McKithen v. Brown

565 F. Supp. 2d 440, 2008 U.S. Dist. LEXIS 55094, 2008 WL 2791852

• Court: District Court, E.D. New York
• Decided: July 21, 2008
• Docket: 1:02-mj-01670
• Status: Published

Opinion

MEMORANDUM AND ORDER

JOHN GLEESON, District Judge:

This case presents the question whether the United States Constitution requires that a convicted defendant be granted access to evidence for the purpose of DNA testing, an issue of first impression in this circuit. Frank McKithen is in prison for stabbing his wife Rose with a kitchen knife on August 21, 1992. McKithen wishes to subject the knife to DNA testing, which he contends will demonstrate that his wife, who survived, fabricated McKithen’s involvement in the incident. After moving unsuccessfully for this relief under New York’s post-conviction DNA testing statute, N.Y.Crim. Proc. Law § 440.30(l-a), McKithen brought this action under 42 U.S.C. § 1983, claiming that Queens County District Attorney Richard Brown has injured him in violation of the Due Process Clause of the Fourteenth Amendment by refusing to grant him access to the knife.

On March 27, 2003,1 adopted Magistrate Judge Lois Bloom’s report and recommendation and granted Brown’s motion to dismiss McKithen’s claim on procedural grounds. On March 13, 2007, the United States Court of Appeals for the Second Circuit remanded for a determination of the merits, specifically directing me to consider whether the Constitution protects a post-conviction right of access to evidence for DNA testing, and if so, what the contours of that right are and whether McKi- then’s claim that he is entitled to access to the knife is precluded by the state court’s decision that he was not entitled to testing under § 440.30(l-a). On the parties’ cross-motions for summary judgment, 1 I conclude that the Due Process Clause of the Fourteenth Amendment grants a convicted offender access to physical evidence for the purpose of DNA testing if it can be performed with negligible cost to the state and exculpatory results would undermine confidence in the outcome of trial; that the issue of McKithen’s entitlement to DNA testing is not precluded; and that McKi-then is entitled to access the knife for the purpose of DNA testing. Accordingly, for the reasons stated below, McKithen’s motion is granted and Brown’s is denied.

BACKGROUND

A. The Offense Conduct

McKithen stands convicted of attempted murder in the second degree, intimidating a victim or witness in the first degree, reckless endangerment in the first degree, criminal possession of a weapon in the fourth degree, assault in the first degree, and resisting arrest based on a stabbing incident on August 21, 1992 and his subsequent arrest on September 1,1992.

The government’s evidence at trial established that on the morning of August 21, 1992, McKithen’s wife Rose testified before a Queens County grand jury that on August 16, 1992, McKithen had menaced her with a firearm in their shared home at 150-18 Yates Road in Queens. At approximately 11:00 PM on August 21, 1992, Rose McKithen, 2 having returned from her grand jury appearance, was watching television with her friend Linda Jones in the living room of 150-18 Yates Road. McKithen climbed in a bedroom window and appeared in the living room. He argued with Rose McKithen and struck her with an open hand, and then retrieved a large kitchen knife from the cutlery drawer. Holding Rose McKithen at knife-point, McKithen ushered Jones out of the apartment.

After Jones was expelled from the apartment, Rose McKithen broke free of McKithen’s grasp and attempted to crawl out a bedroom window. As she tried to flee, McKithen stabbed her once in the back and then immediately fled the apartment. When the police responded, they recovered a knife with a spot or two of blood on it, which Rose McKithen identified as the knife her husband had used to stab her. Emergency medical technicians responded to the scene and asked her who to contact in case of an emergency. She gave McKithen’s name.

McKithen was arrested on September 1, 1992 and resisted arrest in circumstances not at issue here. After arresting him, McKithen’s arresting officer asked him if he knew why he was being arrested, and McKithen responded, “Yes, I should have killed the bitch.”

B. The Procedural History

1. Conviction, Sentence and Posh-Conviction Proceedings Challenging Conviction and Sentence

McKithen was tried by jury in New York Supreme Court, Queens County. The government introduced a knife into evidence during its case. Rose McKithen identified the knife as the weapon McKi-then had stabbed her with, and a police officer identified the knife as having been found on a table in Rose McKithen’s bedroom. Two police officers testified that there may have been a small amount of blood on the knife at the time that it was recovered, but an officer wrapped it in a paper towel to prevent the blade from causing accidental injury, which had the effect of removing the blood. McKithen did not request that the knife be subject to fingerprint or DNA testing.

McKithen argued that his wife had fabricated his involvement in the incident and that she had actually been stabbed by her boyfriend. In support of this theory, he noted that Rose McKithen had given McKithen’s name as an emergency contact. McKithen was convicted of attempted murder in the second degree, intimidating a victim or witness in the first degree, reckless endangerment in the first degree, criminal possession of a weapon in the fourth degree, assault in the first degree, and resisting arrest. On May 19, 1993, McKithen was sentenced as a second violent felony offender to consecutive terms of imprisonment of twelve and one-half to 25 years on the attempted murder conviction, eight to 16 years on the witness intimidation conviction, and two and one-half to five years on the reckless endangerment conviction. He was also sentenced to concurrent terms of imprisonment of seven to fourteen years on the assault conviction, one year on the weapon possession conviction, and one year on the resisting arrest conviction. His aggregate sentence was 23 to 46 years.

McKithen appealed his conviction, arguing that the charges related to his arrest should have been severed from the charges related to his stabbing; the prosecutor committed misconduct in summation; the evidence was insufficient to support the verdict convicting him of reckless endangerment; and his sentences on the attempted murder and witness intimidation convictions should have run concurrently. On November 13, 1995, the Appellate Division, Second Department affirmed his conviction but modified his sentence to run his attempted murder and witness intimidation convictions concurrently, resulting in an aggregate sentence of 15 to 30 years. People v. McKithen (McKithen I), 221 A.D.2d 476, 634 N.Y.S.2d 128, 129-30 (2d Dep’t 1995). McKithen requested leave to appeal to the New York Court of Appeals, which denied his application on May 8, 1995. People v. McKithen (McKithen II), 88 N.Y.2d 881, 645 N.Y.S.2d 456, 668 N.E.2d 427 (1996).

On July 27, 1994, McKithen filed a pro se motion to set aside his sentence under N.Y.Crim. Proc. Law § 440.20, which was denied on September 8, 1994. On November 6,1996, McKithen filed a pro se motion to vacate his conviction under N.Y.Crim. Proc. Law § 440.20, which was denied on February 10, 1997. On March 12, 1997, McKithen filed a pro se motion for leave to appeal the denial of his second § 440.20 motion. This application was denied by the Second Department on April 16, 1997.

2. Proceedings Seeking DNA Testing

On August 21, 2001, McKithen filed a pro se motion in the New York Supreme Court, Queens County, pursuant to N.Y.Crim. Proc. Law § 440.30(l-a), seeking court-ordered DNA testing of the knife to determine if Rose McKithen’s blood was on the knife, as well as testing for fingerprints. On November 8, 2001, the Queens Supreme Court denied his application, finding his request for fingerprint examination untimely, concluding that there was not a reasonable probability that the results of DNA testing would have resulted in a more favorable verdict, noting that McKithen did not dispute that Rose McKi-then was stabbed, and concluding that the presence or absence of her blood on the knife would have little or no probative value in determining whether her wounds were inflicted by McKithen. People v. McKithen (McKithen III), No. 3964/92 (N.Y. Sup.Ct., Queens County Nov. 8,’ 2001). On December 13, 2001, McKithen moved for leave to appeal to the Second Department, which denied his application on February 13, 2002.

On March 1, 2002, McKithen commenced this action before me, filing a pro se complaint under 42 U.S.C. § 1983, seeking access to the knife for the purpose of DNA testing. In his complaint, McKithen alleged that Rose McKithen’s blood is not on the knife, claiming that this finding would establish that the knife introduced as evidence at trial was not the knife used to stab Rose McKithen, and that this would in turn demonstrate that he did not stab her. 3 McKithen argued that Brown’s refusal to produce the knife for testing (1) deprived him of due process of law; (2) deprived him of meaningful access to the courts; (3) deprived him of the opportunity to make a conclusive showing of actual innocence in violation of the Eighth Amendment; and (4) deprived him of his right to present evidence of his innocence in violation of the Confrontation Clause.

Brown moved to dismiss on August 2, 2002, claiming that (1) the Rooker-Feld-man doctrine deprived me of subject matter jurisdiction to hear this action; (2) the Queens Supreme Court’s denial of McKi-then’s § 440.30(l-a) motion precluded McKithen from litigating the issue of his entitlement to the knife; (3) Heck v. Humphrey, 512 U.S. 477, 114 S.Ct. 2364, 129 L.Ed.2d 383 (1994), barred McKithen from seeking DNA testing except by way of a habeas corpus petition, 4 and (4) McKi-then’s complaint failed to state a claim upon which relief could be granted. I referred this motion to Magistrate Judge Bloom for report and recommendation, and on March 27, 2003 she recommended dismissing the action pursuant to the Rooker-Feldman doctrine and not reaching the other grounds. McKithen v. Queens County Disk Atty’s Office (McKithen V), No. 02-CV-1670 (JG)(LB) (E.D.N.Y. Mar. 27, 2003) (report and recommendation). 5 I adopted this report and recommendation on April 15, 2003. McKithen v. Queens County Dist. Atty’s Office (McKithen VI), No. 02-CV-1670 (JG)(LB) (E.D.N.Y. Apr. 15, 2003).

McKithen appealed my dismissal of the action and was appointed counsel, and on March 13, 2007, the Second Circuit reversed and remanded. McKithen v. Brown (McKithen VII), 481 F.3d 89, 93 (2d Cir.2007), cert. denied sub nom. Brown v. McKithen (McKithen VIII), — U.S. -, 128 S.Ct. 1218, 170 L.Ed.2d 59 (2008). It held that (1) the Rooker-Feld-man doctrine was inapplicable due to an intervening change in the law, McKithen VII, 481 F.3d at 99; (2) Heck does not bar post-conviction actions seeking access to evidence for DNA testing from being brought under § 1983, McKithen VII, 481 F.3d at 99; (3) the defense of claim preclusion — which Brown had argued on appeal — had been waived, and declined to raise it sua sponte, id, at 105; and (4) it was impossible to determine whether issue preclusion applied without determining whether the contours of any federal right to post-conviction DNA testing differed from those of the state right, id. at 106. It remanded with instructions to consider whether the Due Process Clause safeguards either a procedural or a substantive right to post-conviction access to evidence for DNA testing; and if so, what the contours of that right are and whether the issue is precluded in McKithen’s case; and if there is a right and the issue is not precluded, whether the right was infringed in McKithen’s case. McKithen VII, 481 F.3d at 93, 106-08 & n. 17.

Brown petitioned the United States Supreme Court for a writ of certiorari, which was denied on February 19, 2008. McKithen VIII, — U.S. -, 128 S.Ct. 1218.

DISCUSSION

A. The Summary Judgment Standard

Under Federal Rule of Civil Procedure 56(c), a moving party is entitled to summary judgment “if the pleadings, the discovery and disclosure materials on file, and any affidavits show that there is no genuine issue as to any material fact and that the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c); see also Gallo v. Prudential Residential Servs., 22 F.3d 1219, 1223 (2d Cir.1994) (“[T]he burden is on the moving party to demonstrate that no genuine issue respecting any material fact exists.”). A fact is “material” within the meaning of Rule 56 when its resolution “might affect the outcome of the suit under the governing law.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986).

An issue is “genuine” when “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Id. In determining whether an issue is genuine, “[t]he inferences to be drawn from the underlying affidavits, exhibits, interrogatory answers, and depositions must be viewed in the light most favorable to the party opposing the motion.” Cronin v. Aetna Life Ins. Co., 46 F.3d 196, 202 (2d Cir.1995). Therefore, although a court “should review the record as a whole, it must disregard all evidence favorable to the moving party that the jury is not required to believe.” Reeves v. Sanderson Plumbing Prods. Inc., 530 U.S. 133, 151, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000).

However, the nonmoving party cannot survive summary judgment by casting mere “metaphysical doubt” upon the evidence produced by the moving party. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). Summary judgment is proper when the moving party can show that “little or no evidence may be found in support of the nonmoving party’s case.” Gallo, 22 F.3d at 1223-24 (citations omitted).

B. Claim Preclusion

Brown argues that the Second Circuit incorrectly decided that he had waived his argument of claim preclusion, and suggests that I consider this defense anew. Brown did not move to dismiss McKithen’s § 1983 action on the grounds of claim pre- elusion, and raised this defense for the first time on McKithen’s appeal of my grant of Brown’s motion to dismiss. Brown, erroneously believing that an answer had been filed, 6 conceded that the defense could be deemed waived, and the Second Circuit concluded that it was waived. McKithen VII, 481 F.3d at 105 (“Brown has waived the defense of claim preclusion, and, given the circumstances of the case, we decline to invoke the defense nostra sponte.”). In fact, however, due to the unusual procedural posture of the case, no answer had been filed. Thus, Brown’s concession that the claim preclusion defense had been waived was in error, and he seeks to renew the claim on remand as a result.

However, under the “mandate rule,” which is a “branch of the law-of-the-case doctrine,” Burrell v. United States, 467 F.3d 160, 165 (2d Cir.2006), cert. denied, - U.S. -, 127 S.Ct. 2031, 167 L.Ed.2d 773 (2007), I am obliged to follow the decision of the Second Circuit finding the defense of claim preclusion waived. See id. (“‘[W]here issues have been explicitly or implicitly decided on appeal, the district court is obliged, on remand, to follow the decision of the appellate court.’ ” (quoting United States v. Minicone, 994 F.2d 86, 89 (2d Cir.1993))). While there may be cases where superfluous dicta in an appellate decision forms no part of the court’s mandate, cf. New England Ins. Co. v. Healthcare Underwriters Mut. Ins. Co., 352 F.3d 599, 605-06 (2d Cir.2003) (concluding that dicta in a prior footnote did not address the point being litigated on remand), this is not one of them. The Second Circuit in this case was obliged to resolve the defense of claim preclusion in order to arrive at any of its holdings. The reasoning by which it resolved this defense is therefore part of its holding. Even if the Second Circuit’s finding that the defense was waived and its decision not to raise the defense sua sponte did not preclude me from raising it sua sponte, but see McKithen VII, 481 F.3d at 105 (“[W]e conclude that it would still be inappropriate, in this case, to invoke claim preclusion nostra sponte .... ” (emphasis added)), I cannot obey the Second Circuit’s mandate to consider the existence and contours of a constitutional right to DNA testing if I apply the defense of claim preclusion.

