Gennero Galtieri v. Louie L. Wainwright, Director, Division of Corrections, John Matera v. Louie L. Wainwright, Director, Division of Corrections

GOLDBERG, Circuit Judge

with whom TUTTLE, Circuit Judge, joins dissenting.

The en banc court today lets pass an opportunity to discard an anomalous and *367 unfortunate limitation on the availability of federal habeas corpus relief for state prisoners. Mesmerized by a specter of its own creation, a litigious prisoner so diabolically clever that he may be counted on to outwit state’s attorneys and federal district judges, the court today condemns real men of flesh and blood, untutored and unlettered in law, to years of unconstitutional confinement. Perhaps we as judges cannot be expected to understand the full meaning of imprisonment and to respond to the full measure of injustice visited upon one whose unlawful incarceration is prolonged for a year or more. We can, however, be expected to understand and respond to the historic office of the Great Writ, provision of “a swift and imperative remedy in all cases of illegal restraint or confinement.” Braden v. 30th Judicial Circuit Court, 410 U.S. 484, 490 , 93 S.Ct. 1123, 1127 , 35 L.Ed.2d 443 (1973), quoting Secretary of State for Home Affairs v. O’Brien, [1923] A.C. 603, 609 (H.L.). With all due respect, today the majority disappoints even this minimal expectation. For the unwitting, naive, or optimistic inclusion of a single unexhausted claim in a prisoner’s petition, the majority would deny a federal hearing on a thousand exhausted claims. This is a new species of poisonous tree whose bitter fruit, I am convinced, is the frustration of justice. Respectfully, I dissent.

I.

Judges Roney and Thornberry properly reject the majority’s rigid rule that the district courts are required to dismiss mixed petitions. I fully agree with my separately dissenting brethren on this point. 1 Indeed, Judge Roney’s opinion takes us most of the way to the position I endorse. In going somewhat further than Judge Roney, I tread a well-beaten path and conclude that the district courts should be required to consider exhausted claims in mixed petitions unless these claims are' substantially interrelated with unexhausted claims. This rule has commended itself to the judgment of five of the six circuits which have faced directly the problem of mixed petitions. Miller v. Hall, 536 F.2d 967 (1st Cir. 1976); United States ex rel. Levy v. McMann, 394 F.2d 402 (2d Cir. 1968); United States ex rel. Boyance v. Myers, 372 F.2d 111 (3d Cir. 1967) (per curiam); Hewett v. North Carolina, 415 F.2d 1316 (4th Cir. 1969); Tyler v. Swenson, 483 F.2d 611 (8th Cir. 1973), r’aff’d in Triplett v. Wyrick, 549 F.2d 57 (8th Cir. 1977). 2 Only the Ninth Circuit adheres to the view that the majority espouses today. Gonzales v. Stone, 546 F.2d 807 (9th Cir. 1976). While the views of our sister circuits are, of course, not binding upon us, such a coincidence of considered opinion is striking. A lonely sibling bears a heavy burden of justification and explanation when it discovers new imperatives of *368 federalism not apparent to five courts of dignity co-equal with our own. That burden has not been carried.

II.

The critical proposition in Judge Tjoflat’s argument for the rule of complete exhaustion is that the enforceable norm in federal habeas corpus proceedings is a single federal proceeding in which all of a petitioner’s potential claims are considered. This norm is supposedly derivable from the policies and provisions of Rule 9 of the Rules Governing Section 2254 Cases in United States District Courts, 28 U.S.C.A. foil. § 2254 (1977). The rule of complete exhaustion is asserted to be the instrument by which the norm is to be achieved.

