Landmark Study Shows the Unreliability of Capital Trial Verdicts

By Hans Sherrer

Justice:Denied magazine, Vol. 2, Issue 2


The unreliability of guilty verdicts and sentences in capital cases has been revealed by what may be the most important study in the history of American jurisprudence. Released to the public on June 12, 2000, the 9-year study of capital cases was conducted by Columbia University School of Law Professor James S. Liebman. The study examined every capital conviction and appeal in the United States between 1973 and 1995. One noteworthy finding of the study is that at least 68% of all the thousands of capital convictions in this country finalized during the study's 23-year period were reversed on appeal.

In summarizing the study titled, "A Broken System: Error Rates in Capital Cases," Professor Liebman said, "American capital sentences are persistently and systematically fraught with serious error. Indeed, capital trials produce so many mistakes that it takes three judicial inspections to catch them, leaving grave doubt as to whether we do catch them all."1

Professor Liebman is an eminent legal scholar who, among his extensive writings, coauthored a leading American treatise on habeas corpus, the two-volume "Federal Habeas Corpus Practice and Procedure" (3rd ed. 1998). He has argued four major habeas corpus or capital cases in the U.S. Supreme Court. The study, begun in 1991 at the request of the Chair of the U.S. Senate Judiciary Committee, was coauthored by Jeffrey Fagan and Valerie West.


The fact that more detailed records of capital cases are created than for lesser criminal prosecutions has made this study possible. This occurs for three reasons: a much higher percentage of capital cases go to trial than ones involving lesser charges; the rules governing pretrial and trial proceedings are more carefully adhered to than in cases involving a lesser penalty; and defendants have a legal right to one round of a three-tiered inspection process that reviews a capital conviction for possible errors.

The three judicial inspections in the order they are conducted are a direct review, a post-conviction review, and a federal habeas corpus review. The purpose of these multiple reviews is to inspect the record of the case for errors serious enough to have substantially undermined the reliability of the guilty verdict or the sentence.2

Direct review for prejudicial trial court errors is by a state's high court. The U.S. Supreme Court suggested in Furman vs. Georgia (408 US 238, 1972) that a state's high court is required to review all capital convictions and all states that impose the death penalty have implemented a direct review process. If a capital conviction isn't reversed on direct review, a defendant's allegations of serious constitutional errors that warrant reversal of his or her conviction are subject to post-conviction review.

Typically, constitutional claims are filed with the trial court and are successively reviewed as necessary by the state's intermediate appellate court and the state's high court as a precursor to seeking federal habeas corpus review of those claims. When state post-conviction review of constitutional claims doesn't result in a reversal, a federal habeas corpus petition that includes those claims is filed with the U.S. District Court in the state where a defendant was convicted.

The finality of carrying out a capital sentence is why the review of a conviction is so extensive, takes so long and is so expensive. Judicial review of a capital case averages almost eleven years to complete at an estimated total cost of $2.5 to $5 million.3


Professor Liebman's study included findings from all 4,578 capital appeals finalized in the U.S. between 1973 and 1995. He is continuing research related to the study and its findings. The single most remarkable fact disclosed in the study is that nationwide, over 68% of all capital convictions are found by the three-tiered judicial review process to have errors so prejudicial to a defendant's rights that reversal of his or her conviction is mandated. The error rate is so constant that in only one of the 23 years covered by the study did it deviate by more than 14% from the 68% average.4

The overall error rate by state from 1973 to 1995 is listed below.5

1. Kentucky 100%
2. Maryland 100%
3. Tennessee 100%
4. Mississippi 91%
5. Wyoming 89%
6. California 87%
7. Montana 87%
8. Idaho 82%
9. Georgia 80%
10. Arizona 79%
11. Alabama 77%
12. Indiana 75%
13. Oklahoma 75%
14. Florida 73%
15. North Carolina 71%
16. Arkansas 70%
17. Nevada 68%
18. South Carolina 67%
19. Utah 67%
20. Illinois 66%
21. Nebraska 65%
22. Louisiana 64%
23. Pennsylvania 57%
24. Texas 52%
25. Missouri 32%
26. Virginia 18%

The national reversal rate would be well over 70% if not for the two anomalous states of Missouri, with a detected error rate of 32%, and Virginia, with a detected error rate of 18%. The list further shows that more than 92% of the capital punishment states have an error rate of over 50%; 85% have an error rate over 64%; and 35% have an error rate over 80%.

Clearly, trial court level errors serious enough to require reversal of a capital conviction are the rule and not the exception. The error detection rate for each of the three stages of judicial review is indicative of how crucial each step of the inspection process is to protect a defendant from an erroneous conviction. On state direct appeal, 41% of capital cases were reversed. The state post-conviction review process overturned more than 10% of the remaining cases, while federal habeas corpus proceedings reversed an additional 40% of the cases that weren't overturned during their review of state courts.

