Discover more from Discourse
From Finality to Fallibility
Thousands of wrongful convictions should make us realize that juries’ decision-making isn’t as reliable as we think it might be
This piece is the eighth and final in a series on defects in our criminal justice system that lead to wrongful convictions. The first piece introduces these seven critical breakdowns in the rules of constitutional criminal procedure, the second examines the problem of eyewitness misidentification, the third shows how the Supreme Court is perpetuating the problem of false confessions, the fourth examines prosecutorial misconduct, the fifth looks at overly favorable treatment for prosecutors, the sixth discusses the importance of good counsel, and the seventh examines failures in forensic science.
In our criminal justice system, juries decide whether a crime has been committed and, if so, whether the defendant committed the crime. These decisions are designed to be final: Victims and communities shaken by crimes would never fully heal if every jury verdict was tentative, and taxpayers would resent funding endless litigation.
However, the thousands of wrongful convictions over the past few decades have shown that juries are far less reliable than we thought. That’s turned the finality of jury decisions into a problem: The near-unreviewable status of jury verdicts often makes the path to exoneration for innocent inmates a prolonged, painful, often futile slog. Innocent people have died in prison because of it.
The Roots of the Jury Verdict Problem
Two landmark Supreme Court cases, Chapman v. California and Herrera v. Collins, both decades old, are largely to blame for the jury verdict problem. Chapman affirmed and extended what’s known as the “harmless error rule.” This rule requires judges to distinguish between “reversible” trial errors, which may have steered jurors toward the wrong decision, and “harmless” ones, which, though they violate the rights of the accused, likely didn’t affect the trial’s outcome. Herrera entrenched the finality of jury decisions into the post-conviction review system by prohibiting federal courts from hearing habeas corpus petitions (petitions regarding potential violations of an inmate’s constitutional rights) from inmates who’ve exhausted their state appellate opportunities but, due to newly discovered evidence, claim they can prove themselves innocent.
To innocent death row inmates, the importance of these rules can hardly be overstated. The average capital case spends 18 years in appellate and post-conviction litigation, and most inmates eventually exhaust all of their state court opportunities and attempt to get a hearing. But getting a federal habeas court to take a fresh look at a case can be lifesaving: According to one leading death penalty casebook, between 1973 and 1995, “40% of the death sentences that survived state direct and post-conviction review were overturned on federal habeas corpus.” But many inmates never get the chance to literally have their day in court.
When the rules from Chapman and Herrera are considered together, innocent inmates are left with a Kafkaesque procedural nightmare that can keep them behind bars indefinitely. By refusing to overturn or revise its rulings in these cases, the Supreme Court has maintained a constitutional appellate and post-conviction system that has refused to adapt to the realities of the criminal justice system exposed by the DNA revolution in criminal forensics.
Not So Harmless
Ruth Chapman and her boyfriend, Robert Teale, were convicted of kidnapping, robbing and brutally murdering bartender Billy Dean Adcock in Lodi, California, in October 1962. Chapman was sentenced to life in prison, while Teale was given the death penalty.
There’s no question that one of their core constitutional rights—the right to choose not to testify—was violated during their trial. During his closing argument, the prosecutor repeatedly drew the jury’s attention to the fact that, instead of taking the stand to tell their sides of the story and protest their innocence, Chapman and Teale chose to exercise their right to silence. The prosecutor was deploying the age-old tactic of equating silence with evidence of guilt: This has the effect of deterring defendants from asserting their constitutional rights, which they’re entitled to exercise freely.
Shortly before agreeing to hear this case, the Supreme Court had ruled that this tactic was strictly forbidden under the Constitution, so there was never a doubt whether there had been an error in Chapman’s and Teale’s cases. They both asked the Supreme Court to reverse their convictions and affirm the position the court had taken in two previous cases, Payne v. Arkansas and Lynumn v. Illinois, by clearly announcing “that denial of a federal constitutional right ... should automatically result in reversal of a conviction, without regard to whether the error is considered harmless.”
The court reversed Chapman’s and Teale’s convictions because it believed the prosecutor’s statements likely affected the verdict. However, it refused to adopt the rule they’d requested: Instead, the court announced the constitutional “harmless error rule,” under which convictions are to be upheld even when trial rights enshrined in the constitution are violated—so long as appellate courts believe beyond a reasonable doubt that the error would not have affected the outcome.