Congress has by no means accepted the majority’s critical proposition. I will quite readily concede the propriety of divining from Rule 9 the existence of a general policy favoring the resolution of all federal claims in a single federal proceeding. Insofar as the majority relies on Rule 9 for this much and no more, the court’s reasoning is unexceptionable. However, in going beyond this modest conclusion to suggest that “rule 9 forces a petitioner to assert his claims promptly and in one petition,” majority opinion at 357, (emphasis in original), and to adopt the rule of complete exhaustion in order to give “full effect to the objective of rule 9,” majority opinion at 357, the majority has turned Rule 9 on its head. For the rule of Rule 9(b) governing successive petitions is the rule of Sanders v. United States, 373 U.S. 1, 18 , 83 S.Ct. 1068 , 10 L.Ed.2d 148 (1963): a federal court must consider on the merits a second or successive petition asserting previously unasserted grounds for relief unless the failure to assert the new grounds in the earlier petition constituted abuse of the writ. Abuse of the writ may be found only if the failure to litigate the new claim on the prior application constituted a deliberate bypass or waiver within the meaning of Fay v. Noia, 372 U.S. 391 , 83 S.Ct. 822 , 9 L.Ed.2d 837 (1963). Sanders v. United States, supra. Thus, while the Sanders standard strikes a balance between competing interest, the interest in considering all claims in a single proceeding and the interest in correcting constitutional error, the balance struck is one which precludes dismissal of second or successive petitions except in rare and extraordinary circumstances.

Rule 9(b) authorizes the district courts to dismiss a second or successive petition asserting new grounds for relief if the judge “finds that the failure of the petitioner to assert those grounds in a prior petition constituted abuse of the writ." (Emphasis added.) The language suggested by the Supreme Court would have permitted dismissal if failure to assert the new grounds was “not excusable.” Congress rejected this language and substituted the “abuse of the writ” standard of Sanders v. United States, supra. Congress feared that the “ ‘not excusable’ language created a new and undefined standard that gave a judge too broad a discretion to dismiss a second or successive petition. The ‘abuse of the writ’ standard brings Rule 9(b) into conformity with existing law.” H.R.Rep. No. 94-1471, 94th Cong., 2d Sess. 5 (1976); reprinted in [1976] U.S. Code Cong. & Admin. News pp. 2478, 2482. Thus, Rule 9(b) reflects a deliberate and considered choice by Congress to reinforce the Sanders rule. In relying on Rule 9(b) as primary support for an enforceable norm of a single federal proceeding adjudicating a petitioner’s entire case, the majority wholly disregards the balance struck by Sanders and expressly validated by Congress in Rule 9(b).

Rule 9(b), then, rejects the very norm which the majority attempts to derive from it. On the majority’s own premises, the rule of complete exhaustion finds support in Rule 9(b) only because it is the instrument for enforcing the norm of a single federal proceeding. 3 Since Rule 9(b) rejects the *369 end, it provides no justification for the means.

Quite apart from all of the foregoing, the majority’s preoccupation with the litigious prisoner has blinded it to a second fundamental inconsistency of the Sanders standard incorporated in Rule 9(b) and the rule of complete exhaustion. The “abuse of the writ” standard, as Sanders itself amply demonstrates, encompasses a highly individualized and fact-sensitive inquiry designed to insure that only individuals whose conduct is truly abusive are denied the opportunity to prove a case of entitlement to the Great Writ. In contrast, the rule of complete exhaustion which the majority purports to derive from that standard operates across-the-board without any reference to abusive behavior. Thus, while the majority at one point states that its concern is “with the petitioner who, prior to seeking federal relief, knows that he has several claims . ,” majority opinion at 358, this claim rings hollow indeed. By reading Rule 9(b) as somehow endorsing a rule of complete exhaustion, the majority transforms a scalpel into a meat axe. Most objectionably, the axe blows will fall most heavily upon the necks of untutored and uncounselled prisoners, too naive or ignorant to circumvent the majority’s rule by deleting unexhausted claims from their petitions. For as the majority itself concedes, apart from encountering a highly dubious claim of laches, the sophisticated litigious prisoner intent upon a strategy of piecemeal litigation, the only end of which is harassment, can carry out his program while avoiding dismissal simply by bringing successive petitions each of which asserts only claims which have been exhausted. Majority opinion at 358. 4

Rule 9(a), authorizing dismissal of habeas petitions on laches grounds, is even more tenuously related to a rule of complete exhaustion than is Rule 9(b). At the outset, it is worth noting that the petitioner who brings his first federal petition the day after his appeal is decided adversely to him cannot be guilty of laches. Yet if he ignorantly, mistakenly, or deliberately includes in his federal petition an unexhausted claim, the majority requires the district court to dismiss the entire petition. On the other hand, a petitioner may be guilty of laches even if he has exhausted every claim in his petition, but waited twenty years to do it. Here, where a genuine laches problem may exist, the rule of complete exhaustion provides no sanction. The rule of complete exhaustion is a most inefficient instrument for advancing Rule 9(a)’s policy favoring the prompt presentation of claims.

