March 12, 2018
Who Killed Habeas Corpus?
District judge in the United States District Court, Eastern District of Wisconsin
*This piece was originally published in the Winter 2018 issue of Dissent Magazine.
On November 4, 1995, Leandro Andrade walked into a K-Mart in Ontario, California, and attempted to shoplift five children’s videotapes. He was caught by a security guard and promptly arrested. Two weeks later, he walked into another K-Mart in nearby Montclair and tried to steal four more videotapes. Again, he was caught and arrested on the spot. This time, he was tried and convicted in a California state court of two counts of petty theft with a prior conviction. His sentence for stealing $153 worth of VHS tapes? Fifty years in prison.
The staggering sentence was largely a result of California’s three-strikes law: Andrade had been in and out of prison since the early 1980s, including on charges of (nonviolent) burglary, and the shoplifting incidents were the final straw. Prosecuted as felonies, each demanded a minimum sentence of twenty-five years under the California law, adopted just the year before.
Andrade’s appeals in the California courts were unsuccessful, so he filed a petition for a writ of habeas corpus in federal court challenging the constitutionality of his sentence. Ultimately, the case went to the Supreme Court, and on March 5, 2003, the Court ruled against him. Andrade remains in prison today, with no possibility of parole until he’s eighty-seven years old.
Andrade’s case offers a window onto what may well be the most tragic development of the modern legal era: the destruction of habeas corpus as an effective remedy for individuals who are imprisoned as a result of a violation of their constitutional rights by a state court. Once known as the Great Writ of Liberty, habeas corpus has been so extensively diminished that it is no longer a protection against unlawful imprisonment but rather an empty procedure that enables and may actually encourage state courts to disregard constitutional rights.
The writ of habeas corpus came to the United States by way of English common law and is explicitly recognized in the Constitution. In the Judiciary Act of 1789, Congress conferred habeas jurisdiction on the newly created federal courts, thereby enabling prisoners to challenge the legality of their custody in such courts. Two great events in American history established the reach and power of the writ. The first was Reconstruction. Among the important constitutional amendments and statutes passed by the Reconstruction Congress was the Habeas Corpus Act of 1867 in which the benefit of the writ was extended to formerly enslaved people and others convicted in state courts. Thus, for the first time, federal courts were authorized to monitor state criminal proceedings to ensure that defendants’ constitutional rights were respected. The second was a series of decisions by the Warren Court in the 1960s extending the procedural protections in the Bill of Rights to criminal defendants in state courts. Many law-enforcement officials resisted these decisions—including, for example, the rulings requiring the exclusion of evidence resulting from unlawful searches or from interrogations unaccompanied by Miranda warnings—and many state court judges did not enforce them. Through habeas corpus, state prisoners could go to federal court to vindicate their rights.
As the Warren Court era came to an end, however, an increasingly conservative Supreme Court, led by Warren Burger (1969–1986) and then by William Rehnquist (1986–2005), came up with a variety of ways to limit federal habeas review of state prisoners’ constitutional claims. The Court prohibited federal courts from reviewing claims of unlawful search and seizure, barred consideration of claims not properly raised in state court, required federal courts to presume that state courts had found facts correctly, and ruled that federal courts could not consider habeas petitions if they contained even a single claim that had not been exhausted in state court. The Court also prohibited federal courts from granting habeas relief based on a new rule and broadly defined when a rule was new. It barred consideration of second petitions, made it more difficult for federal courts to conduct evidentiary hearings, and required federal courts to find that certain state court errors were harmless. In sum, the Court created numerous new obstacles for habeas petitioners.
Nevertheless, even with the impediments the Court created, a state prisoner generally had the right to have a federal court independently review the merits of her or his constitutional claim. And a federal court had the authority and, in fact, the duty, to grant a writ of habeas corpus if a prisoner was in custody as a result of a constitutional violation. In the mid-1990s, however, things took another bad turn. Congressional Republicans led by Newt Gingrich made habeas corpus “reform” part of their “Contract with America” and drafted a bill containing a hodgepodge of restrictions. They declined to discuss the bill with habeas experts, refused to hold hearings, and fast-tracked the legislation onto the floors of both houses. And although the bill had little to do with terrorism and affected all state prisoners, not just those subject to the death penalty, they named it the Antiterrorism and Effective Death Penalty Act (AEDPA) and, in the wake of the bombing of the federal building in Oklahoma City, passed it with an overwhelming majority.
