Wrongful Convictions

Marvin Zalman. 21st Century Criminology: A Reference Handbook. Editor: J Mitchell Miller. 2009. Sage Publications.

Wrongful convictions occur when innocent defendants are found guilty in criminal trials, or when defendants feel compelled to plead guilty to crimes they did not commit in order to avoid the death penalty or extremely long prison sentences. The term wrongful conviction can also refer to cases in which a jury erroneously finds a person with a good defense guilty (e.g., self-defense), or where an appellate court reverses a conviction (regardless of the defendant’s factual guilt) obtained in violation of the defendant’s constitutional rights. This chapter deals with the first type of wrongful convictions, or wrong person convictions. Note also that the verdict of acquittal in American law is “not guilty” rather than “innocent,” meaning that an acquitted person might not be factually innocent. For the sake of clarity, the term actual or factual innocence is used to refer to persons who did not commit the crime. Miscarriage of justice (a legal term in England) is also used to describe wrongful convictions.

A wrongful conviction is a terrible injustice that is magnified when an actually innocent person spends years in prison or on death row. This has always been recognized by the U.S. legal system. The rising number of exonerations, however, and growing awareness that such injustices occur every day in American courts, raises profound doubts about the accuracy and fairness of the criminal justice system. This understanding is supported by considerable recent research. This surge in awareness and budding research has motivated a growing number of innocence projects, which work to exonerate wrongly convicted prisoners, to also propose justice policy reforms designed to reduce the number of wrongful convictions or to alleviate their effects. This chapter explains why wrongful conviction has become a prominent issue, the scope of the problem, its causes, and reform proposals.

The injustice of being convicted and imprisoned for a crime one did not commit is intuitively apparent. Research and anecdotal evidence shows that a high proportion of wrongfully convicted prisoners suffer severe psychological consequences, including posttraumatic stress disorder and anxiety disorders, which is not typical among actually guilty prisoners in the absence of life-threatening experiences in prison. This complicates the ability of exonerated prisoners to return to a normal life after release.

More than half the states do not legally authorize financial compensation for persons who were victimized by the criminal justice system in this way, although the number of states with compensation laws has grown in recent years. Moreover, exonerated prisoners do not receive the services provided to prisoners released on parole. Newer compensation laws provide for health and restorative services, as well as financial compensation, to help exonerated prisoners. A person who has been exonerated does not have automatic grounds to sue and recover money damages against police or prosecutors. A number of such cases have been successful in recent years, but they are infrequent and successful only when specific wrongdoing by criminal justice agencies can be proven and immunity defenses overcome.

The Rise of the Innocence Movement

Prior to 1990, wrongful convictions generated only slight interest. The famous writer of the “Perry Mason” legal thrillers, Erle Stanley Gardner, created an informal “court of last resort” in the 1950s to investigate and correct miscarriages of justice. For the most part, however, the public, as well as most judges and criminal lawyers, was convinced that very few innocent people were ever convicted. When the Supreme Court expanded defendants’ trial rights in the 1960s, for example, the reason given was not to make the criminal justice system more accurate in determining guilt and innocence but to prevent government oppression.

Some pre-1990 scholarship did raise issues of trial accuracy. First, a group of cognitive psychologists began to conduct eyewitness identification experiments in the 1970s. By 1990, they had amassed a wealth of information showing that eyewitnesses were often mistaken and that lineup and identification procedures could significantly increase or decrease eyewitness accuracy. Next, a survey of criminal justice officials by criminologists C. Ronald Huff, Arye Rattner, and Edward Sagarin in the 1980s estimated that thousands of wrongful convictions occurred every year (Huff, Rattner, & Sagarin, 1996). Finally, philosopher Hugo Adam Bedau and sociologist Michael Radelet published a survey in a prestigious law journal in 1988 asserting that 350 innocent persons were convicted of capital and potentially capital crimes in the 20th century and that 23 were executed. Although a handful of these 350 might have been factually guilty, the study’s overall correctness raised awareness in the legal community that an innocent person could be executed. This scholarship did not, despite occasional news stories about wrongful convictions, create widespread concern about miscarriages of justice.