I note as well that the fact that Brown erred in conceding that the defense of claim preclusion had been waived does not imply that the Second Circuit erred in concluding that it was waived. The Second Circuit may have concluded that even though Brown did not waive the defense by omitting it from his motion to dismiss, he waived it by erroneously conceding on appeal that it was waived. In any event, the mandate rule compels me to decline Brown’s suggestion that I reverse the Second Circuit’s decision that claim preclusion does not bar McKithen’s action.

C. The Statute of Limitations

Brown raises the statute of limitations for the first time in this motion. As noted in Section B, supra, due to the unusual procedural posture of the case, no answer has yet been filed and his statute of limitations claim has not been waived. See Santos v. Dist. Council, 619 F.2d 963, 967 (2d Cir.1980) (“The assertion of the limitations defense in the defendant’s answer, rather than in its prior motion for dismissal and summary judgment, was both timely and sufficient as a matter of pleading.”). 7

The statute of limitations for § 1983 actions brought in New York is three years. E.g., Patterson v. County of Oneida, N.Y., 375 F.3d 206, 225 (2d Cir.2004). McKithen suggests that § 1983 actions brought for equitable relief only are subject only to the doctrine of laches, and that § 1983 actions brought to compel DNA testing borrow the generous statute of limitations of the federal Innocence Protection Act of 2004, 18 U.S.C. § 3600, the federal post-conviction DNA testing statute. These mutually incompatible arguments fly in the face of the Supreme Court’s holding that 42 U.S.C. § 1988 constitutes “a directive to select, in each State, the one most appropriate statute of limitations for all § 1983 claims.” Wilson v. Garcia, 471 U.S. 261, 275, 105 S.Ct. 1938, 85 L.Ed.2d 254 (1985) (emphasis added), superseded by statute in part, Judicial Improvements Act of 1990, Pub.L. No. 101-650, § 313(a), 104 Stat. 5114 (providing a general federal statute of limitations for statutes enacted after December 1, 1990), as recognized in Jones v. R.R. Donnelley & Sons Co., 541 U.S. 369, 377-80, 124 S.Ct. 1836, 158 L.Ed.2d 645 (2004); see also Owens v. Okure, 488 U.S. 235, 242-51, 109 S.Ct. 573, 102 L.Ed.2d 594 (1989) (extending Wilson to states with multiple personal injury statutes of limitations and concluding that in all § 1983 actions brought in New York, the three-year statute governing general personal injury actions applies); United Paperworkers Int'l Union v. Specialty Paperboard, Inc., 999 F.2d 51, 56 (2d Cir.1993) (noting that Wilson holds “state tort law statutes of limitations appropriate for all claims under 42 U.S.C. § 1983” (emphasis added)). Moreover, in determining that this action was properly brought under' § 1983 as opposed to § 2254, the Second Circuit noted that in contrast to the one-year limitations period applicable to § 2254 claims, “ ‘the statute of limitations applicable to claims brought under § 1983 in New York is three years.’ ” McKithen VII, 481 F.3d at 100 n. 12 (ellipsis omitted) (quoting Patterson, 375 F.3d at 225).

A cause of action accrues when the plaintiff knows or has reason to know of the injury forming the basis of the action. Pearl v. City of Long Beach, 296 F.3d 76, 80 (2d Cir.2002). In finding the Rooker-Feldman doctrine inapplicable, the Second Circuit held that the injury of which McKi-then complained existed prior in time to the § 440.30(l-a) proceedings. McKithen VII, 481 F.3d at 98. 8 Brown argues that if McKithen’s injury preexisted the § 440.30(l-a) motion, his claim must have accrued more than three years before the filing of this action. McKithen’s § 440.30(l-a) motion was filed on August 27, 2001, and the instant action was filed on March 21, 2002. Therefore, it is technically possible for his injury to have been sustained before the filing of his § 440.30(l-a) action but within the limitations period; Brown simply argues that any plausible date of McKithen’s injury preexisting the state court action — such as the admissibility of DNA evidence in New York courts or the widespread availability of modern DNA testing techniques — must have occurred prior to 1999, and thus is outside of the limitations period.

However, Brown elides the distinction between the date McKithen sustained his injury, which the Second Circuit held precedes his § 440.30(l-a) action, and the date he knew or had reason to know of his injury, which is the date on which his claim accrues, Pearl, 296 F.3d at 80. I assume without deciding that McKithen actually sustained his injury in 1996, when Brown represents that modern DNA testing techniques reached widespread availability. However, the injury giving rise to this action is not the mere fact that DNA testing was not performed, but the infringement of McKithen’s constitutional rights this fact entails. 9 Due to the complexity of the legal issues regarding the scope of these rights and the novelty of this factual context, I find that whenever McKithen actually first sustained his injury, the earliest he had reason to know of the constitutional deprivation was on August 27, 2001, the date on which a federal court first announced a federal constitutional right to DNA testing. See Godschalk v. Montgomery County Dist. Att’y’s Office, 177 F.Supp.2d 366, 370 (E.D.Pa.2001) (finding due process right of access to evidence for DNA testing); see also Wade v. Brady, 460 F.Supp.2d 226, 235 n. 17 (D.Mass.2006) (suggesting that this is the appropriate date of accrual for a cause of action for post-conviction access to evidence for DNA testing).

Accordingly, McKithen’s cause of action accrued, at the earliest, on August 21, 2001, and this action is thus not time-barred.

D. Procedural Due Process

1. The Mathews Test

Section 1 of the Fourteenth Amendment to the United States Constitution states, in pertinent part, that a state may not “deprive any person of life, liberty, or property, without due process of law.” U.S. Const, amend. XIV, § 1. In addition to a set of substantive rights, see Section E, infra, the Due Process Clause grants the right that appropriate procedures — “due process” in a literal sense— shall be used to determine whether any deprivation of life, liberty, or property is justified. See Wilkinson v. Austin, 545 U.S. 209, 224, 125 S.Ct. 2384, 162 L.Ed.2d 174 (2005) (noting that the Supreme Court employs a flexible “framework to evaluate the sufficiency of particular procedures” under the Due Process Clause); see also Conn. Bd. of Pardons v. Dumschat, 452 U.S. 458, 463, 101 S.Ct. 2460, 69 L.Ed.2d 158 (1981) (“A state-created right can, in some circumstances, beget yet other rights to procedures essential to the realization of the parent right.”). This procedural com ponent of the Due Process Clause does not guarantee the use of any procedures to govern any decision that does not result in a deprivation of life, liberty, or property. Wilkinson, 545 U.S. at 221, 125 S.Ct. 2384 (“The Fourteenth Amendment’s Due Process Clause protects persons against deprivations of life, liberty, or property; and those who seek to invoke its procedural protection must establish that one of these interests is at stake.”). Therefore, the first step in determining whether or not a government action or policy violates these so-called “procedural due process” 10 rights is determining whether the action or policy deprives a person of life, liberty, or property.

Property is of course a legal creation, and although liberty could be said, like life, to have entirely natural components, the due process clause is concerned only with legal entitlements to life, liberty, and property. See, e.g., Dumschat, 452 U.S. at 465, 101 S.Ct. 2460 (finding an inmate has no legal entitlement to a pardon or commutation); Bd. of Regents of State Colls. v. Roth, 408 U.S. 564, 570-71, 92 S.Ct. 2701, 33 L.Ed.2d 548 (1972) (“Undeniably, the respondent’s re-employment prospects were of major concern to him — concern that we surely cannot say was insignificant. ... But, to determine whether due process requirements apply in the first place, we must look not to the ‘weight’ but to the nature of the interest at stake.”). A conditional entitlement is called an “interest” for the purposes of procedural due process analysis, so that a “liberty interest” in taking some action is a legal entitlement to take that action under certain conditions; a “property interest” in an item is a legal entitlement to exercise property rights (such as use and exclusion) with respect to that item under certain conditions; and a “life interest” is the legal entitlement not to be killed under certain conditions. See, e.g., Wolff v. McDonnell, 418 U.S. 539, 557, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974) (“[T]he State having created the right to good time [credit] and itself recognizing that its deprivation is a sanction authorized for major misconduct, the prisoner’s interest has real substance and is sufficiently embraced within Fourteenth Amendment ‘liberty’ to entitle him to those minimum procedures appropriate under the circumstances and required by the Due Process Clause to insure that the state-created right is not arbitrarily abrogated.”). Procedural due process rights are the rights to appropriate procedures to determine if the individual possessing an interest satisfies the conditions for the entitlement.

A liberty interest is thus a substantive right related to personal freedom to which one has an entitlement on some conditions. For the purposes of the Fourteenth Amendment, a liberty interest can be created by state statute — when a statute provides a conditional entitlement to take certain actions, it creates a liberty interest. See, e.g., id. (holding that the establishment of a good-time credit scheme created a liberty interest); see also Sandin v. Conner, 515 U.S. 472, 483-84, 115 S.Ct. 2293, 132 L.Ed.2d 418 (1995) (finding that the presence of mandatory language in a correctional statute does not create a conditional entitlement if it does not pertain to “atypical and significant hardship in relation to the ordinary incidents of prison life”). A liberty interest can also come directly from the Constitution, as where some constitutional provision guarantees a substantive right, or where the right is implicit in the concept of liberty. Wilkinson, 545 U.S. at 221, 125 S.Ct. 2384 (“A liberty interest may arise from the Constitution itself, by reason of guarantees implicit in the word ‘liberty’.... ”). Regardless of its source, a liberty interest must be a conditional entitlement, not a mere desire or hope. See, e.g., Dumschat, 452 U.S. at 465, 101 S.Ct. 2460 (“In terms of the Due Process Clause, a Connecticut felon’s expectation that a lawfully imposed sentence will be commuted or that he will be pardoned is no more substantial than an inmate’s expectation, for example, that he will not be transferred to another prison; it is simply a unilateral hope.” (footnote omitted)); cf. Roth, 408 U.S. at 577, 92 S.Ct. 2701 (“To have a property interest in a benefit, a person clearly must have more than an abstract need or desire for it. He must have more than a unilateral expectation of it. He must, instead, have a legitimate claim of entitlement to it.”).

If an individual is found to possess a liberty interest infringed by a governmental action, a court must then determine what process that individual is due in determining whether she is entitled to be free of the government’s interference. E.g., Wilkinson, 545 U.S. at 224, 125 S.Ct. 2384 (“A liberty interest having been established, we turn to the question of what process is due an inmate whom Ohio seeks to place in OSP [a ‘supermax’ security prison].”). States have wide latitude to prescribe rules of criminal procedure within the bounds of “the specific guarantees enumerated in the Bill of Rights,” and judicial review under the Due Process Clause is thus highly deferential under Medina v. California, 505 U.S. 437, 443, 112 S.Ct. 2572, 120 L.Ed.2d 353 (1992) (internal quotation marks omitted). However, the framework applicable in most other contexts is not deference but a careful balancing of interests under Mathews v. Eldridge, 424 U.S. 319, 96 S.Ct. 893, 47 L.Ed.2d 18 (1976). See Parham v. J.R., 442 U.S. 584, 599, 99 S.Ct. 2493, 61 L.Ed.2d 101 (1979) (noting that the Mathews test provides “a general approach for testing challenged state procedures under a due process claim”). Under Mathews, once a liberty interest is identified:

[Identification of the specific dictates of due process generally requires consideration of three distinct factors: First, the private interest that will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards; and finally, the Government’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.

Mathews, 424 U.S. at 335, 96 S.Ct. 893.

Despite Brown’s argument that analysis of the procedures governing post-conviction access to evidence for DNA testing is governed by Medina, these procedures are distinct from “ ‘the process afforded during criminal proceedings themselves,’ ” McKithen VII, 481 F.3d at 107 (quoting Krimstock v. Kelly, 464 F.3d 246, 254 (2d Cir.2006)), and thus, as the Second Circuit has specifically stated, “Mathews applies, rather than the more demanding Medina v. California.” McKithen VI, 481 F.3d at 107.

2. The Residual Posh-Conviction Liberty Interest

“The Supreme Court has made clear that prisoners lawfully deprived of their freedom retain substantive liberty interests under the Fourteenth Amendment.” McKithen VII, 481 F.3d at 106; see also Youngberg v. Romeo, 457 U.S. 307, 315, 102 S.Ct. 2452, 73 L.Ed.2d 28 (1982) (“The mere fact that [the plaintiff] has been committed under proper procedures does not deprive him of all substantive liberty interests under the Fourteenth Amendment.”). I am obliged to “consider whether this residual post-conviction liberty interest encompasses an interest in accessing or possessing potentially exonera-tive biological evidence.” McKithen VII, 481 F.3d at 106-07.

