Montana state prisoner Leland F. Docken brings this federal habeas petition challenging, as here pertinent, the Montana parole board's refusal to provide him with annual review of his suitability for parole. The district court dismissed this claim as not properly cognizable under the federal habeas statute, 28 U.S.C. § 2254. Because such parole-based claims — which may, but will not necessarily, affect the duration of a prisoner's confinement if meritorious — are cognizable via habeas, we reverse the district court's dismissal of Docken's petition and remand for further proceedings not inconsistent with this opinion.

In September 1984, Docken pleaded guilty in a Jefferson County, Montana, court to committing "deliberate homicide" by murdering his father-in-law. See MONT. CODE ANN. § 45-5-102. Docken was sentenced to 100 years in prison and designated, for purposes of parole eligibility, a "dangerous offender." See id. § 46-18-404 (repealed 1995). Docken's plea and sentence were affirmed by the Montana Supreme Court. See State v. Docken, 222 Mont. 58, 720 P.2d 679 (1986).

In late 2001, the Montana Board of Pardons and Parole ("Board") denied parole in Docken's first post-conviction hearing. The Board set the date for his next hearing for September 2006. Soon thereafter, Docken filed a state habeas petition in the Montana Supreme Court, naming Doug Chase, Sheriff of the Missoula County Detention Center, as Respondent.1 Docken's petition argued, among other claims, that the Board violated his constitutional rights under the Ex Post Facto Clause when it changed the period between his reviews for parole from one year to five. The Montana Supreme Court denied Docken's petition in a written order, finding none of his state or federal claims persuasive. See Docken v. Chase, 309 Mont. 531, 43 P.3d 984 (2002) (unpublished table decision).

Docken then filed this habeas petition in the Montana federal district court, raising four principal contentions. The magistrate judge recommended dismissing the first three claims with prejudice, and the fourth, the only claim before us, without prejudice, on the ground that it is not cognizable via habeas but must rather be brought as a civil rights action under 42 U.S.C. § 1983. Citing the Ninth Circuit's decision in Neal v. Shimoda, 131 F.3d 818 (9th Cir.1997), the magistrate judge concluded that "claims involving parole eligibility, as compared to parole decisions," may only be brought under § 1983, for, "[i]f Docken's claim succeeded, he would not be entitled to release." The district court adopted the magistrate judge's findings. We granted a Certificate of Appealability limited to the question of "whether ... appellant's claim of a right to annual review of his suitability for parole is not cognizable under 28 U.S.C. § 2254 and must rather be brought under 42 U.S.C. § 1983."

We review de novo the district court's dismissal of a habeas petition under 28 U.S.C. § 2254. See Hunt v. Pliler, 384 F.3d 1118, 1123 (9th Cir.2004).

Traditionally, challenges to prison conditions have been cognizable only via § 1983, while challenges implicating the fact or duration of confinement must be brought through a habeas petition. Looked at in hindsight, the characterization of the problem of parole review frequency is not difficult: For an inmate who will ultimately be paroled at his second parole hearing, whether that hearing is in one year or five indisputably implicates the duration of his confinement. But where the claim is, as here, purely prospective, hindsight is of little use. We cannot know in advance how the parole hearing will turn out. Also, substantively, a timing-of-parole-hearing claim necessarily turns not on the outcome of any hearing but on the asserted legal right to an opportunity to be heard and to have the chance of a shorter confinement. The question this case raises is whether such a prospective challenge is cognizable under § 2254 or § 1983 — or both.

The landmark precedent for determining whether inmate suits challenging particular aspects of prison conditions may be brought under the federal habeas statute is Preiser v. Rodriguez, 411 U.S. 475, 93 S. Ct. 1827, 36 L. Ed. 2d 439 (1973). Preiser held that inmates challenging the deprivation of good-conduct time credits must sue under the federal habeas statute, 28 U.S.C. § 2254, and not under 42 U.S.C. § 1983, because such a suit necessarily challenges the duration of the inmates' confinement. See id. at 487-88, 93 S. Ct. 1827; see also id. at 490, 93 S. Ct. 1827 ("Congress has determined that habeas corpus is the appropriate remedy for state prisoners attacking the validity of the fact or length of their confinement, and that specific determination must override the general terms of § 1983.").2 Even when prisoners only sought to shorten their sentences, the Preiser Court held, such a challenge falls within the "core" of habeas corpus by "attacking the very duration of their physical confinement itself." Id. at 487-88, 93 S. Ct. 1827; see also id. at 498, 93 S. Ct. 1827 (defining suits "challenging the fact or duration of ... physical confinement" and seeking "immediate release or a speedier release from that confinement" as the "heart of habeas corpus").