It was DNA testing, used to prove guilt with near certainty and to absolutely exclude suspects or defendants, that caused a sea change in attitudes about wrongful convictions. Previously, blood testing based on group types and other blood factors could not exclude suspects whose blood factors matched the crime sample; even though a large percentage of the population also shared those factors, prosecutors placed these “matches” in evidence. In forensic DNA testing, 13 loci (sites) in a suspect’s DNA strand that vary among people are analyzed to create a distinct DNA profile (or DNA fingerprint). The profile is compared with that of the same 13 loci in the biological sample linked to the crime (e.g., semen or blood deposited during a rape or assault). If the profiles match, based on population genetics studies, the probability that the suspect was the source of the crime scene DNA is astronomically high. If only 1 of the loci does not match, the suspect is absolutely excluded.

The first DNA exoneration in the United States occurred in 1989 and showed how DNA transformed a confusing tale of innocence or guilt into one of absolute clarity. Gary Dotson was convicted of rape in Illinois on a teenage girl’s eyewitness identification. In fact, she made up the rape story to cover her fear and shame after consensual sex with a boyfriend. Six years later she was married, got religion, and recanted her story. The police and a judge refused to believe that the recantation was true, despite her pastor supporting her truthful state of mind and the former boyfriend admitting to the consensual sex. Dotson was released on parole by the governor of Illinois in 1985, who inconsistently said that he did not believe the recantation. Dotson was reimprisoned for a parole violation in 1987. Finally, with the support of journalists and a determined defense lawyer, a DNA test was performed on the semen in the rape kit. Dotson was absolutely cleared and formally exonerated. His case became a template for tens and then hundreds of thousands of police rape investigations, which exonerated suspects in the early stages of crime investigations. By the early 1990s, the FBI laboratory reported that one quarter of all rape kit samples from police around the country were exclusions. This meant that in thousands of cases, accusations based on eyewitness identifications were wrong.

Soon, prisoners who knew they were innocent and serving time or sitting on death row for crimes that did not happen or were committed by someone else began to petition for DNA testing. Most were denied testing because of prosecutors’ resistance based on legal technicalities. However, a sufficient number of exonerations occurred by the mid-1990s to generate significant happenings. Newspapers prominently reported DNA exonerations. In New York, two enterprising law school clinical professors, Barry Scheck and Peter Neufeld, started the first law school innocence project at Cardozo Law School to pursue cases of inmates claiming innocence. Janet Reno, then attorney general of the United States, commissioned a report highlighting the weakness of eyewitness identification. The report raised the profile of the wrongful convictions issue in criminal justice and legal circles. By the late 1990s, several powerful documentaries, such as Errol Morris’s Thin Blue Line, brought the issue to moviegoers and television audiences.

In 2001, Scheck and Neufeld, together with reporter Jim Dwyer, published Actual Innocence, recounting several of their exoneration cases in gripping detail. Each case listed a specific way in which the criminal justice system had failed. This list, along with previous studies, created catalogues of what are considered causes of wrongful convictions. Although the book was well received, the major event in 2000 that did more to put wrongful convictions on the map was Illinois Governor George Ryan’s moratorium on executions. Between 1990 and 2000, Illinois had executed 12 prisoners while 13 on death row had been exonerated and freed. This so shocked Ryan that he halted all executions and set up a commission to review capital punishment in Illinois. The commission recommended many reforms, and several were enacted. Ryan’s continuing concern with unreliable death sentences led him to commute the sentences of all 167 death row prisoners and pardon 4 on the grounds of actual innocence before he left office in 2003. This led other states to impose moratoria or to end the death penalty. Exonerations have weakened support for capital punishment and raised general public awareness about wrongful convictions.