Despite the novelty of the technologies at issue from a constitutional perspective, it is important to use traditional methods of legal analysis in determining what substantive rights are protected by the Constitution. Because the Constitution supersedes laws passed by democratic legislatures and signed by elected executive officials, rights set forth in the Constitution limit the ability of the citizens of the United States to make decisions regarding the policies of their government, and they should not lightly be implied. See, e.g., In re Winship, 397 U.S. 358, 384-85, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970) (Black, J., dissenting) (“It can be, and has been, argued that when this Court strikes down a legislative act because it offends the idea of ‘fundamental fairness’ it furthers the basic thrust of our Bill of Rights by protecting individual freedom. But that argument ignores the effect of such decisions on perhaps the most fundamental individual liberty of our people — the right of each man to participate in the self-government of his society.”). However, although constitutional rights represent infringements on the ideal of pure majoritarianism, they are part of our law. As such, these rights must trump raw appeals to majori-tarian values. Cf. Harvey v. Horan (Harvey IV), 285 F.3d 298, 303 (4th Cir.2002) (Wilkinson, J., respecting the denial of rehearing en banc) (“To constitution-alize this area [i.e., post-conviction access to evidence for DNA testing] ... in the face of all this legislative activity and variation is to evince nothing less than a loss of faith in democracy. It is to believe that democratic processes are incapable of rising to the challenge, and that federal courts must do the governing for us.”). As constitutional rights derive their force from being law, they must be identified through traditional methods of legal analysis, taking into account the unusual breadth of the textual guarantees. Thus, I will consider several potential grounds for concluding that prisoners might retain a liberty interest, after conviction, in acquiring physical evidence for the purpose of DNA testing, and assess to what extent each is consistent with the Constitution’s text as interpreted by binding precedent.

i. The Interest of an Innocent Prisoner in Release From Custody

An overriding concern for innocence is a central preoccupation of our constitutional tradition. See, e.g., Winship, 397 U.S. at 363, 90 S.Ct. 1068 (“The standard [of proof beyond a reasonable doubt] provides concrete substance for the presumption of innocence — that bedrock axiomatic and elementary principle whose enforcement lies at the foundation of the administration of our criminal law.” (internal quotation marks omitted)); id. at 372, 90 S.Ct. 1068 (Harlan, J., concurring) (“[I]t is far worse to convict an innocent man than to let a guilty man go free.”); McKithen VII, 481 F.3d at 91-92 (“ ‘[0]ur procedure has been always haunted by the ghost of the innocent man convicted ....’” (quoting United States v. Garsson, 291 F. 646, 649 (S.D.N.Y.1923) (Hand, J.))). Furthermore, “ ‘[f]reedom from bodily restraint has always been at the core of the liberty protected by the Due Process Clause from arbitrary governmental action.’ ” Foucha v. Louisiana, 504 U.S. 71, 80, 112 S.Ct. 1780, 118 L.Ed.2d 437 (1992) (quoting Youngberg, 457 U.S. at 316, 102 S.Ct. 2452). It might then seem natural that inmates retain a liberty interest in being released from custody, conditioned on their innocence.

There is some support for the idea that prisoners possess a liberty interest, cognizable in procedural due process, in undoing their convictions if they are innocent. In recognition of the tremendous importance of innocence, the Supreme Court has stated that “a prisoner retains an overriding ‘interest in obtaining his release from custody if he is innocent of the charge for which he was incarcerated.’ ” Schlup v. Delo, 513 U.S. 298, 321, 115 S.Ct. 851, 130 L.Ed.2d 808 (1995) (quoting Kuhlmann v. Wilson, 477 U.S. 436, 452, 106 S.Ct. 2616, 91 L.Ed.2d 364 (1986) (Powell, J.) (plurality opinion)). On the strength of this, and on the Supreme Court’s other cases finding convicted inmates to retain substantive liberty interests, including freedom from bodily restraint, Judge Nancy Gertner of the United States District Court for the District of Massachusetts has concluded that inmates retain a liberty interest, for procedural due process purposes, in access to biological evidence for post-conviction DNA testing, and several other district courts have found this reasoning persuasive. Wade, 460 F.Supp.2d at 247 & n. 37 (citing, among others, Schlup, Youngberg and Foucha); see also Bryson v. Macy, No. 05-CV-1150 (F), 2007 WL 682030, at *5 (W.D.Okla. Mar.1, 2007) (“The court specifically concurs with the- analysis in Wade as to the viability of a claim based on that right [i.e., the right to post-conviction access to evidence for DNA testing], under due process principles.” (citation omitted)); Breest v. N.H. Att’y Gen., 472 F.Supp.2d 116, 120-21 (D.N.H.2007) (citing Wade and recognizing right of access).

As initially compelling as this theory is, I find it difficult to reconcile with Supreme Court precedent, at least for noncapital defendants. The Supreme Court has never held that inmates convicted after a valid trial have an interest cognizable as a matter of procedural due process in being released from confinement if they are innocent. The remark in Schlup that innocent inmates have an overriding interest in release from custody was in the context of an exception to the rules governing procedurally defaulted petitions for habeas corpus, and did not appear to use the term “interest” in the procedural due process sense of “conditional legal entitlement.” 513 U.S. at 321-22, 115 S.Ct. 851. The liberty interests that the Supreme Court has definitively held convicted inmates to possess either concern the conditions of the inmate’s confinement, see Sandin, 515 U.S. at 483-84, 115 S.Ct. 2293 (recognizing a liberty interest in freedom from restraint which “imposes atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life”); Youngberg, 457 U.S. at 315-16, 102 S.Ct. 2452 (finding that an involuntarily committed detainee had liberty interests in safety and freedom from bodily restraint within a mental institution); Vitek v. Jones, 445 U.S. 480, 491-94, 100 S.Ct. 1254, 63 L.Ed.2d 552 (1980) (finding a liberty interest in avoiding involuntary psychiatric treatment and commitment in a mental institution); or have been created by statute or regulation, see Greenholtz v. Inmates of Neb. Penal & Corr. Complex, 442 U.S. 1, 11-12, 99 S.Ct. 2100, 60 L.Ed.2d 668 (1979) (finding that the use of mandatory language in the Nebraska parole statute conveyed a liberty interest entitled to “some measure of constitutional protection”); Wolff, 418 U.S. at 556-58, 94 S.Ct. 2963 (finding a liberty interest in retention of good-time credits under a mandatory-state system). 11

The Supreme Court has thus never held that for procedural due process purposes, an inmate has a liberty interest in overcoming the fact of confinement pursuant to a valid conviction. Indeed, in a line of cases surprisingly unmentioned by Brown, it has said almost exactly the opposite. In Greenholtz, the Court stated, “There is no constitutional or inherent right of a convicted person to be conditionally released before the expiration of a valid sentence. The natural desire of an individual to be released is indistinguishable from the initial resistance to being confined. But the conviction, with all its procedural safeguards, has extinguished that liberty right: Given a valid conviction, the criminal defendant has been constitutionally deprived of his liberty.” 442 U.S. at 7, 99 S.Ct. 2100 (internal brackets and quotation marks omitted). This statement formed the basis for the Court’s holding that there is no liberty interest in being released on parole absent a state-created entitlement conferred by the particular statute. Id. at 11, 99 S.Ct. 2100. 12 This holding was reaffirmed and extended to commutations in Connecticut Board of Pardons v. Dumschat. 452 U.S. at 463-64, 101 S.Ct. 2460.

The view that inmates’ liberty interest in release is “extinguished” by a valid conviction was most recently restated, with slight modifications, by two separate plurality opinions representing a total of seven justices in Ohio Adult Parole Authority v. Woodard, 523 U.S. 272, 118 S.Ct. 1244, 140 L.Ed.2d 387 (1998). Chief Justice Rehnquist, writing for himself and three other justices, extended the holding that an “individual’s interest in release or commutation is indistinguishable from the initial resistance to being confined, and that interest has already been extinguished by the conviction and sentence,” id. at 280, 118 S.Ct. 1244 (Rehnquist, C.J.) (plurality opinion) (internal quotation marks omitted), to apply not only to the liberty interest but also to the life interest of a capital prisoner seeking clemency, 13 id. at 281, 118 S.Ct. 1244 (Rehnquist, C.J.). 14 Writing for herself and three other justices, Justice O’Connor agreed that “[w]hen a person has been fairly convicted and sentenced, his liberty interest, in being free from such confinement, has been extinguished,” id. at 289, 118 S.Ct. 1244 (O’Connor, J.) (plurality opinion), but found that a prisoner’s life interest was subject to “some minimal procedural safeguards,” justifying judicial intervention in extreme cases, such as “in the face of a scheme whereby a state official flipped a coin to determine whether to grant clemency, or in a case where the State arbitrarily denied a prisoner any access to its clemency process,” id. at 289, 118 S.Ct. 1244 (O’Connor, J.). 15 As Justice Stevens also found that a capital prisoner has a life interest deserving of procedural due process protection, id. at 292, 118 S.Ct. 1244 (Stevens, J., concurring in part and dissenting in part), a majority of the Court held that capital defendants do retain a life interest sufficient to impose at least minimal safeguards on clemency proceedings. 16 However, the liberty interests of noncapital defendants appear not to include an interest in release before the expiration of a valid sentence. 17

The cases concerning an inmate’s interest in clemency, however, do not distinguish between clemency based on mercy and clemency based on doubts regarding culpability. See, e.g., Woodard, 523 U.S. at 284, 118 S.Ct. 1244 (Rehnquist, C.J.) (“Clemency proceedings are not part of the trial — or even of the adjudicatory process. They do not determine the guilt or innocence of the defendant, and are not intended primarily to enhance the reliability of the trial process.”). They do not, therefore, explicitly address the possibility that an inmate might have a liberty interest in release on the condition that she is innocent. However, to say that the Con stitution gives an inmate a liberty interest in release if she is innocent, for the purposes of procedural due process, is to say that an inmate who is innocent is entitled, by the Constitution, to be released. This in turn is to say that an inmate who is innocent is in “custody in violation of the Constitution” within the meaning of 28 U.S.C. § 2254. That possibility — that a petition for a writ of habeas corpus can be brought merely on the basis of a freestanding claim of innocence — was conspicuously left unresolved in Herrera v. Collins, 506 U.S. 390, 113 S.Ct. 853, 122 L.Ed.2d 203 (1993), where the Supreme Court assumed “for the sake of argument” that “in a capital case a truly persuasive demonstration of ‘actual innocence’ made after trial would render the execution of a defendant unconstitutional.” Id. at 417, 113 S.Ct. 853; see also id. at 427, 113 S.Ct. 853 (O’Connor, J., concurring) (“Nowhere does the Court state that the Constitution permits the execution of an actually innocent person.”); House v. Bell, 547 U.S. 518, 555, 126 S.Ct. 2064, 165 L.Ed.2d 1 (2006) (once more declining to resolve the question).

Though I must determine the contours of the right, if it exists, to post-conviction access to evidence for DNA testing, I find it unnecessary to resolve the difficult question whether actual innocence is a freestanding ground for habeas relief. The Supreme Court in Herrera made clear that any claim of actual innocence would require an “extraordinarily high” threshold showing of innocence. 506 U.S. at 417, 113 S.Ct. 853. I conclude in Section D.4, infra, that a prisoner’s liberty interest in meaningful access to existing clemency mechanisms requires the disclosure of evidence that has significantly less probative value than what would be needed to meet an “extraordinarily high” threshold. Therefore, the existence vel non of a constitutional right to release based on an extraordinarily high threshold showing of innocence does not affect the amount of evidence to which inmates are entitled; any disclosure that would be required to vindicate a liberty interest in release based on innocence would already be required to vindicate the liberty interest in meaningful access to existing clemency mechanisms. 18

In sum, although prisoners retain liberty interests regarding the conditions of their confinement, and although capital prisoners retain some degree of life interests in receiving clemency, prisoners do not have liberty interests in release from custody before the end of valid sentences. I need not decide whether prisoners have liberty interests, for the purposes of procedural due process, in release from custody on the condition of their innocence. 19

ii. The Interest in Meaningful Access to Existing Clemency Mechanisms

As discussed in Section D.2.i, supra, noncapital prisoners have no liberty inter est in receiving clemency in the form of a pardon, commutation or parole. However, in his concurrence in the Fourth Circuit’s denial of rehearing en banc in Harvey IV, which concerned post-conviction access to evidence for DNA testing, Judge J. Michael Luttig suggested that prisoners retain a “residual, substantive liberty interest in meaningful access to existing executive mechanisms of clemency.” 285 F.3d at 314 (Luttig, J., respecting the denial of rehearing en banc); see also Wade, 460 F.Supp.2d at 247 n. 37 (same, citing Harvey IV). That is, although a prisoner has no conditional entitlement to receive any form of clemency, she retains an entitlement to take actions seeking clemency through existing clemency mechanisms. This entitlement is conditional; she is not entitled to take absolutely any action in pursuit of clemency. Instead, she is entitled to take actions seeking clemency on the condition that her liberty to take those actions is necessary for her to enjoy meaningful access to the clemency mechanisms. I agree with Judge Luttig and hold that prisoners possess such a liberty interest.

a. The Existence of an Interest in Meaningful Access to Existing Clemency Mechanisms

The existence of a liberty interest in access to existing clemency mechanisms is clear, and it appears to encompass meaningful, not merely formal, access. The First Amendment guarantees the right to “petition the Government for a redress of grievances.” U.S. Const. amend. I. This right “has been recognized as one of the most precious of the liberties safeguarded by the Bill of Rights,” City of N.Y. v. Beretta U.S.A. Corp., 524 F.3d 384, 397 (2d Cir.2008) (internal quotation marks omitted), and survives incarceration, see Turner v. Safley, 482 U.S. 78, 84, 107 S.Ct. 2254, 96 L.Ed.2d 64 (1987) (“[prisoners retain the constitutional right to petition the government for the redress of grievances.”).

Among several other clauses, the Petition Clause has been seen as the basis for the constitutional right of access to the courts. See, e.g., Bill Johnson’s Restaurants, Inc. v. NLRB, 461 U.S. 731, 741, 103 S.Ct. 2161, 76 L.Ed.2d 277 (1983) (“[T]he right of access to the courts is an aspect of the First Amendment right to petition.”); Beretta, 524 F.3d at 397-98 (finding Petition Clause to undergird right of access to the courts); cf. Christopher v. Harbury, 536 U.S. 403, 415 n. 12, 122 S.Ct. 2179, 153 L.Ed.2d 413 (2002) (noting several other constitutional provisions that have been found by the Supreme Court to undergird the right of access to the courts, citing cases).

The right to petition guaranteed by the First (and, as against the states, the Fourteenth) Amendment “applies with equal force to a person’s right to seek redress from all branches of government.” Franco v. Kelly, 854 F.2d 584, 589 (2d Cir.1988). The Second Circuit has found a “clear relationship between the right of access to the courts and the right to petition for redress of grievances.” Id. As Franco concerned a § 1983 claim of retaliation for pursuing a prison grievance, the Second Circuit had no occasion to decide whether the right to petition other branches of government shares the same contours as the right of access to the courts. However, it strongly suggested that the two rights were closely analogous. In light of the “clear relationship” it found between the two rights, Franco, 854 F.2d at 589, the Second Circuit found the case “controlled,” id., by Morello v. James, 810 F.2d 344 (2d Cir.1987), an earlier case regarding the right of access to the courts where a prisoner had complained not of retaliation for accessing the courts but of confiscation of his legal papers, Morello, 810 F.2d at 345. Indeed, the Second Circuit held that “Franco should not be entitled to any less relief under section 1983 because he was addressing his complaints to a state administrative agency rather than to a court of law.” Franco, 854 F.2d at 589-90.

