission would overburden the courts does not appear to have empirical support, at least with respect to the British experience.

• Law enforcement officials who (a) knowingly alter, conceal, fabricate, or otherwise distort evidence, (b) commit perjury, (c) intimidate or bias witnesses or suggest information to them in order to strengthen cases, (d) knowingly enter into plea agreements in bad faith, ignoring the existence of potentially exculpatory or mitigating evidence, or (e) fail to make available to the defense in a timely manner all information or evidence of a potentially exculpatory or mitigating nature should be removed from their positions of public trust and subjected to severe sanctions. For prosecutors, this should include disbarment proceedings. Such unethical behavior by officials of the justice system is incompatible with public confidence in the administration of justice.

• To assist in implementing the above proposal, state and local bar associations, prosecutors, and state attorneys general should routinely evaluate the performance of law enforcement officers, prosecutors, and defense attorneys involved in every case of known wrongful conviction. Also, judicial performance in such cases should be reviewed by the appropriate body in each jurisdiction.

• The appropriate legislative bodies should enact legislation requiring the financial compensation and socioeconomic integration of wrongfully convicted persons, including both monetary compensation (based on a formula prescribed by the legislation) and the provision of social services to assist in the reintegration process.

• In recognition of the possibility of error, those jurisdictions that permit the death penalty should abolish it and replace it with life sentences for those convicted of aggravated murder or equally heinous crimes. This would enable the state to release and compensate the wrongfully convicted upon the discovery of error.

  The implementation of a number of these recommendations would require a strong commitment from the legal profession—bar associations, judges, prosecutors, and defense attorneys. The challenge is for the legal profession to continue its assessment of the problem of wrongful conviction, as dramatically represented by the recent spate of DNA exonerations, and decide how best to implement needed reforms to reduce our rate of error. Such reforms, in my judgment, are necessary to preserve the integrity of our criminal justice system, restore/maintain the public’s (and jurors’) faith in the fairness of the system, and protect public safety by reducing the number of serious offenders who continue victimizing citizens while the wrongfully convicted go to prison.

Wrongful Convictions Bibliography

Due process generally:

• About Guilt and Innocence: The Origins, Development, and Future of Criminal Procedure by Donald A. Dripps (2003).

Overview of systemic causes:

• Report of the Governor’s Commission on Capital Punishment (Illinois, April 2002).

• Wrongly Convicted: Perspectives on Failed Justice (eds. Saundra D. Westervelt and John A. Humphrey 2001).

• Beyond Repair? America’s Death Penalty (ed. Stephen P. Garvey 2003).

Identification procedures:

• New Jersey Attorney General Guidelines for Preparing and Conducting Photo and Live Lineup Identification Procedures (April 18, 2001).

• "Eyewitness Identification Procedures: Recommendations for Lineups and Photospreads" by Gary Wells et al., in the journal, Law and Human Behavior, volume 22 (1998).

• Mistaken Identification: The Eyewitness, Psychology and the Law by Brian L. Cutler and Steven Penrod (1995).

Confessions:

• Miranda’s Waning Protections: Police Interrogation Practices After Dickerson by Welsh White (2001).

• "The Consequences of False Confessions: Deprivations of Liberty and Miscarriages of Justice in the Age of Psychological Interrogation" by Richard Leo and Richard Ofshe in the Journal of Criminal Law and Criminology, volume 88, page 429 (1998).

Jailhouse informants:

• "Cooperation with Federal Prosecutors: Experiences of Truth Telling and Embellishment" by Ellen Yaroshefsky in the Fordham Law Review, volume 68, page 917 (1999).

• Ratting: The Use and Abuse of Informants in the American Justice System by Robert Bloom (2002).

Ineffective assistance of counsel:

• "The Criminal Defense Lawyer: Zealous Advocate, Double Agent, or Beleaguered Dealer" by Rodney Uphoff in the Criminal Law Bulletin, volume28, page 419 (1992).

• "Neither Equal Nor Just: The Rationing and Denial of Legal Services to the Poor When Life and Liberty Are at Stake" by Stephen Bright, in the 1997 New York University Annual Survey of American Law, page 783.

Junk science:

• Forensic Science by David L. Faigman (2002).

• "Lab Fraud: Lessons Learned from the ‘Fred Zain‘ Affair" by George Castelle in The Champion, May 1999, page 12.

Suppression of exculpatory evidence:

• "The Prosecutors’ Ethical Duty to Seek Exculpatory Evidence in Police Hands: Lessons from England" by Stanley Z. Fisher in the Fordham Law Review, volume 68, page 1379 (2000).

• "A Defendant’s Right to Exculpatory Evidence: Does the Constitutional Duty to Disclose Exculpatory Evidence Extend to New Evidence Discovered Post-Conviction?" by Todd A. Jaworsky in the St. Thomas Review, volume 15, page 245 (2002).

Innocence projects and commissions:

• "Toward the Formation of ‘Innocence Commissions’ in America" by Barry Scheck and Peter J. Neufeld in Judicature, volume 85, page 98 (Sept. 2002).

• "The Hurricane Meets the Paper Chase: Innocence Project’s New Emerging Role in Clinical Legal Education", in Wrongful Convictions, A Call To Action by Justin Brooks et al., 655 (ed. Criminal Justice Institute of Harvard Law School 2002).

Racial bias:

• "Discrimination, Death, and Denial: The Tolerance of Racial Discrimination in the Infliction of the Death Penalty" by Stephen Bright in the Santa Clara Law Review, volume 35, page 533 (1995).

• Negrophobia and Reasonable Racism: The Hidden Costs of Being Black in America by Jody David Armour (1997).

Websites:

• www.innocenceproject.com

Selected case law:

• U.S. v. Quiones, 205 F. Supp. 2d 256 (S.D. N.Y. 2002) (holding Federal Death Penalty Act, 18 U.S.C. §§ 3591–3598 to be unconstitutional because its "result can only be the fully foreseeable execution of numerous innocent persons") (updates the Honorable Jed S. Rakoff’s tentative earlier decision in United States v. Quiones, 196 F. Supp. 2d 416 (S.D.N.Y. 2002).)

• United States v. Diabate, 90 F. Supp. 2d 140 (D. Mass. 2000) (prosecutor has duty to learn of any exculpatory evidence known to others acting on the government’s behalf in a case, including the police).

• Minnesota v. Scales, 518 N.W.2d 587 (Minn. 1994) (supervisory power of state courts authorized mandatory videotaping of all stages of suspect interrogation).

• United States v. Mathis, 264 F.3d 321, 341 (3d Cir. 2001) (usefulness to the fact-finder of an expert witness on factors affecting the accuracy of eyewitness identification).

• Commonwealth v. Bowie, 243 F.3d 1109 (9th Cir. 2001) (independent prosecutor duty to investigate the accuracy of an informant’s tip).

For a more in-depth look at the topic of wrongful convictions, the author offers the following resources:

Return to Table of Contents - Spring 2003

Return to Criminal Justice magazine home page