Size and Scope of the Wrongful Conviction Problem

If wrongful convictions were rare, they could be downplayed as inevitable failings of a complex human system. If they are frequent and are linked to systemic problems, then they pose a challenge to the fairness and accuracy of the justice system that calls for a public response. The issue is controversial. Some prosecutors and judges believe the number of wrongful convictions to be vanishingly small and have offered an estimate of approximately 260 a year or an error rate of 0.027% (or 0.00027). This figure is a mistaken interpretation of a study conducted by Proffessor Samuel Gross and colleagues that counted 340 known exonerations between 1989 and 2003 (Gross, Jacoby, Matheson, Montgomery, & Patil, 2005). Critics fail to note that an exoneration is not the same as a wrongful conviction (although the terms are loosely used as equivalents). Gross et al. (2005) defined an exoneration as an official act declaring a previously convicted defendant not guilty, by means of (a) a governor’s pardon on the basis of evidence of innocence; (b) a court dismissing criminal charges on the basis of new evidence of innocence (e.g., DNA); (c) a defendant being acquitted on retrial after an appeal, on evidence of factual innocence; or (d) a state’s posthumous acknowledgment that a defendant who had died in prison was innocent. The study, however, demonstrated that the 340 exonerations it catalogued were the tip of an iceberg, with the number of wrongful convictions probably reaching into the thousands.

Most known exonerations have occurred in murder and rape cases rather than more numerous crimes, such as robbery, for which unreliable eyewitness identification is the only evidence. This is so partly because DNA evidence is available in most rape cases (although 60% of known exonerations were revealed by means other than DNA testing). High-stakes capital cases also generate greater assistance to avoid executions. It is likely that more errors occur in assault, robbery, and burglary convictions based on erroneous eyewitness identification and circumstantial evidence. Studies of wrongful convictions in death penalty cases since 1973 (when the modern era of capital punishment began), as to which careful statistics are kept by the government, have estimated “wrong person” wrongful convictions at 1% to 3.5%. In Illinois, of the 289 persons sentenced to death between 1973 and 2003, 17 (or 5.9%) were exonerated and released.

Surveys of state judges, prosecutors, defense lawyers, and police officials have provided an alternate estimate of wrongful convictions—of about 1%. Although police and prosecutors give lower estimates than judges and defense lawyers, in light of what is now known about wrongful convictions a 1% felony error rate is plausible. On the basis of slightly more than 1 million state court adult felony convictions in 2004, and prison and jail rates of 40% and 30%, respectively, this rate translates to an estimated 10,790 adults wrongly convicted, of whom 4,316 were sent to prison and 3,237 wrongly jailed in 2004.

The number of wrongly convicted persons cannot be known with certainty, because no federal or state agency keeps track of exonerations, let alone wrongful convictions. Many news stories, reports, and books fairly describe wrongful convictions in detail, although not all of these wrongful convictions resulted in formal exonerations. In some of these cases, prosecutors insisted that the original verdict was accurate despite strong new evidence of factual innocence, further clouding an understanding of wrongful convictions.

Causes of Wrongful Convictions

Studies reveal several factors related to miscarriages of justice, labeled “causes,” although they are not so in a scientific sense. Typically, more than one factor is found in each wrongful conviction. Although a few wrongful convictions are caused only by honest witness error, most involve some level of negligence or malfeasance by criminal justice officers or defense lawyers. A troubling minority of cases involve perjury or knowingly dishonest action by forensic examiners, prosecutors, and police. Brief descriptions of the major factors related to wrongful convictions follow. Note that the list that follows is not comprehensive.

Eyewitness Identification

Mistaken eyewitness identification is the leading cause of wrongful convictions. It was involved in 79% of the first 200 DNA exonerations. Although an overall error rate for eyewitnesses is not well established, some experts place it at about 25%. This figure is the same as the proportion of DNA tests in rape cases conducted by the Federal Bureau of Investigation laboratory in which the DNA did not match the mistakenly identified suspect. The human memory does not record all information like a video recorder; it drops most information out of short-term memory and stores the central, but not peripheral, elements of those events in long-term memory. This makes facial recall somewhat uncertain. Events during a crime, such as extreme stress or focus on a weapon, decreases facial recall by victims and witnesses. In addition, unconscious transference can lead witnesses to superimpose the face of someone previously observed but not well-known onto the memory of the perpetrator. Memory is dynamic and can change during the recall stage from what was observed. Memory is also malleable and can change under the influence of suggestion. These and other factors show that eyewitness identification should be received with caution, and yet police, prosecutors, and especially jurors tend to rarely disbelieve eyewitness evidence.

Problems with eyewitness identification are made worse by flawed police procedures. Police have relied on “showups,” which is showing the suspect or the suspect’s photograph alone to the witness without a lineup. Courts rule that showups are suggestive and will exclude showup evidence unless one of several easy-to-produce factors is present. The showup exception factors are whether the witness paid attention, had a good opportunity to view the perpetrator, gave an accurate description, was certain, and viewed the showup shortly after the crime.

