The Withered Writ

The writ of habeas corpus, until not long ago, was a mysterious yet potent safeguard of liberty in American law. It worked like an incantation to break an evil spell. A prisoner petitions a court for a writ. “Habeas corpus” means “May you produce the body,” spoken from the point of view of a judge. He orders whoever is depriving the prisoner of his freedom to bring him to court—a warden confining a prisoner, the secretary of defense holding a detainee, or a magistrate who has denied bail to someone jailed but not convicted—and to justify the detention. The judge then decides whether the petitioner is being detained in breach of the Constitution or some other law.

In 1963, Justice William Brennan Jr. wrote that “government must always be accountable to the judiciary for a man’s imprisonment.” The prisoner is “entitled to his immediate release,” the justice emphasized, if the government violates the law in putting him behind bars. Brennan was one of the most influential justices of the 20th century. A champion of the individual versus the government—in what he called “the unceasing contest between personal liberty and government oppression”—he led the Supreme Court to strengthen American democracy by strengthening various roles of citizens, to minimize those clashes. That included bolstering the First Amendment so people could be confident about criticizing officials without getting sued, adopting the principle of “one person, one vote” so each vote would have equal weight, and enhancing the power of federal courts and citizens’ access to them to protect these and many other rights.

He embraced habeas corpus as part of that enhancement—as a crucial tool for the protection of constitutional rights. Following his lead, the Warren Court expanded habeas law to redress appalling treatment by many states of criminal defendants, especially minorities and the poor, who had often experienced unfair arrests, been coerced into confessions, and received unjust trials.

Habeas became, above all, a tool for challenging death sentences. Robust habeas law protects death-row inmates from a kind of vigilantism carried out in state courts—particularly inmates whose crimes were experienced as an outrage not just against the victim but also against the community. The idea was to give the petitioner access to the safe harbor of a federal court.

By expanding habeas law, the Supreme Court empowered the hundreds of federal trial courts to play an essential part in making sure state and local governments were complying with the Constitution. Since the Judiciary Act of 1867 made habeas corpus available in federal court to state prisoners who claimed their rights had been violated, the writ has been pushed and pulled in the persistent American debate about the balance of power between the national government and the states.

In Brennan’s view, the federal courts had the ultimate responsibility to defend the Constitution. No state’s denial to an individual of constitutional rights could become final without that review—especially state criminal decisions that deprived someone of his liberty and possibly his life. No statute of limitations cuts off an individual’s access to a habeas writ because, the Supreme Court said, there is no higher duty than to maintain the writ unimpaired. For that reason, a petitioner’s failure to persuade a federal trial court to grant a writ did not keep him from applying for one again.

The Constitution explicitly limits the power of Congress to suspend the right to habeas. In doing so, Brennan believed, the Constitution underscores the importance of the writ and its role as a check on government. He declared that “the Constitution invites, if it does not compel, a generous construction of the power of the federal courts” to grant what he, like others going back to this country’s founding, reverentially called the Great Writ.


Today, federal law governing habeas corpus makes Brennan’s view seem antique. The writ has withered and is in a shabby state. The Supreme Court occasionally lets petitions for writs go forward in federal courts, as it did this past term in a pair of 5-4 rulings. But far more often, the Court thwarts such intrusion, impatiently overturning federal appeals courts that have granted a state prisoner relief based on a habeas petition. These reversals matter because they almost always involve life-and-death stakes: The habeas story is about capital punishment. More to the point, it is about the radical cutback in habeas law as a reliable means of challenging unconstitutional death sentences.

Many reversals involve no dispute between the conservatives who control the Roberts Court with five seats and the four moderate liberals. The decisions are often per curiam—by the court—which is to say, collectively and unanimously. Per curiam opinions are supposed to be for uncontroversial rulings, but that is not how the current Court regularly uses them.

Last year, for example, with a per curiam opinion, the Court made a typical reversal of a grant of a habeas writ, this one by the Sixth Circuit, which encompasses Michigan, Ohio, Kentucky, and Tennessee. The appeals court’s majority opinion described “flagrant prosecutorial misconduct” at the trial of the petitioner, concluding that the prosecutor’s comments “so infected the trial with unfairness as to make the conviction a denial of due process.” The Court rebuffed that and other well-delineated reasons for granting habeas as “the flimsiest of rationales.”

