Full opinion text
Chief Justice Roberts delivered the opinion of the Court. DNA testing has an unparalleled ability both to exonerate the wrongly convicted and to identify the guilty. It has the potential to significantly improve both the criminal justice system and police investigative practices. The Federal Government and the States have recognized this, and have developed special approaches to ensure that this evidentiary tool can be effectively incorporated into established criminal procedure — usually but not always through legislation. Against this prompt and considered response, the respondent, William Osborne, proposes a different approach: the recognition of a freestanding and far-reaching constitutional right of access to this new type of evidence. The nature of what he seeks is confirmed by his decision to file this lawsuit in federal court under 42 U. S. C. § 1983, not within the state criminal justice system. This approach would take the development of rules and procedures in this area out of the hands of legislatures and state courts shaping policy in a focused manner and turn it over to federal courts applying the broad parameters of the Due Process Clause. There is no reason to constitutionalize the issue in this way. Because the decision below would do just that, we reverse. I A This lawsuit arose out of a violent crime committed 16 years ago, which has resulted in a long string of litigation in the state and federal courts. On the evening of March 22, 1993, two men driving through Anchorage, Alaska, solicited sex from a female prostitute, K. G. She agreed to perform fellatio on both men for $100 and got in their ear. The three spent some time looking for a place to stop and ended up in a deserted area near Earthquake Park. When K. G. demanded payment in advance, the two men pulled out a gun and forced her to perform fellatio on the driver while the passenger penetrated her vaginally, using a blue condom she had brought. The passenger then ordered K. G. out of the car and told her to lie face-down in the snow. Fearing for her life, she refused, and the two men choked her and beat her with the gun. When K. G. tried to flee, the passenger beat her with a wooden axe handle and shot her in the head while she lay on the ground. They kicked some snow on top of her and left her for dead. 521 F. 3d 1118,1122 (CA9 2008) (case below); Osborne v. State, 163 P. 3d 973, 975-976 (Alaska App. 2007) (Osborne II); App. 27, 42-44. K. G. did not die; the bullet had only grazed her head. Once the two men left, she found her way back to the road, and flagged down a passing car to take her home. Ultimately, she received medical care and spoke to the police. At the scene of the crime, the police recovered a spent shell easing, the axe handle, some of K. G.’s clothing stained with blood, and the blue condom. Jackson v. State, No. A-5276 etc. (Alaska App., Feb. 7, 1996), App. to Pet. for Cert. 117a. Six days later, two military police officers at Fort Richardson pulled over Dexter Jackson for flashing his headlights at another vehicle. In his car they discovered a gun (which matched the shell casing), as well as several items K. G. had been carrying the night of the attack. Id., at 116a, 118a-119a. The car also matched the description K. G. had given to the police. Jackson admitted that he had been the driver during the rape and assault, and told the police that William Osborne had been his passenger. 521 F. 3d, at 1122-1123; 423 F. 3d 1050, 1051-1052 (CA9 2005); Osborne v. State, 110 P. 3d 986, 990 (Alaska App. 2005) (Osborne I). Other evidence also implicated Osborne. K. G. picked out his photograph (with some uncertainty) and at trial she identified Osborne as her attacker. Other witnesses testified that shortly before the crime, Osborne had called Jackson from an arcade, and then driven off with him. An axe handle similar to the one at the scene of the crime was found in Osborne’s room on the military base where he lived. The State also performed DQ Alpha testing on sperm found in the blue condom. DQ Alpha testing is a relatively inexact form of DNA testing that can clear some wrongly accused individuals, but generally cannot narrow the perpetrator down to less than 5% of the population. See Dept, of Justice, National Comm’n on the Future of DNA Evidence, The Future of Forensic DNA Testing 17 (NCJ 183697, 2000) (hereinafter Future of Forensic DNA Testing); Dept, of Justice, National Comm’n on the Future of DNA Evidence, Post-conviction DNA Testing: Recommendations for Handling Requests 27 (NCJ 177626, 1999) (hereinafter Postconvietion DNA Testing). The semen found on the condom had a genotype that matched a blood sample taken from Osborne, but not ones from Jackson, K. G., or a third suspect named James Hunter. Osborne is black, and approximately 16% of black individuals have such a genotype. App. 117-119. In other words, the testing ruled out Jackson and Hunter as possible sources of the semen, and also ruled out over 80% of other black individuals. The State also examined some pubic hairs found at the scene of the crime, which were not susceptible to DQ Alpha testing, but which state witnesses attested to be similar to Osborne’s. App. to Pet. for Cert. 117a. B Osborne and Jackson were convicted by an Alaska jury of kidnaping, assault, and sexual assault. They were acquitted of an additional count of sexual assault and of attempted murder. Finding it “‘nearly miraculous’” that K. G. had survived, the trial judge sentenced Osborne to 26 years in prison, with 5 suspended. Id., at 128a. His conviction and sentence were affirmed on appeal. Id., at 113a-130a. Osborne then sought postconviction relief in Alaska state court. He claimed that he had asked his attorney, Sidney Billingslea, to seek more discriminating restriction-fragment-length-polymorphism (RFLP) DNA testing during trial, and argued that she was constitutionally ineffective for not doing so. Billingslea testified that after investigation, she had concluded that further testing would do more harm than good. She planned to mount a defense of mistaken identity, and thought that the imprecision of the DQ Alpha test gave her “ ‘very good numbers in a mistaken identity, cross-racial identification case, where the victim was in the dark and had bad eyesight.’” Osborne 1, 110 P. 3d, at 990. Because she believed Osborne was guilty, “‘insisting on a more advanced . . . DNA test would have served to prove that Osborne committed the alleged crimes.’” Ibid. The Alaska Court of Appeals concluded that Billingslea’s decision had been strategic and rejected Osborne’s claim. Id., at 991-992. In this proceeding, Osborne also sought the DNA testing that Billingslea had failed to perform, relying on an Alaska postconviction statute, Alaska Stat. §12.72 (2008), and the State and Federal Constitutions. In two decisions, the Alaska Court of Appeals concluded that Osborne had no right to the RFLP test. According to the court, § 12.72 “apparently” did not apply to DNA testing that had been available at trial. Osborne I, 110 P. 3d, at 992-993. The court found no basis in our precedents for recognizing a federal constitutional right to DNA evidence. Id., at 993. After a remand for further findings, the Alaska Court of Appeals concluded that Osborne could not claim a state constitutional right either, because the other evidence of his guilt was too strong and RFLP testing was not likely to be conclusive. Osborne II, 163 