Preiser did not stop there. In response to the prisoners' argument that numerous challenges to prison conditions had been sustained under § 1983, the Court concluded that "a § 1983 action is a proper remedy for a state prisoner who is making a constitutional challenge to the conditions of his prison life, but not to the fact or length of his custody." Id. at 499, 93 S. Ct. 1827. The Court went on to issue a significant caution about the reach of its ruling: "This is not to say that habeas corpus may not also be available to challenge such prison conditions." Id. (emphasis added). Instead, the Court reiterated that its holding spoke only to the limits of § 1983 as a remedy: "[W]e need not in this case explore the appropriate limits of habeas corpus as an alternative remedy to a proper action under § 1983. That question is not before us. What is involved here is the extent to which § 1983 is a permissible alternative to the traditional remedy of habeas corpus." Id. at 500, 93 S. Ct. 1827.

Dissenting, Justice Brennan expanded on the reserved question: "[E]ven under the [Preiser majority's] approach, there are undoubtedly some instances where a prisoner has the option of proceeding either by petition for habeas corpus or by suit under § 1983." Id. at 504, 93 S. Ct. 1827 (Brennan, J., dissenting); see also id. at 506, 93 S. Ct. 1827 ("[S]ome instances remain ... where an action may properly be brought in habeas corpus, even though it is somehow sufficiently distant from the `core of habeas corpus' to avoid displacing concurrent jurisdiction under[§ 1983]."); id. at 503-04, 93 S. Ct. 1827 ("[T]he two statutes necessarily overlap."). Thus, both the majority and the dissent in Preiser suggested that there are some circumstances concerning prison conditions in which both habeas corpus and § 1983 suits may lie — that is, that the two remedies are not always mutually exclusive so long as the "core" or "heart" of habeas corpus is not implicated.

Like Preiser, Heck v. Humphrey, 512 U.S. 477, 114 S. Ct. 2364, 129 L. Ed. 2d 383 (1994), was also a § 1983 suit brought by an inmate. Unlike Preiser, however, in Heck the inmate sought only damages as a remedy for alleged constitutional violations committed during his initial arrest and confinement, not injunctive relief or release from custody. See id. at 479, 114 S. Ct. 2364. Heck concluded that a damages action is not cognizable under § 1983 if "a judgment in favor of the plaintiff would necessarily imply the invalidity of his conviction or sentence ... unless the plaintiff can demonstrate that the conviction or sentence has already been invalidated." Id. at 487, 114 S. Ct. 2364. Enunciating what has since become known as the "favorable termination" rule, Heck held that, when a § 1983 claim would necessarily implicate the validity of the plaintiff's conviction or sentence, "a § 1983 plaintiff must prove that the conviction or sentence has been reversed on direct appeal, expunged by executive order, declared invalid by a state tribunal authorized to make such determination, or called into question by a federal court's issuance of a writ of habeas corpus." Id. at 486-87, 114 S. Ct. 2364; see also Muhammad v. Close, 540 U.S. 749, 124 S. Ct. 1303, 158 L. Ed. 2d 32 (2004) (per curiam) (summarizing the relationship between Preiser and Heck).

Also like Preiser, Heck addressed only the limited reach of § 1983 with regard to prisoner suits. It did not set out any concomitant limitation on habeas jurisdiction or hold that the habeas and § 1983 causes of action are mutually exclusive. See also Nelson v. Campbell, 541 U.S. 637, 124 S. Ct. 2117, 2122, 158 L. Ed. 2d 924 (2004) ("[C]onstitutional claims that merely challenge the conditions of a prisoner's confinement, whether the inmate seeks monetary or injunctive relief, fall outside of [Preiser's] core and may be brought pursuant to § 1983 in the first instance." (emphasis added)); cf. Bell v. Wolfish, 441 U.S. 520, 527 n. 6, 99 S. Ct. 1861, 60 L. Ed. 2d 447 (1979) ("[W]e leave to another day the question of the propriety of using a writ of habeas corpus to obtain review of the conditions of confinement, as distinct from the fact or length of the confinement itself.").

