III. Purposes of Collateral Review.

There have been a number of proposals to restrict the availability of collateral review and some vehement assertions that we should leave it untouched. Each proposal for change and each opposition to change is based on assumptions about the purposes of habeas corpus.

A. Forcing courts to toe the mark.

The traditional view of the principal purpose of habeas corpus as a collateral attack was summarized by Justice Harlan in Mackey v. United States⁽³⁸⁾: "The primary justification given by the Court for extending the scope of habeas to all alleged constitutional errors is that it provides a quasi-appellate review function, forcing trial and appellate courts . . . to toe the constitutional mark."

As applied to federal habeas for state convicts, this notion rests on the idea that state court judges will reverse convictions they would otherwise affirm simply to avoid having them set aside later on federal habeas corpus. There are a number of problems with this idea. First, it assumes state judges lack integrity, a point discussed further below. Second, a mark cannot be "toed" unless it is visible. There is no dispute in state courts today that the Constitution must be obeyed or that the precedents of the United States Supreme Court must be followed.⁽³⁹⁾ The problem lies in figuring out what these precedents actually require in a given case. While the Bill of Rights itself consists of only a few, relatively simple rules, the vast, complex body of caselaw that has been grafted onto it⁽⁴⁰⁾ necessarily produces honest, good-faith disagreements over what the law requires.

In state courts today, it would be an extremely rare occurrence for a court to simply refuse to follow a clear precedent. Congress could rely on the Supreme Court's certiorari jurisdiction to deal with such cases, or it could, with carefully limited wording, authorize federal habeas relief when the state courts are wrong beyond question. Assuming for the sake of argument that there is a continuing need to insure that some state courts "toe the mark," de novo review of every borderline question of law is an intrusion far beyond the need.

B. Correcting fundamentally unjust incarcerations.

The oldest and most important purpose of habeas corpus is to set free those prisoners who have done nothing illegal. The celebrated Bushell's Case,⁽⁴¹⁾ for example, freed a juror held in contempt for bringing in the "wrong" verdict. The concept of jurisdiction was first expanded beyond its usual meaning to cover cases where the defendant had been convicted of violating an unconstitutional statute and was therefore actually innocent of any crime.⁽⁴²⁾ The early habeas cases on procedural claims involved circumstances raising grave doubt whether the defendants were actually guilty.⁽⁴³⁾

Opponents of reform like to emphasize the role of habeas corpus in saving the innocent from imprisonment or execution,⁽⁴⁴⁾ but the reality is that habeas is a very poor fit to that function. Until fairly recently, the actual guilt or innocence of the petitioner was virtually irrelevant to habeas corpus cases,⁽⁴⁵⁾ and even today it serves only as a "safety valve" to open the door to procedural claims which are otherwise defaulted.⁽⁴⁶⁾

If the question is who committed the act (as opposed to mental defenses), the reality is that relatively few criminal cases and only a tiny handful of capital cases involve substantial questions of innocence. In a study by the Criminal Justice Legal Foundation of 100 capital habeas cases from the Eleventh Circuit, only one case involved a significant claim of mistaken identity.⁽⁴⁷⁾ The Supreme Court recently confirmed that independent research shows that cases of federal habeas petitioners who can make a showing of actual innocence by any standard are very rare.⁽⁴⁸⁾ The few cases of innocent people being sentenced to death have generally been resolved in state proceedings, not on federal habeas corpus.