In some ways, this rule made sense, especially coming, as it did, in an era before the problem of wrongful convictions was widely known. But the rule placed undue confidence in the ability of reviewing judges to assess the value of evidence presented during trials. Furthermore, data on DNA wrongful conviction cases and harmless error has since proved the rule to be anything but harmless. We now know that by rejecting the reasonable balance struck between the state’s interest in finality and defendants’ rights in Payne and Lynumn, the court has increased the number of inmates whose trial rights have been violated and are also innocent of the crimes for which they were convicted.
However, the true impact of the harmless error rule is better understood by considering the individual cases of those harmed by it. After Ron Williamson was sentenced to death for the rape and murder of Debra Sue Carter in 1982, the Criminal Court of Appeals of Oklahoma acknowledged that the prosecution had illegally introduced evidence that Williamson had “made a pass” at the victim and that she “did not feel comfortable around him.” However, citing the Supreme Court’s ruling in Chapman and convinced the other evidence against Williamson had proved him guilty beyond a reasonable doubt, the court upheld his conviction and death sentence. Williamson spent more than a decade behind bars before DNA testing exonerated him in 1999. His story was later the subject of a book by John Grisham and of a Netflix documentary.
Under the Payne-Lynumn understanding of harmless error, convictions would be upheld only in cases where there is either no error or the error is both unconstitutional and doesn’t impact the verdict. While implementing that understanding wouldn’t completely solve the problems caused by the harmless error rule, it would more reasonably balance the state’s interest in finality with an inmate’s right to a fair opportunity to win his liberty.
The Story of Leo Herrera
Leo Herrera’s last words before his execution by lethal injection on May 13, 1993, were clear and emphatic. “I am innocent, innocent, innocent,” he said. “Make no mistake about this. I owe society nothing ... I am an innocent man, and something very wrong is taking place tonight.”
He might have been telling the truth, as we now know his fellow Texans Carlos DeLuna (executed four years before him) and Cameron Todd Willingham (executed 11 years after him) were. Herrera had been convicted under suspicious and sensational circumstances of killing two police officers, Enrique Carrisalez and David Rucker, during a 1981 shootout at the Texas-Mexico border, where drug smugglers routinely moved back and forth. Cocaine trafficking was the Herrera family business: Leo worked with his father and brother as a “cutter” who divided large quantities of cocaine into small ones for distribution. According to his attorneys, Texas law enforcement named Herrera ‘“public enemy number one,’ announced the largest manhunt in the history of the area to track him down and, upon his arrest, beat him into a coma and left him handcuffed in the floor of his cell to die.”
Herrera’s trial for the murder of Officer Carrisalez took place in an atmosphere dominated by the police. A police detective, gun at his side, even served as a voting member of the jury. Prosecutors presented what seemed like a strong case: An eyewitness identified Herrera, acting alone, as the shooter. Blood consistent with one of the victims was found on Herrera’s pants, and “most compelling” of all, according to the Supreme Court, he was carrying a written confession with him at the time of his arrest. As anticipated, Herrera was convicted and sentenced to death. Seemingly resigned to his fate, and fearful and broken from abuse, he pleaded guilty to the murder of Officer Rucker.
While Herrera was on death row, his brother Raul allegedly began confessing that Leo was innocent and that he was the one who had murdered the two officers. Officer Rucker was on the take, he claimed. Raul had killed him when an argument broke out between them about the drug operation they were both involved in. Shortly after shooting Rucker, Raul was pulled over by Officer Carrisalez, and to avoid arrest, Raul killed him too.
After Raul’s death a few years later, some of those who’d been loyal to Raul came forward to file sworn statements that the wrong Herrera was on death row. Raul’s lawyer, a former judge, stated that Raul had confessed to him. Raul’s son claimed that he himself had watched his father kill the officers. Two other sworn witnesses pointed the finger at Raul as well.
Federal district court judge Ricardo Hinojosa found this new evidence credible enough to warrant a hearing that would enable him to determine whether Herrera should be removed from death row. However, the circuit court of appeals, sitting above Judge Hinojosa, intervened to abort the hearing, and the Supreme Court affirmed the appellate court’s decision. Herrera had already had his day in court, the court ruled. Even if the new evidence might show him to be factually innocent, he was legally guilty: “Federal habeas courts do not sit to correct errors of fact,” the court explained in Herrera v. Collins, “but to ensure that individuals are not imprisoned in violation of the Constitution.”
Leo Herrera was executed without a judge ever considering his new evidence. As a result, we may never know which of the two Herrera brothers squeezed the trigger that night in 1981. And Leo Herrera may have been executed for the entirely arbitrary reason that evidence of his innocence emerged after, rather than before, his conviction.