Thus, although Supreme Court case law makes clear that § 1983 is not available where a prisoner's claim "necessarily" implicates the validity or duration of confinement, it does not set out any mirror-image limitation on habeas jurisdiction. The Court's central concern, in all of the cases cited above, has been with how far the general remedy provided by § 1983 may go before it intrudes into the more specific realm of habeas, not the other way around. At the same time, though the Court has so suggested, it has never squarely held that there is an area of overlap between state habeas and § 1983 prisoner suits. Instead, it has policed the distinction between the two remedies solely by defining the limits of § 1983, as in Heck, and by defining those classes of claims that must be brought through habeas, as in Preiser. Put simply, when the Supreme Court has concerned itself with the interaction between § 1983 and habeas, it has looked in only one direction.

Our own precedents have more directly addressed the interplay between § 1983 and habeas in the parole context and are somewhat illuminating with regard to the reach of habeas jurisdiction over parole-related prisoner suits. In Bostic v. Carlson, 884 F.2d 1267 (9th Cir.1989), for example, we held that "[h]abeas corpus jurisdiction ... exists when a petitioner seeks expungement of a disciplinary finding from his record if expungement is likely to accelerate the prisoner's eligibility for parole." Id. at 1269 (citing McCollum v. Miller, 695 F.2d 1044, 1047 (7th Cir.1982)) (emphasis added). As we recently explained, "Bostic thus holds that the likelihood of the effect on the overall length of the prisoner's sentence... determines the availability of habeas corpus." Ramirez v. Galaza, 334 F.3d 850, 858 (9th Cir.2003) (emphasis added), cert. denied, ___ U.S. ___, 124 S. Ct. 2388, 158 L. Ed. 2d 963 (2004).

Importantly, in speaking of claims only "likely to accelerate" eligibility for parole, Bostic defined a class of suits outside the "core" habeas claims identified in Preiser. Success on the merits in such cases would not "necessarily" implicate the fact or duration of confinement. Instead, such claims have, at best, only a possible relationship to the duration of a prisoner's confinement, as eligibility for parole is distinct from entitlement to parole.

Following Bostic, we have also held that challenges to the procedures used in denying parole are only cognizable via habeas. See Butterfield v. Bail, 120 F.3d 1023, 1024 (9th Cir.1997) ("Few things implicate the validity of continued confinement more directly than the allegedly improper denial of parole. This is true whether that denial is alleged to be improper based upon procedural defects in the parole hearing or upon allegations that parole was improperly denied on the merits."). Yet a change in the procedures governing parole review will not necessarily affect the duration of confinement, although it certainly could.

Five months after Butterfield, this court handed down a decision looking in a somewhat different direction on an issue similar to that raised in Butterfield. See Neal v. Shimoda, 131 F.3d 818 (9th Cir.1997). In Neal, two prisoners challenged their administrative placement in the Hawaii "Sex Offender Treatment Program" (SOTP), a placement that made them ineligible for parole. The prisoners styled their claims as § 1983 actions against the responsible prison officials. The court affirmed the propriety of § 1983 as an avenue of relief:

If Neal and Martinez are successful in their challenge of the SOTP and their labeling as sex offenders, that decision will not undermine the validity of their convictions or continuing confinement at all. The only benefit that a victory in this case would provide Neal and Martinez, besides the possibility of monetary damages, is a ticket to get in the door of the parole board, thus only making them eligible for parole consideration according to the terms of their sentences. If Neal and Martinez win, it will in no way guarantee parole or necessarily shorten their prison sentences by a single day. The parole board will still have the authority to deny the inmates' requests for parole on the basis of any of the grounds presently available to it in evaluating such a request. A victory in this case would not alter the calculus for the review of parole requests in any way. Because the inmates' challenge in this case does not necessarily imply the invalidity of their convictions or continuing confinement, it is properly brought under § 1983.

Id. at 824 (footnote omitted). If the district court in this case was correct in reading this passage of Neal for the proposition that § 1983 was the only appropriate remedy (thus foreclosing habeas), then Neal arguably created an intra-circuit split with Bostic, as Bostic held that habeas was appropriate so long as success on the merits would "likely" accelerate eligibility for parole.3

One way to reconcile these cases would be to construe Neal as applying only to eligibility for parole vel non, and not to cases — like this one — implicating the timing of parole hearings. Arguably, whether or not an inmate is eligible for parole consideration is less closely related to the duration of his confinement than is the amount of time between the hearings before the parole board for an already-eligible inmate. The timing of parole review is also more closely tied to "duration" than to "conditions" of confinement. This distinction would not resolve the apparent conflict between Neal and Bostic, however, because Bostic held that the availability of habeas turns on the likelihood that a successful claim will accelerate a prisoner's eligibility for parole, not his release.