Even lineups are often flawed. Police are not always scrupulous in ensuring that the suspect does not stand out from the lineup fillers. Laboratory research shows that more errors occur when fillers are selected on the basis of their similarity to the suspect than on the basis of the victim’s description of the perpetrator. Police or prosecutors have at times suppressed the uncertain identification or nonidentification by one lineup witness while promoting the testimony of another. These and other elements often make lineups a less-than-optimal method of an accurate identification.

Forensic Science Error or Misconduct

Problems with expert evidence presented by forensic scientists or forensic examiners is the second leading cause of wrongful convictions; erroneous forensic evidence supported the convictions of 57% of the first 200 DNA exonerations. Forensic error and misconduct take a variety of forms, including problems inherent in the method, incompetent or untruthful experts, and substandard forensic laboratories.

Some expert evidence is based not on scientific testing but on comparisons that rely ultimately on the experts’ subjective evaluations. Some of these methods, such as fingerprints, bullet and tool mark examinations, and footprint and tire impressions, are relatively credible and accurate, but known errors have nevertheless occurred. If such expert evidence goes unchallenged by defense attorneys (by having other experts evaluate it), it is possible that honest but mistaken conclusions will lead to false convictions. Other kinds of expert comparison, such as handwriting analysis, are more subjective and require closer scrutiny. Even lower on the reliability scale are comparison methods that are so tentative that some label it “junk science.” Two such methods, microscopic hair analysis and bite mark impressions on skin, have caused numerous wrongful convictions. Hair analysis has now been largely replaced by DNA analysis, and bite mark evidence, although accepted in courts, has been subject to strong criticism.

Examiners have been known to err even where evidence is based on forensic science, which includes blood analysis (serology, which has been replaced by DNA analysis), drug analysis, forensic toxicology (the science of poisons), and organic and inorganic analysis of crime scene trace evidence. Even worse, in a few notorious cases forensic examiners have been exposed as pathological liars who always testified to benefit the prosecution, even when no tests were conducted. In addition to outright falsification, forensic experts can mislead courts and juries by overstating the strength of their findings, reporting inconclusive reports as conclusive, failing to report conflicting results, and the like. When expert witness perjury has been exposed, state criminal justice systems have had to reopen hundreds of cases to ensure that they did not result in wrongful convictions. Some specialized arson investigators have relied on incorrect or outdated fire science to report that fire and burn patterns were evidence of arson when this was not true.

Finally, even the most reliable methods can produce incorrect results if the forensic laboratories are substandard. As DNA testing becomes more sensitive, the risks of contamination rises unless the laboratories are in pristine condition. Testing in some inferior laboratories has even led to several people being wrongly convicted on the basis of erroneous DNA analysis. Among the worst cases was the Houston, Texas, police laboratory. A few years ago conditions were so poor that the laboratory’s roof leaked, contaminating samples with excess moisture.

False Confessions

Most people cannot understand why innocent persons confess, especially as the “third degree” (beating and torture to get confessions) has mostly disappeared from American law enforcement. Yet, false confessions were obtained in about 20% of exonerations, and at least 125 false confessions have been documented. What people do not know is that police interrogation is a “guilt presumptive” process designed to extract a confession from the guilty person who is reluctant to confess. As such, it uses powerful psychological techniques to get suspects, even innocent suspects, to talk and to confess.

When police interrogate a suspect, they are usually not trying to solve a crime because they are already convinced that the suspect is guilty, even if the investigation has not been completed. Police conduct pre-interrogation interviews to ascertain whether a suspect is truthful, but the ability of police to detect lies is no better than chance. Despite Miranda warnings, most suspects waive their rights. Laboratory experiments and case studies have shown that innocent persons waive more frequently because they know they have nothing to hide.