The main advocate for habeas retrenchment after the heyday of the Warren Court was Justice William Rehnquist, who ardently believed that states should be free to run their systems of criminal justice without federal interference, and the best way to avoid that was by drastically constraining the federal habeas writ.

After Rehnquist became chief justice in 1986, the Supreme Court moved to the right about habeas by narrowing the grounds on which the writ could be used. The restrictions were procedural but had serious substantive consequences. They blocked federal habeas claims, for example, when a state court denied them on non-federal grounds, even if the petitioner’s constitutional rights might well have been violated.

As part of his campaign to restrict habeas, Rehnquist in 1988 persuaded the American Bar Association (ABA) to recommend steps for reform. A commission of ten experts with divergent outlooks—from a law-and-order state chief justice and a pro-death-penalty state deputy attorney general, to a leading anti-death-penalty trio including a lawyer, a scholar, and a state court justice—made a comprehensive review of habeas.

Everyone on the commission agreed that reform was vital. But the members disagreed about whether they needed to concentrate on speeding up the habeas process or making it fairer. In 1986 and 1987, when a total of 43 death-row inmates were executed, the average delay between conviction and execution was about seven years. Habeas critics blamed the delay on frivolous appeals and the inexplicably long time it took to complete each step of a capital case, from trial through final appeal. Habeas defenders attributed the delay to a process that was admittedly arcane yet should not be rushed since a life hung in the balance, and to the time it took to discover the many ways an inmate was squashed by American criminal justice.

A book-length report laid out the evidence the commission had gathered and concluded that one would be “grossly naïve to think that delay arises simply from repetitious filings in federal court at the eleventh hour by unethical lawyers who hope to get stays of execution from anti-death-penalty judges using their discretion as a subterfuge.” Delays were caused by a host of reasons: from incompetent counsel; from reviewing the elaborate record in a capital case; from drawn-out state court contemplation about the meaning of cruel and unusual punishment; from uncertainty about when it was OK to petition for federal habeas review; from the time it took to investigate new facts; from assessing how new developments in the law affected a case; and, finally, “from the understandable inclination of both litigants and their attorneys to postpone the ultimate sanction”—executing an inmate. The commission also reported that “many state supreme court judges and justices either feel no resentment when federal courts address the merits of state death-row inmates’ claims or actually invite federal consideration.”

The commission made 16 recommendations for habeas reform. Six were about providing competent counsel at every stage of challenging a capital conviction and death sentence. (A former Supreme Court law clerk told the commission what he had learned during that experience: “The death penalty frequently results from nothing more than poverty and poor lawyering.”) Other recommendations were intended to speed up the habeas process by adding time constraints, limiting challenges following an appeal of the conviction and sentencing, and narrowing the scope of habeas in most instances to claims that the petitioner was innocent rather than that he suffered a violation of his constitutional rights.

The ABA’s endorsement of the recommendations suggested that the commission had forged a solid consensus, with compromises on both sides. When the House and Senate got around to passing a new statute to address imperfections of habeas law, it was hoped they could adopt the ABA model already accepted by conservatives and liberals.


The dark age of federal habeas law began in 1996, with the passage of a federal statute called the Antiterrorism and Effective Death Penalty Act (AEDPA). In the words of the leading treatise about habeas law, this was the moment “Congress dropped the atomic bomb of AEDPA on the federal judiciary, shattering the preexisting structure of habeas corpus law.” The statute seemed to take the association’s recommendations and twist them into more restrictive versions.

The basics of the new habeas provisions were part of Newt Gingrich’s 1994 Republican “Contract with America.” They went from being right-wing pipe dreams to stringent law-and-order options in 1995, after the bombing of the Murrah Federal Building in Oklahoma City killed 168 people, including 19 children in a day-care center.

Senator Orrin Hatch, a Republican from Utah, took the position that federal habeas proceedings gave convicted criminals the chance “to relitigate claims already considered and rejected by State courts.” Habeas law, he argued, allowed a single federal trial judge to undo three levels of time-consuming state court work: the trial, a regular appeal, and a post-conviction review. In death-penalty cases, he insisted, delays due to habeas had “seriously eroded the public’s confidence in our criminal justice system.”