Criminal law and procedure were at a crucial turning point when the court issued its Herrera decision in 1993. The DNA revolution in criminal forensics had begun four years before, when genetic fingerprinting proved that a young Illinois man named Gary Dotson couldn’t have committed the rape for which he’d been convicted. The Innocence Project, the vanguard of the revolution, had been founded in 1992 and would go on to exonerate hundreds of inmates, including many from death row. “Thousands” more innocent inmates were languishing in U.S. prisons, Innocence Project co-founder Barry Scheck explained to PBS a few years later. By 1993 the whole world—and certainly the court—knew that wrongful convictions were a significant problem.
But in Herrera, the court turned a blind eye to that problem. By sending Leo Herrera to his death, the court made it clear that old rules, long predating DNA testing, were not going to adapt to new realities.
Herrera’s case bitterly divided the Supreme Court. Chief Justice William Rehnquist, writing for a slim 5-4 majority, rejected Herrera’s claims completely. In response, Justice Harry Blackmun penned a dissent alive with moral outrage. Blackmun had been appointed to the court by President Richard Nixon in 1970 to help restore law and order to the nation after the civil unrest of the 1960s, but after struggling with death penalty cases for decades, he had lost faith in the government’s ability to fairly implement it. Rehnquist’s opinion against Herrera magnified Blackmun’s concerns about fairness by introducing the possibility of the most unfair outcome imaginable: the execution of someone whom the court itself has prevented from proving his own innocence. “The execution of a person who can show that he is innocent comes perilously close to simple murder,” Blackmun wrote.
There are passages in Rehnquist’s opinion in which he seemed to express discomfort with his own decision—almost as if he was preemptively defending himself against a future wrongful execution. For instance, Rehnquist imagined a case involving a death row inmate whose exonerating evidence was much stronger than Herrera’s. Rehnquist “assum[ed]” “for the sake of argument,” that this inmate, who could “present a truly persuasive post-trial demonstration of innocence,” would have a right to a hearing. Herrera’s evidence, however, fell short of that “extraordinarily high” standard. In the 30 years since its ruling, the court has never shone light on what kind of evidence might satisfy this hypothetical standard. An inmate who discovers a video of someone else committing the crime, maybe? We can’t be sure. Lower courts have been left to speculate as to what this cryptic passage means.
Executive Clemency Is Insufficient
Elsewhere in the opinion, Rehnquist emphasized that, while his opinion closed the habeas courthouse door to all actual innocence claims that weren’t combined with constitutional arguments, he was not leaving inmates without a potential remedy: They could always petition their state governors for a pardon or commutation. “Executive clemency has provided the ‘fail-safe’ in our criminal justice system ... History shows that the traditional remedy for claims of innocence based on new evidence, discovered too late in the day to file a new trial motion, has been executive clemency.”
But the prospect of executive clemency alone is simply not sufficient to deal with the wrongful convictions problem. Freeing factually innocent inmates isn’t the original purpose of the clemency power, and, though executives occasionally pardon the innocent, it’s never been typically used that way. Historically, acts of executive clemency have been either been political—as when Jimmy Carter pardoned Vietnam draft dodgers and Donald Trump pardoned Roger Stone—or acts of mercy—as when President Harding commuted the sentence of free speech martyr Eugene Debs, who had become a broken old man after years of hard time in a Georgia federal prison.
The most defining feature of the clemency power, when compared to a judicial proceeding, is arbitrariness. The federal Constitution, for instance, gives the president a clemency power that is complete and unreviewable. The president can pardon anyone, even prior to conviction, for any reason or no reason at all without oversight from Congress or the courts. A pardon is an “act of grace,” Chief Justice Marshall wrote for a majority of the court in 1833—and this is a concept altogether different from justice. Pardons are based in presidential power, not judicial reasoning, and are issued more as a gift than as something deserved.
In 1981, the court ruled that convicts have no constitutional right to be considered for executive clemency, and in 1998, the court further explained that if an executive chooses to consider clemency petitions, he can create virtually any process he wants for doing so. These are hardly the forums to which factually innocent death row inmates who can prove their innocence should be sent. Moreover, these inmates have committed no act that needs to be “pardoned.” Rather, they deserve the kind of justice that comes after the truth-testing process of a judicial hearing, not the same “act of grace” Bill Clinton granted his half-brother, Roger Clinton, simply because the two are related.
More than 30 years into the DNA revolution in criminal forensics, law books continue to thicken with cases of inmates who, years after conviction, are proved innocent by newly discovered evidence. It’s past time for the Supreme Court’s emphasis on jury finality to adjust to the reality of jury fallibility.