Both Franco and Morello, in turn, defined the right of access to the courts by citing generally to the canonical court access case of Bounds v. Smith, 430 U.S. 817, 97 S.Ct. 1491, 52 L.Ed.2d 72 (1977), overruled in part by Lewis v. Casey, 518 U.S. 343, 116 S.Ct. 2174, 135 L.Ed.2d 606 (1996), see Franco, 854 F.2d at 588 (citing Bounds); Morello, 810 F.2d at 346-47 (same), and Morello specifically cited Bounds for the proposition that the right of access to the courts imposes affirmative obligations on the government, Morello, 810 F.2d at 347. 20 Bounds, in turn, held that prisoners’ access to the courts must be “adequate, effective and meaningful,” and required the provision of some legal assistance to inmates. Bounds, 430 U.S. at 822, 97 S.Ct. 1491. Bounds reached this conclusion by drawing on a number of disparate cases involving court access, which had proceeded on differing legal theories, and considering them all to be cases involving the right of access to the courts. Id. at 821-25, 97 S.Ct. 1491. Bounds itself considered the right of access to the courts simply under the Fourteenth Amendment, without specifying which clause. Id. at 818 & n. 1.

I take the Second Circuit’s description in Franco of the relationship between the two rights, its view that Morello controlled the decision in Franco, and both cases’ reli-anee on Bounds to define the right of court access to indicate that the right to petition the executive branch has contours that are analogous to the right of access to the courts. The Supreme Court has, albeit in general language, similarly suggested a fairly tight relationship between the contours of the right of access to the courts and those of the right to petition the other branches. Cal. Motor Transp. Co. v. Trucking Unlimited, 404 U.S. 508, 510, 92 S.Ct. 609, 30 L.Ed.2d 642 (1972) (“The same philosophy governs the approach of citizens or groups of them to administrative agencies (which are both creatures of the legislature, and arms of the executive) and to courts, the third branch of government. Certainly the right to petition extends to all departments of the government. The right of access to the courts is indeed but one aspect of the right to petition.”).

Given the wide variety of contexts in which one might petition the government, it might at first seem incongruous to find an analogy between the contours of a right to access court proceedings and a right to petition other branches. However, Bounds itself suggests a way in which the “clear relationship” seen by the Second Circuit in Franco can cause court access cases to “control[]” cases involving petitions addressed to other branches of government. Bounds recognized the unifying feature among various court access cases was the requirement that the access be “adequate, effective, and meaningful,” 430 U.S. at 822, 97 S.Ct. 1491. I conclude that similarly, access to existing mechanisms for relief in other branches of government must also be “adequate, effective, and meaningful,” though the wide variation be tween such mechanisms means that what constitutes meaningful access can vary significantly. I thus take the cases regarding meaningful access to the courts presumptively to provide guidance regarding what constitutes meaningful access, but only insofar as they address analogous situations.

It is a difficult question what constitutes meaningful access to a discretionary system such as an executive clemency system. I begin by noting acknowledging again that prisoners possess no liberty interest in receiving clemency. See Section D.2.i, supra. However, whether access to courts is “meaningful” turns in large part on the prospects that a prisoner with a certain degree of access to existing mechanisms and with a meritorious claim can obtain relief. See, e.g., Bounds, 430 U.S. at 826, 97 S.Ct. 1491 (noting, in finding that access to a law library is part of meaningful access to the courts, that “[e]ven the most dedicated trial judges are bound to overlook meritorious cases without the benefit of an adversary presentation”); cf. Lewis, 518 U.S. at 356, 116 S.Ct. 2174 (“When any inmate, even an illiterate or non-English-speaking inmate, shows that an actionable claim of this nature lie., challenging his conviction, sentence, or conditions of confinement] which he desired to bring has been lost or rejected, or that the presentation of such a claim is currently being prevented, because this capability of filing suit has not been provided, he demonstrates that the State has failed to furnish adequate law libraries or adequate assistance from persons trained in the law.” (emphasis omitted)). Similarly, I take the question whether access to existing clemency mechanisms is “meaningful” to depend in part on the prospects that a prisoner with that level of access has of obtaining clemency. 21

b. The Existence of Affirmative Obligations to Render Access Meaningful

The right of meaningful access to the courts imposes affirmative obligations on the government. Bounds, 430 U.S. at 824, 97 S.Ct. 1491 (“[0]ur decisions have consistently required States to shoulder affirmative obligations to assure all prisoners meaningful access to the courts.”). Prison officials must provide writing and mailing materials, along with notarial materials and an adequate law library. Id. at 824-25, 97 S.Ct. 1491. Though the Supreme Court has made clear that the state need not “enable the prisoner to discover grievances, [or] to litigate effectively once in court,” Lewis, 518 U.S. at 354, 116 S.Ct. 2174 (emphasis omitted), it has required state prisons to enable prisoners to effectively challenge the fact or conditions of their confinement, id. at 355, 116 S.Ct. 2174. Furthermore, “the State is obligated to furnish prisoners not otherwise able to obtain it, with a transcript or equivalent recordation of prior habeas corpus hearings for use in further proceedings,” Johnson v. Avery, 393 U.S. 483, 486, 89 S.Ct. 747, 21 L.Ed.2d 718 (1969), demonstrating that the state’s duties extend to providing at least some inmate-specific material, not merely general supplies. See generally United States v. Sliker, 751 F.2d 477, 489-92 (2d Cir.1985) (tracing Supreme Court jurisprudence on provision of transcripts). 22

However, it is not clear that the right of meaningful access to clemency proceedings imposes affirmative obligations on the government. I am unaware of any binding authority holding that the Petition Clause imposes affirmative obligations on the government as opposed to merely preventing impediments to, or retaliation for, an individual’s exercise of her right to petition. See, e.g., Smith v. Ark. State Highway Emp., Local 1315, 441 U.S. 463, 464-65, 99 S.Ct. 1826, 60 L.Ed.2d 360 (1979) (“The First Amendment right to associate and advocate provides no guarantee that a speech will persuade or that advocacy will be effective. The public employee surely can associate and speak freely and openly, and he is protected by the First Amendment from retaliation for doing so. But the First Amendment does not impose any affirmative obligation on the government to listen, to respond or, in this context, to recognize the association and bargain with it.” (citations and quotation marks omitted)). Further, given the wide variety of contexts in which an individual can petition the government for redress of grievances, and the wide range of mechanisms available to hear such petitions, 23 it seems that, as a general matter, mere noninterference constitutes meaningful access to any mechanism set up to hear a petition other than a court.

However, even assuming that meaningful access to executive clemency mechanisms generally consists of mere noninterference, the prosecutorial duty to seek justice, recognized in Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), and subsequent cases, provides a limited exception to this general rule. As the Supreme Court made clear in Brady and its progeny, prosecutors have a special obligation to seek justice with respect to their defendants, and this obligation mandates the disclosure of material favorable evidence for the defendant’s use at trial. 24 This duty to disclose material favorable evidence, though it continues after a jury verdict, see Leka v. Portuondo, 257 F.3d 89, 100 (2d Cir.2001) {“Brady requires disclosure of information that the prosecution acquires during the trial itself, or even afterward.”), implicates the Due Process Clause only insofar as it affects the fairness of a defendant’s trial, see United States v. Ruiz, 536 U.S. 622, 629, 122 S.Ct. 2450, 153 L.Ed.2d 586 (2002) (characterizing Brady right as a trial right waived by guilty plea); see also United States v. Bagley, 473 U.S. 667, 675-76, 105 S.Ct. 3375, 87 L.Ed.2d 481 (1985) (finding that nondisclosure does not violate “ ‘the prosecutor’s constitutional duty’ ” unless “ ‘the omission deprived the defendant of a fair trial’ ” (quoting United States v. Agurs, 427 U.S. 97, 108, 96 S.Ct. 2392, 49 L.Ed.2d 342 (1976))). Therefore, I agree with the courts that have held Brady inapplicable after judicial proceedings, at least to the extent that the defendant’s right to disclosure of all material favorable evidence does not persist after judicial proceedings have run their course. See, e.g., Gibson v. Sup’t of N.J. Dep’t of Law & Pub. Safety — Div. Of State Police, 411 F.3d 427, 444 (3d Cir.2005) (“[The plaintiff] has pointed to no constitutional duty to disclose potentially exculpatory evidence to a convicted criminal after the criminal proceedings have concluded and we decline to conclude that such a duty exists.”); Warney v. City of Rochester, 536 F.Supp.2d 285, 295 (2008) (finding that Brady does not apply after “the conclusion of trial and/or post-trial procedures,” but concluding that the principles underlying Brady continue to have force); see also Harvey v. Horan (Harvey III), 278 F.3d 370, 378-79 (4th Cir.2002) (rejecting claim that Brady allows post-conviction DNA testing when the defendant was able to test the evidence at trial “using the best technology available at the time”); Grayson v. King, 460 F.3d 1328, 1339-40 (11th Cir.2006) (similar), ce rt. denied, — U.S. -, 127 S.Ct. 1005, 166 L.Ed.2d 712 (2007); cf. Wade v. Brady, 460 F.Supp.2d at 243 (finding post-conviction access to DNA testing to be guaranteed by a due process right “analogous to Brady, not a literal application of its pre-trial guarantee”). 25

However, even if the failure to disclose all material favorable evidence discovered after the conclusion of judicial proceedings does not deprive a defendant of due process, the prosecutor’s duty to seek justice survives in some form. The Court has recognized that “the prosecutor’s role transcends that of an adversary,” Bagley, 473 U.S. at 675 n. 6, 105 S.Ct. 3375, and that the prosecutor is “ ‘the representative not of an ordinary party to a controversy, but of a sovereignty ... whose interest ... in a criminal prosecution is not that it shall win a case, but that justice shall be done,’ ” id. (quoting Berger v. United States, 295 U.S. 78, 88, 55 S.Ct. 629, 79 L.Ed. 1314 (1935)). Making clear that “justice” in this context means not merely the absence of unfair advantages, but also a substantively just outcome, the Supreme Court has observed that a prosecutor is “ ‘the servant of the law, the twofold aim of which is that guilt shall not escape or innocence suffer.’ ” Agurs, 427 U.S. at 111, 96 S.Ct. 2392 (emphasis added) (quoting Berger, 295 U.S. at 88, 55 S.Ct. 629); see also Strickler v. Greene, 527 U.S. 263, 281, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999) (noting that Brady and its progeny “illustrate the special role played by the American prosecutor in the search for truth in criminal trials” (emphasis added)); California v. Trombetta, 467 U.S. 479, 486, 104 S.Ct. 2528, 81 L.Ed.2d 413 (1984) (noting that the purpose of Brady and its progeny is to deliver “exculpatory evidence into the hands of the accused, thereby protecting the innocent from erroneous conviction and ensuring the integrity of our criminal justice system” (emphasis added)). Most tellingly, the Supreme Court has noted that “after a conviction the prosecutor also is bound by the ethics of his office to inform the appropriate authority of after-acquired or other information that casts doubt upon the correctness of the conviction.” Imbler v. Pachtman, 424 U.S. 409, 427 n. 25, 96 S.Ct. 984, 47 L.Ed.2d 128 (1976). 26 Of course, after the conclusion of judicial proceedings, the obligation is not “enforced by the requirements of due process,” id., in the sense that there is no more process for the prosecutor’s omissions to taint, but this does not mean that the obligation does not exist.

At least in certain circumstances, it would be inconsistent with this duty to seek justice for a prosecutor intentionally to thwart a prisoner’s attempt to obtain evidence of innocence for use in clemency proceedings. If potential evidence of innocence of sufficient probative value is available at sufficiently low cost, a prosecutor cannot intentionally prevent its discovery and still remain “the servant of the law, the twofold aim of which is that guilt shall not escape or innocence suffer.” Agurs, 427 U.S. at 111, 96 S.Ct. 2892 (internal quotation marks omitted). I do not presume that the Supreme Court used this language lightly or unseriously, and prosecutors are obligated to strive to comply with it.

The Brady cases are replete with references to the prosecutor’s duty to seek justice, and those statements are not mere verbal flourishes. In Agurs, the Supreme Court found that the prosecutor’s duty to do justice required courts to be more ready to reverse a conviction based on after-discovered evidence that a prosecutor had withheld than based on after-discovered evidence simply discovered from a neutral source. 427 U.S. at 111, 96 S.Ct. 2392 (“If the standard applied to the usual motion for a new trial based on newly discovered evidence were the same when the evidence was in the State’s possession as when it was found in a neutral source, there would be no special significance to the prosecutor’s obligation to serve the cause of justice.”). That is, the duty to seek justice was a central part of the reasoning supporting the Court’s articulation of the standard for reversing a conviction based on a Brady violation. I am thus bound to conclude that after a conviction becomes final the prosecutor’s duty to seek justice, though no longer enforceable through the due process right to a fair trial, does not entirely evaporate. It is true that after conviction, a defendant is legally presumed guilty, not legally presumed innocent, and accordingly retains fewer rights. See Herrera, 506 U.S. at 399-400, 113 S.Ct. 853 (“[I]n the eyes of the law, petitioner does not come before the Court as one who is ‘innocent,’ but, on the contrary, as one who has been convicted by due process of law of two brutal murders.”). However, the prosecutor’s duty is to seek justice, not to seek what is legally presumed to be justice.