P. 3d, at 979-981. Two of the three judges wrote separately to say that “[i]f Osborne could show that he were in fact innocent, it would be unconscionable to punish him,” and that doing so might violate the Alaska Constitution. Id., at 984-985 (Mannheimer, J., concurring). The court relied heavily on the fact that Osborne had confessed to some of his crimes in a 2004 application for parole— in which it is a crime to lie. Id., at 978-979, 981 (majority opinion) (citing Alaska Stat. §11.56.210 (2002)). In this statement, Osborne acknowledged forcing K. G. to have sex at gunpoint, as well as beating her and covering her with snow. 163 P. 3d, at 977-978, n. 11. He repeated this confession before the parole board. Despite this acceptance of responsibility, the board did not grant him discretionary parole. App. to Pet. for Cert. 8a. In 2007, he was released on mandatory parole, but he has since been rearrested for another offense, and the State has petitioned to revoke this parole. Brief for Petitioners 7, n. 3. Meanwhile, Osborne had also been active in federal court, suing state officials under 42 U. S. C. § 1983. He claimed that the Due Process Clause and other constitutional provisions gave him a constitutional right to access the DNA evidence for what is known as short-tandem-repeat (STR) testing (at his own expense). App. 24. This form of testing is more discriminating than the DQ Alpha or RFLP methods available at the time of Osborne’s trial. The District Court first dismissed the claim under Heck v. Humphrey, 512 U. S. 477 (1994), holding it “inescapable” that Osborne sought to “set the stage” for an attack on his conviction, and therefore “must proceed through a writ of habeas corpus.” App. 207 (internal quotation marks omitted). The United States Court of Appeals for the Ninth Circuit reversed, concluding that § 1983 was the proper vehicle for Osborne’s claims, while “expressing] no opinion as to whether Osborne ha[d] been deprived of a federally protected right.” 423 F. 3d, at 1056. On cross-motions for summary judgment after remand, the District Court concluded that “there does exist, under the unique and specific facts presented, a very limited constitutional right to the testing sought.” 445 F. Supp. 2d 1079, 1081 (2006) (some emphasis deleted). The court relied on several factors: that the testing Osborne sought had been unavailable at trial, that the testing could be accomplished at almost no cost to the State, and that the results were likely to be material. Id., at 1081-1082. It therefore granted summary judgment in favor of Osborne. The Court of Appeals affirmed, relying on the prosecutorial duty to disclose exculpatory evidence recognized in Pennsylvania v. Ritchie, 480 U. S. 39 (1987), and Brady v. Maryland, 373 U. S. 83 (1963). While acknowledging that our precedents “involved only the right to pre-trial disclosure,” the court concluded that the Due Process Clause also “extends the government’s duty to disclose (or the defendant’s right of access) to post-conviction proceedings.” 521 F. 3d, at 1128. Although Osborne’s trial and appeals were over, the court noted that he had a “potentially viable” state constitutional claim of “actual innocence,” id., at 1130, and relied on the “well-established assumption” that a similar claim arose under the Federal Constitution, id., at 1131; cf. Herrera v. Collins, 506 U. S. 390 (1993). The court held that these potential claims extended some of the State’s Brady obligations to the postconviction context. The court declined to decide the details of what showing must be made to access the evidence because it found “Osborne’s case for disclosure ... so strong on the facts” that “[wjherever the bar is, he crosses it.” 521 F. 3d, at 1134. While acknowledging that Osborne’s prior confessions were “certainly relevant,” the court concluded that they did not “necessarily trum[p] . . . the right to obtain post-conviction access to evidence” in light of the “emerging reality of wrongful convictions based on false confessions.” Id., at 1140. We granted certiorari to decide whether Osborne’s claims could be pursued using § 1983, and whether he has a right under the Due Process Clause to obtain postconviction access to the State’s evidence for DNA testing. 555 U. S. 992 (2008); Pet. for Cert. i. We now reverse on the latter ground. II Modern DNA testing can provide powerful new evidence unlike anything known before. Since its first use in criminal investigations in the mid-1980s, there have been several major advances in DNA technology, culminating in STR technology. It is now often possible to determine whether a biological tissue matches a suspect with near certainty. While of course many criminal trials proceed without any forensic and scientific testing at all, there is no technology comparable to DNA testing for matching tissues when such evidence is at issue. Postconviction DNA Testing 1-2; Future of Forensic DNA Testing 13-14. DNA testing has exonerated wrongly convicted people, and has confirmed the convictions of many others. At the same time, DNA testing alone does not always resolve a case. Where there is enough other incriminating evidence and an explanation for the DNA result, science alone cannot prove a prisoner innocent. See House v. Bell, 547 U. S. 518, 540-548 (2006). The availability of technologies not available at trial cannot mean that every criminal conviction, or even every criminal conviction involving biological evidence, is suddenly in doubt. The dilemma is how to harness DNA’s power to prove innocence without unnecessarily overthrowing the established system of criminal justice. That task belongs primarily to the legislature. “[T]he States are currently engaged in serious, thoughtful examinations,” Washington v. Glucksberg, 521 U. S. 702, 719 (1997), of how to ensure the fair and effective use of this testing within the existing criminal justice framework. Forty-six States have already enacted statutes dealing specifically with access to DNA evidence. See generally Brief for State of California et al. as Amici Curiae 3-13; Garrett, Claiming Innocence, 92 Minn. L. Rev. 1629, 1719 (2008) (surveying state statutes); see also An Act to Improve the Preservation and Accessibility of Biological Evidence, Mississippi S. 2709 (enacted March 16, 2009); An Act to Provide for DNA Testing for Certain Inmates for the Purposes of Determining Whether They May Have Been Wrongfully Convicted, South Dakota H. R. 1166 (enacted March 11, 2009). The State of Alaska itself is considering joining them. See An Act Relating to Post-conviction DNA Testing, H. 174, 26th Leg., 1st Sess. (2009) (proposed legislation similar to that enacted by the States). The Federal Government has also passed the Innocence Protection Act of 2004, §411, 118 Stat. 2278, codified in part at 18 U. S. C. § 3600, which allows federal prisoners to move for court-ordered DNA testing under certain specified conditions. That Act also grants money to States that enact comparable statutes, §413, 118 Stat. 2285, note following 42 U. S. C. § 14136, and as a consequence has served as a model for some state legislation. At oral argument, Osborne agreed that the federal statute is a model for how States ought to handle the issue. Tr. of Oral Arg. 33, 38-39; see also Brief for United States as Amicus Curiae 19-26 (defending constitutionality of Innocence Protection Act). These laws recognize the value of DNA evidence but also the need for certain