The majority only connects the complete exhaustion rule to a laches defense by postulating a litigious, vexatious, prisoner of quite exceptional talent. This prisoner, intent on pursuing a strategy of piecemeal litigation rather than obtaining his release, might circumvent the rule of complete exhaustion by withholding his unexhausted claims. Should he do so, however, he runs the risk of encountering a laches defense when he presents the new claim in his second petition. In order to avoid encountering this defense, the majority tells us, the petitioner will bring all his federal claims forward in the initial federal proceeding. Thus, Rule 9(a) appears to provide an incentive to bringing all possible claims in a single federal proceeding.

With all due respect, it is this argument, and not the approach of our five sister circuits, which is “unrealistic.” Majority opinion at 356. If there are indeed petitioners who sequence their trips to state and federal court with an eye to foreclosing defenses potentially available to the state in a subsequent proceeding, they are indeed *370 rarae aves. 5 To argue, as the majority appears to do, that a sanction which may alter the conduct of these few justifies a rule penalizing the many, is to allow a very small tail to wag a very large and important dog. 6

The majority, thus, has woven a garment of exceptionally ill-matched cloth from the threads of Rule 9. Rule 9 provides nothing more than the emperor’s newest set of clothes for the rule of complete exhaustion. Stripped of any pretense that Congress has commanded the rule of complete exhaustion, that rule can now be considered on its own merits.

III.

Neither principle nor precedent counsels continued adherence to the rule of complete exhaustion. Instead, both point towards a rule requiring the district courts to entertain mixed petitions and to consider the merits of all exhausted claims not substantially interrelated with unexhausted claims. This rule protects the petitioner’s interest in speedy consideration of his federal claim while at the same time giving full effect to the policies underlying the exhaustion requirement. In contrast, the rule of complete exhaustion purchases gains that are largely illusory at a price which is entirely unacceptable.

At the outset, I note that the statutory provision requiring exhaustion bars a petitioner from obtaining federal relief if state law permits him to raise “the question presented.” 28 U.S.C. § 2254 (c). On its face, the statute appears to require the exhaustion of claims, not cases. Miller v. Hall, 536 F.2d 967, 969 (1st Cir. 1976).

The conclusion suggested by the statutory language finds firm support in an analysis of the structure of federal habeas corpus relief under Brown v. Allen, 344 U.S. 443 , 73 S.Ct. 397 , 97 L.Ed. 469 (1953), and Sanders v. United States, supra. Brown established the rule that a petitioner was required to present his federal claim only once to the state courts in order to exhaust state remedies. Even if state courts would entertain further petitions asserting the grounds already considered, federal relief would not be postponed until these state remedies had also been pursued to conclusion. Implicit in the rule of Brown is the proposition that a petitioner need not exhaust the capacity of the state system to discharge him from custody prior to invoking the federal forum. See Roberts v. LaVallee, 88 S.Ct. 195 (1967) (per curiam). The exhaustion requirement is intended to give the state courts a fair opportunity to pass on a federal claim before a federal court considers it. It follows that “once the federal claim has been fairly presented to *371 the state courts, the exhaustion requirement is satisfied.” Picard v. Connor, 404 U.S. 270 , 92 S.Ct. 509, 512 , 30 L.Ed.2d 438 (1971).