The interrogation setting and process create psychological pressure designed to extract a confession. The suspect is isolated and confined in a small, uncomfortable space. The interrogator forcefully asserts the suspect’s guilt and cuts off any denials or objections. In the United States police may lawfully lie to a suspect during interrogation, by, for example, falsely asserting that his or her fingerprint or DNA profile was found at the scene. The interrogation creates a sense of hopelessness in the suspect. The interrogator then develops themes, such as minimizing the seriousness of the crime, that make it psychologically easier for the suspect to admit guilt. Once an admission is made, the process moves on to generating detailed oral and written admissions or confessions.

Police interrogation produces incriminating statements or full confessions two thirds of the time. The techniques and subterfuges are so powerful that interrogation also induces innocent persons to confess. Research suggests that teens, mentally impaired individuals, and people with personality deficits are more likely to falsely confess than normal adults. Compliant false confessions are made in order to end the psychological pressure of interrogation. Some who confess naively believe that they will be released, led to that belief by subtle police statements that do not amount to the clear promises banned by the rule against coerced confessions. Others think that once they get out of the interrogation room they will be able to explain their case to a judge and have their case dismissed. Less frequently, cases of internalized false confessions occur, where the innocent suspect comes to doubt himself, after extensive and insistent police persuasion that includes false statements presented as fact, and admits that he “must have” committed the crime while in a blackout state. Even when such confessions are retracted, they play a strong role in convicting innocent suspects.

Perjury: Perpetrators, Informants, Jailhouse Snitches, and Criminal Justice Personnel

Perjury in various forms is common in wrongful convictions. Informant perjury was a factor in 49% of the first 111 death penalty exonerations. In capital cases the real perpetrator is often a suspect, and in several cases where police focused on an innocent person, the actual killers led investigators further astray with false testimony. Witnesses frequently lie to police for a number of reasons, and although police tend to believe that they are proficient at detecting witnesses’ deception, scientific studies show that investigators do no better than chance at detecting liars. In laboratory studies, all groups (whether police or students) could identify falsehoods about half the time.

Miscarriages of justice happen often when police pay informants to supply incriminating information about suspects, which the informants then fabricate. Informants are often criminals and can be paid in many ways: money, dropped criminal charges, leniency when they are charged with serious crimes, or favors to friends or family. This gives the most untrustworthy people incentives to lie, and police handlers often fail to properly screen their stories. A pernicious type of informant is the jailhouse snitch. Numerous false convictions are obtained in part on the testimony of jailed snitches who claim that the innocent suspect confessed the crime to them or made incriminating statements. It is not all that difficult for a clever snitch to get enough information about a case to make up a plausible story for the prosecution to use. In some jails the use of snitches has become so routine as to suggest willful blindness on the part of police and prosecutors.

Unfortunately, there are several recent notorious cases in which rogue police officers have framed innocent people for drug and weapons possession. Although this kind of corruption is rare, when it happens it requires officials to reinvestigate hundreds of convictions. Police and prosecutors have great discretion in conducting investigations and trying cases. Although the overwhelming majority are honest, their opportunity to cover the truth requires internal vigilance on the part of these agencies.

Ineffective Defense Counsel

Most defendants are poor (indigent) and rely on government-paid assigned counsel or public defenders rather than retained lawyers. Indigent defense is chronically underfunded, making it difficult for competent attorneys to routinely provide adequate defense. Studies in several states have shown a higher proportion of defense lawyers in exoneration cases with poorer disciplinary records than average, offering proof that substandard lawyering is a cause of wrongful convictions. The U.S. Constitution requires effective assistance of counsel for defendants, but the Supreme Court’s standards for determining ineffective assistance are weak and require proof that attorney negligence caused a verdict. Only 38 of the first 200 (29%) DNA exonerees raised ineffective-assistance claims on appeal, reflecting the difficulty of making this kind of challenge, and only 4 received a reversal on ineffective-assistance grounds.

Egregious cases of defense attorney misconduct in court have ranged from sleeping or total unpreparedness to drunkenness and being high on drugs. Even ordinarily competent defense lawyers have failed to prevent the conviction of innocent clients in ways too numerous to catalogue. Among the most serious underlying problems are failures to adequately investigate case facts and failing to properly challenge prosecutors’ witnesses, including forensic experts. Although the wrongful conviction literature does not list ineffective assistance as the highest cause, in a sense there is a failure by the defense in every wrongful conviction.

