concurring:
Today, more than thirteen years after a state court levied an unconstitutional death sentence against Dewey Coleman, a federal court has invalidated that punishment. While the majority properly considers only one of Montana’s unlawful acts, the fact remains that the state’s prosecutors and courts committed a series of errors that are extraordinary both for their breadth and their egregiousness.1 The history of Montana’s unrelenting effort to hang Dewey Coleman illustrates not only the failings of our legal system but also its saving graces. In a more perfect world, Dewey Coleman would not have lived under a death sentence for over a decade, and protracted litigation would not have sapped the limited resources of state and federal courts. In a less perfect world, a court system that had grown impatient with his numerous appeals would already have overseen Dewey Coleman’s execution.
I write separately today not to repeat any of the arguments thoughtfully presented for the court by Judge Thompson. I concur without reservation in his opinion. I add my additional comments only in order to point out that the case of Dewey Coleman illustrates the fact that curtailing the federal habeas corpus procedures in death penalty cases would seriously undermine our system of justice and our commitment to constitutional values.
I.
In 1975, Coleman was sentenced to death for the crime of aggravated kidnapping. Constitutional error riddled the proceedings.2 Despite glaring deficiencies, it was *1293not until after thirteen years and thirteen court proceedings that we finally granted relief.3 Dewey Coleman’s experience is not atypical for a death row inmate seeking constitutional relief. Many prisoners spend more than a decade on death row before federal courts vindicate their years of litigation. See infra § III. These peripatetic passages through our legal system have raised serious questions about both habeas corpus and the practicality of the death penalty. Critics of the former have argued that the extended process undermines judicial finality and threatens the efficient functioning of the federal courts.4 Some have even suggested that the writ be streamlined or abolished.
I do not think that ... [the Supreme Court] ... can continue to evade some responsibility for this mockery of our criminal justice system. Perhaps out of a desire to avoid even the possibility of a “Bloody Assizes,” this Court and the lower federal courts have converted the constitutional limits upon imposition of the death penalty by the States and the Federal Government into arcane niceties which parallel the equity court practices described in Charles Dickens’ “Bleak House”.
Coleman v. Balkcom, 451 U.S. 949, 958, 101 S.Ct. 2994, 2995, 68 L.Ed.2d 834 (1981) (Rehnquist, J., dissenting from denial of certiorari). I agree with Chief Justice Rehnquist that there are lessons to be gleaned from the federal habeas experience in death penalty cases; but because I believe that the substantial constitutional issues raised by defendants such as Dewey Coleman are much more than “arcane nice*1294ties”, I would conclude that the mockery of our criminal justice system lies not in repetitive federal review but in the persistent disregard by our courts of fundamental constitutional rights.
II.
No analysis of the habeas process is complete without consideration of its historical background. The story of the Writ of Ha-beas Corpus begins with the birth of the English Common Law. See C. Antieau, The Practice of Extraordinary Remedies 1 (1987). The Great Writ “is perhaps the most important writ known to the constitutional law of England, affording as it does a swift and imperative remedy in all cases of illegal restraint or confinement.” Secretary of State for Home Affairs v. O’Brien, 1923 A.C. 603, 609 (H.L.). Its lineage in American jurisprudence is no less august, extending from the earliest days of colonial law through the Constitution5 to modern times. Although in form simply a method of procedure, the writ of habeas corpus has long stood as a bulwark against arbitrary and illegal imprisonment; “its history is inextricably intertwined with the growth of fundamental rights of personal liberty.” Fay v. Noia, 372 U.S. 391, 401, 83 S.Ct. 822, 828, 9 L.Ed.2d 837 (1963). In many ways, the history of the Great Writ is the history of constitutional liberty in this country.
The historical role of federal habeas review of state proceedings has been more limited. The contours of federal habeas jurisdiction were sketched in the first days of the new country but were not significantly expanded until the Judiciary Act of 1867.6 The reach of the writ into state prisons has varied with the ebb and flow of Supreme Court jurisprudence. The Noia Court extended the Great Writ deep into state court adjudication, but recent cases have invoked procedural doctrine to bar certain claims in federal court. See, e.g., Wainwright v. Sykes, 433 U.S. 72, 97 S.Ct. 2497, 53 L.Ed.2d 594 (1977) (adopting the cause and prejudice test for unlitigated state claims). These erosions of the Great Writ, however, have not robbed it of its essential value. “If the States withhold effective remedy, the federal courts have the power and the duty to provide it.” Noia, 372 U.S. at 441, 83 S.Ct. at 850. Habeas corpus process over state incarceration still stands as a basic safeguard of our liberties.7
III.