Sanders adds a second structural element. A federal court is to entertain a second federal petition presenting new grounds for relief unless the court finds that the prisoner has abused the writ. Sanders v. United States, supra, 83 S.Ct. at 1078. The predicate for invoking this doctrine of “rare and extraordinary application,” Simpson v. Wainwright, 488 F.2d 494, 495 (5th Cir. 1973), is a finding that the petitioner deliberately bypassed the opportunity to present his available claim in the first proceeding. See Turnbow v. Beto, 464 F.2d 527, 528 (5th Cir. 1972). It follows a fortiori from Sanders that the federal courts are, as a general matter surely, to be open to petitions presenting grounds for relief not assertable in the first petition because not exhausted at that time. Note, Habeas Petitions with Exhausted and Unexhausted Claims: Speedy Release, Comity and Judicial Efficiency, 57 B.U.L.Rev. 864, 871 (1977). As we stated in Simpson v. Wainwright, supra, the abuse of the writ doctrine, “could not apply to bar a return to the federal court by one who has been remitted to the state courts to exhaust and reappears alleging that he has done just that.” 488 F.2d at 495 . See Tannehill v. Fitzharris, 451 F.2d 1322, 1324 (9th Cir. 1971).

Sanders itself demonstrates just how narrowly limned are the policies disfavoring successive petitions. Sanders’ first petition asserted three grounds for relief and was summarily dismissed by the district court. Subsequently, Sanders filed a second petition alleging for the first time that he was under the influence of narcotics at the time he entered his plea of guilty. The district court dismissed this petition without a hearing and the Court of Appeals affirmed, stating:

Where, as here, it is apparent from the record that at the time of filing the first motion the movant knew the facts on which the second motion is based, yet in the second motion set forth no reason why he was previously unable to assert the new ground and did not allege that he had previously been unaware of the significance of the relevant facts, the district court may, in its discretion, decline to entertain the second motion.

Sanders v. United States, supra, 83 S.Ct. at 1072, quoting 9th Cir., 297 F.2d 735 at 736-37 . The Supreme Court, of course, reversed and remanded for a hearing to determine whether Sanders had abused the writ by deliberately bypassing the opportunity to present the claim in his first petition. Id., 83 S.Ct. at 1080.

The Sanders Court cited Wong Doo v. United States, 265 U.S. 239 , 44 S.Ct. 524 , 68 L.Ed. 999 (1924) as a case where abuse of the writ had been established. Wong Doo, the petitioner, asserted two grounds for relief in his first petition, but offered no proof in support of the second ground at the hearing held to consider his petition. The Court held that Wong Doo’s subsequent application, relying solely on the second ground, was properly dismissed for abuse of the writ. In Price v. Johnston, 334 U.S. 266 , 68 S.Ct. 1049 , 92 L.Ed. 1356 (1948) where the Supreme Court reversed the district court for refusing to consider a new claim appearing for the first time in Price’s fourth petition, the Court characterized Wong Doo as a case which, “involved a situation where one has properly raised an issue in an earlier petition, has received a full opportunity at a hearing to present evidence on the point, and has refused to avail oneself of that opportunity.” Id., 68 S.Ct. at 1062 . And since Sanders , this court has on many occasions required consideration of grounds for relief asserted for the first time in a second or subsequent petition. See, e. g., Simpson v. Wainwright, supra (three previous petitions); Turnbow v. Beto, supra (three prior petitions); Goins v. Allgood, 391 F.2d 692 (5th Cir. 1968).

While I concede that neither Sanders nor Brown is logically inconsistent with a rule of complete exhaustion, together they certainly make highly dubious the majority’s supposition that the enforceable norm of federal habeas corpus jurisdiction is a single *372 federal proceeding following exhaustion of all possible claims for release from custody. From Sanders we learn that there is no norm of a single federal proceeding; from Brown we learn that the exhaustion requirement does not mean that the federal remedy is to be unavailable to a prisoner until he has exhausted the capacity of the state system to release him from custody. 7

A general rule of partial exhaustion is entirely consistent with the purposes of the exhaustion doctrine. Those purposes were stated in a student note quoted approvingly by the Supreme Court in Braden v. 30th Judicial Circuit of Kentucky, 410 U.S. 484 , 93 S.Ct. 1123 , 35 L.Ed. 443 (1973):

the doctrine preserves the role of the state courts in the application and enforcement of federal law. Early federal intervention in state criminal proceedings would tend to remove federal questions from the state courts, isolate those courts from constitutional issues, and thereby remove their understanding of an hospitality to federally protected interests. Second, [the doctrine] preserves orderly administration of state judicial business, preventing the interruption of state adjudication by federal habeas proceedings. It is important that petitioners reach state appellate courts, which can develop and correct errors of state and federal law and most effectively supervise and impose uniformity on trial courts.