Other judges have recognized that the logic behind Brady still applies to some extent after conviction. See, e.g., Harvey IV, 285 F.3d at 316-17 (Luttig, J.) (arguing that the principles of fairness underlying Brady continue after conviction); Wade, 460 F.Supp.2d at 247 (“[T]he same interest that motivated Brady — preventing wrongful conviction of the innocent — -still applies in the post-conviction setting....”). However, recognizing that the duty to disclose all material favorable evidence without a specific request presumably does not continue indefinitely after trial, they have struggled to articulate the extent to which the disclosure obligations are diminished. See Harvey IV, 283 F.3d at 317 (Luttig, J.) (acknowledging that full Brady disclosure is unwarranted after conviction but arguing that the principle of “elemental fairness” requires disclosure of evidence which could prove the defendant’s innocence beyond any doubt). In my view, the answer is that in all circumstances, a prosecutor violates her duty to seek justice when she fails to provide the defendant with evidence that, if possessed by the defendant and erroneously excluded from pending or possible proceedings, would deprive him of a constitutional right relating to his guilt or innocence.

Therefore, at and before trial the prosecutor’s duty is simply to disclose evidence which, if it had erroneously been excluded from the proceedings, would deprive the defendant of a fair trial. After trial, the prosecutor’s duty is to disclose only such evidence which, if the defendant had access to it but was erroneously prevented from using it, would deprive the defendant of either of his remaining rights relating to guilt or innocence. One is meaningful access to the courts at the level of post-conviction proceedings, which requires far less access than does the first appeal as of right. Compare Douglas v. California, 372 U.S. 353, 355, 83 S.Ct. 814, 9 L.Ed.2d 811 (1963) (finding a right to counsel on appeal), with Pennsylvania v. Finley, 481 U.S. 551, 555, 107 S.Ct. 1990, 95 L.Ed.2d 539 (1987) (finding no right to counsel for post-conviction proceedings). 27 The other is meaningful access to clemency proceedings. Violations of the prosecutor’s duty are enforced, in each case, by the right that erroneous deprivation would violate. One caveat is that the prosecutor’s attention must be drawn to the evidence for this duty to attach; at trial, the proceedings themselves draw the prosecutor’s attention, but in the case of DNA evidence, a specific request would be needed. 28

This analysis flows from the Supreme Court’s reasoning in Agurs and Bagley, which defined the evidence that must be produced as that evidence the withholding of which deprived the defendant of a fair trial. Bagley, 473 U.S. at 675-76, 105 S.Ct. 3375 (“ ‘[U]nless the omission deprived the defendant of a fair trial, there was no constitutional violation re quiring that the verdict be set aside; and absent a constitutional violation, there was .no breach of the prosecutor’s constitutional duty~to~d!sa0serrvH3^he-prosecutor will not have violated his constitutional duty of disclosure unless his omission is of sufficient significance to result in the denial of the defendant’s right to a fair trial.’ ” (quoting Agurs, 427 U.S. at 108, 96 S.Ct. 2392)); see also United States v. Coppa, 267 F.3d 132, 142 (2d Cir.2001) (“Bagley makes the extent of the disclosure required by Brady dependent on the anticipated remedy for violation of the obligation to disclose .... ”). It is no coincidence that a prosecutor’s duty to disclose is limited to that evidence of sufficient materiality that the defendant would be deprived of a fair trial if she possessed it but defense counsel unprofessionally failed to present it, see Strickland v. Washington, 466 U.S. 668, 694, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984) (setting out the materiality standard used upon a finding that counsel committed an unprofessional error), or the trial court erroneously excluded it, see, e.g., Wade v. Mantello, 333 F.3d 51, 59 (2003) (describing the materiality standard used upon a finding that a state court erroneously excluded evidence). 29 Taken together, these cases all suggest that the prosecutor’s duty to seek justice obligates him to disclose evidence which is sufficiently probative that, if the defendant possessed it, the erroneous deprivation of its use would deprive the defendant of a fair trial. 30

Given that the prosecutor’s duty to seek justice requires disclosure of evidence the erroneous deprivation of which would deprive a defendant of a fair trial, and that the prosecutor’s duty to seek justice survives in some form after conviction, I find that this duty requires disclosure of evidence the erroneous deprivation of which would deprive a prisoner of meaningful access to courts or clemency proceedings. The fact that the rights as to which this duty applies after conviction are far less generous than the fair trial right accounts for the intuition, shared by other courts, that post-conviction Brady duties are drastically diminished, and provides guidance as to how and why the duties are diminished. It also harmonizes with Imbler’s statement that the prosecutor’s obligation, though not enforced by due process in the absence of a trial, continues in some form. It is true that the Court described this duty as merely an ethical obligation and did not indicate that it was enforceable by the right of access to post-conviction proceedings or to clemency proceedings, but Imbler was decided long before the advent of forensic DNA analysis, which, for the reasons described in Section D.2.f, infra, may be the only type of evidence the absence of which could deny meaningful access to such proceedings. It is thus not surprising that the Court in Imbler did not anticipate that post-conviction withholding of evidence could violate constitutional rights relating to guilt or innocence.

Thus, in my view, when evidence of innocence is so probative that, if it were in the defendant’s possession, erroneous interference with the defendant’s use of the evidence would deprive the defendant of a constitutional right relating to guilt or innocence, the prosecutor’s duty to seek justice imposes an affirmative obligation to provide a defendant with that evidence. Accordingly, even if the Petition Clause generally imposes no affirmative obligations on the government, the duty to seek justice imposes an obligation on the prosecutor to provide evidence of such significance that, if a prisoner had it and was erroneously deprived of its use, the prisoner would be deprived of his right to meaningful access to existing executive clemency mechanisms.

c. Whether the Right of Access Depends on the Access of Others

The cases involving affirmative obligations to provide meaningful access to the courts all involve the provision of resources to those who cannot attain the same level of access as other litigants. Many of the early cases, in particular, make clear their concern with equality of access. See, e.g., Griffin v. Illinois, 351 U.S. 12, 16-18, 76 S.Ct. 585, 100 L.Ed. 891 (1956) (finding denial of free transcripts to the indigent to be invidious discrimination in light of the greater access enjoyed by the affluent); Long v. Dist. Court, 385 U.S. 192, 194, 87 S.Ct. 362, 17 L.Ed.2d 290 (1966) (similar). It is possible to conclude simply that the baseline level of meaningful access is simply whatever access others enjoy, and thus to avoid the difficult task of determining what level of access is meaningful. This would effectively defeat any claim that meaningful access to clemency proceedings could require the state to provide evidence of innocence, because others may not have evidence of innocence to present. However, the Supreme Court has not taken this course, instead focusing on the adequacy of the litigant’s ability to make an effective claim, not merely on the comparison between the litigant’s access and the access of others.

In Evitts v. Lucey, 469 U.S. 387, 105 S.Ct. 830, 83 L.Ed.2d 821 (1984), the Supreme Court held that the Due Process Clause alone, absent any equal protection principles, guaranteed the right to effective assistance of appellate counsel. Evitts, 469 U.S. at 391-92, 105 S.Ct. 830. Crucially, the Supreme Court found that the aspect of the Due Process Clause that extended this guarantee was the line of cases guaranteeing access to the courts, beginning with the first of the cases requiring the provision of transcripts. Id. at 392-93, 105 S.Ct. 830 (noting that safeguards such as transcripts or provision of appellate counsel have been found necessary to make appeals “adequate and effective” (quoting Griffin, 351 U.S. at 20, 76 S.Ct. 585, citing Douglas, 372 U.S. 353, 83 S.Ct. 814)). 31 The Court in Evitts rejected the notion that these court access cases rested solely on equal protection concerns and conveyed no substantive rights beyond the right to access enjoyed by others. 469 U.S. at 403-05, 105 S.Ct. 830. 32

More recently, the Supreme Court has adhered to the methodology of assessing meaningful access to the courts not by reference to the access enjoyed by others but instead to a baseline level of capability. In Lewis, the Supreme Court made clear that the affirmative obligation recognized in Bounds to provide meaningful access to the courts was not defined by reference to the level of access others had, Lewis, 518 U.S. at 355, 116 S.Ct. 2174 (noting the fact that incarceration permissibly reduces inmates’ litigating capacity on certain subjects compared to the non-incarcerated); but instead by whether an individual’s claim was “hindered,” id. at 351, 116 S.Ct. 2174, “frustrated,” or “impeded” in some way, id. at 353, 116 S.Ct. 2174. Though with a slightly different emphasis — that is, noting that a meaningful level of access may be lower than the level enjoyed by others — Lewis, like Evitts, indicates that meaningful access to the courts does not require analysis of whether others enjoy similar access, but is determined with reference to an independent baseline.

A separate line of cases involving active obstruction of the right of access to the courts further supports the conclusion that the level of access enjoyed by others does not determine what level of access is meaningful. While government officials are not required by the First Amendment affirmatively to come forward with information that could allow private parties to bring a lawsuit, or to provide pertinent information in response to a generalized request, Whalen v. County of Fulton, 126 F.3d 400, 407 (2d Cir.1997) (finding that an adoption agency did not interfere with the right of access to the courts when it did not inform plaintiffs that an infant they had expressed interest in adopting was placed in foster care), they may not “obstruct legitimate efforts to seek judicial redress,” id. at 406 (citing Barrett v. United States, 798 F.2d 565, 575 (1986)). Such obstruction occurs when government employees intentionally destroy or conceal evidence that may allow plaintiffs to bring a claim. See Whalen, 126 F.3d at 406 (citing Barrett and Bell v. City of Milwaukee, 746 F.2d 1205, 1261 (7th Cir.1984), overruled on other grounds by Russ v. Watts, 414 F.3d 783, 791 (7th Cir.2005), both of which involved government employees engaging in cover-ups that deprived plaintiffs of facts necessary to pursue legal claims); see also Christopher, 536 U.S. at 413-14 & n. 9, 122 S.Ct. 2179 (describing this type of action and assuming without deciding the correctness of circuit court cases adjudicating these claims).

While the Second Circuit has often focused upon whether the obstructive conduct was contrary to law, Barrett, 798 F.2d at 575-76, or regulation, Whalen, 126 F.3d at 407, it has also used language suggesting that the inquiry revolves not around whether the obstructive actions are formally prohibited but rather whether they violate “basic principles.” Barrett, 798 F.2d at 575 (“Unconstitutional deprivation of a cause of action occurs when government officials thwart vindication of a claim by violating basic principles that enable civil claimants to assert their rights effectively.” (citing Goldberg v. Kelly, 397 U.S. 254, 267-71, 90 S.Ct. 1011, 25 L.Ed.2d 287 (1970), and Greene v. McElroy, 360 U.S. 474, 496-97, 79 S.Ct. 1400, 3 L.Ed.2d 1377 (1959))); accord Whalen, 126 F.3d at 406-07 (quoting Barrett, 798 F.2d at 575); see also Whalen, 126 F.3d at 406 (“ ‘[T]he Constitution protects causes of action from arbitrary interference’ by government officials.” (quoting Barrett, 798 F.2d at 575)).

These cover-up cases support the proposition that the level of access which counts as meaningful is not determined by reference to the degree of access others have. Some members of the public are aware of facts giving rise to legal claims, while some are unaware of facts which would if known give rise to claims. To whatever extent the knowledge of facts giving rise to a claim affects one’s level of access to the courts, two individuals ignorant of facts giving rise to a claim have the same level of access, even if one is ignorant by chance and the other is ignorant due to a coverup. If the baseline level of meaningful access to the courts was simply the level of access enjoyed by others, the cover-up victim would not have a cause of action, as she has no less access than the accidentally ignorant individual. However, under Second Circuit law, a governmental coverup can in fact deprive an individual of meaningful access to the courts. That is, these cover-up cases indicate that the baseline level of meaningful access is not the level of access enjoyed by others, but rather the access an individual would enjoy if government officials performed their duties.

This method of determining the baseline level of meaningful access, in conjunction with the Second Circuit’s statement that the relevant duties may be as abstract as “basic principles,” suggests that meaning ful access may require access to evidence of innocence in at least some circumstances even though the general public is not in a similar position. Denial of evidence of innocence may not be contrary to any criminal law or formal regulation, or, in the absence of a judicial proceeding, the “ ‘basic principles that enable civil claimants to assert their rights effectively.’ ” Whalen, 126 F.3d at 406-07 (quoting Barrett, 798 F.2d at 575). However, intentionally preventing a prisoner from discovering what may be highly probative evidence of innocence, even if its only use would be to support a plea for clemency, does conflict with the “basic principles” undergird-ing the prosecutor’s role in the criminal justice system. 33 As discussed in Section D.2.ii.b, supra, the prosecutor’s obligation to seek justice persists after a conviction. To the extent that the Second Circuit recognizes violation of anything short of a formal regulation as impeding the right of access to the courts, a violation of the Brady duty to seek justice — certainly a basic pillar of our criminal justice system, ef. Section E.4 — would seem to qualify.

Therefore, while a number of cases involving meaningful access to the courts have expressed equal protection con cerns — i.e., the need to ensure access similar to that enjoyed by others — meaningful access is not defined solely by reference to the access of others, and the concealment of evidence in violation of a duty seems in particular to obstruct meaningful access to courts. The fact that other members of the public do not have access to evidence of innocence thus does not mean that a prosecutor’s decision to withhold it from a convicted defendant cannot deny the defendant meaningful access to executive clemency mechanisms. I turn, therefore, to whether the nature of the clemency mechanisms themselves defeats the analogy to the right of meaningful access to the courts.

d. The Significance of the Discretionary Nature of Clemency Proceedings

The discretionary nature of clemency proceedings creates a significant difference between the rights prisoners retain to access the courts and the rights to access existing clemency mechanisms. Deprivation of evidence is far more significant in judicial proceedings than clemency proceedings. A grant of clemency is “ ‘an act of grace,’ ” Herrera, 506 U.S. at 413, 113 S.Ct. 853 (quoting United States v. Wilson, 32 U.S. (7 Pet.) 150, 160, 8 L.Ed. 640 (1833) (Marshall, C.J.)), and “depends not simply on objective factfinding, but also on purely subjective evaluations and on predictions of future behavior by those entrusted with the decision,” Dumschat, 452 U.S. at 464, 101 S.Ct. 2460. A particular fact can have tremendous legal significance in judicial proceedings; it is often the difference between stating a claim for relief and not. However, any one piece of evidence or information has a far different role in a clemency proceeding. Unlike a court, the executive may grant clemency without learning of a favorable fact, or deny clemency even if she learns that fact. Thus, a deprivation of information is generally far less likely to interfere with meaningful access to existing clemency mechanisms than with meaningful access to the courts. 34

However, the right of meaningful access to the courts can also be impaired even in circumstances where the proceeding has significant discretionary components. The Supreme Court has held that the right of meaningful access to courts does not confer a right to counsel for a traditional discretionary appeal, Ross v. Moffitt, 417 U.S. 600, 94 S.Ct. 2437, 41 L.Ed.2d 341 (1974), but it did so in rebanee on the fact that the prisoner seeking such an appeal had access to the appellate brief prepared by his constitutionally-required appellate counsel. Id. at 615, 94 S.Ct. 2437. Thus, Ross indicated that the right of access to the courts implies a right to significant assistance even for an appeal to a court of purely discretionary jurisdiction.