While the historical role of the writ of habeas corpus illustrates its significance in American law, modem practice underscores the need for its continued vitality. Dewey Coleman’s passage through the Montana judicial system symbolizes a problem plaguing death penalty litigation generally. Between 1976 and 1983, of the 41 death penalty cases decided by the Courts of Appeals on the merits, the prisoner prevailed 30 times, or almost 75% of the time. Barefoot v. Estelle, 463 U.S. 880, 915, 103 S.Ct. 3383, 3406, 77 L.Ed.2d 1090 (1983) (Marshall, J., dissenting). “This record establishes beyond any doubt that a very large proportion of federal habeas corpus appeals by prisoners on death row are meritorious, even though they present claims that have been unsuccessful in the state courts, that this Court in its discretion has decided not to review on certiorari, and that a federal district judge has rejected.” Id. To protect the rights of capital defendants, the Supreme Court has erected a complex *1295structure of procedural and substantive rules. However, these protections, often casually treated by state courts, would be rendered virtually meaningless if federal habeas were to disappear. The statistics show convincingly and the experience of Dewey Coleman illustrates that any curtailment of the writ of habeas corpus would be tantamount to federal collaboration in a scheme to deny death row inmates their constitutional rights.
Critics have charged that the high rate of successful habeas appeals signals not an inability of state courts to adjudicate constitutional rights but rather heightened sensitivity of federal courts to death row inmates. While it is true that the federal courts scrutinize death penalty appeals more closely than other cases, the judiciary is doing nothing more than following established constitutional doctrine. “Death, in its finality, differs more from life imprisonment than a 100-year prison term differs from one of only a year or two. Because of that qualitative difference, there is a corresponding difference in the need for reliability in the determination that death is the appropriate punishment in a specific case.” Woodson v. North Carolina, 428 U.S. 280, 305, 96 S.Ct. 2978, 2991, 49 L.Ed.2d 944 (1976) (plurality opinion). I find it troubling that the most determined attacks on the habeas process have come in an area of litigation where the stakes are so high, and the cost of error equals a man’s life.
It is difficult to disagree with the Chief Justice that the results of death penalty litigation threaten to make a mockery of the criminal justice system. However, it is not frivolous appeals or complicitous judges that shake confidence in fair adjudication; rather, “it is difficult to avoid the suspicion that our criminal justice system impeaches its own integrity by producing reversible errors in between half and three-quarters of its [death penalty] cases.” Burt, Disorder in the Court: The Death Penalty and the Constitution, 85 Mich.L. Rev. 1741, 1793 (1987). When state court judges ignore fundamental principles of constitutional law,8 the basic premises of the judicial system are shaken; the vast array of errors encourages speculation about the impartiality and detachment necessary to fair adjudication. This case is a prime example. The Montana Supreme Court had a number of opportunities to correct what amounts to a primer of constitutional error: race and equal protection, due process, cruel and unusual punishment. Yet, the majority of the court failed to do so and experienced little difficulty in rejecting Coleman’s claims.9 Given the unwillingness or inability of some state courts to vindicate federal constitutional rights, ha-beas review of their judgments remains a necessary, as well as desirable, element of our federal system.
IY.