The high rate of error detection at the third tier, federal habeas corpus review, lends support to Professor Liebman's suggestion that errors serious enough to mandate reversal of a conviction continue to go undetected. The prevalence of undetected errors perhaps warrants the imposition, as the report suggests, of additional safeguards against erroneous convictions, up to and including a fourth level of judicial inspection.


The reliability of Professor Liebman's findings was confirmed by a U.S. Department of Justice study of all death sentences imposed in 1989.6 State and federal courts overturned 76% of the cases from 1989 that had completed the appeal process. The Department of Justice's study confirmed the truth of Professor Liebman's observations that if someone asked what would happen after a defendant was sentenced to death, a responsible answer would be:

"The capital conviction or sentence will probably be overturned due to serious error. It'll take nine or ten years to find out, given how many other capital cases being reviewed for likely errors are lined up ahead of this one. If the judgment is overturned, a lesser conviction or sentence will probably be imposed."7 REASONS FOR REVERSAL

The errors causing reversal of a capital conviction are not mere technicalities. They are only considered serious enough to support a reversal if they may have undermined the reliability of a guilty verdict or the sentence imposed. The most common error cited by appellate courts when they overturn a capital case is incompetent defense lawyers. This occurs in 37% of cases. Faulty jury instructions cause an additional 29% of reversals. In 19%, the third leading reason cited for a reversal is prosecutorial misconduct. This often involves the concealment of potentially exculpatory evidence of an accused's innocence or mitigating factors eliminating their qualification for the death penalty. These three factors accounted for a total of 76% of all capital cases overturned during the 23 years covered by the study.8


The report suggests that a fourth level of judicial review may be required to cull through capital cases to find errors that are currently not being detected. Although it isn't cited as a reason in the review, the need for an additional review process is compellingly indicated by the study's finding that more than 7% of all overturned capital cases involve people innocent of the crime they were convicted of committing. It is also indicated by the more than 82% of capital defendants who were sentenced to less than death after their retrial.9 These two groups account for nearly 90% of the defendants affected by overturned capital cases, so only one out of ten reversed death penalty cases results in a new capital sentence after a retrial.

That raises two unanswered questions: how many erroneous capital cases are slipping through the cracks in the current review process, and how many defendants are not having their innocence detected? Recent cases of men being carried to the execution chamber while proclaiming their innocence lends anecdotal support to the study's finding of an implied need for an additional level of independent review to catch currently undetected errors in a timely manner. Freddie Lee Wright is a case in point. Hours before his execution in March 2000, a justice of the Alabama Supreme Court futilely argued in support of a last minute stay that the state of Alabama was going to execute an innocent man.


Not only does Virginia have the dubious distinction of ranking last in the reversal of capital cases at the state level, but it also ranks last among states with three or more capital cases reviewed by federal courts. As the study notes: "Virginia is a district outlier here, falling almost literally 'off the charts' on the low side of error detection. Virginia's overall rate of detected error is barely half that of the next closest state (Missouri, which itself is much lower than all the other states), and barely a quarter of the national rate."10

Although Virginia's known error rate of almost one out of five cases is significant, the rate might be comparable to the average 76% error rate of other states in the Federal Fourth Circuit Court of Appeals if Virginia provided enough money to pay for competent legal aid at the trial and direct appeal stage of a capital prosecution. As it is, the amount Virginia provides for the defense of indigents is barely adequate to cover the costs of negotiating an inequitable plea bargain, much less the extensive legal preparations and investigation necessary for a defendant to present a credible defense to a capital charge and create a record for appeal.

Virginia's inadequate funding of a capital defendant's defense ends entirely at the conclusion of the direct appeal stage. Without pro bono assistance, an indigent defendant is unable to pursue the critical state post-conviction or federal habeas corpus review of their conviction that accounts for almost half of all reversed capital convictions.11

Virginia also has the distinction of executing the highest percentage of people condemned to death, which it does at five times the national average. So given the nationally known error rate of wrongful capital convictions, there is a very real possibility that innocent people have been, and will continue to be executed in Virginia due to inadequate funding of legal representation and review of the compelling issues in their cases. This is a pressing concern given the 7% of capital cases nationwide that are reversed because the condemned person was found to be innocent.


In January 2000, the governor of Illinois declared a moratorium on executions after 13 men were released from death row between 1976 and 1999. Twelve men were executed during that same period of time. There has been speculation by people, including the governors of Florida and Texas, that Illinois may have such a high rate of exoneration because it has a higher than average rate of wrongful convictions.12 Professor Liebman's study proves that is a fallacious argument. Illinois' error detection rate ranks in the bottom 25% of death penalty states.13

Since its 66% rate of overturning capital convictions is slightly below the national average, it is possible that what is anomalous about Illinois is that cases of wrongful conviction in its courts are being detected by means overlooked in other states. One of those methods is investigation of questionable cases by students at Northwestern University under the direction of Professor David Protess.