The Supreme Court subsequently made clear that its reliance in Ross on the prisoner’s access to an appellate brief was not mere dicta. In Halbert v. Michigan, 545 U.S. 605, 125 S.Ct. 2582, 162 L.Ed.2d 552 (2005), the Supreme Court considered Michigan’s system for handling appeals of convictions based on guilty pleas, under which the initial appeal to the intermediate court of appeals is discretionary, and not as of right. Id. at 609, 125 S.Ct. 2582. Antonio Halbert, convicted based on a plea of nolo contendere, argued that he had a right to counsel on this initial appeal, even though it was discretionary. Id. He sought the appointment of counsel under the right of meaningful access to the courts. See Halbert, 545 U.S. at 610, 125 S.Ct. 2582 (citing Griffin, 351 U.S. at 24, 76 S.Ct. 585, and Douglas, 372 U.S. at 357, 83 S.Ct. 814); see also Halbert, 545 U.S. at 624 n. 8, 125 S.Ct. 2582 (citing Douglas as a case holding that “the State must affirmatively ensure that poor defendants receive the legal assistance necessary to provide meaningful access to the judicial system”); see also Mayle v. Felix, 545 U.S. 644, 664 n. 8, 125 S.Ct. 2562, 162 L.Ed.2d 582 (2005) (describing Halbert as a case where counsel was needed to ensure an indigent litigant’s “meaningful access to justice”).

The Supreme Court held that even though the first-tier appeal was discretionary, the right of meaningful access to courts entitled Halbert to counsel. Halbert, 545 U.S. at 616-17, 125 S.Ct. 2582. The holding was based on two factors. First, the Court found that the state court of appeals sat as an “error-correction instance,” id. at 617, 125 S.Ct. 2582, and made its discretionary determination whether to grant leave to appeal based on “some evaluation of the merits of the applicant’s claims,” id. at 618, 125 S.Ct. 2582. This stood in contrast to traditional courts with discretionary appellate jurisdiction, such as Michigan’s Supreme Court, which sit “not to correct errors in individual cases, but to decide matters of larger public import.” Id.

Second, the Supreme Court stressed that “indigent defendants pursuing review in the Court of Appeals [were] generally ill equipped to represent themselves.” Id. at 617, 125 S.Ct. 2582. Part of this difficulty came from the procedural complexity of the appeals process and from the fact that the indigent defendants often had limited educational backgrounds. Id. at 621-22, 125 S.Ct. 2582. Importantly, however, some of the difficulty came from the fact that because the court of appeals was a first-tier body, the applicants did not have access to an appellate lawyer’s brief. Id. at 619, 125 S.Ct. 2582. This absence was significant in light of Ross, and led the Court in Halbert to conclude that in the absence of such a brief, “a pro se applicant’s entitlement to seek leave to appeal to Michigan’s intermediate court may be more formal than real.” Halbert, 545 U.S. at 620, 125 S.Ct. 2582.

Halbert provides strong support for my conclusion that the right of meaningful access to existing clemency mechanisms entails the right to certain evidence of innocence. Though clemency proceedings are not exclusively or even primarily “error-correction” proceedings, and often turn not on a revisitation of the facts underlying a conviction, but on an analysis of a defendant’s contrition and personality, 35 they nevertheless have one significant, even if discretionary, error-correcting function: they are the last resort for the wrongfully convicted. The Supreme Court in Herrera, after strongly suggesting that freestanding claims of actual innocence were not cognizable grounds for habeas relief, 506 U.S. at 400-01, 113 S.Ct. 853, emphasized at length the vital role that clemency has historically played as the “fail safe” “relief mechanism” for innocent prisoners, id. at 415, 113 S.Ct. 853; see also id. at 411-12, 113 S.Ct. 853 (“Clemency is deeply rooted in our Anglo-American tradition of law, and is the historic remedy for preventing miscarriages of justice where judicial process has been exhausted.” (footnote omitted)); id. at 412, 113 S.Ct. 853 (noting that clemency in England-“was the only means by which one could challenge his conviction on the ground of innocence”); id. at 415, 113 S.Ct. 853 (“Recent authority confirms that over the past century clemency has been exercised frequently in capital cases in which demonstrations of ‘actual innocence’ have been made.”); id. at 417, 113 S.Ct. 853 (“History shows that the traditional remedy for claims of innocence based on new evidence, discovered too late in the day to file a new trial motion, has been executive clemency.”); see also id. at 428, 113 S.Ct. 853 (Scalia, J., concurring) (arguing that claims of actual innocence are not cognizable on habeas review and stating, “With any luck, we shall avoid ever having to face this embarrassing question again, since it is improbable that evidence of innocence as convincing as today’s opinion requires would fail to produce an executive pardon.”); Ex parte Grossman, 267 U.S. 87, 120, 45 S.Ct. 332, 69 L.Ed. 527 (1925) (“Executive clemency exists to afford relief from undue harshness or evident mistake in the operation or enforcement of the criminal law.”).

With respect to evaluating claims of actual innocence, clemency proceedings are not fully error-correcting, but they come close enough for Halbert to provide some guidance. Even when fulfilling their historic function of “preventing miscarriages of justice where judicial process has been exhausted,” Herrera, 506 U.S. at 412, 113 S.Ct. 853, clemency proceedings need not focus on trial errors but on a reevaluation of the actual guilt of the defendant. However, they do focus closely on the individual factors personal to each defendant and serve as the “fail safe” against wrongful imprisonment. Id. at 415, 113 S.Ct. 853. Unlike most courts of discretionary appellate jurisdiction, clemency proceedings are not concerned with “the general importance of the questions presented,” Halbert, 545 U.S. at 619, 125 S.Ct. 2582, and so resemble error-correcting proceedings more closely than does anything besides direct judicial review.

I acknowledge that the similarity between clemency proceedings and the Michigan Court of Appeals’ review of convictions based on guilty pleas is imprecise, but even so, Halbert strongly implies that access to some types of evidence is necessary for meaningful access to clemency proceedings. If clemency proceedings were exact analogues of the appeals at issue in Halbert, the right of meaningful access to existing clemency mechanisms would imply a right to counsel, which it clearly does not. However, the right at issue here is nothing so costly as the right to counsel. It is the far less burdensome right of access to evidence in the state’s possession, which is perhaps closer to the right to at least an appellate brief, a right that applies even to purely discretionary appeals.

One might object that even though leave to appeal in Halbert was discretionary, it was not standardless. See Halbert, 545 U.S. at 618, 125 S.Ct. 2582 (“[T]he court’s response to the leave application by any of the specified alternatives — including denial of leave — necessarily entails some evaluation of the merits of the applicant’s claims.”). However, while some aspects of a clemency application, such as the decision to grant mercy for a crime that was in fact committed by the applicant, may be standardless, see Dumschat, 452 U.S. at 464, 101 S.Ct. 2460 (noting that a decision whether to commute generally depends in part on “purely subjective evaluations and on predictions of future behavior by those entrusted with the decision”), a decision to grant clemency based on actual innocence, though entirely discretionary, is not. Though not enforceable, at least in noncap-ital cases, the standard for determining whether to grant clemency based on innocence is obviously to free those who are known, to a sufficient degree of certainty, to be innocent. See Herrera, 506 U.S. at 398, 113 S.Ct. 853 (“[T]he central purpose of any system of criminal justice is to convict the guilty and free the innocent.”). That different clemency mechanisms may prescribe different degrees of certainty is no more troubling than the fact that judges on the Michigan Court of Appeals may hold widely different views regarding what showing of error warrants a grant of leave to appeal.

Thus, the discretionary nature of clemency proceedings does not defeat the analogy between the right of meaningful access to courts and meaningful access to clemency mechanisms, though, as noted above and described in more detail below, it does significantly lessen the value of any evidence to the prisoner.

e. The Necessity of the Evidence of Innocence

Though as described in Section D.2.ii.d, supra, the discretionary nature of clemency proceedings does not extinguish the analogy between meaningful access to courts and meaningful access to clemency mechanisms, it does substantially diminish the value to the prisoner of any evidence of innocence. Even overwhelming evidence of innocence would not compel the grant of clemency, meaning that even the most persuasive evidence of innocence is not sufficient to give a prisoner relief. And it is possible — though not likely — for a prisoner without significant evidence of innocence to receive clemency, rendering access to evidence of innocence not strictly necessary to give a prisoner relief.

However, this does not mean that deprivation of access to evidence of innocence can never deprive a prisoner of meaningful access to existing clemency mechanisms. The right of meaningful access to courts can entitle prisoners to resources which are neither sufficient nor strictly necessary to allow a litigant to obtain relief. For similar reasons, the right of meaningful access to clemency mechanisms can entitle prisoners to access to evidence of innocence, in some circumstances, even though that evidence is neither sufficient nor strictly necessary nor sufficient to allow a prisoner to obtain clemency.

I find the analogy to transcripts of ha-beas corpus proceedings instructive. The transcript is very unlikely to be sufficient to allow a prisoner to obtain relief. And while very useful, it is not absolutely indispensable to further proceedings; the petitioner is aware of everything that happened at the proceeding. Though the difficulty of working from memory makes a transcript practically necessary, it is thus not strictly necessary. Because a transcript is practically necessary, a prisoner’s right of meaningful access to the courts gives the state an affirmative obligation to provide it — even though it is neither sufficient nor strictly necessary for relief — at least when it is readily available to the state and the prisoner has no other means of acquiring it. See Long, 385 U.S. at 194-95, 87 S.Ct. 362 (requiring provision of a transcript of a habeas corpus proceeding when it was readily available to the state, even though state law permitted filing an appeal without the transcript); Gardner v. California, 393 U.S. 367, 369-70, 89 S.Ct. 580, 21 L.Ed.2d 601 (1969) (specifically rejecting the argument that the prisoner’s memory of a prior habeas corpus hearing removed the need to provide him with the transcript); Johnson, 393 U.S. at 485-86, 89 S.Ct. 747 (noting that the obligation to provide a transcript of a habeas corpus proceeding when it is otherwise unavailable derives from the right of access to the courts). By analogy, a prisoner’s right of meaningful access to clemency proceedings gives the state an affirmative obligation to provide access to at least some types of evidence of innocence — evidence that, even though not sufficient or strictly necessary, is practically necessary to allow' a prisoner to receive clemency — when it is readily available to the state and the prisoner has no other means of acquiring it. 36

The Supreme Court has at one point indicated that a transcript need not be provided if it is not necessary to decide a nonfrivolous claim, but context and subsequent precedent makes clear that it referred to practical necessity, not strict necessity. In United States v. MacCollom, 426 U.S. 317, 319, 96 S.Ct. 2086, 48 L.Ed.2d 666 (1976), the Supreme Court upheld 28 U.S.C. § 753(f), which provides indigent prisoners filing post-conviction motions under 28 U.S.C. § 2255 with access to transcripts of prior proceedings only if a judge certifies “that the suit or appeal is not frivolous and that the transcript is needed to decide the issue presented by the suit or appeal,” § 753(f).

It is fairly clear that § 753(f) uses “needed” to refer not to strict necessity but to at most practical necessity. Justice Blackmun’s opinion concurring in the judgment in MacCollom, which is controlling due to its narrow scope, see 426 U.S. at 329, 96 S.Ct. 2086 (Blackmun, J., concurring in the judgment) (“I write separately, however, to emphasize the narrowness of the constitutional issue that is before us and the ease of its resolution.”), construed § 753(f) to require substantially less than a showing of strict necessity, stating:

In order for [the prisoner] to obtain a transcript of his trial, he was required to show only that his claim was not frivolous and that there was a basis, grounded on some articulable facts, for believing that transcript would assist him in his § 2255 proceeding. Clearly, there is no constitutional requirement that the United States provide an indigent with a transcript when that transcript is not necessary in order for him to prove his claim, or when his claim is frivolous on its face.

Id. at 329-30, 96 S.Ct. 2086 (Blackmun, J.). 37 Thus, Justice Blackmun construed § 753(f)’s requirement that a judge find the transcript was “needed” to require not strict necessity but only a showing of “a basis, grounded on some articulable facts, for believing th[e] transcript would assist [the prisoner] in his § 2255 proceeding.” MacCollom, 426 U.S. at 329-30, 96 S.Ct. 2086 (Blackmun, J.). And the fact that the second quoted sentence uses language mirroring the first suggests that Justice Blackmun was using “necessary” to refer to the statutory requirement, not to strict necessity.