conditions on access to the State’s evidence. A requirement of demonstrating materiality is common, e. g., 18 U. S. C. § 3600(a)(8), but it is not the only one. The federal statute, for example, requires a sworn statement that the applicant is innocent. § 3600(a)(1). This requirement is replicated in several state statutes. E. g., Cal. Penal Code Ann. §§ 1405(b)(1), (c)(1) (West Supp. 2009); Fla. Stat. §925.11(2)(a)(3) (2007); N. H. Rev. Stat. Ann. § 651-D:2(I)(b) (West 2007); S. C. Code Ann. §17-28-40 (Supp. 2008). States also impose a range of diligence requirements. Several require the requested testing to “have been technologically impossible at trial.” Garrett, supra, at 1681, and n. 242. Others deny testing to those who declined testing at trial for tactical reasons. E. g., Utah Code Ann. § 78B-9-301(4) (Lexis 2008). Alaska is one of a handful of States yet to enact legislation specifically addressing the issue of evidence requested for DNA testing. But that does not mean that such evidence is unavailable for those seeking to prove their innocence. Instead, Alaska courts are addressing how to apply existing laws for discovery and postconviction relief to this novel technology. See Osborne I, 110 P. 3d, at 992-993; Patterson v. State, No. A-8814, 2006 WL 573797, *4 (Alaska App., Mar. 8, 2006). The same is true with respect to other States that do not have DNA-specific statutes. E.g., Fagan v. State, 957 So. 2d 1159 (Ala. Crim. App. 2007). Cf. Mass. Rule Crim. Proc. 30(c)(4) (2009). First, access to evidence is available under Alaska law for those who seek to subject it to newly available DNA testing that will prove them to be actually innocent. Under the State’s general postconviction relief statute, a prisoner may challenge his conviction when “there exists evidence of material facts, not previously presented and heard by the court, that requires vacation of the conviction or sentence in the interest of justice.” Alaska Stat. § 12.72.010(4) (2008). Such a claim is exempt from otherwise applicable time limits if “newly discovered evidence,” pursued with due diligence, “establishes by clear and convincing evidence that the applicant is innocent.” § 12.72.020(b)(2). Both parties agree that under these provisions of § 12.72, “a defendant is entitled to post-conviction relief if the defendant presents newly discovered evidence that establishes by clear and convincing evidence that the defendant is innocent.” Osborne I, supra, at 992 (internal quotation marks omitted). If such a claim is brought, state law permits general discovery. See Alaska Rule Crim. Proc. 35.1(g) (2008-2009). Alaska courts have explained that these procedures are available to request DNA evidence for newly available testing to establish actual innocence. See Patterson, supra, at *4 (“If Patterson had brought the DNA analysis request as part of his previous application for [postconviction] relief ... he would have been able to request production of evidence”). In addition to this statutory procedure, the Alaska Court of Appeals has invoked a widely accepted three-part test to govern additional rights to DNA access under the State Constitution. Osborne II, 163 P. 3d, at 974-975. Drawing on the experience with DNA evidence of State Supreme Courts around the country, the Court of Appeals explained that it was “reluctant to hold that Alaska law offers no remedy to defendants who could prove their factual innocence.” Osborne 1, 110 P. 3d, at 995; see id., at 995, n. 27 (citing decisions from other state courts). It was “prepared to hold, however, that a defendant who seeks post-conviction DNA testing ... must show (1) that the conviction rested primarily on eyewitness identification evidence, (2) that there was a demonstrable doubt concerning the defendant’s identification as the perpetrator, and (3) that scientific testing would likely be conclusive on this issue.” Id., at 995. Thus, the Alaska courts have suggested that even those who do not get discovery under the State’s criminal rules have available to them a safety valve under the State Constitution. This is the background against which the Federal Court of Appeals ordered the State to turn over the DNA evidence in its possession, and it is our starting point in analyzing Osborne’s constitutional claims. Ill The parties dispute whether Osborne has invoked the proper federal statute in bringing his claim. He sued under the federal civil rights statute, 42 U. S. C. § 1983, which gives a cause of action to those who challenge a State’s “deprivation of any rights . . . secured by the Constitution.” The State insists that Osborne’s claim must be brought under 28 U. S. C. § 2254, which allows a prisoner to seek “a writ of habeas corpus ... on the ground that he is in custody in violation of the Constitution.” While Osborne’s claim falls within the literal terms of § 1983, we have also recognized that § 1983 must be read in harmony with the habeas statute. See Preiser v. Rodriguez, 411 U.S. 475, 500 (1973); Heck, 512 U. S., at 487. “Stripped to its essence,” the State says, “Osborne’s § 1983 action is nothing more than a request for evidence to support a hypothetical claim that he is actually innocent. . . . [T]his hypothetical claim sounds at the core of habeas corpus.” Brief for Petitioners 19. Osborne responds that his claim does not sound in habeas at all. Although invalidating his conviction is of course his ultimate goal, giving him the evidence he seeks “would not necessarily imply the invalidity of [his] confinement.” Brief for Respondent 21. If he prevails, he would receive only access to the DNA, and even if DNA testing exonerates him, his conviction is not automatically invalidated. He must bring an entirely separate suit or a petition for clemency to invalidate his conviction. If he were proved innocent, the State might also release him on its own initiative, avoiding any need to pursue habeas at all. Osborne also invokes our recent decision in Wilkinson v. Dotson, 544 U. S. 74 (2005). There, we held that prisoners who sought new hearings for parole eligibility and suitability need not proceed in habeas. We acknowledged that the two plaintiffs “hope[d]” their suits would “help bring about earlier release,” id., at 78, but concluded that the §1983 suit would not accomplish that without further proceedings. “Because neither prisoner’s claim would necessarily spell speedier release, neither l[ay] at the core of habeas corpus.” Id., at 82 (internal quotation marks omitted). Every Court of Appeals to consider the question since Dotson has decided that because access to DNA evidence similarly does not “necessarily spell speedier release,” ibid., it can be sought under § 1983. See 423 F. 3d, at 1055-1056; Savory v. Lyons, 469 F. 3d 667, 672 (CA7 2006); McKithen v. Brown, 481 F. 3d 89, 103, and n. 15 (CA2 2007). On the other hand, the State argues that Dotson is distinguishable because the challenged procedures in that case did not affect the ultimate “exercise of discretion by the parole board.” Brief for Petitioners 32. It also maintains that Dotson does not set forth “the exclusive test for whether a prisoner may proceed under § 1983.” Brief for Petitioners 32. While we granted certiorari on this question, our resolution of Osborne’s claims does not require us to resolve this difficult issue. Accordingly, we will assume without deciding that the Court of Appeals was correct that Heck does not bar Osborne’s § 1983 claim. Even