Note, Developments in the Law —Federal Habeas Corpus, 83 Harv.L.Rev. 1038, 1094 (1970), quoted in Braden v. 30th Judicial Circuit Court of Kentucky, supra, 93 S.Ct. at 1127 . These policies are given full effect by a rule of partial exhaustion. State courts fully participate in the constitutional dialogue promoted by the exhaustion requirement when they pass on a claim prior to its being considered by a federal court. The federal court is informed and enlightened by the views of the state court when it has those views before it on any claim considered on the merits. Thus, a rule of complete, rather than partial, exhaustion does not further advance the interest in protecting the role of state courts in deciding constitutional claims. And since a rule of partial exhaustion fully insures that state courts will have an opportunity to rule upon any federal ground for relief before it is considered by the federal courts, a rule of partial exhaustion does not promote “early federal intervention” removing federal questions from state courts and isolating state courts from constitutional issues. As the First Circuit put it, “The very fact of exhaustion means that the state courts have ruled, or have had the chance to rule on the federal claim in the first instance.” Miller v. Hall, supra, 536 F.2d at 969 . Insofar as this policy underlying the exhaustion requirement is concerned, the rule of complete exhaustion is productive of very little but delay in the consideration of federal claims.

As to the second interest, preservation of orderly administration of state judicial business, a rule of partial exhaustion does not provide any incentive for the interruption of state adjudication or for bypassing the earliest opportunity for correction of trial errors, i. e., appeal within the state system. A prisoner who is interested in speedy release has every reason for pursuing his state remedies promptly, whether the rule governing reception of his federal claims is one of partial or complete exhaustion. 8

*373 Indeed, the majority grudgingly concedes that “comity in a narrow sense has been served by the prior state consideration of the exhausted claim.” Majority opinion at 359. In order to avoid the effect of this concession, however, the majority offers three insubstantial arguments. First, the majority argues that “the state has an interest in the resolution of the entire case” as well as in each of a petitioner’s constitutional claims. Majority opinion at 359. If there is such an interest, it is the state’s own, to promote or not as it sees fit, by its own rules of procedure. Should the state assert such an interest by requiring that all claims be brought in a single state proceeding, deliberate bypass of that proceeding will bar subsequent federal habeas corpus relief. Murch v. Mottram, 409 U.S. 41 , 93 S.Ct. 71 , 34 L.Ed.2d 194 (1972) (per curiam). 9

Even less plausible is the majority’s suggestion that the rule of complete exhaustion is justified by an interest in avoiding intermittent interruption of a petitioner’s imprisonment consequent upon repeated trips to federal court. Quite apart from the fact that in many cases a prisoner’s claim can be ruled upon without his once seeing the outside of the prison walls, it hardly seems a matter for judicial notice that greater interruption of custody occurs when a prisoner makes a trip to the federal courthouse than when he makes a similar trip to the state courthouse to exhaust the unexhausted claims in his mixed petition.

Finally, the majority offers the “specter” of virtually simultaneous adjudications in state and federal court resulting in inconsistent relief on the claims. 10 Initially, it seems rather odd to justify a rule which places additional barriers in the path of the habeas petitioner by positing the case of a petitioner with two or more meritorious constitutional claims. Yet this is precisely the thrust and effect of the majority’s argument. Moreover, should the problem of inconsistent relief be of serious concern in any given case, the district courts have ample power to shape or modify the remedy as needed.

In point of fact, a rule of complete exhaustion has significant adverse impacts upon comity between state and federal courts. Remitting a prisoner to state court to assert transparently frivolous new claims which must be addressed by a state court before the federal court will consider a serious, already-exhausted, claim, is decidedly not the best way to show respect for state courts. The state judge is put in the position of knowing that he is expending his time and the state’s resources in consideration of a transparently frivolous claim which will undoubtedly prove to be a complete irrelevance in the subsequent federal *374 proceeding which will focus on the serious exhausted claim.