Prosecutorial Misconduct

Prosecutorial misconduct, whether or not it leads to wrongful convictions, is common. In-court misconduct includes making inflammatory comments or mischaracterizing evidence to the jury, allowing witness perjury (suborning perjury), or permitting snitches to lie about their payoffs for testifying. Prosecutors have even been known to destroy evidence. The suppression of exculpatory evidence (that which points to innocence), in violation of Supreme Court rules, appears in many wrongful conviction cases. Suppressing exculpatory evidence is a cloudy issue because it is up to the prosecutor to determine in the first instance whether the evidence is exculpatory.

When DNA testing became standard in the 1990s, a large proportion of prosecutors, all of whom welcomed DNA as an investigation tool, strongly resisted postconviction, postappeal petitions by prisoners seeking to test DNA crime scene samples in storage. Such resistance added to the frustration and tragedy of actually innocent prisoners, and it delayed justice. In a few cases, any chance of getting to the truth was terminated when existing DNA samples in evidence lockers were deliberately destroyed after prisoners petitioned for testing.

Prosecutorial misconduct is especially significant because prosecutors are the most powerful figures in the criminal justice process, with great discretion as to whether to charge suspects or to dismiss cases. Before prosecutions are formally initiated, prosecutors have a judge-like role. They dismiss one quarter of all cases filed by police, often because they believe that the suspect is innocent. Two theories guide prosecutors in their discretionary decisions and in the way they prosecute their cases: the (1) adversary role and (2) the minister of justice role. In the adversary role a prosecutor can go forward with a case in which the evidence is equivocal, on the theory that it is up to the jury to decide whether a defendant is guilty beyond a reasonable doubt. As a minister of justice a prosecutor must be personally convinced that the defendant is guilty beyond a reasonable doubt. A problem with the adversary role is that prosecutors holding that view will tend to have a win-at-any-cost attitude, likely resulting in fewer dismissed cases, more aggressive trial tactics, more instances of misconduct, and greater opportunities of generating wrongful convictions. The minister of justice role requires the difficult human and institutional ability to balance vigorous prosecution with fairness and decency. This balance was captured in a 1940 speech to federal prosecutors made by Attorney General Robert Jackson, later a U.S. Supreme Court justice:

Your positions are of such independence and importance that while you are being diligent, strict, and vigorous in law enforcement you can also afford to be just. Although the government technically loses its case, it has really won if justice has been done.

Police Investigation

At its best, police investigation is the patient, systematic, and dispassionate search for, discovery of, and evaluation of all relevant facts of a suspected crime. The goal is to establish whether a crime was committed and to identify and apprehend the perpetrator(s). This complex and sensitive task requires solid understanding of criminal law, extensive knowledge about criminal behavior and crime patterns, complete familiarity with the methods of evidence collection and analysis of forensic evidence, skill in interviewing witnesses, and good analytic and writing abilities.

Although it is inevitable that some crimes will not be solved and some innocent suspects will be mistakenly identified, several factors increase the probability of error. One is tunnel vision, which in fact affects all people and all criminal justice system participants. This term encompasses well-established psychological cognitive mechanisms such as confirmation bias, or the human tendency to seek and interpret new information in ways that confirm preconceptions and avoid information and interpretations that contradict prior beliefs. When applied to police investigation, tunnel vision is the tendency to focus on the first suspect and then to select and filter evidence that builds a case for guilt, while ignoring or suppressing exculpatory evidence. Tunnel vision is an unconscious or “natural” and entirely nonmalicious process. However, its worst effects are amplified by aspects of police investigation such as the nature of interrogation, discussed earlier.

Another problem is the nature of the police investigation report. This document is tremendously important, because in most cases there is limited or no investigation by the defense (in large part because of severely limited funds) and so the police case, found in the report, becomes the official facts in the case. The police report is relied on heavily by the prosecutor in deciding whether and what crimes to charge, by the magistrate in setting bail and ordering detention for psychological evaluation, and even by defense attorneys who do not independently investigate their client’s cases for plea bargaining and trial purposes. It is the basis on which an officer testifies at trial and can influence sentencing decisions.