The high rate of exonerations in Illinois may mean the rate of conviction of innocent people in capital cases nationwide may be significantly higher than the 7% documented by the study. The possibility is at least in part indicated by the fact that some of the capital cases overturned in Illinois on grounds of innocence, such as the Anthony Porter case, were only detected after their conviction was initially affirmed by the state appellate courts.

Illinois has a below average rate of detecting erroneous capital convictions. The problems that plague trial courts nationwide are indicated by a newspaper exposé of Illinois courts that revealed "a system so riddled with faulty evidence, unscrupulous trial tactics and legal incompetence that justice has been forsaken. The findings reveal a system so plagued by lack of professionalism, imprecision and bias that they have rendered the state's ultimate form of punishment its least credible."14

The report points out that the human cost of an erroneous capital conviction goes far beyond the financial price tag of $2.5 to $5 million. It exacts a steep price in different ways from all the people involved. During the average of more than a decade that passes from a trial to the overturning of a guilty verdict, the death row inmate is confined under brutal conditions and subject to stresses to which the average person can't relate. The injustice of that person's treatment is shared to one degree or another by everyone concerned about his or her fate. Furthermore, that treatment continues if the person is retried and the judicial process begins once again. The family and friends of the victim also experience the void of not having closure as the legal process completes the procedures necessary to uncover reversible errors.


One of the primary justifications advanced by proponents of the death penalty is that it acts as a deterrent to violent crime. Yet this isn't borne out by research related to the study. Liebman found that in the 34 states which have issued and appealed death sentences, the average homicide rate is 9.3 per 100,000 people. The sixteen other states have a rate of 7.6 homicides per 100,000.15 This difference can't be explained away by racial demographics because the non-white population of the non-death penalty states is proportionately higher than in the death penalty states.

The inexactitude of capital convictions and the length of time necessary to complete the three stage review process that confirms less than one-third of convictions as valid are sometimes cited as a reason for the death penalty's lack of deterrence value. These are reasons that capital punishment advocates have suggested that executing people would be more of a deterrent if the review process of capital cases was speeded up. That idea was used as a political justification for passing the Anti-terrorism and Effective Death Penalty Act in 1994 that placed restrictions on the federal appeal rights of state prisoners, even though there was no supporting data on its effectiveness or its impact on the quality of review that cases would receive.

The study's findings imply that any impairment of a defendant's access to the courts may contribute to an increase in the number of undetected errors, while having no reductive effect on the crime rate. That assessment is supported by hundreds of years of scientific and anecdotal evidence that at best indicates the death penalty has an effect of promoting violence in society. The ineffectiveness of the death penalty as a deterrent was documented in a study of capital punishment in European countries that was published in the 1840s. There are also eyewitness accounts of people committing property crimes, such as pickpocketing during the public execution of people in England convicted of committing those very crimes.

Professor Liebman's findings reinforce the retributive emotions underlying support for capital punishment in the absence of facts justifying its use. The report states, "If there is any relationship at all between homicide and capital-sentencing rates, it is weak and inverse."16 It further notes there is no "relationship between a state's serious crime rate and its willingness or capacity to execute its citizens."17


The report reveals some of the incongruities that contribute to the legal system's consistent production of serious errors. One of these is the apparent causal relationship between wrongful capital conviction and the judicial and prosecutorial environment that consistently generates serious errors. The report suggests the possibility that high reversal rates in capital cases may be attributed in part to judges and prosecutors cutting corners to obtain convictions that "curry favor" with the public.18 The conviction bias of trial courts is compounded by political pressures impelling the imposing of capital sentences which, a large majority of the time, are found after many years of appellate review to be faulty.19 Furthermore, the report points out that the rate of errors in capital cases appears to increase as the number of cases grow.20

In general terms, the study reveals that a large majority of capital defendants are erroneously tried; an elaborate and expensive multi-tiered inspection process is required to find the egregious errors in their cases; many thousands of men and women are warehoused under costly death row conditions for an average of almost eleven years; and more than two out of three capital cases have to be retried, a prisoner's innocence established after years awaiting execution for something he or she did not do. As Professor Liebman wrote of his findings: "by this measure, the capital punishment system revealed by our 23-year study is not a success, and is not even minimally rational."21 Perhaps the best summary of the report's meaning was its observation, "If what were at issue here was the fabrication of toasters (to return to our prior example), or the processing of social security claims, or the pre-takeoff inspection of commercial aircraft -- or the conduct of any other private or public-sector activity -- neither the consuming and the taxpaying public, nor managers and investors, would for a moment tolerate the error-rates and attendant costs that dozens of states and the nation as a whole have tolerated in their capital punishment system for decades. Any system with this much error and expense would be halted immediately, examined, and either reformed or scrapped."22