To the extent that there was any doubt, Bounds subsequently made clear that MacCollom dealt with at most practical necessity, and certainly not strict necessity. In justifying its holding that prisons had to provide law libraries' — a holding that rested on the fact that the libraries, though not strictly necessary to enable a prisoner to file legal papers, were practically necessary' — -the Court in Bounds distinguished MacCollom as a case where the prisoner, among other defaults, “demonstrated no need for the transcript.” Bounds, 430 U.S. at 828 n. 16, 97 S.Ct. 1491. The use of “no need” suggests that the Court in Bounds interpreted § 753(f)’s “need” requirement to refer not to strict necessity but to something less — possibly to some degree of need; possibly, as Justice Blackmun construed it, to a mere basis for believing that the transcript would assist; but in any case to no more than practical necessity. And of course Bounds itself held that the state was affirmatively obligated to provide not merely those materials that were strictly necessary to file lawsuits, such as paper and pens, but those that were practically necessary as well— “adequate law libraries or adequate assistance from persons trained in the law.” 430 U.S. at 828, 97 S.Ct. 1491. 38

Therefore, the fact that evidence of innocence is neither sufficient nor strictly necessary for a favorable outcome does not indicate that it can never be required by the right of meaningful access to existing clemency mechanisms. Transcripts of court proceedings are closely analogous, and it is difficult to see why they would be required if sufficiently overwhelming evidence of innocence is not. 39

f. What Evidence Must Be Provided In Order to Accord Meaningful Access

The next issue I address is under what circumstances the interest in meaningful access to existing mechanisms of executive clemency entitles a prisoner to obtain evidence. As discussed in Section D.2.ii.d, supra, since executive clemency is purely discretionary, evidence — even evidence of innocence — -is not as valuable to an inmate as it would be if it conferred a right to relief under law. As the right of access to the courts does not require the state to make prisoners able to “litigate effectively once in court,” Lewis, 518 U.S. at 354, 116 S.Ct. 2174, it is inescapable that a prisoner can have meaningful access to executive clemency mechanisms without receiving all favorable evidence. Like a transcript of a habeas corpus proceeding the prisoner attended, it need neither be sufficient nor strictly necessary for a successful plea of clemency, but like such a transcript, the evidence of innocence must be forceful enough to be practically necessary in a plea for clemency based on innocence. Cf. Johnson, 393 U.S. at 485-86, 89 S.Ct. 747 (noting that the right of access to the courts requires transcripts to be furnished to those prisoners who cannot otherwise obtain them).

In evaluating what type of evidence would be practically necessary, there are two factors to consider. One is the reliability of the evidence, and the other is the relevance the evidence would have to the trial if true. I separate out these two issues, instead of simply considering the probative value of the evidence taken as a whole, because the Supreme Court has placed great emphasis on the unreliability that taints evidence over the passage of time. See Herrera, 506 U.S. at 403-04, 113 S.Ct. 853 (discussing the difficulty of ascertaining the reliability of newly discovered evidence and the likelihood that a retrial would be less accurate due to the length of time that had elapsed). Along with loss of memory and unavailability of witnesses upon further proceedings, one factor the Supreme Court appeared to find especially problematic in considering new evidence in Herrera was the possibility that the late-discovered evidence was fabricated. See id. at 417-18, 113 S.Ct. 853 (“No satisfactory explanation has been given as to why the affiants waited until the 11th hour — and, indeed, until after the alleged perpetrator of the murders himself was dead — to make their statements”); see also id. at 418, 113 S.Ct. 853 (“ ‘[I]t is ... reasonable to presume that there is something suspect about a defense witness who is not identified until after the 11th hour has passed.’ ” (quoting Taylor v. Illinois, 484 U.S. 400, 414, 108 S.Ct. 646, 98 L.Ed.2d 798 (1988))). 40 In light of these concerns, and of the fact that a clemency proceeding is not a traditional factfinding proceeding, cf. Dumschat, 452 U.S. at 464, 101 S.Ct. 2460 (“A decision whether to commute a long-term sentence generally depends not simply on objective factfind-ing, but also on purely subjective evaluations and on predictions of future behavior by those entrusted with the decision.”), evidence of less than the utmost reliability is likely to be disregarded, no matter how relevant it purports to be. Thus, evidence of innocence that is not of essentially unimpeachable reliability is not practically necessary for meaningful access to clemency mechanisms.

The final question to address in defining the liberty interest at stake is how relevant evidence of innocence must be in order to be practically necessary. I conclude that evidence of innocence that is of unimpeachable reliability is practically necessary if it “ ‘undermines confidence in the outcome of the trial,’ ” Kyles, 514 U.S. at 434, 115 S.Ct. 1555 (quoting Bagley, 473 U.S. at 678, 105 S.Ct. 3375). 41 Though this standard is equivalent to Bagley’s reasonable probability standard, I find the “undermines confidence” formulation easier to apply. See Strickler, 527 U.S. at 297-301, 119 S.Ct. 1936 (Souter, J., concurring in part and dissenting in part) (criticizing the “reasonable probability” formulation and urging an alternative shorthand while noting that the “touchstone” of a court’s inquiry is whether the evidentiary suppression undermines confidence in the outcome). 42

In reaching this conclusion, I am mindful that a clemency proceeding is very different from a trial and that the executive can take account of a number of factors not cognizable at law. See McKithen VII, 481 F.3d at 106 (noting that DNA’s availability in clemency proceedings might necessitate a deviation from the “trial-focused” reasonable probability standard). However, if evidence, whatever its reliability, is not relevant enough to undermine confidence in the outcome of a trial, it is hard to see how that evidence is practically necessary for an inmate pleading for clemency on the grounds of innocence. Conversely, if evidence of unimpeachable reliability does undermine confidence in the outcome of trial, then it is at least as indispensable as a transcript of a relevant court proceeding. A prisoner denied a transcript of a court proceeding relevant to a meritorious claim is not wholly thwarted in seeking relief, but his likelihood of success is significantly lower. A prisoner denied evidence of unimpeachable reliability which undermines confidence in the outcome of trial is not technically without any hope of receiving clemency, but his likelihood of success is virtually negligible. A prisoner forced to rely on his memory of a court proceeding is surely no worse off than one forced to rely solely on his own powers of persuasion, without any evidence, to convince the executive of his innocence. Thus, it would be inconsistent with the logic of the Supreme Court’s transcript cases not to recognize an entitlement to evidence of innocence under such circumstances. 43 The state’s decision to prevent a prisoner from obtaining evidence of innocence of unimpeachable reliability that is sufficient to undermine confidence in the outcome of the trial does not literally bar the prisoner from making a plea for clemency, but it so effectively thwarts his plea as to deny him meaningful access to the clemency mechanism.

In sum, prisoners retain a liberty interest in meaningful access to existing executive mechanisms of clemency. In any state with a clemency mechanism, this interest in meaningful access includes an interest in obtaining evidence of innocence when that evidence is of essentially unimpeachable reliability and undermines confidence in the outcome of the trial. Before proceeding to consider the right of access to courts as an independent basis for a liberty interest, I will pause briefly and address potential objections.

g. Potential Objections

Woodard, Dumschat and Greenholtz might at first glance seem in conflict with my conclusion that a prisoner retains an entitlement to meaningful access to existing clemency mechanisms. However, each of those cases concerned plaintiffs who claimed conditional entitlement to release, not to access, and who sought to actually alter the clemency mechanisms themselves in order to provide the plaintiffs with additional procedural protections against “erroneous” determinations of whether they should receive clemency. See Woodard, 523 U.S. at 277, 118 S.Ct. 1244 (noting that the plaintiff sought a greater notice period and the presence of counsel at clemency interviews); Dumschat, 452 U.S. at 461, 101 S.Ct. 2460 (noting that the plaintiff sought a written statement of reasons for denying commutation); Greenholtz, 442 U.S. at 6, 99 S.Ct. 2100 (noting that the court of appeals ordered a full formal hearing with advance notice, the right to present documentary evidence, a record of proceedings, and written explanations of the parole board’s facts and reasons supporting any adverse determination). The liberty interest I conclude exists is an interest only in access to the clemency mechanisms that exist, not in the outcome of clemency proceedings, and thus could not require that the clemency mechanisms themselves be changed to provide additional process. Cf. Woodard, 528 U.S. at 285, 118 S.Ct. 1244 (Rehnquist, C.J.) (“[T]he executive’s clemency authority would cease to be a matter of grace committed to the executive authority if it were constrained by the sort of procedural requirements the respondent urges.”).

Similarly, the Supreme Court has made clear that “[p]rocess is not an end in itself,” and does not create a liberty interest when there is no underlying substantive “interest for process to protect.” Olim v. Wakinekona, 461 U.S. 238, 250, 103 S.Ct. 1741, 75 L.Ed.2d 813 (1983). However, this merely means that in situations where, as here, clemency procedures have been established even though the prisoner has no liberty interest in their outcome, the prisoner also has no liberty interest in the state following those procedures. See id. at 243, 103 S.Ct. 1741 (noting that the plaintiff alleged that a prison transfer committee deprived him of due process when it failed to follow its own rule). And it may be true that states need not establish clemency mechanisms at all. See Herrera, 506 U.S. at 414, 113 S.Ct. 853 (“Of course, although the Constitution vests in the President a pardon power, it does not require the States to enact a clemency mechanism.”). However, when states do establish such mechanisms, the First Amendment guarantees prisoners a right of meaningful access to them. See Harvey IV, 285 F.Sd at 314 (Luttig, J.) (“[The interest in meaningful access to existing executive clemency mechanisms] exists, I believe, even if there is no independent liberty interest in these mechanisms themselves; in the particular processes by which the executive exercises his discretion to grant or deny clemency; or in the freedom that would result in favorable executive action obtained through these mechanisms, such as would entitle one under the Constitution to a clemency procedure, to particular processes within a clemency procedure, or to actual release pursuant to a clemency procedure.”).

A possible response might be that there is no difference between requiring a prosecutor’s office to provide evidence and requiring a state to adopt procedural changes in its clemency mechanism. As the prosecutor’s office is part of the same state executive that established the clemency mechanism, the argument might go, any obligations imposed on the prosecutor’s office are the equivalent of obligations imposed on the clemency mechanism. This argument proves too much, as it would if credited authorize the executive to interfere freely with prisoners’ rights of access to existing clemency mechanisms. Further, the structure of the Petition Clause, in guaranteeing a right to petition but not imposing on the government any duty to act on citizens’ petitions, supports my conclusion that a requirement that the government allow prisoners access to evidence for use in a petition does not impair the government’s discretion to consider and decide the petitions in any way it sees fit.

The case for considering prosecutor’s offices to be subject to special duties regarding the retention and production of evidence, duties that are distinct from the discretion accorded the executive branch in administering clemency, is particularly strong. At least while judicial proceedings are pending, prosecutors are obliged to disclose material favorable evidence, e.g., Bagley, 473 U.S. at 682, 105 S.Ct. 3375, and retention of evidence is a core prose-cutorial function, invoking absolute immunity from civil liability for damages, Parkinson v. Cozzolino, 238 F.3d 145, 152 (2d Cir.2001). The fact that a prisoner’s right to meaningful access to clemency mechanisms may require prosecutor’s offices to provide evidence does not in any way affect the clemency mechanism itself; it merely requires prosecutors to perform their traditional functions regarding disclosure and retention of evidence in a slightly different setting. 44

A different objection is that my conclusion conflicts with United States v. Qui nones, 313 F.3d 49 (2d Cir.2002), which held that “the continued opportunity to exonerate oneself throughout the natural course of one’s life is not a right ‘so rooted in the traditions and conscience of our people as to be ranked as fundamental.’” Id. at 62 (quoting Rochin v. California, 342 U.S. 165, 169, 72 S.Ct. 205, 96 L.Ed. 183 (1952)). However, Quinones addressed a different question than I consider. In Quinones, the Second Circuit overturned the district court’s conclusion that the Federal Death Penalty Act deprived prisoners of the substantive right to seek exoneration throughout the natural course of their lives. See Quinones, 313 F.3d at 61 n. 9 (“[T]he claim is properly framed as one of substantive due process.”). The Petition Clause entitles prisoners to seek clemency while they remain alive; Qui-nones implies only that the state does not violate the Petition Clause by executing them, at least if it uses proper procedures to do so.

Therefore, Woodard, Dumschat, Green-holtz, Olim, and Quinones do not conflict with my conclusion that a prisoner retains a liberty interest in meaningful access to executive mechanisms of clemency, including an interest in access to evidence of innocence where that evidence is of essentially unimpeachable reliability and undermines confidence in the outcome of trial.

iii. The Entitlement to Post-Conviction Relief and the Interest in Meaningful Access to the Counts

As discussed in Section D.2.ii.a, supra, prisoners retain a liberty interest in meaningful access to the courts. However, the right of access to the courts is infringed only if an underlying action is impeded by government conduct. Christopher, 536 U.S. at 415, 122 S.Ct. 2179 (“[T]he right is ancillary to the underlying claim, without which a plaintiff cannot have suffered injury by being shut out of court.”).

In New York, newly-discovered evidence of innocence can warrant the grant of a new trial in circumstances where there is “a probability that had such evidence been received at trial the verdict would have been more favorable to the defendant.” N.Y.Crim. Proc. Law § 440.10(1)(g). The “probability” standard is interpreted by New York courts as requiring that the evidence be “ ‘such as will probably change the result if a new trial is granted.’ ” People v. Tankleff, 49 A.D.3d 160, 848 N.Y.S.2d 286, 300 (2d Dep’t 2007) (quoting People v. Salemi, 309 N.Y. 208, 216, 128 N.E.2d 377 (1955)). As any evidence entitling McKithen to relief under § 440.10(g) would necessarily undermine confidence in the outcome of trial and thus be necessary for meaningful access to existing clemency mechanisms, I will consider the contours of the right of meaningful access to clemency mechanisms instead of the right of meaningful access to courts for relief under § 440.10(1)(g). 45

In the absence of a statute such as New York’s, a prisoner may still have several causes of action which could be impeded by frustration of access to DNA testing. Although many DNA exoneration cases have involved mistaken identification, there might be some situations where DNA evidence provides the factual predicate for a claim that perjury at trial deprived the defendant of due process of law. 46 If a freestanding claim of actual innocence is a cognizable ground for habe-as relief, denial of access to evidence for DNA testing may frustrate the ability to bring this claim. Cf. Osborne v. Dist. Att’y’s Office, 521 F.3d 1118, 1130-31 (9th Cir.2008) (basing entitlement to DNA testing in part on its assumption that a freestanding claim of actual innocence is cognizable on habeas review). Similarly, a prisoner who wishes to excuse a procedural default on the grounds of actual innocence, and to bring a challenge to her underlying conviction, would also have a claim which turns on the availability of evidence for DNA. 47

Even when there is a possible underlying action, however, the obstructive conduct must frustrate any attempts to bring the action or acquire equivalent relief in the future. Christopher, 536 U.S. at 415, 122 S.Ct. 2179 (“There is, after all, no point in spending time and money to establish the facts constituting denial of access when a plaintiff would end up just as well off after litigating a simpler case without the denial-of-access element.”). Accordingly, if any viable causes of action entitle the prisoner to obtain access to evidence for DNA testing as part of the adjudication of the claim, the prisoner’s access to the courts is not thwarted by the state’s refusal to allow testing beforehand. Cf. Thomas v. Goldsmith, 979 F.2d 746, 749-50 (9th Cir.1992) (requiring disclosure of biological evidence under Brady when relevant to the threshold showing of actual innocence in a habeas proceeding). In such situations, the prisoner has a liberty interest in her underlying cause of action without reference to the right of access to the courts.