What little empirical evidence is available suggests that the efficiency gains achieved by a rule of complete exhaustion are very small. Professor Shapiro’s study of federal habeas corpus in Massachusetts is noteworthy for the light it sheds on the functioning and malfunctioning of the exhaustion requirement in a jurisdiction applying the rule of partial exhaustion for which I contend. See Shapiro, Federal Habeas Corpus: A Study in Massachusetts, 87 Harv.L.Rev. 321 (1973). Of the 257 petitions filed in the U.S. District Court for Massachusetts during the three-year study period, only 34 were filed by prisoners who had filed any previous federal petitions. Even if we were to indulge the incredibly generous assumption that a rule of complete exhaustion would induce each of these repeat petitioners to bring every one of their claims in a single federal proceeding, it is obvious that the repeat petition is of rather minor significance in the habeas corpus caseload, to say nothing of the overall federal caseload. Since a strict accounting of the theoretical gains achieved by a rule of complete exhaustion would have to include a debit for the time spent by state courts reviewing unexhausted claims which would never be presented under a rule of partial exhaustion, it is clear that the net gains would be very small indeed.

Professor Shapiro’s study also points out that failure to exhaust state remedies was by far the most common basis for dismissal of petitions filed in Massachusetts during the study period; over half of all petitions filed were dismissed for failure to exhaust. Id. at 356. These figures suggest that even in a jurisdiction applying a rule of partial exhaustion, the exhaustion requirement continues to play a critical function in the allocation of responsibility between state and federal courts. 11 A rule of partial exhaustion is in no sense an implied repeal of the exhaustion requirement.

Finally, even assuming the ultimate desirability of the majority’s goal of a single federal proceeding adjudicating all a petitioner’s federal claims, the rule of complete exhaustion is ineffectual in achieving that end. So long as a petitioner with both exhausted and unexhausted claims refrains from including the unexhausted claims in his petition, the petition cannot be dismissed under the majority’s rule. His second petition, asserting only newly exhausted grounds, likewise cannot be dismissed. The majority concedes this. Majority opinion at 358. A petitioner clever and sophisticated enough to understand and be influenced in his behavior by the majority’s rule will also be clever and sophisticated enough to avoid its effect. 12 The specter which haunts the majority opinion proves in the end too elusive to grasp. Meanwhile, the unsophisticated petitioner who appends to his federal petition some new notion not previously presented to the state courts, is denied a federal hearing on the merits of his exhausted claims. 13 Simultaneously under-inclusive and over-inclusive, the majority’s requirement of complete exhaustion has none of the characteristics of a good rule. In sum, the efficiency gains achieved by adoption of the rule of complete exhaustion are quite possibly illusory, certainly small, and undeniably achieved by penalizing the wrong people.

*375 IV.

There is one additional interest which must be considered today. It is an interest as remarkable for the recognition given it by our sister circuits as for the majority’s apparent indifference to it. I speak of the prisoner’s interest in speedy release from unconstitutional confinement.

There is no demand more urgent, more insistent, more deserving of our time and attention and than the demand of one who claims to be unlawfully confined. To be imprisoned is to be denied the society of one’s fellow men. To be imprisoned is to be forcibly separated from loved ones, spouse, children, parents. To be imprisoned is to be subjected to a numbing, relentless, regimentation. To be imprisoned is, far too often, to be subjected to barbaric and unconscionable conditions of confinement. To be imprisoned unlawfully is to be denied the essence of life and liberty without just cause.

This is the moral burden we bear when we defer consideration of a petition for the Great Writ. This burden cannot possibly be supported by considerations of judicial economy and efficiency; an Atlas, not an Anchises, of a justification is needed to shoulder the burdens of the rule of complete exhaustion. A judge’s time is precious, to be sure, but precious only in relation to the tasks the judge performs. In a habeas corpus case, we are dealing with human life and human liberty, precious commodities even in today’s world of depreciated traditional values. I will not participate in the process of depreciating further human life and liberty by accepting the proposition that saving an hour or two of a judge’s time justifies keeping a man locked behind the bars of a state penitentiary for a year or more.