While disagreements over the death penalty habeas process continue to fester, both proponents and opponents of the death penalty agree on at least one issue, that death penalty litigation threatens effective administration of the law. Unlike many ha-*1296beas cases which can be disposed of on the pleadings, see supra n. 4, the gravity of capital cases coupled with the startling high rate of state error mandates intensive federal scrutiny. Dewey Coleman’s case is again illustrative. Over seven years has passed since a habeas petition was docketed with the federal district court for Montana; both a three judge and an en banc panel of this court ultimately subjected his claims to intense review.10 This single case tied up significant federal resources over the last seven years, and there are two hundred more potential death penalty litigants living on death row in California alone. Across the country, the number of potential petitioners has grown rapidly. At the beginning of 1985, there were 1420 inmates on death row in the United States; by the end of the year, 171 prisoners had been added to the executioner’s ledger. By March, 1987, 1,874 inmates languished on death row, an increase of approximately 32% over a 2lk year period. See Bureau of Justice Statistics, U.S. Department of Justice, Sourcebook of Criminal Justice Statistics 1986 at 428-29 (1987). This increase in the number of death row inmates will be reflected in the number of habeas petitions. In fiscal year (FY) 1988, the number of new death penalty cases entering the federal court system is estimated to be approximately 300. In FY 1989, an estimated 345 more death row inmates will file habeas petitions in the federal courts. In FY 1990, we can expect another 425 habeas petitions to flood the district courts. See generally Spangenberg Group, Time and Expense Analysis in Post-Conviction Death Penalty Cases (1988); Spangenberg Group, Caseload and Cost Projections for Federal Ha-beas Corpus Death Penalty Cases in FY 1988 and FY 1989 (1988).11 Because of California’s frequent invocation of the death penalty, the Ninth Circuit will bear a substantial part of the burden. An estimated 76 new death penalty cases will confront this court in 1989, and the number of new entrants will rise to approximately 95 in 1990. Id. If we properly review these habeas petitions, we will be unable to handle our ordinary calender of civil and criminal cases in an efficient and orderly manner.
Since it takes an average of over seven years from the date of sentencing to properly adjudicate a death penalty claim, collateral attacks on capital sentences will create a massive backlog in the federal system. Given the nature of the punishment and the high rate of state court errors, the federal courts must continue to scrutinize these cases with utmost care. But the costs of the fair and accurate adjudication mandated by the Constitution are extremely high; the limited capacities of the district and circuit courts will be challenged, and the ability of the federal system to handle the pressing business of other litigants will be diminished. As long as capital punishment is condoned in our country, extensive review of the death penalty must remain a priority of the federal courts, but, as the figures indicate, this mandatory review will exact a price in the impaired administration of our civil and criminal dockets.
V.
In 1975, Dewey Coleman was sentenced to death; while state and federal courts debated the merits of his claim, he languished on death row for over thirteen years. A great deal of time, effort, and money, both public and private, has been expended, but the fact that this case has finally been adjudicated properly makes the process worthwhile.12 I realize that there *1297are other values — such as finality — that are important to the judicial process, but when the stakes are a man's life, these values pale in comparison to accurate and fair adjudication. If cumbersome administration of the death penalty threatens efficient handling of all other civil and criminal matters, and some changes must therefore be made with respect to death penalty cases, the answer lies not in restricting legitimate appeals but in rethinking the social utility of the death penalty. Until legislatures reassess the wisdom of capital punishment13, exacting scrutiny of capital cases will continue to be the duty of the federal courts. And as long as state courts unconstitutionally sentence defendants to death, the only choice allowed by our laws is for the federal courts to put their judgments to the highest tests of the Constitution.14
. See Coleman v. Risley, 839 F.2d 434, 465 (9th Cir.1988) (Reinhardt, J., dissenting) (discussing those errors in detail).
. The constitutional problems can be roughly divided into four categories of error: the Equal Protection Clause, sentencing procedures, due process, and cruel and unusual punishment. First, Montana's decision to refuse plea bargaining and seek a death sentence raises serious questions of racial bias and discriminatory intent concerning which Coleman has been unable to obtain an evidentiary hearing. While the State offered Coleman’s white codefendant, a hardened criminal, a life sentence, Montana refused to negotiate in good-faith with Coleman — who is black — despite his lack of a criminal record or a violent past, the difficulty in prosecuting a case built almost entirely on the testimony of a confessed murderer, and substantial doubts as to his guilt. Second, during the capital sentencing phase, Coleman was denied an opportunity to present oral argument. The trial court, by formulating, writing, and distributing its final order prior to the sentencing hearing, abdicated its constitutional duty to provide the defendant a fair hearing. The trial court also unconstitutionally based Coleman’s sentence on an unadjudicated offense. Third, Coleman was forced, by statute, to carry the burden of persuasion on the existence of mitigating circumstances and on the issue of whether these mitigating circumstances outweighed the aggravating circumstances, turning the normal method of proof on its head. Fourth, Coleman was ultimately sentenced to death under a new death penalty statute that was passed after he had been tried, convicted, and sentenced under an unconstitutional statute. See Maj. op. passim. Finally, an adjudication of guilt based only upon the dubious and self-interested testimony of a confessed murderer and the minimal physical evidence present here is constitutionally insufficient to support a capital sentence. See Gardner v. Florida, 430 U.S. 349, 97 S.Ct. 1197, 51 L.Ed.2d 393 (1977) (plurality opinion). In sum, serious constitutional error affected al*1293most every aspect of this case, from the passage of the initial Montana death penalty statute to the imposition of the current death sentence.