Sadly, over the objections of habeas scholars, civil libertarians, and his own counsel, Abner Mikva, President Clinton signed the bill in April 1996. One would have hoped that Clinton, who had taught constitutional law, would not have signed a bill weakening the Great Writ. Clinton was well aware of the harm that AEDPA would cause to people whose state court convictions resulted from the deprivation of constitutional rights, but he was willing to sacrifice them for his own political gain. He wanted to be perceived as being “tough on crime,” and habeas corpus had no politically significant constituency. A terrible bill thus became the law of the land.
Two provisions of AEDPA are particularly objectionable. The first, as interpreted by the Supreme Court, requires federal courts to defer to state court rulings that are based on incorrect interpretations of federal constitutional law as long as such interpretations, however wrong, can be regarded as “reasonable.” The second bars federal courts from granting relief based on any authority other than clearly established Supreme Court precedent. This provision bars lower federal courts analyzing constitutional claims from relying on the case law in their circuit. Thus, even if circuit precedent clearly supports relief, if the Supreme Court has not spoken, the writ must be denied. The effect of these provisions is to prevent federal courts from granting relief to many prisoners whose convictions are obtained unconstitutionally.
Although AEDPA was misguided from its inception because it elevates flawed state court adjudications of guilt over individuals’ constitutional rights, the Supreme Court has made it much worse. Interpreting the language of the statute in an astonishingly one-sided and restrictive way, the Court has exacerbated the worst aspects of AEDPA and essentially barred independent federal court review of state prisoners’ constitutional claims. The Supreme Court’s disturbing interpretations of AEDPA began in 2000 in Williams v. Taylor, a decision written by Justice Sandra Day O’Connor, an avid states’ rights enthusiast. In Williams, the Court held that even if a state court misinterprets Supreme Court precedent and thus reaches an incorrect legal conclusion, that error does not require the state court’s decision to be overturned. Then, in Leandro Andrade’s 2003 case involving the stolen videotapes, Lockyer v. Andrade, the Court went further, holding that even if a state court’s resolution of an issue governed by Supreme Court precedent is clearly erroneous, that is not enough to warrant habeas relief.
Since then, the Court has become even more extreme. In 2011, in Harrington v. Richter, it said that habeas corpus was appropriate only for addressing “extreme malfunctions” in state criminal justice systems, namely “where there is no possibility that fair-minded jurists could disagree.” This interpretation is not supported even by the language of AEDPA and, taken literally, would mean that a federal court could never grant relief because to do so it would have to find that none of the state court judges who denied the claim was a “fair-minded jurist.” In Richter, the Court also determined how the statute applied to a state court decision denying relief without giving any reason for its decision. The deference to state courts required by AEDPA applies only to claims “adjudicated on the merits.” Because the state court’s decision in Richter was totally devoid of any explanation, it was unclear that the court had even considered the claim’s merits. Nevertheless, the Supreme Court created a presumption that unexplained decisions were decisions on the merits entitled to AEDPA deference.
As I mentioned earlier, under AEDPA, a federal court can grant habeas relief only if the state court decision unreasonably applies or is contrary to “clearly established federal law, as determined by the Supreme Court.” The Supreme Court’s interpretation of this phrase has also been horribly one-sided. In Andrade, the California courts rejected the defendant’s argument that a fifty-year sentence over $150 worth of videotapes violated the Eighth Amendment’s prohibition of cruel and unusual punishment. A federal habeas court disagreed, ruling that the punishment was “grossly disproportionate”—a principle created by the Supreme Court to determine when punishment is excessive. The Supreme Court reversed, holding that its own decisions regarding gross disproportionality were not clear enough to constitute clearly established federal law.
Then in 2006, in Carey v. Musladin, the Court made things even worse. Mathew Musladin was tried for murder in California, and the family of the deceased, Tom Studer, sat in the front row behind the prosecutor, in view of the jury, wearing large buttons with Studer’s photograph on them. Musladin claimed self-defense, and argued that by allowing Studer’s family to portray the deceased so sympathetically, the court violated his right to be tried by an unbiased jury. Musladin was convicted, and the question on habeas review was whether his due-process right to a fair trial was violated. Supreme Court precedent established that a defendant was entitled to a trial free from impermissible influences on the jury such as being required to appear in prison clothing. The lower federal court granted habeas relief, but the Supreme Court reversed, holding that the state court did not violate clearly established federal law. This was so, the Court said, because the prison-apparel case involved a government-sponsored influence, unlike the “private communications” from Studer’s family. Thus, almost any factual distinction between a prior Supreme Court case and a new habeas case now justifies denying relief.