In the case of Schlup v. Delo, Centurion Ministries filed an amicus brief along with five former death row inmates who had since been freed. Yet far from showing a compelling need for federal habeas for the wrongly convicted, the stories of these five men show exactly the opposite. Four of the five had obtained their relief through direct appeal, state collateral review, executive clemency, or some combination of these remedies.⁽⁴⁹⁾

Only one case, that of Joseph Brown, was resolved on federal habeas corpus.⁽⁵⁰⁾ After the grant of federal habeas relief, the prosecution decided that it would not be able to retry this then 13-year-old case.⁽⁵¹⁾ That hardly constitutes a finding that Brown was actually innocent, for practicing attorneys are well aware that stale cases are harder to try than fresh cases, and this is especially so for a party that must surmount the formidable barrier of proof beyond a reasonable doubt. As Justice O'Connor has noted, "Retrial becomes very difficult, and sometimes impossible, when many years have passed since the original trial. Witnesses and evidence become difficult to relocate; memories fade."⁽⁵²⁾

The Arizona case of Mitchell Blazak illustrates the point. Blazak was convicted of two counts of first-degree murder in 1974. Nineteen years later, in a splintered decision, the Ninth Circuit upheld a grant of habeas corpus. In preparing the case for retrial, however, the prosecution discovered that critical evidence had been lost, possibly by the federal habeas court itself. They had to settle for a no-contest plea to second-degree murder, and Blazak was released six months later. Far from preventing a miscarriage of justice, federal habeas corpus has put a double murderer back on the street.⁽⁵³⁾

Opponents of reform assume that a different result on retrial means that an injustice has been prevented.⁽⁵⁴⁾ That assumption is unjustified. The kinds of marginal errors that typically result in disputed questions of law today are rarely more harmful to the search for truth than is the passage of time.

Before leaving the area of actual innocence, it is important to correct a misstatement previously laid before Congress by the "Emergency Committee to Save Habeas Corpus":

"In the Coleman v. Thompson case in 1991, a death row inmate with strong new evidence of actual innocence--evidence so powerful and disturbing that Time magazine featured it as a cover story--was denied an opportunity to even have his new evidence heard in federal court, because his lawyer had unwittingly missed a filing deadline by three days. . . . Mr. Coleman was executed.

"Such rulings devalue the protections of the Bill of Rights. No person should pay with his life for the neglect or ignorance of his lawyer."⁽⁵⁵⁾

There is no need to mince words here. This is a patent falsehood, as even the most cursory review of the Coleman case would have revealed.⁽⁵⁶⁾

Coleman's case went to the Supreme Court twice on federal habeas corpus. In the 1991 case, to which the statement refers, the high court expressly acknowledged its prior holding that a strong showing of actual innocence would exempt the prisoner from the procedural default rule, but noted that Coleman had not made such a claim in the proceeding then before it.⁽⁵⁷⁾

The claim of actual innocence came later. The "Emergency Committee's" statement that the evidence was not heard in federal court is false. Coleman filed a second habeas petition seeking to invoke the "actual innocence" exception. Federal District Judge Glen M. Williams did consider this evidence and wrote an extensive, published opinion reviewing that evidence.⁽⁵⁸⁾ Judge Williams concluded that Coleman had not even made a "colorable showing" of actual innocence.⁽⁵⁹⁾ He found particularly compelling the result of a DNA PCR test ordered by the state court at Coleman's request and done by the expert chosen by Coleman's lawyers. That test "significantly bolster[ed] the jury's finding of guilt."⁽⁶⁰⁾ The Court of Appeals affirmed, and the Supreme Court denied a stay of execution, premised squarely on the district judge's finding that Coleman did not even have a colorable claim of innocence, much less a "powerful" one as the "Emergency Committee" claims.⁽⁶¹⁾

While it is true that Time magazine featured the case as a cover story, that choice says more about Time than it does about Coleman. While extensively reviewing the evidence claimed by Coleman to "prove" his innocence, Time simply omitted any mention of the inconvenient DNA evidence.⁽⁶²⁾

In short, Coleman was not executed because of his lawyer's error. Coleman was executed because he raped and murdered Wanda Fay McCoy.

The bottom line on innocence is that de novo habeas corpus relitigation of all federal constitutional questions is a meat-axe approach to a problem that requires a scalpel. If actual innocence is the problem, an answer crafted specifically for that problem can and should be devised.