A more sensible reconciliation of the case law, and the one we adopt here, is to regard Neal as holding only that § 1983 was an appropriate remedy in that case, without reaching the issue of whether it was the exclusive remedy. Certainly, nothing in the Neal opinion suggests that § 1983 was available to the prisoners in Neal in lieu of habeas. As in Preiser and Heck, the inmates in Neal were attempting to bring suit under § 1983. The court therefore only reached whether the claim was cognizable under § 1983, without passing on the relationship between § 1983 and habeas.4

A review of other circuits' approaches indicates that only the Seventh Circuit has even implicitly suggested that habeas and § 1983 can be mutually exclusive, and that court did so with little analysis and in a different context. See Moran v. Sondalle, 218 F.3d 647, 650-51 (7th Cir.2000) (per curiam) (noting when prisoners "must" use § 1983 and when they "must" use habeas, but not addressing where challenges to the frequency of parole review would fit).5 The other circuits have been indistinct in delineating the line between § 1983 and habeas, especially in the parole context, and some have suggested (though none have held) that the two statutory remedies may not always be mutually exclusive.6

As outlined above, the question of the relationship between habeas and § 1983 relief has only explicitly come up before in converse form: whether claims are not cognizable under § 1983 because their resolution will necessarily impact the fact and duration of confinement. In the only instance where the Supreme Court addressed whether habeas and § 1983 are necessarily mutually exclusive, the suggestion was that they are not. See Preiser, 411 U.S. at 499, 93 S. Ct. 1827; id. at 503-04, 93 S. Ct. 1827 (Brennan, J., dissenting). We agree.

Freed from the erroneous constraints of deciding whether Docken's claim is cognizable via habeas as an either/ or proposition, we must still resolve whether it falls into the area of overlap suggested in Preiser. In light of Bostic, however, this question largely answers itself. It is certainly at least possible that Docken's suit would impact the duration of his confinement if the Board's actions in changing the frequency of his parole review violated the Ex Post Facto Clause.

Of course, it is not a certainty that annual review would affect the duration of Docken's confinement, especially given Docken's designation as a "dangerous offender." We are ill-inclined, however, to substitute our substantive analysis of the likely outcome of Docken's parole hearings for that of the Board. And, in any event, we find nothing in Ninth Circuit or Supreme Court precedent foreclosing habeas in such a case.

Instead, we understand Bostic's use of the term "likely" to identify claims with a sufficient nexus to the length of imprisonment so as to implicate, but not fall squarely within, the "core" challenges identified by the Preiser Court. Such a reading follows from Bostic itself, which spoke of claims that are "likely to accelerate the prisoner's eligibility for parole," 884 F.2d at 1269 (emphasis added), rather than those likely to accelerate the prisoner's release. Docken's central contention — that he is entitled to annual review — is even more related to the duration of his confinement than eligibility for parole in the abstract, and therefore appears at least as viable as the subject of a habeas petition as that which was before the court in Bostic and Butterfield.

Ultimately, though Docken's claim may not be the kind of "core" challenge the Preiser Court had in mind, the potential relationship between his claim and the duration of his confinement is undeniable. In such a case, we are reluctant to unnecessarily constrain our jurisdiction to entertain habeas petitions absent clear indicia of congressional intent to do so. See, e.g., INS v. St. Cyr, 533 U.S. 289, 121 S. Ct. 2271, 150 L. Ed. 2d 347 (2001); Flores-Miramontes v. INS, 212 F.3d 1133 (9th Cir.2000).

We therefore hold that when prison inmates seek only equitable relief in challenging aspects of their parole review that, so long as they prevail, could potentially affect the duration of their confinement, such relief is available under the federal habeas statute. Whether such relief is also available under § 1983 depends on the application of Heck's favorable termination rule in this case, an issue not before us and one that we do not decide.

Because we hold Docken's challenge to the timing of his parole review properly cognizable under the federal habeas statute, the district court's dismissal of Docken's habeas petition is REVERSED, and this case is REMANDED to the district court for further proceedings on the merits of Docken's claim.7

IT IS SO ORDERED.