under this assumption, it was wrong to find a due process violation. IV A “No State shall . . . deprive any person of life, liberty, or property, without due process of law.” U. S. Const., Amdt. 14, § 1; accord, Amdt. 5. This Clause imposes procedural limitations on a State’s power to take away protected entitlements. See, e. g., Jones v. Flowers, 547 U. S. 220, 226-239 (2006). Osborne argues that access to the State’s evidence is a “process” needed to vindicate his right to prove himself innocent and get out of jail. Process is not an end in itself, so a necessary premise of this argument is that he has an entitlement (what our precedents call a “liberty interest”) to prove his innocence even after a fair trial has proved otherwise. We must first examine this asserted liberty interest to determine what process (if any) is due. See Board of Regents of State Colleges v. Roth, 408 U. S. 564, 570-571 (1972); Olim v. Wakinekona, 461 U. S. 238, 250-251 (1983). In identifying his potential liberty interest, Osborne first attempts to rely on the Governor’s constitutional authority to “grant pardons, commutations, and reprieves.” Alaska Const., Art. Ill, §21. That claim can be readily disposed of. We have held that noncapital defendants do not have a liberty interest in traditional state executive clemency, to which no particular claimant is entitled as a matter of state law. Connecticut Bd. of Pardons v. Dumschat, 452 U. S. 458, 464 (1981). Osborne therefore cannot challenge the constitutionality of any procedures available to vindicate an interest in state clemency. Osborne does, however, have a liberty interest in demonstrating his innocence with new evidence under state law. As explained, Alaska law provides that those who use “newly discovered evidence” to “establis[h] by clear and convincing evidence that [they are] innocent” may obtain “vacation of [their] conviction or sentence in the interest of justice.” Alaska Stat. §§ 12.72.020(b)(2), 12.72.010(4). This “state-created right can, in some circumstances, beget yet other rights to procedures essential to the realization of the parent right.” Dumschat, supra, at 463; see also Wolff v. McDonnell, 418 U. S. 539, 556-558 (1974). The Court of Appeals went too far, however, in concluding that the Due Process Clause requires that certain familiar preeonviction trial rights be extended to protect Osborne’s postconviction liberty interest. After identifying Osborne’s possible liberty interests, the court concluded that the State had an obligation to comply with the principles of Brady v. Maryland, 373 U. S. 83. In that case, we held that due process requires a prosecutor to disclose material exculpatory evidence to the defendant before trial. The Court of Appeals acknowledged that nothing in our precedents suggested that this disclosure obligation continued after the defendant was convicted and the case was closed, 521 F. 3d, at 1128, but it relied on prior Ninth Circuit precedent applying “Brady as a post-conviction right,” ibid, (citing Thomas v. Goldsmith, 979 F. 2d 746, 749-750 (1992)). Osborne does not claim that Brady controls this case, Brief for Respondent 39-40, and with good reason. A criminal defendant proved guilty after a fair trial does not have the same liberty interests as a free man. At trial, the defendant is presumed innocent and may demand that the government prove its case beyond reasonable doubt. But. “[o]nee a defendant has been afforded a fair trial and convicted of the offense for which he was charged, the presumption of innocence disappears.” Herrera, 506 U. S., at 399. “Given a valid conviction, the criminal defendant has been constitutionally deprived of his liberty.” Dumschat, supra, at 464 (internal quotation marks and alterations omitted). The State accordingly has more flexibility in deciding what procedures are needed in the context of postconviction relief. “[W]hen a State chooses to offer help to those seeking relief from convictions,” due process does not “dictat[e] the exact form such assistance must assume.” Pennsylvania v. Finley, 481 U. S. 551, 559 (1987). Osborne’s right to due process is not parallel to a trial right, but rather must be analyzed in light of the fact that he has already been found guilty at a fair trial, and has only a limited interest in postconviction relief. Brady is the wrong framework. Instead, the question is whether consideration of Osborne’s claim within the framework of the State’s procedures for postconviction relief “offends some principle of justice so rooted in the traditions and conscience of our people as to be ranked as fundamental,” or “transgresses any recognized principle of fundamental fairness in operation.” Medina v. California, 505 U. S. 437, 446, 448 (1992) (internal quotation marks omitted); see Herrera, supra, at 407-408 (applying Medina to postconviction relief for actual innocence); Finley, supra, at 556 (postconviction relief procedures are constitutional if they “compor[t] with fundamental fairness”). Federal courts may upset a State’s postconviction relief procedures only if they are fundamentally inadequate to vindicate the substantive rights provided. We see nothing inadequate about the procedures Alaska has provided to vindicate its state right to postconviction relief in general, and nothing inadequate about how those procedures apply to those who seek access to DNA evidence. Alaska provides a substantive right to be released on a sufficiently compelling showing of new evidence that establishes innocence. It exempts such claims from otherwise applicable time limits. The State provides for discovery in postconviction proceedings, and has — through judicial decision — specified that this discovery procedure is available to those seeking access to DNA evidence. Patterson, 2006 WL 573797, *4. These procedures are not without limits. The evidence must indeed be newly available to qualify under Alaska’s statute, must have been diligently pursued, and must also be sufficiently material. These procedures are similar to those provided for DNA evidence by federal law and the law of other States, see, e. g., 18 U. S. C. § 3600(a), and they are not inconsistent with the “traditions and conscience of our people” or with “any recognized principle of fundamental fairness,” Medina, supra, at 446, 448 (internal quotation marks omitted). And there is more. While the Alaska courts have not had occasion to conclusively decide the question, the Alaska Court of Appeals has suggested that the State Constitution provides an additional right of access to DNA. In expressing its “reluctante] to hold that Alaska law offers no remedy” to those who belatedly seek DNA testing, and in invoking the three-part test used by other state courts, the court indicated that in an appropriate case the State Constitution may provide a failsafe even for those who cannot satisfy the statutory requirements under general postconviction procedures. Osborne I, 110 P. 3d, at 995-996. To the degree there is some uncertainty in the details of Alaska’s newly developing procedures for obtaining postconviction access to DNA, we can hardly fault the State for that. Osborne has brought this §1983 action without ever using these procedures in filing a state or federal habeas claim relying on actual innocence. In other words, he has not tried to use the process provided to him by the State or attempted to vindicate the liberty interest that is now the centerpiece of his