A bipartisan Emergency Committee to Save Habeas Corpus, led by former Democratic and Republican attorneys general from the Johnson, Nixon, Ford, and Carter administrations, presented a different view. Nicholas Katzenbach, a former Johnson attorney general, testified before the Senate that “undoubtedly the writ does raise problems of federalism.” The habeas process should and could be streamlined, the committee argued, but Congress would create a greater problem by compromising the Bill of Rights, as the legislation would clearly do.

Republicans in Congress attached the habeas provisions to an anti-terrorism bill of the Clinton administration. Bill Clinton, running to the right for re-election as president, beat back opposition to the bill in both houses of Congress, insisting that it be passed. A year and a week after the bombing, it was. That was the year that annual death sentences in the United States peaked at 315.

For the first time, the law imposed a statute of limitations on habeas petitions, with a complexity that ensured most defendants would not be able to comply with the limit; most defendants do not have a lawyer, since the right to counsel generally does not apply in habeas cases. The law also made it nearly impossible to qualify for filing a second or subsequent petition for a writ.

Most significantly, it adopted a novel concept that the emergency committee considered unconstitutional. It said that a federal court cannot grant a writ of habeas corpus to an inmate in state prison solely because a state court misapplied the Constitution. Instead, the inmate must prove that the state court’s decision was “contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” by the time the inmate was convicted and sentenced. In other words, even if a federal court is certain a state court decision is wrong and there are excellent reasons that view is correct, that does not make it “unreasonable.” This draconian provision attacked the writ at its core as a check on government.


In 2000, James Liebman of Columbia Law School and colleagues published “A Broken System,” a groundbreaking study of every American capital appeal in the 23 years before AEDPA was passed. It found precisely the opposite of what Congress said was needed. In that span, 34 states imposed 5,760 death sentences and executed 313 inmates. Liebman and his colleagues learned that, in almost seven out of every ten cases, a reviewing court decided that the trial court had made a serious, reversible error. State courts found errors in almost half the capital cases. Federal courts found serious errors in 40 percent or more of the cases they reviewed—after state courts had upheld those convictions and death sentences.

To the authors, the study affirmed the view that many capital appeals took too long, not for the reasons AEDPA supporters claimed but “because American capital sentences are so persistently and systematically fraught with error that seriously undermines their reliability.” There were two persistent reasons for errors: “(1) egregiously incompetent defense lawyers who didn’t even look for—and demonstrably missed—important evidence that the defendant was innocent or did not deserve to die; and (2) police or prosecutors who did discover that kind of evidence but suppressed it, again keeping it from the jury.” The study proved that many on death row should not be there and that their principal avenue for relief—filing a writ of habeas corpus—had been closed off.

“A Broken System” helps explain why the American Law Institute, the country’s most prestigious nonpartisan group of scholars, judges, and lawyers, voted “overwhelmingly” (its word) in 2009 to support a resolution against the death penalty. The institute reported on the “major reasons why many thoughtful and knowledgeable individuals doubt whether the capital-punishment regimes in place in three-fourths of the states, or in any form likely to be implemented in the near future, meet or are likely ever to meet basic concerns of fairness in process and outcome.”

One reason is the huge costs of trials for people charged with capital crimes, of the intricate appeals process, of keeping prisoners on death row, and of maintaining a corrections unit that meets constitutional standards for carrying out executions. Another reason is the near impossibility of expunging racial bias from a criminal-justice system that disproportionately imposes death sentences on blacks who kill whites. Still another is the long-unmet challenge of providing capital defendants with lawyers who have the skill, experience, and expertise to represent them ably.

The institute’s concerns didn’t end there. In the 32 states that now have the death penalty, 28 have elected judges who are more likely to feel pressed by politics to stand up for the death penalty. There is also the likelihood that, after some people have been executed, evidence will then surface proving they were innocent: Eighteen people who served time on death row for an average of 11 years have been exonerated since 1993 with DNA evidence.

Six states in the past six years have abolished the death penalty, bringing the total to 18 that do not allow capital punishment. There is a de facto moratorium in a dozen more states that still have the death penalty: None of these has executed anyone for at least seven years. That makes 30 states that have abolished capital punishment or have long paused from carrying out executions. This death-penalty-free zone is even larger, since prosecutors pursue most death-penalty cases in counties where just one-eighth of the American population lives.