Professor Liebman's study is significant because it provides documentary proof of the magnitude of the legal system's failure at performing its adjudicatory function. Its ineffectiveness at the trial court level is so great that even after two full-scale independent reviews that result in a 47% failure (i.e., reversal) rate, the third appellate review finds serious fault with 40% of the remaining cases it inspects. That 40% error rate is not just indicative of the competence prevailing at the trial court level, but also of the previous appellate court(s) that overlooked those reversible errors.

The report's analogy that a 68% error rate in the "manufacture and distribution of toasters or any other product or service provided wouldn't be tolerated" is borne out by the recent furor over the apparent failure of some Firestone tires installed on Ford SUVs. The known failure rate of the estimated 6 million tires involved is somewhat less that 1/10th of 1% of the tires installed on those vehicles. Yet, within a matter of weeks congressional hearings were advocated to get to the bottom of how and why the tires continued to be used by Ford and Firestone after they became aware there might be a problem with them.

However, even though the legal system has a consistent error rate thousands of times greater than the failure rate of the suspect Firestone tires, there is no nationwide outrage over the legal processes' systematic and predictable production of erroneous capital convictions that is believed to have contributed to the execution of dozens of innocent people.23 This deafening silence conceals the devastating personal and financial impact on everyone involved, that can exceed the impact of the much less common random automobile crash caused by a defective tire.

Although the report doesn't attempt to do so, the study's findings are applicable to prosecutions that involve a lesser penalty. This is because the same factors contributing to erroneous capital convictions also adversely affect non-capital cases. However, those factors can be even more pervasive in a non-capital case because the quality of defense representation may be lower, and there is less public scrutiny of the behavior of the judge, prosecutor and government witnesses involved.

These factors can coalesce to cause incompetent representation for a defendant, prosecutorial misconduct and faulty jury instructions with greater frequency than in much more high-profile capital cases. Unfortunately, when these injustices occur in non-capital cases, they are typically concealed from adequate scrutiny on appeal. Without the filing of adequate pretrial motions and making timely objections to preserve the record of appeal, even a competent appellate lawyer may not prevail, except on grounds of incompetent trial representation.

The effect of the foregoing is that the report's estimate based on admittedly sketchy data that 15% of non-capital cases are reversed on appeal vastly understates the percentage of cases that would be overturned if competent counsel was available at the trial and appellate level.24 It is reasonable to surmise that if competent attorneys were available to non-capital defendants, the reversal rate of non-capital cases would be at least comparable to that of capital cases.

The study covering almost a quarter of a century reveals a judicial process that can't be relied on to correct all the errors it creates, and its increasingly apparent lack of credibility is of pressing concern to everyone who might be affected by it.


1. "Landmark Study Finds Capital Punishment System "Fraught with Error," "Fenton Communications, NY, June 12, 2000 Back to Story

2. "A Broken System: Error Rates in Capital Cases," Columbia Law school, James S. Liebman, Jeffrey Fagan and Valerie West, June 12, 2000, Section II Back to Story

3. Ibid. Footnotes 57, 74-75 Back to Story

4. Ibid. Section VII Back to Story

5. Ibid. Table 10' Back to Story

6. "Capital Punishment 1998," U.S. Department of Justice Bureau of Justice Statistics, Dec. 1999, NCJ 179012, pp. 12-13 Back to Story

7. "A Broken System" Section IV Back to Story

8. Ibid. Section II Back to Story

9. Ibid. Executive Summary Back to Story

10. Ibid. Section VIII Back to Story

11. Ibid. fn. 36 and Section VII. Back to Story

12. Ibid. Section VIII. Back to Story

13. Ibid. Table 10 Back to Story

14. Footnote 78, quoting from "Death Row Justice Derailed," Ken Armstrong & Steve Mills, Chicago Tribune, Nov. 14, 1999, p.1. Back to Story

15. "A Broken System" Section VII. Back to Story

16. Ibid. Section VIII, see fig. 23. Back to Story

17. Ibid. Section VIII, see fig. 24 Back to Story

18. Ibid. Section VIII Back to Story

19. Ibid. Section VIII. Back to Story

20. Ibid. Section X. Back to Story

21. Ibid. Section X. Back to Story

22. Ibid. Section IV Back to Story

23. "Miscarriages of Justice in Potentially Capital Cases," Hugo Adam Bedau & Michael Radelet, 40 Stanford Law Review 21 (1987) Back to Story

24. "A Broken System," Section II. Back to Story