claim. When Osborne did request DNA testing in state court, he sought RFLP testing that had been available at trial, not the STR testing he now seeks, and the state court relied on that fact in denying him testing under Alaska law. Id., at 992 (“[T]he DNA testing that Osborne proposes to perform on this evidence existed at the time of Osborne’s trial”); Osborne II, 163 P. 3d, at 984 (Mannheimer, J., concurring) (“[T]he DNA testing [Osborne] proposes would not yield ‘new evidence’ for purposes of . . . [Alaska Stat. §12.72.010]” because it was “available at the time of Osborne’s trial”). His attempt to sidestep state process through a new federal lawsuit puts Osborne in a very awkward position. If he simply seeks the DNA through the State’s discovery procedures, he might well get it. If he does not, it may be for a perfectly adequate reason, just as the federal statute and all state statutes impose conditions and limits on access to DNA evidence. It is difficult to criticize the State’s procedures when Osborne has not invoked them. This is not to say that Osborne must exhaust state-law remedies. See Patsy v. Board of Regents of Fla., 457 U. S. 496, 500-501 (1982). But it is Osborne’s burden to demonstrate the inadequacy of the state-law procedures available to him in state postconviction relief. Cf. Medina, supra, at 453. These procedures are adequate on their face, and without trying them, Osborne can hardly complain that they do not work in practice. As a fallback, Osborne also obliquely relies on an asserted federal constitutional right to be released upon proof of “actual innocence.” Whether such a federal right exists is an open question. We have struggled with it over the years, in some cases assuming, arguendo, that it exists while also noting the difficult questions such a right would pose and the high standard any claimant would have to meet. House, 547 U. S., at 554-555; Herrera, 506 U. S., at 398-417; see also id., at 419-421 (O’Connor, J., concurring); id., at 427-428 (SCALIA, J., concurring); Friendly, Is Innocence Irrelevant? Collateral Attack on Criminal Judgments, 38 U. Chi. L. Rev. 142, 159, n. 87 (1970). In this case too we can assume without deciding that such a claim exists, because even if so there is no due process problem. Osborne does not dispute that a federal actual innocence claim (as opposed to a DNA access claim) would be brought in habeas. Brief for Respondent 22-24. If such a habeas claim is viable, federal procedural rules permit discovery “for good cause.” 28 U. S. C. §2254 Rule 6; Bracy v. Gramley, 520 U. S. 899, 908-909 (1997). Just as with state law, Osborne cannot show that available discovery is facially inadequate, and cannot show that it would be arbitrarily denied to him. B The Court of Appeals below relied only on procedural due process, but Osborne seeks to defend the judgment on the basis of substantive due process as well. He asks that we recognize a freestanding right to DNA evidence untethered from the liberty interests he hopes to vindicate with it. We reject the invitation and conclude, in the circumstances of this case, that there is no such substantive due process right. “As a general matter, the Court has always been reluctant to expand the concept of substantive due process because guideposts for responsible decisionmaking in this unchartered area are scarce and open-ended.” Collins v. Harker Heights, 503 U. S. 115, 125 (1992). Osborne seeks access to state evidence so that he can apply new DNA-testing technology that might prove him innocent. There is no long history of such a right, and “[t]he mere novelty of such a claim is reason enough to doubt that ‘substantive due process’ sustains it.” Reno v. Flores, 507 U. S. 292, 303 (1993). And there are further reasons to doubt. The elected governments of the States are actively confronting the challenges DNA technology poses to our criminal justice systems and our traditional notions of finality, as well as the opportunities it affords. To suddenly constitutionalize this area would short circuit what looks to be a prompt and considered legislative response. The first DNA-testing statutes were passed in 1994 and 1997. Act of Aug. 2, 1994, ch. 737, 1994 N. Y. Laws 3709 (codified at N. Y. Crim. Proc. Law Ann. § 440.30(l-a) (West 2005)); Act of May 9, 1997, Pub. Act No. 90-141, 1997 Ill. Laws 2461 (codified at 725 Ill. Comp. Stat., ch. 725, § 5/116-3(a) (West 2007)). In the past decade, 44 States and the Federal Government have followed suit, reflecting the increased availability of DNA testing. As noted, Alaska itself is considering such legislation. See supra, at 64. “By extending constitutional protection to an asserted right or liberty interest, we, to a great extent, place the matter outside the arena of public debate and legislative action. We must therefore exercise the utmost care whenever we are asked to break new ground in this field.” Glucksberg, 521 U. S., at 720 (internal quotation marks omitted). “[J]udieial imposition of a categorical remedy . . . might pretermit other responsible solutions being considered in Congress and state legislatures.” Murray v. Giarratano, 492 U. S. 1, 14 (1989) (Kennedy, J., concurring in judgment). If we extended substantive due process to this area, we would cast these statutes into constitutional doubt and be forced to take over the issue of DNA access ourselves. We are reluctant to enlist the Federal Judiciary in creating a new constitutional code of rules for handling DNA. Establishing a freestanding right to access DNA evidence for testing would force us to act as policymakers, and our substantive due process rulemaking authority would not only have to cover the right of access but a myriad of other issues. We would soon have to decide if there is a constitutional obligation to preserve forensic evidence that might later be tested. Cf. Arizona v. Youngblood, 488 U. S. 51, 56-58 (1988). If so, for how long? Would it be different for different types of evidence? Would the State also have some obligation to gather such evidence in the first place? How much, and when? No doubt there would be a miscellany of other minor directives. See, e. g., Harvey v. Horan, 285 F. 3d 298, 300-301 (CA4 2002) (Wilkinson, C. J., concurring in denial of rehearing). In this case, the evidence has already been gathered and preserved, but if we extend substantive due process to this area, these questions would be before us in short order, and it is hard to imagine what tools federal courts would use to answer them. At the end of the day, there is no reason to suppose that their answers to these questions would be any better than those of state courts and legislatures, and good reason to suspect the opposite. See Collins, supra, at 125; Glucksberg, supra, at 720. * * * DNA evidence will undoubtedly lead to changes in the criminal justice system. It has done so already. The question is whether further change will primarily be made by legislative revision and judicial interpretation of the existing system, or whether the Federal Judiciary must leap ahead— revising (or even discarding) the system by creating a new constitutional right and taking over responsibility for refining it. Federal courts should not presume that state criminal procedures will be inadequate to deal with technological change. The criminal justice system has historically accommodated new types of evidence, and is a time-tested means of carrying out society’s interest in convicting the guilty while respecting individual rights. That system, like any human endeavor, cannot be perfect. DNA evidence shows that it has not been. But there is no basis for Osborne’s approach of assuming that because DNA has shown that these procedures are not flawless, DNA evidence must be treated as categorically outside the process, rather than within it. That is precisely what his § 1983 suit seeks to do, and that is the contention we reject. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. RFLP testing, unlike DQ Alpha testing, “has a high degree of discrimination,” although it is sometimes ineffective on small samples. Postconviction DNA Testing 26-27; Future of Forensic DNA Testing 14-16. Billingslea testified that she had no memory of Osborne making such a request, but said she was “ ‘willing to accept’ ” that he had. Osborne I, 110 P. 3d 986, 990 (Alaska App. 2005). It is not clear whether the Alaska Court of Appeals was correct that Osborne sought only forms of DNA testing that had been available at trial, compare id., at 992, 995, with 521 F. 3d 1118, 1123, n. 2 (CA9 2008), but it resolved the ease on that basis. STR testing is extremely discriminating, can be used on small samples, and is “rapidly becoming the standard.” Future of Forensic DNA Testing 18, n. 9. Osborne also sought to subject the pubic hairs to mitochondrial DNA testing, a secondary testing method often used when a sample cannot be subjected to other tests. See Postconviction DNA Testing 28. He argues that “[a]ll of the same arguments that support access to the condom for STR testing support access to the hairs for mitochondrial testing as well,” Brief for Respondent 11, n. 4, and we treat the claim accordingly. The dissent asserts that our position “resembles” Justice Harlan’s dissent in Miranda v. Arizona, 384 U. S. 436 (1966). Post, at 101, n. 10 (opinion of Stevens, J.). Miranda devised rules to safeguard a constitutional right the Court had already recognized. Indeed, the underlying requirement at issue in that case that confessions be voluntary had “roots” going back centuries. Dickerson v. United States, 530 U. S. 428, 432-433 (2000). In contrast, the asserted right to access DNA evidence is unrooted in history or tradition, and would thrust the Federal Judiciary into an area previously left to state courts and legislatures.
Justice Alito, with whom Justice Kennedy joins, and with whom Justice Thomas joins as to Part II, concurring. Respondent was convicted for a brutal sexual assault. At trial, the defense declined to have DNA testing done on a semen sample found at the scene of the crime. Defense counsel explained that this decision was made based on fear that the testing would provide further evidence of respondent’s guilt. After conviction, in an unsuccessful attempt to obtain parole, respondent confessed in detail to the crime. Now, respondent claims that he has a federal constitutional right to test the sample and that he can go directly to federal court to obtain this relief without giving the Alaska courts-a full opportunity to consider his claim. I agree with the Court’s resolution of respondent’s constitutional claim. In my view, that claim also fails for two independent reasons beyond those given by the majority. First, a state prisoner asserting a federal constitutional right to perform such testing must file a petition for a writ of habeas corpus, not an action under Rev. Stat. § 1979,42 U. S. C. § 1983, as respondent did here, and thus must exhaust state remedies, see 28 U. S. C. § 2254(b)(1)(A). Second, even though respondent did not exhaust his state remedies, his claim may be rejected on the merits, see § 2254(b)(2), because a defendant who declines the opportunity to perform DNA testing at trial for tactical reasons has no constitutional right to perform such testing after conviction. I As our prior opinions illustrate, it is sometimes difficult to draw the line between claims that are properly brought in habeas and those that may be brought under 42 U. S. C. § 1983. See Preiser v. Rodriguez, 411 U. S. 475 (1973); Heck v. Humphrey, 512 U. S. 477 (1994); Wilkinson v. Dotson, 544 U. S. 74 (2005). But I think that this case falls on the habeas side of the line. We have long recognized the principles of federalism and comity at stake when state prisoners attempt to use the federal courts to attack their final convictions. See, e. g., Darr v. Burford, 339 U. S. 200, 204 (1950); Braden v. 30th Judicial Circuit Court of Ky., 410 U. S. 484, 490-491 (1973); Preiser, supra, at 491-492; Rose v. Lundy, 455 U. S. 509, 518-519 (1982); Rhines v. Weber, 544 U. S. 269, 273-274 (2005). We accordingly held that “ ‘it would be unseemly in our dual system of government for a federal district court to upset a state court conviction without an opportunity to the state courts to correct a constitutional violation.’ ” Dandy, supra, at 518 (quoting Darr, supra, at 204). Congress subsequently codified Dandy’s exhaustion requirement in the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), 28 U. S. C. § 2254(b)(1)(A). We also have long recognized the need to impose sharp limits on state prisoners’ efforts to bypass state courts with their discovery requests. See, e. g., Wainwright v. Sykes, 433 U. S. 72, 87-90 (1977); Keeney v. Tamayo-Reyes, 504 U. S. 1, 8-10 (1992); Williams v. Taylor, 529 U. S. 420, 436 (2000). For example, we have held that “concerns of finality, comity, judicial economy, and channeling the resolution of claims into the most appropriate forum” require a state prisoner to show “cause-and-prejudice” before asking a federal habeas court to hold an evidentiary hearing. Keeney, supra, at 8. That result reduces opportunities for “ ‘sandbagging’ on the part of defense lawyers,” Sykes, supra, at 89, and it “reduces the ‘inevitable friction’ that results when a federal habeas court ‘overturns either the factual or legal conclusions reached by the state-court system,’” Keeney, supra, at 9 (quoting Sumner v. Mata, 449 U. S. 539, 550 (1981); brackets omitted). Congress subsequently codified Keeney’s cause-and-prejudice rule in AEDPA, 28 U. S. C. § 2254(e)(2). The rules set forth in our cases and codified in AEDPA would mean very little if state prisoners could simply evade them through artful pleading. For example, I take it as common ground that a state prisoner’s claim under Brady v. Maryland, 373 U. S. 83 (1963), must be brought in habeas because that claim, if proved, would invalidate the judgment of conviction or sentence (and thus the lawfulness of the inmate’s confinement). See Heck, supra, at 481. But under respondent’s view, I see no reason why a Brady claimant could not bypass the state courts and file a § 1983 claim in federal court, contending that he has a due process right to search the State’s files for exculpatory evidence. Allowing such a maneuver would violate the principles embodied in Lundy, Keeney, and AEDPA. Although respondent has now recharacterized his claim in an effort to escape the requirement of proceeding in habeas, in his complaint he squarely alleged that the State “deprived [him] of access to exculpatory evidence in violation of Brady[, supra], and the Due Process Clause of the Fourteenth Amendment to the U. S. Constitution.” App. 37. That allegedly “exculpatory” evidence — which Brady defines as “evidence favorable to [the] accused” and “material either to guilt or to punishment,” 373 U. S., at 87 — would, by