In contrast to European countries, where police investigation is supervised by investigative magistrates who are well-trained judicial officers in national ministries (departments) of justice, American police investigation (aside from federal cases) is almost entirely under the control of local police departments. European police reports are highly detailed, part of the official dossier, and geared to ascertaining the truth; American reports are internal documents that serve functions other than informing the prosecution, such as evaluating personnel. Police are not specifically trained to include exculpatory information in reports, despite orders to include “all” information. Severe time pressure makes it difficult for police to write comprehensive reports, and police are trained to not report information that could lead to civil suits against themselves or their departments. Studies indicate that police reports do not normally contain exculpatory facts, and in fact are often deficient in reporting inculpatory facts, to the discomfit of prosecutors. A few cases have uncovered deliberate fabrication and the exclusion of exculpatory evidence in police reports that framed innocent suspects. It is not known how pervasive these behaviors are. What is probably more common is for overworked police officers to focus on the initial suspect and enter facts in their field notes and follow-up reports that confirm their initial suspicion. Once written, there is no regular procedure to incorporate contradictory or exculpatory evidence in police reports.

Other Causes, Root Causes

This list of wrongful conviction causes is not comprehensive. Race may play a part, either by blatant discrimination, subtle bias, or as a result of the weaknesses of cross-racial identification. This last results not from bias but from familiarity that allows people to see subtle facial features among people with whom they are familiar. A disproportionate number of wrongful convictions have occurred against African American men convicted of raping white women.

The death penalty may also generate a higher proportion of wrongful convictions. Police, under extreme pressure to solve capital murders, rely on marginal evidence to fasten their attention on suspects. The American war on crime and its hyperimprisonment (at five to eight times the levels found in other advanced democracies with comparable crime levels except for homicide) has created a proprosecution atmosphere, leading police, prosecutors, judges, juries, and appellate courts away from balanced decision making, probably resulting in more miscarriages of justice.

In addition, pervasive root causes of wrongful convictions exist. One, discussed earlier, is tunnel vision. Another is the lack of resources for all actors (police, prosecutors, and courts, as well as defense lawyers). This creates extreme pressure to investigate cases within limited time periods. Overlapping with pressure is the system’s normal bureaucratic functioning and production demands, which channel the work of justice officials into routines that make it difficult to slow case processing for more careful examination where called for. These bureaucratic imperatives are aggravated by structural factors that may be impossible to change. The American governmental and criminal justice system is the most fragmented of any modern nation. There are substantial differences in quality among the 16,000 local police departments and 3,000 prosecutors’ offices in the nation. The election of prosecutors and judges, virtually unheard of anywhere else in the world, injects a level of partisanship into criminal justice that often undermines rational action. These factors may in turn create a culture of impunity among investigators and prosecutors in which errors are seldom restrained and misconduct rarely punished. The adversary system of trial, which imposes a large burden on the defense to counter the prosecution with its own evidence, is fatally flawed when criminal defendants almost never have the ability to independently gather evidence.

Reforms: Reducing the Number of Wrongful Convictions

Research and systematic thinking about the proximate causes of wrongful convictions have suggested a number of feasible reforms likely to reduce miscarriages of justice. At this early stage in the innocence movement, no comparable thought has been given to dealing with the far more intractable root causes. The partial list and descriptions that follow do not explain the research bases for the proposed reforms, but there is good reason to believe that the widespread adoption and systematic application of these reforms will reduce the number of wrongful convictions.

Recommended lineup reforms are grounded in laboratory research findings that show they will reduce the number of false identifications without significantly reducing accurate identifications. Witnesses in all lineups (live and photo) should be instructed that the perpetrator may not be present, to reduce the tendency to pick anyone. All lineup fillers should be selected on the basis of the victim’s verbal description, and not on similarity to the suspect. Lineups should contain only one suspect and should be fair in that there are similarities of race, height, general appearance, facial hair, photograph characteristics, and the like between the suspect and fillers. It is best that the lineup administrator not know who the suspect is (blind administration), to ensure that there is no unconscious influence on the witness (as is done for subjects in medical and pharmaceutical trials). If lineup administration is blind, the lineup participants (live or photo) should be presented one at a time (sequentially) rather than as a group (simultaneously). This helps to prevent the exercise of relative judgment, by which a witness picks a person out of the lineup who looks most like the memory of the perpetrator rather than recognizing the perpetrator. A witness should be asked for a confidence statement immediately after making an identification, to prevent his or her inflation of confidence as the case proceeds.