Finally, there was the Court’s 2011 decision in Cullen v. Pinholster, in which Scott Lynn Pinholster was sentenced to death in state court and sought federal habeas relief based on ineffective assistance of counsel at the penalty phase of his trial. The federal court held an evidentiary hearing at which Pinholster offered powerful mitigating evidence not presented to the jury that tended to explain his aggressive behavior and organic personality disorder. The federal court overturned the death sentence, but the Supreme Court reversed, construing AEDPA to say that a federal habeas court can look only at the record that was before the state courts. Pinholster is a catastrophe for death penalty defendants, whose best claim often is that their lawyer failed to present mitigating evidence that might have caused the jury to impose a less harsh sentence. Evidence supporting such claims is almost never available in the state court record.
Through this series of rulings, the Supreme Court turned a terrible statute, AEDPA, into an utter and unmitigated disaster. The problem arises from the Court’s obvious antipathy to federal court review of state court convictions and its resulting tendency to read every provision in the statute as altering habeas law to the disadvantage of petitioners. The Court often expresses its antipathy in terms of federalism and clearly regards having a federal court overturn a flawed state court conviction as more harmful than imprisoning someone who has been unlawfully convicted. What the Supreme Court refuses to recognize, however, is that state courts, as presently constituted, are institutionally incapable of providing adequate review of their own criminal convictions.
In many states, post-conviction review is very hard for prisoners to come by. Even where such review is accessible, however, state courts are not well positioned to enforce the constitutional rights of criminal defendants. Whereas federal judges have life tenure, judges in most states are elected or subject to appointment or reappointment by officials who are themselves subject to election. For a prisoner claiming she or he has been deprived of a constitutional right, there is a substantial difference between having the claim heard by a judge or judges with life tenure and a judge or judges who must be reelected or reappointed. This problem has gotten worse as judicial elections have become increasingly contested, increasingly involve substantial spending by special-interest groups, and increasingly result in the defeat of incumbents.
Conservative organizations now spend millions of dollars on judicial elections. Although many such organizations are primarily interested in economic-oriented issues such as tort law, their TV ads typically focus on criminal law issues, which have higher salience with voters. My home state of Wisconsin provides an example. In 2008, with nearly $3 million worth of help from the state’s major business lobby, Wisconsin Manufacturers and Commerce, and a similarly oriented right-wing group, Wisconsin Club for Growth, a little-known trial court judge, Michael Gableman, ousted a respected state supreme court justice, Louis Butler, by contrasting his own law-and-order philosophy with the incumbent’s supposedly soft-on-crime stance. Almost all of the corporate money behind Gableman was spent on sham issue ads—attack ads that do not use the words “vote for” or “vote against” and that, therefore, are not considered political under the Supreme Court’s campaign finance jurisprudence and do not trigger disclosure requirements. The ads called Justice Butler “Loophole Louie,” highlighted his dissent in a highly publicized murder case and urged voters to tell him “to stand up for victims—not technicalities.” They resulted in the first defeat of an incumbent Wisconsin supreme court justice in almost half a century.
No matter how impartial and well-intentioned its members, a court made up of elected judges can struggle to maintain balance in a high-profile criminal case. Consider, for example, the case of Ted Oswald, who was convicted of the murder of a police officer in a conservative, suburban Wisconsin county in 1995. Because of the publicity, the state trial judge sequestered the jury pool while he and the lawyers questioned jurors individually in the courtroom. On the fourth day of questioning, a juror testified that several others had repeatedly told the jury pool that Oswald was guilty and that a trial would be a waste of time and money. Naturally, Oswald’s lawyer asked the judge to permit further questioning of the jurors to determine if they had been prejudiced. In an unusual and arbitrary decision, the judge denied the request, clearly fearing that more questioning would result in his having to dismiss the entire jury pool and that county taxpayers would not look favorably on the expense involved in empaneling another pool.
The decisions of the appellate courts in the Oswald case were also affected by concern about public opinion. Both the state court of appeals and the state supreme court attempted to avoid deciding the case, the former by certifying the case directly to the state supreme court, and the state supreme court by declining to accept the certification. Ultimately, after a long delay, the court of appeals affirmed Oswald’s conviction, conveniently overlooking the trial court’s errors in jury selection. (Full disclosure: even under AEDPA, as a federal judge in the Eastern District of Wisconsin, I was able to grant a writ of habeas corpus in this case.)