C. Giving every defendant a federal forum.

One reason often asserted against limitation of habeas corpus is the contention that every defendant should have the opportunity to litigate his federal questions in a federal forum.⁽⁶³⁾ This position is founded on a deep distrust of state courts. It asserts, in essence, that state courts are incompetent to adjudicate federal questions.

Our judicial system has not historically been structured that way. It has been understood from the very beginning that state courts could pass on federal questions.⁽⁶⁴⁾ Indeed, the Constitution itself expressly directs them to do so.⁽⁶⁵⁾ If a state court decides a question of civil rights law in the course of deciding a civil case, for example, that decision is binding on the parties and cannot be relitigated in federal court.⁽⁶⁶⁾ The desire for a federal forum, by itself, is similarly insufficient to justify collateral attack on final judgments in criminal cases. A federal forum may be necessary where state remedies are grossly inadequate, as they may have been decades ago.⁽⁶⁷⁾ The contention that federal court resolution of every federal constitutional issue in every criminal case in every state is essential today, however, is unsupportable.⁽⁶⁸⁾

     1. Elected judges.

The need for a federal forum is sometimes premised on the fact that most state judges do not have life tenure. Congress has already rejected that argument. As part of "home rule" for the District of Columbia, Congress established a local court system without life tenure.⁽⁶⁹⁾ This court system has the authority to finally adjudicate federal constitutional questions, subject only to review on certiorari by the United States Supreme Court; habeas review in the federal district court is prohibited, unless the local remedy proves "inadequate or ineffective."⁽⁷⁰⁾ This limitation is constitutional.⁽⁷¹⁾

In the state judiciaries, a few have life tenure, some have retention by a means other than election, a plurality have "retention elections," in which the justice stands for a yes/no vote without a named opponent on the ballot, some have nonpartisan elections, with a named opponent but no party label, and some have partisan elections. Of the 50 states, the number with each type breaks down as follows:⁽⁷²⁾

Only 19 states require appellate justices to face named opponents on the ballot. Nineteen others have a system which requires an enormous groundswell to unseat a sitting justice. In California, for example, such a movement has only succeeded once in 60 years, after a court majority personally opposed to the death penalty had repeatedly abused its power, applying "a standard of review so strict . . . that it virtually guarantee[d] that the law [would] not be enforced."⁽⁷³⁾ The impact of elections on the state appellate judiciary is vastly overestimated.

The essence of the "federal forum" argument is a picture of justices who face reelection as abandoning the Constitution and fleeing in terror from a horde of club-wielding, Neanderthal voters. Justice Brennan's dissent in Butler v. McKellar⁽⁷⁴⁾ is typical of the bottomless disdain for state judges that underlies this argument:

"Because state courts need not fear federal habeas review so long as they avoid clearly unreasonable constructions of existing doctrine, they will have no incentive to reflect carefully about existing legal principles and thereby to develop novel and more sophisticated understandings of constitutional guarantees."⁽⁷⁵⁾

State judges have exactly the same incentives as federal judges: respect for the Constitution, fidelity to their oaths of office,⁽⁷⁶⁾ and dedication to the rule of law and to the rights of people. If Justice Brennan's view of state judges were correct, there would never be any cases of state courts expanding protections for defendants on independent state grounds, nor would there be any cases of state courts overestimating the extent of federal protections and being reversed by the U. S. Supreme Court. Yet both types of cases occur regularly, even in the much-maligned South. Here are just a few of many examples:

• In California v. Ramos,⁽⁷⁷⁾ the Supreme Court held that a state may constitutionally inform the jury of the governor's commutation powers, reversing a state court decision in favor of the defendant. In State v. Jones,⁽⁷⁸⁾ the Louisiana Supreme Court declined to follow Ramos and struck down a similar state statute on state constitutional grounds, although there was no relevant difference in the language of the state constitution.