We are heirs to the most precious legacy of Lord Coke, the power to discharge from custody even one imprisoned by order of the King. Today, the majority sells that legacy for a mess of pottage, a gruel composed of questionable notions of efficiency and vague notions of federalism. With all due respect to my brethren in the majority, judges are not good systems analysts or traffic controllers in the administration of justice. We do not have the training; we do not have the data. While we must be conscious of the state of our docket, the allure of a faddish modernity, of a stopwatch style of jurisprudence, will prove as treacherous as the Sirens’ Song. Today’s decision sacrifices on the alter of a spurious and specious science of efficiency the consti-' tutional guarantees which we profess to hold most dear. For the sake of a ritualized obeisance to the supposed demands of comity, today’s decision defers the vindication of constitutional rights. I would abjure this new legal religion. For us, the Great Writ should be as Holy Writ.

Since I am firmly convinced that the majority today has used the exhaustion doctrine “as a blunderbuss to shatter the attempt at litigation of constitutional claims without regard to the purposes that underlie the doctrine and that called it into existence,” Braden v. 30th Judicial Circuit Court of Kentucky, supra, 93 S.Ct. at 1127, I dissent.

. As Judge Roney notes, this case was taken en banc for reconsideration of the panel’s holding that this court was required to vacate the district court’s grant of habeas relief on a properly exhausted claim if petitioner’s federal habeas petition contained unexhausted claims for relief. 1 agree with the apparently unanimous conclusion of this court that our review on the merits of district court determinations respecting exhausted habeas claims is not precluded where the district court has entertained a mixed petition. Indeed, 1 believe such review is obligatory. Since the majority, in the course of reaching this result, has seen fit to reopen the question of the treatment of mixed petitions by the district courts, I take this opportunity to set forth my views on that question in some detail.

I, like my Brother Roney, would remand this case to the panel for consideration of the merits of petitioner’s exhausted claims.

. Where exhausted and unexhausted claims are substantially interrelated as a factual or legal matter, the district courts should be afforded discretion to dismiss the petition. Relevant factors, in addition to the degree of interrelation among the claims, might include the amount of delay consequent upon deferring consideration of the exhausted claims, the apparent substantiality of the exhausted claims, and the investment of the court’s own resources in determining which claims had been exhausted. While the instant case does not raise the question of the appropriate scope of the district courts’ discretion in such circumstances, 1 note that the “flexible rule” advocated by Judge Roney might find appropriate application in this area.

. The majority does not purport to find in the language of Rule 9 any reference to the exhaustion requirement whatsoever. The Rule contains no such language. Nor does the majority purport to find any support for the rule of complete exhaustion in the legislative history *369 of Rule 9. So far as 1 have been able to determine, the legislative history contains no such support. See H.R.Rep. No. 94-1471, 94th Cong., 2d Sess. 5 (1976), reprinted in [1976] U.S. Code Cong. & Admin. News, pp. 2478, 2482.

. As we shall see infra in text at n. 12, this is only one of the respects in which the majority’s rule of complete exhaustion fails to advance the goal of a single federal proceeding while simultaneously producing predictable injustice.

. The majority also suggests that the rule of complete exhaustion will have the effect of denying a petitioner the opportunity to pursue an unfair trial strategy (!). The Clarence Darrow Joneses who apparently populate the state penitentiaries in the Fifth Circuit might otherwise include unexhausted claims in their petitions in the calculated hope of influencing the district court’s resolution of the exhausted claims. Majority opinion at 359. I suppose it is possible to imagine a petitioner capable of formulating such a strategy. I cannot, however, imagine that the district judges of this circuit would be taken in by such a strategem.