definition, undermine respondent’s “guilt” or “punishment” if his allegations are true. Such claims should be brought in habeas, see Heck, supra, at 481, and respondent cannot avoid that result by attempting to bring his claim under § 1983, see Dotson, supra, at 92 (Kennedy, J., dissenting). It is no answer to say, as respondent does, that he simply wants to use § 1983 as a discovery tool to lay the foundation for a future state postconviction application, a state clemency petition, or a request for relief by means of “prosecutorial consent.” See Brief for Respondent 23. Such tactics implicate precisely the same federalism and comity concerns that motivated our decisions (and Congress’) to impose exhaustion requirements and discovery limits in federal habeas proceedings. If a petitioner can evade the habeas statute’s exhaustion requirements in this way, I see no reason why a state prisoner asserting an ordinary Brady claim — i. e., a state prisoner who claims that the prosecution failed to turn over exculpatory evidence prior to trial — could not follow the same course. What respondent seeks was accurately described in his complaint — the discovery of evidence that has a material bearing on his conviction. Such a claim falls within “the core” of habeas. Preiser, supra, at 489. Recognition of a constitutional right to postconviction scientific testing of evidence in the possession of the prosecution would represent an expansion of Brady and a broadening of the discovery rights now available to habeas petitioners. See 28 U. S. C. § 2254 Rule 6. We have never previously held that a state prisoner may seek discovery by means of a §1983 action, and we should not take that step here. I would hold that respondent’s claim (like all other Brady claims) should be brought in habeas. II The principles of federalism, comity, and finality are not the only ones at stake for the State in cases like this one. To the contrary, DNA evidence creates special opportunities, risks, and burdens that implicate important state interests. Given those interests — and especially in light of the rapidly evolving nature of DNA testing technology — this is an area that should be (and is being) explored “through the workings of normal democratic processes in the laboratories of the States.” Atkins v. Virginia, 536 U. S. 304, 326 (2002) (Rehnquist, C. J., dissenting). A As the Court notes, DNA testing often produces highly reliable results. See ante, at 62. Indeed, short tandem repeat (STR) “DNA tests can, in certain circumstances, establish to a virtual certainty whether a given individual did or did not commit a particular crime.” Harvey v. Horan, 285 F. 3d 298, 305 (CA4 2002) (Luttig, J., respecting denial of rehearing en banc). Because of that potential for “virtual certainty,” Justice Stevens argues that the State should welcome respondent’s offer to perform modern DNA testing (at his own expense) on the State’s DNA evidence; the test will either confirm respondent’s guilt (in which case the State has lost nothing) or exonerate him (in which case the State has no valid interest in detaining him). See post, at 97-98. Alas, it is far from that simple. First, DNA testing — even when performed with modern STR technology, and even when performed in perfect accordance with protocols — often fails to provide “absolute proof” of anything. Post, at 98 (Stevens, J., dissenting). As one scholar has observed: “[F]orensic DNA testing rarely occurs [under] idyllic conditions. Crime scene DNA samples do not come from a single source obtained in immaculate conditions; they are messy assortments of multiple unknown persons, often collected in the most difficult conditions. The samples can be of poor quality due to exposure to heat, light, moisture, or other degrading elements. They can be of minimal or insufficient quantity, especially as investigators push DNA testing to its limits and seek profiles from a few cells retrieved from cigarette butts, envelopes, or soda cans. And most importantly, forensic samples often constitute a mixture of multiple persons, such that it is not clear whose profile is whose, or even how many profiles are in the sample at all. All of these factors make DNA testing in the forensic context far more subjective than simply reporting test results ....” Murphy, The Art in the Science of DNA: A Layperson’s Guide to the Subjectivity Inherent in Forensic DNA Typing, 58 Emory L. J. 489, 497 (2008) (footnotes omitted). See also R. Michaelis, R. Flanders, & P. Wulff, A Litigator’s Guide to DNA 341 (2008) (hereinafter Michaelis) (noting that even “STR analyses are plagued by issues of suboptimal samples, equipment malfunctions and human error, just as any other type of forensic DNA test”); Harvey v. Horan, 278 F. 3d 370, 383, n. 4 (CA4 2002) (King, J., concurring in part and concurring in judgment) (noting that the first STR DNA test performed under Virginia’s postconviction DNA access statute was inconclusive). Such concerns apply with particular force where, as here, the sample is minuscule, it may contain three or more persons’ DNA, and it may have degraded significantly during the 24 or more hours it took police to recover it. Second, the State has important interests in maintaining the integrity of its evidence, and the risks associated with evidence contamination increase every time someone attempts to extract new DNA from a sample. According to Professor John Butler — who is said to have written “the canonical text on forensic DNA typing,” Murphy, supra, at 493, n. 16 — “[t]he extraction process is probably where the DNA sample is more susceptible to contamination in the laboratory than at any other time in the forensic DNA analysis process,” J. Butler, Forensic DNA Typing 42 (2d ed. 2005). Indeed, modern DNA testing technology is so powerful that it actually increases the risks associated with mishandling evidence. STR tests, for example, are so sensitive that they can detect DNA transferred from person X to a towel (with which he wipes his face), from the towel to Y (who subsequently wipes his face), and from Y’s face to a murder weapon later wielded by Z (who can use STR technology to blame X for the murder). See Michaelis 62-64; Thompson, Ford, Doom, Raymer, & Krane, Evaluating Forensic DNA Evidence: Essential Elements of a Competent Defense Review (Part 2), The Champion, May 2003, pp. 25-26. Any test that is sensitive enough to pick up such trace amounts of DNA will be able .to detect even the slightest, unintentional mishandling of evidence. See Michaelis 63 (cautioning against mishandling evidence because “two research groups have already demonstrated the ability to obtain STR profiles from fingerprints on paper or evidence objects”). And that is to say nothing of the intentional DNA-evidence-tampering scandals that have surfaced in recent years. See, e. g., Murphy, The New Forensics: Criminal Justice, False Certainty, and the Second Generation of Scientific Evidence, 95 Cal. L. Rev. 721, 772-773 (2007) (collecting examples). It gives short shrift to such risks to suggest that anyone — including respondent, who has twice confessed to his crime, has never recanted, and passed up the opportunity for DNA testing at trial — should be given a never-before-recognized constitutional right to rummage through the State’s genetic-evidence locker. Third, even if every test was guaranteed to provide a conclusive answer, and even if no one ever contaminated a DNA sample, that