. Coleman was first convicted and sentenced to death by the Sixteenth District Court of Montana in 1975. The Montana Supreme Court vacated that sentence three years later. State v. Coleman, 177 Mont. 1, 579 P.2d 732 (1978) (Coleman I). On remand, Coleman was again sentenced to death. The Montana Supreme Court affirmed. State v. Coleman, 185 Mont. 299, 605 P.2d 1000 (1979) (Coleman II). After the United States Supreme Court’s decision in Sandstrom v. Montana, 442 U.S. 510, 99 S.Ct. 2450, 61 L.Ed.2d 39 (1979), the Montana Supreme Court reheard argument and again affirmed. See Coleman II. The United States Supreme Court denied certiorari. Coleman v. Montana, 446 U.S. 970, 100 S.Ct. 2952, 64 L.Ed.2d 831 (1980). In early 1981, the Sixteenth District Court of Montana refused post-conviction relief. The Montana Supreme Court affirmed. Coleman v. State, 633 P.2d 624 (1981) (Coleman III). The United States Supreme Court denied certiorari. Coleman v. Montana, 455 U.S. 983, 102 S.Ct. 1492, 71 L.Ed.2d 693 (1982). Thirteen months later, Montana's highest court rejected Coleman’s state habeas corpus petition. Coleman v. Risley, 663 P.2d 1154 (1983). On August 9, 1985, the United States District Court denied Coleman’s petition of habeas corpus. A divided three judge panel of this circuit affirmed. Coleman v. Risley, 839 F.2d 434 (9th Cir.1988).
. The genesis of this hostility towards habeas corpus stems in part from a widely shared mis-perception of a habeas explosion. See Smith, Title 28, § 2255 of the U.S.Code, 40 Notre Dame Law. 171, 175-76 (1964) (listing filing statistics to demonstrate ‘abuse’ of the writ). Statistics do not support this picture of a beleagured federal judiciary. Since separate habeas statistics were first compiled in 1971, the number of claims per prisoner has steadily declined. Although growth in the overall prison population has offset this per capita decline, there has also been a steady growth in the number of federal district court judges and magistrates. Over a long-term perspective — since 1944 — the burden on the federal courts of successive habeas petitions has increased, but “the rhetoric of the boom has outlasted the reality_prisoner’s habeas petitions have declined, and that decline began in the early 1970’s long before the major cases and rules restructuring habeas relief were in place." Resnick, Tiers, 57 So.Cal.L.Rev. 837, 950 (1984). In 1971, at their peak, habeas petitions occupied over 12% of the federal docket; that number dwindled to 5% twelve years later. In addition, while 6.1% of all civil cases reach trial, only 2.4% of habeas cases proceed to the trial stage. Id. at 947, citing Annual Report of the Director of the Administrative Office of the United States Courts 60 (1982). Thus, the evidence does not support the portrait of a federal judicial system tottering under the weight of successive habeas papers. On the other hand, death penalty habe-as cases raise questions of a different magnitude. The severity of capital punishment mandates greater scrutiny of the merits of death row appeals. Since questions of death penalty law often involve complex factual and doctrinal inquiries, death penalty petitions — unlike many other habeas cases — are more likely to survive motions to dismiss or other summary motions. Consequently, these complex questions, fueled by recent expansions in the death penalty, demand a significant amount of the federal courts’ attention. See infra § IV.
. “The privilege of the Writ of Habeas Corpus shall not be suspended, unless when in cases of Rebellion or Invasion the public Safety may require it." Art. I, § 9, cl. 2.