In the meantime, AEDPA operates in its alternate universe. This summer, Warren Lee Hill, with an IQ of around 70, waited for the Supreme Court to decide to hear his lawyers’ challenge to his death sentence. While serving a life term in a Georgia prison for murdering his girlfriend, he killed another inmate. A Georgia state court sentenced him to death in 1993. Under a 2002 Supreme Court ruling, it is unconstitutional to execute someone who is intellectually disabled (some courts still use the term “mentally retarded”).

Hill’s state case was a close one, because he was considered mildly disabled. This year, at 52, he functioned at the level of a sixth-grader. Alone among the states, Georgia requires proof of this disability beyond a reasonable doubt. Seven experts were divided about whether Hill met that severe standard: Four experts for the defense concluded he did; three for the prosecution said he did not. A state trial judge said that, if the Georgia standard of proof were the same as in most states—where a claim of being intellectually disabled would have to be shown to be more likely than not—he would have found Hill disabled and exempt from execution. But under Georgia’s more demanding standard, the judge upheld the death sentence. State and federal appeals courts have upheld that sentence again and again.

In February, Hill was within half an hour of being put to death by lethal injection when the federal United States Court of Appeals for the 11th Circuit, which encompasses Alabama, Georgia, and Florida, was persuaded to stay the execution. In response to publicity about the case, the three state experts changed their mind and explained why in sworn affidavits. As a result, all of the experts who examined him were of the unanimous opinion that he was disabled—beyond a reasonable doubt.

In a brief to the federal appeals court, lawyers for Hill wrote that the new evidence proving his disability “was previously unavailable” and that, under these facts, “no reasonable fact finder” would have found Hill “eligible for the death penalty.” His lawyers sought life without parole for their client. In April, a three-judge panel of the 11th Circuit ruled 2-1 against him and vacated the stay on his execution. In the view of Judge Frank M. Hull for the court’s two-judge majority, while Hill presented new evidence, it was in support of the same claim he had made in a previous habeas petition—that his intellectual disability made him ineligible for the death penalty. The court had to reject that old claim under AEDPA, Judge Hull wrote, because the statute “greatly restricts the power of federal courts to award relief to state prisoners” in order “to advance the principles of comity, finality, and federalism.” The court could consider only a new claim asserting Hill’s innocence, the majority said. Since he accepted his own guilt, what business did he have in federal court?

It is a measure of how draconian AEDPA is, and the Court’s application of it, that the country’s leading habeas scholars felt compelled to file a brief with a stern message for the justices: If you don’t grant relief, it will mean that AEDPA has eliminated the power of the federal courts to grant relief as a last resort even in a “rare and extraordinary” case like Hill’s that presents “a patently meritorious, life-or-death constitutional claim.” The state of Georgia scheduled Hill’s execution for July, unless the Court stayed his execution again.

In dissent, Judge Rosemary Barkett wrote, “The perverse consequence of such an application of AEDPA is that a federal court must acquiesce to, even condone, a state’s insistence on carrying out the unconstitutional execution of a mentally retarded person.” She went on, “The idea that courts are not permitted to acknowledge that a mistake has been made which would bar an execution is quite incredible for a country that not only prides itself on having the quintessential system of justice but attempts to export it to the world as a model of fairness.”

Since AEDPA became law, most of the federal courts that have interpreted the statute have said it obliged them to defer to state court rulings. The statute has largely kept federal courts from reaching the merits in countless cases where lawyers made compelling arguments that the prisoner deserved to win under the Constitution; the statute has largely eliminated the federal writ of habeas corpus for state prisoners. In too many AEDPA cases, federal courts have allowed inmates on death row to be executed in models of unfairness.

There now exists an illogical chasm between law and justice—between, on the one hand, the statute’s reflexive antagonism to challenges to the death penalty and, on the other hand, much of the country’s reckoning with the failure of states to administer capital punishment constitutionally.

AEDPA and the Supreme Court all but hide this chasm—and the injustice it regularly leads to—behind byzantine rules and rulings, making it exceedingly hard for our legal system to keep the Constitution’s promise that habeas will be available to prevent the most serious deprivations of liberty. Rather than a grand means to break an evil spell, AEDPA is the opposite. 

You may also like