The threat of being defeated at the polls is real, and it affects judicial behavior. Many other cases and a substantial body of research make this clear. In the years following the 1986 recall of Chief Justice Rose Bird and two colleagues for overturning numerous death sentences, the California Supreme Court upheld death sentences at one of the highest rates in the country. And a 2004 study showed that Pennsylvania sentencing judges became “significantly more punitive the closer they [were] to standing for reelection.” This was so, the authors concluded, even though judges in Pennsylvania serve ten-year terms and face nonpartisan retention rather than contested elections. Another study discovered that elected state supreme courts affirmed 62 percent of death penalty sentences, while supreme courts comprised of judges appointed for life affirmed just 26 percent of such sentences. Further, elected supreme court justices, including justices generally supportive of defendant’s rights, were more likely to affirm death sentences as they approached the end of their terms.
Nor is there anything new about this. In the Federalist No. 78, Alexander Hamilton wrote, “adherence to the rights of the Constitution, and of individuals, which we perceive to be indispensable in the courts of justice, can certainly not be expected from judges who hold their offices by a temporary commission.” In the civil-rights era, elected state court judges often proved unable or unwilling to protect the rights of racial minorities, mirroring the inability of elected state judges to protect the constitutional rights of criminal defendants today. And in 1977, in an influential article, civil-liberties lawyer and scholar Burt Neuborne refuted the notion that state courts provided an adequate forum for the adjudication of Fourth Amendment claims, which was the basis for the Burger Court’s troubling decision that federal habeas review of unlawful search and seizure claims was unnecessary.
An elected judiciary is not the only factor that puts state courts in a poor position to decide difficult issues of constitutional criminal procedure. Many state courts have heavy caseloads, limited staff, little experience in addressing such issues, and extremely conservative legal cultures. As a federal judge, I have observed a considerable number of cases where state courts overlooked clear constitutional violations, including cases where lawyers were grossly ineffective—failing, for example, to contact exculpatory witnesses, obtain a copy of the defendant’s file from prior counsel, or read police reports because the officer’s handwriting was hard to read. On another occasion, Wisconsin trial and appellate courts considered a case six times without noticing that police officers had obtained admissions from a defendant in a homicide case without providing him with Miranda warnings.
For all of these reasons, federal judges are far better positioned than state court judges to protect the constitutional rights of criminal defendants. And protecting such rights, often in cases involving minority defendants, is of course one of the paramount functions of courts. Unfortunately, the destruction of federal habeas corpus by AEDPA and the Supreme Court has made this function virtually impossible for federal courts to perform. Along the way, the Supreme Court’s habeas jurisprudence has not only undermined the constitutional rights of individuals but also the authority of federal courts to decide cases as provided in Article III of the Constitution. The Supreme Court has done this by forcing federal courts to defer to interpretations of federal law by state courts that are incorrect. Federal courts are now compelled to place their stamp of approval on constitutional error.
Two additional points require mention. The first is that the Supreme Court has been extraordinarily relentless in dismantling the writ. As Judge Stephen Reinhardt explains, contrary to its usual practice and to its statements that it should not accept a case for review for the mere purpose of correcting an error, the Court has increasingly granted review in habeas cases solely to reverse a circuit-court decision that the Court considered insufficiently deferential to a state court judgment, even though, as discussed, that judgment may have been wrong. Between 2007 and 2015, the Court summarily reversed some fifteen cases in which circuit courts had granted writs on the ground that the state court’s adjudication, whether or not mistaken, was not unreasonable. In short, procedurally as well as substantively, the Court takes every opportunity it can to eliminate meaningful federal review.
Second, the Clinton and Obama appointees to the Court seem to have lost whatever appetite they once had for contesting the conservative majority on this issue. They appear to have given up and accepted the right-wing justices’ destruction of habeas corpus as a settled matter not to be further disputed. The statistics bear this out. In 2006, as Judge Reinhardt notes, the Court issued five 5-4 decisions regarding whether a state court decision on the merits was contrary to or an unreasonable application of clearly established federal law and in three found that AEDPA did not bar relief. From 2007 to 2013, however, the Court decided twenty-eight such cases and denied relief in twenty-six. And of the twenty-six denials, only six engendered a dissent and only one dissent received four votes.
Inasmuch as deference to state court decisions has emerged as a suffocating force in most habeas cases, a number of scholars have attempted to determine what the appropriate next step might be. Two have suggested substantially abandoning federal oversight of state convictions. Others have proposed different forms of review than federal courts have traditionally performed. The fact is that, short of repealing AEDPA and the Supreme Court’s decisions interpreting it, there is no satisfactory response to what the Court has done and no way to escape its tragic consequences. Except when a state has done something truly shocking in the course of convicting someone (and even then, only sometimes), habeas corpus no longer provides a meaningful avenue for protecting the constitutional rights of criminal defendants.