• In State v. Middlebrooks,⁽⁷⁹⁾ the Tennessee Supreme Court held that the circumstance of killing in the course of a felony could not be both a part of the definition of murder and an aggravating circumstance making the case eligible for capital punishment. That position had previously been accepted by only one federal circuit, which had already reversed itself and rejected that position before the Middlebrooks decision.⁽⁸⁰⁾ When the U. S. Supreme Court granted certiorari in Middlebrooks and seemed poised to reverse, the Tennessee Supreme Court declared that its decision rested on independent state grounds.⁽⁸¹⁾

• In Arizona v. Youngblood,⁽⁸²⁾ the U. S. Supreme Court held that accidental destruction of evidence of no clear exculpatory value does not violate the Due Process Clause. The Alabama Court of Appeals declined to follow Youngblood and reversed a conviction on state constitutional grounds in Gurley v. State.⁽⁸³⁾

• In Kastigar v. United States,⁽⁸⁴⁾ the U. S. Supreme Court held that a witness who claims the privilege against self-incrimination may be compelled to testify with a grant of "use" immunity. In State v. Thrift,⁽⁸⁵⁾ the South Carolina Supreme Court declined to follow Kastigar, and required complete "transactional" immunity.

In each of these cases, there was no significant possibility that a decision against the defendant would have been overturned on habeas corpus. If the view of state courts typified by Justice Brennan's Butler dissent were correct, all of them would have been decided the other way. That view, therefore, is fatally flawed.

     2. The forty percent figure.

An argument commonly made to justify de novo federal review of state capital cases is based on the now-outdated statistic that 40% of habeas petitions had resulted in some grant of relief.⁽⁸⁶⁾ There is less to this statistic than meets the eye.

If one flips a coin twice, the second flip will "reverse" the first flip 50% of the time. Deciding capital cases in the 1970's and 1980's was not quite a coin toss, but it was a very uncertain process, due to the ever-changing state of the law:

• Texas, until fairly recently, had a system for sentencing capital murderers by asking the jury three statutory questions and basing the sentence on their answers. Is that system constitutional? The Supreme Court upheld it in 1976,⁽⁸⁷⁾ struck it down for a seemingly broad class of cases in 1989,⁽⁸⁸⁾ and then sharply narrowed its disapproval in 1993.⁽⁸⁹⁾

• Florida originally enacted a sentencing system in which the jury was given a list of mitigating factors without a "catchall" instruction to include other factors not on the list. The Supreme Court approved this system in Proffitt v. Florida,⁽⁹⁰⁾ and then struck it down in Hitchcock v. Dugger.⁽⁹¹⁾
• After implying that mandatory sentencing was required in Furman v. Georgia⁽⁹²⁾ and then striking it down in Woodson v. North Carolina,⁽⁹³⁾ the Supreme Court expressly noted a possible exception for murders committed by life prisoners, who are otherwise effectively immune from criminal punishment.⁽⁹⁴⁾ The Court took another ten years to finally answer the dangling question in the negative.⁽⁹⁵⁾

Thus, when state court judges fail to anticipate the next hairpin turn in the Supreme Court's "annually improvised" jurisprudence of capital sentencing,⁽⁹⁶⁾ it is not from a lack of diligence or integrity, but merely from a lack of clairvoyance. Decisions made from an honest, thorough examination of the Supreme Court's precedents at any point in time very often turned out to be "wrong" years later. Where that "error" ran in the defendant's favor, it stood uncorrected.⁽⁹⁷⁾ Where it ran against the defendant, it added to the 40% statistic. The judgment was overturned even though the state courts diligently obeyed every rule in effect at the time of the decision.

It stands to reason that if rapid change in the law was the primary reason for the high grant rate, then a settling of the law would dramatically lower the grant rate. That is, in fact, exactly what has happened. Appendix A describes earlier studies of the grant rate and a new study of cases decided in a recent 15-month period. The rate has dropped dramatically over time. Habeas relief is granted as to the guilt determination in only 5% of cases, the same as noncapital cases, and only 15% overall. Outside the often-reversed Ninth Circuit, moreover, the overall grant rate is only 8%, and the guilt phase grant rate is only 4%.

The 40% figure, which has been the cornerstone of the anti-reform argument to date, is no longer valid.