All crime laboratories should be accredited and their examiners certified and required to undergo periodic proficiency testing. Defense attorneys, as well as prosecutors and judges, should be educated in forensic testing techniques, and funding should be sufficient to have challenged forensic evidence retested. Defense attorneys should become aware that comparison testing methods, like fingerprinting, are not infallible. Where standards for comparison testing are weak or even suspect, as with bite mark evidence, special caution must be taken in allowing and weighing such evidence. Forensic science research is needed to ensure that methods and findings are valid. Substandard laboratories should be closed and not reopened until all problems are remedied.

The most widely recommended interrogation reform is to videotape entire interrogations, from initiation and before Miranda warnings to the conclusion, and not just the confessions. Videotaping allows pretrial judges to determine whether interrogation was coercive or likely to produce a false confession. Police benefit from videotaped interrogations, because confessions by guilty suspects provide powerful prosecution evidence. Interrogations should be time limited, especially for vulnerable suspects, such as teenagers, to 2 hours, because many false confessions are the product of protracted interrogation. Police in Canada and the United Kingdom are not allowed to use lies to get suspects to confess. This rule should be adopted even though the U.S. Supreme Court has held that lies do not violate a suspect’s constitutional rights. Another valuable reform would require police to provide, before interrogating, stronger evidence of their belief that a suspect is guilty than is now the case. A judicial instruction that informs the jury about the risks associated with nonvideotaped station house confessions creates incentives for police and prosecutors to adopt electronic recording.

If the use of jailhouse snitches is to continue, prosecutors should carefully corroborate their stories and take into account snitches’ characters and past experiences before using their claims that suspects confessed to them. Legal rules should allow defendants extensive discovery to explore the nature of deals made in return for their testimony. Judges should warn juries that jailhouse snitch evidence should be examined with greater care than that of other witnesses.

At present, many groups urge that compensation for indigent defense be raised to reasonable compensation, to allow competent assigned attorneys the time to better represent clients and bring public defenders’ workloads into compliance with established standards. Changes in appellate rulings should allow findings of ineffective assistance without needing to prove that incompetence caused a verdict. Greater bar association scrutiny of appointed counsel and public defenders can enjoin attorneys to do their jobs properly. Defense attorneys should be expected to visit crime scenes and interview all prosecution and defense witnesses. Funding for investigators should increase.

Additional funding for prosecutors and their investigators, by reducing caseloads, will create better understanding of cases and may reduce wrongful convictions. Prosecutors should advise police and forensic laboratories to include exculpatory evidence in their reports. As the leading executive branch participants in the criminal justice system, prosecutors should promote laws and regulations to improve lineups and interrogations in accordance with best practices established by psychological research and should not resist reasonable postconviction requests for reinvestigation of evidence.

Police investigators also need greater resources to make work pressures more manageable. Standards should be rewritten and training revised to educate investigators in wrongful conviction matters, to become aware of the effects of tunnel vision, and to include exculpatory evidence in their reports. This may be very hard to achieve, because it calls for a change in police culture away from proprosecution partisanship and more toward a neutral and scientific attitude toward cases.


In 20 years, wrongful conviction has gone from a littlenoted phenomenon to an important topic within criminal justice. The number of innocence projects working to exonerate prisoners has grown from 1 or 2 in the early 1990s to about 50 today. Partly as a result of their policy advocacy, innocence reforms have been enacted. Congress passed the Innocence Protection Act in 2004, providing funding for state postconviction DNA testing, encouraging states to pass postconviction DNA testing laws, and raising the annual compensation for exonerated federal prisoners to $50,000 for each year of imprisonment. More than 40 states have passed postconviction testing laws. Six states and hundreds of police departments have required videotaping of interrogations. Seven states and a growing number of police departments have established eyewitness identification reforms. North Carolina created the first innocence inquiry commission that reviews wrongful convictions claims and presents successful claims to a special court.

The investigation of wrongful convictions, which challenge the fairness and accuracy of the criminal justice system, are becoming a necessary feature of criminal justice analysis. The adoption of innocence reforms will not only reduce this kind of injustice but will also improve the quality and professionalism of criminal justice participants.