. The extent of this nineteenth century expansion has been hotly debated by courts, compare Noia, 372 U.S. at 415-19, 83 S.Ct. at 836-38 with Stone v. Powell, 428 U.S. 465, 96 S.Ct. 3037, 49 L.Ed.2d 1067 (1976), and by academics, compare Peller, In Defense of Federal Habeas Corpus Litigation, 16 Harv. C.R.-C.L.L.Rev. (1982) (extended to the limits of the Constitution) with Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 Harv.L. Rev. 441 (1963) (limited to attacks on state court jurisdiction).
.Some of the most influential civil rights decisions of our time have resulted from habeas corpus petitions filed by state prisoners. See, e.g., Gideon v. Wainwright, 372 U.S. 335, 83 S.Ct. 792, 9 L.Ed.2d 799 (1963); Malloy v. Hogan, 378 U.S. 1, 84 S.Ct. 1489, 12 L.Ed.2d 653 (1964).
. The high state court error rate stems from several sources. First, despite Justice Powell’s protestations to the contrary, see Stone v. Powell, 428 U.S. at 493 n. 35, 96 S.Ct. at 3052 n. 35, experience suggests that federal courts stand in a better position to adjudicate constitutional rights. This may be a function of greater receptivity of federal courts to Supreme Court dictates, insulation from majoritarian pressures, and even superior technical competence. See generally Neuborne, The Myth of Parity, 90 Harv.L.Rev. 1105 (1977). The recent experience of California’s Supreme Court forcefully shows that the system of direct election of judges can impose public opinion upon ‘politically-neutral’ constitutional interpretations. Second, mere re-dundance of federal review of state imprisonment poses a formidable barrier to high error rates. Each successive decision diminishes the possibility of unconstitutional executions. For the mathematics of redundancy, see Cover & Aleinikoff, Dialectical Federalism: Habeas Corpus and the Court, 86 Yale L.J. 1035, 1045 (1977).
. In Coleman I, the court, squarely faced with a recent controlling United States Supreme Court precedent, was compelled to correct an earlier constitutional violation in the initial sentence 579 P.2d at 741-42. Thereafter, the Montana Supreme Court had three opportunities to correct the fundamental constitutional errors raised in the habeas petition. In all three instances, the majority incorrectly denied relief.
. The majority and dissenting opinions of the three-judge panel cover 90 pages in the federal reporter. 839 F.2d at 434-523. While I am not certain that mere volume is a perfect indicia of the extent of judicial scrutiny, I suspect that there is, on some occasions at least, a basic correlation.
. The Spangenberg Group’s projections of ha-beas corpus petitions in the federal court system derive from a 50 state survey of Attorney General’s Offices and Public Defender’s statistics. The estimates closely match the figures compiled by the NAACP.
.I am confident that the death penalty litigation in Coleman’s case has now drawn to a close. Although the majority opinion properly does not reach the hypothetical question whether a new death sentence could be imposed if *1297Coleman sought and obtained a new trial and was again convicted, I think the answer to the question is plain. As Judge Thompson eloquently writes for the en banc court, “Because Coleman had no reason to suspect that his decisions at trial would come back to haunt him at a sentencing hearing, we must conclude that he was denied due process when he was resen-tenced to death under Montana's revised death penalty statute.” Majority Op. at 1288. The reasoning is necessarily applicable to any future death sentence imposed on Coleman for the crimes on which he has heretofore been tried. The record of the first trial can never be undone. Any future trial decisions Coleman would make would inevitably be affected by the trial record his counsel has already created.
. Among Western nations, retention of the death penalty is a rarity. In Western Europe, eight countries nominally keep death penalty statutes on the books. In five of those nations— Italy, Malta, Spain, Switzerland, and England— the laws permit capital sentences only for exceptional crimes, such as wartime treason. The other three — Belgium, Greece, and Ireland — retain the death penalty for ordinary homicide. But not one of the eight countries has executed a prisoner within the last decade. See Amnesty International, Death Penalty List of Abolitionist and Retentionist Countries (1988). Clearly, the process of reassessment has taken a different turn in other developed societies.
. It would, of course, be inappropriate to comment here on the recently enacted federal legislation which provides for the imposition of the death penalty in certain cases. See Title VII of the Anti-Drug Abuse Act, P.L. 100-690. No case has yet been decided under that statute.