. Although I hesitate to enter an argument which I am firmly convinced has more to do with the court of the Red Queen than it does with the courts of this circuit, I must point out that even on the majority’s own premises, the threat of a laches dismissal is not likely to be perceived as a very serious incentive to join all potential claims in the initial federal proceeding. A petitioner sophisticated enough to worry about the laches problem, but determined to pursue a strategy of piecemeal litigation, will certainly be clever enough to develop the simple strategem which permits him to pursue his plan and avoid a laches defense: file two separate, but simultaneous, petitions, one in federal court containing only exhausted claims, the other, in either federal or state court, containing only unexhausted claims. Filing of the petition containing unexhausted claims will foreclose a laches defense to the extent that mere notice of the existence of such claims has this effect. See, majority opinion at 358. Meanwhile the petition containing only exhausted claims would be considered on the merits. The rule of complete exhaustion is a very blunt and largely ineffective instrument for dealing with the legal prestidigitations of a habeas Houdini.

. In Braden v. 30th Judicial Circuit Court of Kentucky, 410 U.S. 484 , 93 S.Ct. 1123 , 35 L.Ed.2d 443 (1973), the Supreme Court held that a speedy trial claim exhausted in advance of trial was properly before the federal district court notwithstanding the fact that other unexhausted federal claims might be present or arise at trial. Id., 93 S.Ct. at 1127-28. Braden has been read as rejecting the theory implicit in the rule of complete exhaustion, i. e., that the state must get a crack at a petitioner’s entire case before federal relief is appropriate. See Miller v. Hail, supra, 536 F.2d at 969 .

. While it is true that grant of the writ will moot any pending state proceedings on unexhausted claims, thus aborting the state proceeding, the rule of complete exhaustion will produce unnecessary state proceedings on, perhaps frivolous, unexhausted claims. See, 582 F.2d 360 infra. Whether one or the other is the source of greater disruption of state processes is certainly arguable.

. At an earlier point in its opinion, the majority suggests that several policies would be “best served if all a petitioner’s claims are presented to the state court system at one time.” Majority opinion at 356. (Emphasis added.) It should be obvious that the rule of complete exhaustion has not the slightest relation to the number of state court proceedings on a petitioner’s case. So long as the state courts permit a petitioner to raise new or additional claims in a second state petition, a matter of state law, a federal court’s dismissal of a mixed petition will provide no special incentive for a prisoner to consolidate all his potential claims in a single state proceeding. Thus, even if we were to assume, and I do not, that federal courts are in the business of deciding that it is desirable to have all claims presented in a single state proceeding, the rule of complete exhaustion is utterly irrelevant to attaining this objective which is advanced by application of an entirely separate rule.

Likewise, I find incomprehensible the majority’s implication that the rule of complete exhaustion contributes to the perception of a state trial as “a decisive and portentous event.” Majority opinion at 359, quoting Wainwright v. Sykes, 433 U.S. 72, 97 S.Ct. 2497, 2508 , 53 L.Ed.2d 594 (1977). A conviction is no more final under a rule of complete exhaustion than it is under a rule of partial exhaustion. The choice between these rules is entirely independent of the possibility vel non of ultimately asserting a given basis for attack upon a conviction.

. Such a situation might arise where, for example, a federal court, ruling on the exhausted claim that the petitioner had been denied his right to counsel on appeal, directed the state to grant an appeal while a state court, ruling on an unexhausted claim that a confession had been coerced, overturned the conviction and granted a new trial.

. Professor Shapiro draws another conclusion from these statistics, i. e., that the exhaustion requirement ought to be liberalized in order to reduce the number of dismissals for failure to exhaust which, in his view, do not advance the policies underlying the requirement. Id. at 359-61.

. The majority’s only response to this undeniable fact is the laches argument discussed in Part II, supra. As I showed in my discussion of that argument, the laches argument is a good deal less than compelling.

. Habeas corpus petitioners are, like other civil litigants, free to amend their pleadings. Indeed, district judge endeavoring to show an appropriate degree of tolerance for a prisoner’s lack of familiarity with rules of pleading and procedure, see Haines v. Kerner, 404 U.S. 519 , 92 S.Ct. 594 , 30 L.Ed.2d 652 (1972); Price v. Johnston, supra, 68 S.Ct. at 1063 , could certainly inform petitioners presenting mixed petitions of their right to amend and the consequences, i. e., dismissal, of failing to do so.