still would not justify disregarding the other costs associated with the DNA access regime proposed by respondent. As the Court notes, recognizing a prisoner’s freestanding right to access the State’s DNA evidence would raise numerous policy questions, not the least of which is whether and to what extent the State is constitutionally obligated to collect and preserve such evidence. See ante, at 73-74. But the policy problems do not end there. Even without our creation and imposition of a mandatory-DNA-access regime, state crime labs are already responsible for maintaining and controlling hundreds of thousands of new DNA samples every year. For example, in the year 2005, the State of North Carolina processed DNA samples in approximately 1,900 cases, while the Commonwealth of Virginia processed twice as many. See Office of State Budget and Management, Cost Study of DNA Testing and Analysis As Directed by Session Law 2005-267, Section 15.8, pp. 5, 8 (Mar. 1, 2006) (hereinafter North Carolina Study), http://www.osbm.state.nc.us/files/pdf_files/3-l-2006FinalDNA Report.pdf (all Internet materials as visited June 16, 2009, and available in Clerk of Court’s case file); see also id., at 8 (noting that the State of Iowa processed DNA samples in 1,500 eases in that year). Each case often entails many separate DNA samples. See Wisconsin Criminal Justice Study Commission, Position Paper: Decreasing the Turnaround Time for DNA Testing, p. 2 (hereinafter Wisconsin Study), http://www.wcjsc.org/WCJSC_Report_on_DNA_Backlog.pdf (“An average case consists of 8 samples”). And these data— which are now four years out of date — dramatically underestimate the States’ current DNA-related caseloads, which expand at an average annual rate of around 24%. See Wisconsin Dept, of Justice, Review of State Crime Lab Resources for DNA Analysis 6 (Feb. 12, 2007), http://www.doj. state. wi.us/news/files/dnaanalysisplan.pdf. The resources required to process and analyze these hundreds of thousands of samples have created severe backlogs in state crime labs across the country. For example, the State of Wisconsin reports that it receives roughly 17,600 DNA samples per year, but its labs can process only 9,600. Wisconsin Study 2. Similarly, the State of North Carolina reports that “[i]t is not unusual for the [State] Crime Lab to have several thousand samples waiting to be outsourced due to the federal procedures for [the State’s] grant. This is not unique to North Carolina but a national issue.” North Carolina Study 9. The procedures that the state labs use to handle these hundreds of thousands of DNA samples provide fertile ground for litigation. For example, in Commonwealth v. Duarte, 56 Mass. App. 714, 723, 780 N. E. 2d 99, 106 (2002), the defendant argued that “the use of a thermometer that may have been overdue for a standardization check rendered the DNA analysis unreliable and inadmissible” in his trial for raping a 13-year-old girl. The court rejected that argument and held “that the status of the thermometer went to the weight of the evidence, and not to its admissibility,” id., at 724, 780 N. E. 2d, at 106, and the court ultimately upheld Duarte’s conviction after reviewing the testimony of the deputy director of the laboratory that the Commonwealth used for the DNA tests, see ibid. But the case nevertheless illustrates “that no detail of laboratory operation, no matter how minute, is exempt as a potential point on which a defense attorney will question the DNA evidence.” Michaelis 68; see also id., at 68-69 (discussing the policy implications of Duarte). My point in recounting the burdens that postconviction DNA testing imposes on the Federal Government and the States is not to denigrate the importance of such testing. Instead, my point is that requests for postconviction DNA testing are not cost free. The Federal Government and the States have a substantial interest in the implementation of rules that regulate such testing in a way that harnesses the unique power of DNA testing while also respecting the important governmental interests noted above. The Federal Government and the States have moved expeditiously to enact rules that attempt to perform this role. And as the Court holds, it would be most unwise for this Court, wielding the blunt instrument of due process, to interfere prematurely with these efforts. B I see no reason for such intervention in the present case. When a criminal defendant, for tactical purposes, passes up the opportunity for DNA testing at trial, that defendant, in my judgment, has no constitutional right to demand to perform DNA testing after conviction. Recognition of such a right would allow defendants to play games with the criminal justice system. A guilty defendant could forgo DNA testing at trial for fear that the results would confirm his guilt, and in the hope that the other evidence would be insufficient to persuade the jury to find him guilty. Then, after conviction, with nothing to lose, the defendant could demand DNA testing in the hope that some happy accident — for example, degradation or contamination of the evidence — would provide the basis for seeking postconviction relief. Denying the opportunity for such an attempt to game the criminal justice system should not shock the conscience of the Court. There is ample evidence in this case that respondent attempted to game the system. At trial, respondent’s lawyer made an explicit, tactical decision to forgo restriction-fragment-length-polymorphism (RFLP) testing in favor of less reliable DQ Alpha testing. Having forgone more accurate DNA testing once before, respondent’s reasons for seeking it now are suspect. It is true that the STR testing respondent now seeks is even more advanced than the RFLP testing he declined — but his counsel did not decline RFLP testing because she thought it was not good enough; she dedined because she thought it was too good. Osborne I, 110 P. 3d 986,990 (Alaska App. 2005). “[A] defendant should not be allowed to take a gambler’s risk and complain only if the cards [fall] the wrong way.” Osborne v. State, 163 P. 3d 973, 984 (Alaska App. 2007) (Mannheimer, J., concurring) (internal quotation marks omitted). Justice Stevens contends that respondent should not be bound by his attorney’s tactical decision and notes that respondent testified in the state postconviction proceeding that he strongly objected to his attorney’s strategy. See post, at 97-98, n. 8. His attorney, however, had no memory of that objection, and the state court did not find that respondent’s testimony was truthful. Nor do we have reason to assume that respondent was telling the truth, particularly since he now claims that he lied at his parole hearing when he twice confessed to the crimes for which he was convicted. In any event, even assuming for the sake of argument that respondent did object at trial to his attorney’s strategy, it is a well-accepted principle that, except in a few carefully defined circumstances, a criminal defendant is bound by his attorney’s tactical decisions unless the attorney provided constitutionally ineffective assistance. See Vermont v. Britton, 556 U. S. 81, 90-91 (2009). Here, the state postconviction court rejected respondent’s ineffective-assistance claim, Osborne I, swpra, at 991-992; respondent does not challenge that holding; and we must therefore proceed on the assumption that his attorney’s decision was reasonable and binding. * * If a state prison