3. Consideration of the Affected Interests

i. The Private Interest

The private interest in meaningful access to existing clemency mechanisms is powerful, but defeasible. The right to petition the government “has been recognized as one of the most precious of the liberties safeguarded by the Bill of Rights.” Beretta, 524 F.3d at 397 (internal quotation marks omitted). The importance of the right of access to clemency mechanisms in particular is compounded by executive clemency’s status as “the historic remedy for preventing miscarriages of justice where judicial process has been exhausted.” Herrera, 506 U.S. at 412, 113 S.Ct. 853.

On the other hand, despite the importance of the right to petition for clemency, it is not so weighty that it prevents the state from executing a capital prisoner. Quinones, 313 F.3d at 52-53. Additionally, I bear in mind that the decision whether to act on a prisoner’s petition is entirely discretionary, and the undeniable importance of an innocent prisoner’s release and exoneration are not legally cognizable as liberty interests safeguarded by procedural due process rights, at least unless a freestanding claim of actual innocence can be the basis for habeas relief. Section D.2.i, supra. 48

In all, the liberty interest of noncapital prisoners in meaningful access to executive clemency mechanisms, though important, and even compelling in isolation, is not sufficiently weighty to overcome powerful countervailing interests. 49

ii. The Risk of Erroneous Deprivation and the Probable Value of Additional Safeguards

A prisoner’s liberty interest is only in acquiring evidence of innocence, assuming the evidence is of essentially unimpeachable reliability and undermines confidence in the outcome of the trial. The evidence a prisoner seeks in requesting access to a physical item for DNA testing is not the item itself, but the results of the DNA testing. Theoretically, there could be a dispute over whether the results of a DNA test would be of essentially unimpeachable reliability, but in practice the dispute is more likely to revolve around whether the evidence would be evidence of innocence, and if so whether it would be sufficient to undermine confidence in the outcome of the trial.

The results of modern DNA testing are of essentially unimpeachable reliability. Modern forensic DNA testing examines short tandem repeats (“STRs”), which are repeating base pairs, and compiles “profiles” composed of the numbers of STRs at each of 13 predetermined loci on the DNA strand. E.g., President’s DNA Initiative, STR Analysis, http://www.dna.gov/basics/ analysis/str (last visited July 21, 2008). The chance that two different DNA sequences would have the same number of STRs at each of the 13 loci is infinitesimal. 50 As every person has a unique DNA sequence, excluding identical twins, STR analysis can confirm identity, or lack of identity, to a “practical certainty.” Harvey TV, 285 F.3d at 305 n. 1 (Luttig, J.). While laboratory error can never be wholly eliminated, samples can be retested if there is any real doubt about test results. In all, unlike previous methods of DNA testing, see generally David EL Kaye, The Science of DNA Identification: From the Laboratory to the Courtroom (and Beyond), 8 Minn. J.L. Sci. & Tech. 409 (2007) (describing controversies over the accuracy and reliability of earlier DNA testing methods), modern STR analysis (known as PCR-STR testing due to its use of polymerase chain reaction to multiply the sample DNA and enable testing of smaller samples of biological material) provides identification evidence of essentially unimpeachable reliability.

More often in dispute, as here, are the relevance of the evidence and the likelihood that it will be exculpatory. Under the existing procedures set forth in N.Y.Crim. Proc. Law § 440.30(l-a)(a), a court will order DNA testing of evidence if it determines “that if a DNA test had been conducted, and if the results had been admitted in the trial resulting in the judgment, there exists a reasonable probability that the verdict would have been more favorable to the defendant.” Though it is not absolutely clear from the face of the statute, on the most natural reading of this provision the court will order testing if there is a reasonable probability that the results of the testing would be exculpatory and would affect the verdict. One case in the United States District Court for the Southern District of New York reads the statute to allow testing if there is a reasonable probability that the verdict would have been more favorable, assuming exculpatory results, but cites no authority besides the statute itself for this interpretation. See Bryant v. Greiner, No. 02-CV-6121 (RMBXRLE), 2006 WL 1675938, at *5 (S.D.N.Y. June 15, 2006) (“New York courts will allow testing if there is a ‘reasonable probability’ the verdict would have been more favorable to [the defendant] if the DNA tests were performed and were exculpatory.” (emphasis added)). However, the New York State Supreme Court, Appellate Division, First Department has taken the opposite interpretation, denying DNA testing under § 440.30(l-a) in one case because it concluded the testing was unlikely to produce an exculpatory result. People v. Figueroa, 36 A.D.3d 458, 826 N.Y.S.2d 256, 256-57 (1st Dep’t) (“Given the trial testimony, including that of the defense witnesses, there is no basis upon which to suspect that any blood found on the street near the victim’s body came from anyone but the victim. Defendant’s assertion that the blood may have come from the claimed alternate assailant, who allegedly had a fight with the victim in a nearby club prior to the shooting, is highly speculative.”), lv. denied, 9 N.Y.3d 843, 840 N.Y.S.2d 770, 872 N.E.2d 883 (2007). Thus, it is fairly clear that under § 440.30(1-a), New York courts do not assume that the test result will be exculpatory and then determine whether the results would be sufficient to raise a reasonable probability of a different outcome. Rather, they assess whether there is a reasonable probability that the test result will be both exculpatory and sufficiently relevant to cause a different outcome. 51

Existing procedures present a serious risk of erroneously depriving a prisoner of his right to meaningful access to New York’s executive clemency mechanisms. 52 The procedures contemplated in § 440.30(l-a) are, of course, far less likely to result in an erroneous deprivation than the “procedure” of the unilateral decision by the prosecutor whether or not to make evidence available for DNA testing. A court’s assessment of whether an exculpatory test result would be sufficiently significant to raise a reasonable probability that the outcome of the trial would be more favorable to the defendant is not likely to lead to erroneous deprivations of the right to meaningful access to clemency procedures. This is because “a ‘reasonable probability’ is a probability sufficient to undermine confidence in the outcome,” Bagley, 473 U.S. at 682, 105 S.Ct. 3375, in New York as well as in the federal system, People v. Vilardi, 76 N.Y.2d 67, 74, 556 N.Y.S.2d 518, 555 N.E.2d 915 (1990). However, § 440.30(l-a)’s failure to require courts to assume exculpatory results raises a significant possibility of erroneous deprivation.

When a court follows § 440.30(l-a) and assesses the possibility that the results of a DNA test will be exculpatory, it substitutes its speculation for the process of actually conducting a test. Such an inquiry creates a substantial possibility that a court will decide that the defendant was probably guilty based on the evidence produced at trial, and thus conclude that the result of any DNA testing would not be exculpatory. This is a problem because recent history teaches us that courts can err in concluding that defendants are guilty. Courts evaluating the appeals and post-conviction motions of defendants who were later exonerated by DNA evidence greatly overestimated the strength of the government’s evidence. In approximately 50% of the first 133 DNA exonerations involving written decisions, the court commented on the later-exonerated defendant’s likely guilt, and in 10% the court referred to the evidence of guilt as “overwhelming.” Brandon L. Garrett, Judging Innocence, 108 Colum. L.Rev. 55, 107 tbl. 8, 109 (2008); see also id. at 107-09 & 107 tbl. 8 (noting that in 32% of the cases, the court found that any error was harmless beyond a reasonable doubt, and in over 14% the court found that there was no reasonable probability that any error affected the outcome). Given the proven fallibility of reviewing courts in making such assessments, it is puzzling why § 440.30(l-a) calls upon courts to speculate as to what the result of DNA testing is likely to be. A far more accurate way to determine the result of DNA testing is to conduct the testing. If a prisoner would be entitled to the test results in the event that those results are exculpatory, simply opining what the results are likely to be poses a significant and needless risk of erroneous deprivation. This risk of erroneous deprivation can be wholly eliminated if the court, in such cases, simply authorizes the testing.

Thus, the risk of an erroneous deprivation and the probable value of additional safeguards are both substantial and weigh strongly in favor of the conclusion that procedural due process requires that a court assume exculpatory results when assessing whether the results of DNA testing would undermine confidence in the outcome of trial.

iii. The Government’s Interest

The cost of DNA testing is significant. Forensic PCR-STR analysis in this case is estimated to cost $1095 per test, and it is difficult to predict how many tests would need to be administered in order to extract a usable DNA profile. However, in this case it is not necessary to weigh this cost against the liberty interest possessed by the prisoner because the state is not called upon to bear the cost. The Innocence Project has offered to pay the costs of DNA testing in this case, which I understand to include any costs incident to preserving the security and evidentiary value of the knife during the testing. The government has not provided any evidence of administrative or fiscal burdens that McKithen could not compensate.

Similarly, the cost of DNA retention may be significant, but it is not at issue in this case. Retention of biological evidence is governed by Arizona v. Youngblood, 488 U.S. 51, 57-58, 109 S.Ct. 333, 102 L.Ed.2d 281 (1988), under which destruction of biological evidence is not unconstitutional unless it is done in bad faith. Neither of the factors counseling the Court in Young-blood to adopt a bad faith test — the difficulty of determining the significance of destroyed evidence and the burdens caused by a requirement that the state preserve all conceivably relevant evidence indefinitely, id. — apply to the question whether a defendant may access evidence that the state has chosen to preserve.

The government does have an interest in finality, but that interest is not substantially burdened by a prisoner performing DNA testing on physical evidence used to convict him. As Judge Gertner noted in Wade, “[t]he Supreme Court has explained that the interest in finality is, in reality, an amalgam of several interests,” 460 F.Supp.2d at 248 (citing Calderon v. Thompson, 523 U.S. 538, 555-56, 118 S.Ct. 1489, 140 L.Ed.2d 728 (1998)), including retribution, deterrence, the quality of judicial decisionmaking, federalism, and the impact on victims. While such interests are upset by procedures that threaten to actually undo judgments of conviction, they are not appreciably disturbed by mere testing in service of access to existing clemency mechanisms. See Wade, 460 F.Supp.2d at 248-49 (noting the minimal impact on finality caused by mere testing and arguing that even attacking the judgment of conviction does not seriously harm finality when DNA evidence raises concerns about the accuracy of the initial verdict). States have uniformly found the value of clemency mechanisms to outweigh whatever impacts on finality they may have. Having set up clemency mechanisms specifically for the purpose of overriding the finality of judgments, and having reserved entirely to itself the discretion to do so, the state cannot argue that finality concerns weigh heavily against providing prisoners a meaningful opportunity to access those mechanisms. 53

4. The Procedural Due Process Requirements Regarding Post-Conviction Access to Evidence for the Purpose of DNA Testing

For the reasons stated above, I conclude that a convicted prisoner retains a liberty interest entitling him to meaningful access to existing executive clemency mechanisms. Meaningful access includes an interest in access to evidence of innocence when that evidence is of essentially unimpeachable reliability and is sufficient to undermine confidence in the outcome of trial. After balancing this interest against the government’s interests in light of the risk of erroneous deprivation and the probable value of additional safeguards, I conclude that in states possessing some clemency mechanism (including a parole system) which can reduce or undo a prisoner’s sentence based on the ground that the prisoner is actually innocent, and at least where, as here, the tests can be performed effectively without imposing non-negligible fiscal or other burdens on the government or subjecting an individual to nonconsensual DNA testing not otherwise authorized by law, 54 a prisoner has a right to access physical evidence for the purpose of DNA testing if: the physical evidence is in the possession of the government; the testing is nonduplicative; and assuming exculpatory results, the results of the testing would undermine confidence in the outcome of trial. 55

E. Substantive Due Process

1. Analyzing Substantive Due Process Constraints on Executive Action

In addition to guaranteeing that appropriate procedures will be used to determine whether a person has substantive rights to take actions free from government interference in certain circumstances, the Due Process Clause of the Fourteenth Amendment has also been interpreted to confer substantive rights. While the substantive component of the Fourteenth Amendment’s Due Process Clause technically includes such relatively determinate rights as the right of meaningful access to court proceedings and, as against the states, most of the same protections the Bill of Rights secures as against the federal government, the oxy-moronic term “substantive due process” 56 is generally reserved for a residuum of substantive rights resisting easy categorization besides centering around a protection of individual liberty from arbitrary government action.

Mindful of the danger of unduly limiting the freedom of the political branches and the voters they represent, the Supreme Court has cautioned that “[wjhere a particular Amendment provides an explicit textual source of constitutional protection against a particular sort of government behavior, that Amendment, not the more generalized notion of substantive due process, must be the guide for analyzing these claims.” County of Sacramento v. Lewis, 523 U.S. 833, 842, 118 S.Ct. 1708, 140 L.Ed.2d 1043 (1998) (internal quotation marks omitted). In Section D, supra, I concluded that the First Amendment’s Petition Clause provides explicit textual protection to the liberty interest in meaningful access to clemency proceedings and courts, that this interest encompasses a qualified right of access to evidence for DNA testing, and that the procedural component of the Due Process Clause requires the provision of such evidence in certain circumstances. Thus, I ordinarily would not reach the question whether the “more generalized notion of substantive due process” applies. Sacramento, 523 U.S. at 842, 118 S.Ct. 1708.

However, the Second Circuit has specifically instructed me that my analysis should not end with procedural due process rights, and in particular has suggested substantive due process as a possible source of a right to access evidence for DNA testing. McKithen VII, 481 F.3d at 107 n. 17. Accordingly, and in case I have erred in my conclusions regarding the requirements of procedural due process, I turn to the substantive due process inquiry. 57 For convenience, for the remainder of Section E I assume that my conclusions in Section D are in error and that no “particular Amendment provides an explicit textual source of constitutional protection,” Sacramento, 523 U.S. at 842, 118 S.Ct. 1708 (internal quotation marks omit ted), covering access to evidence for the purpose of DNA testing.

Rights are protected by substantive due process only if they are “fundamental,” and it is essential that a court begin with a “ ‘careful description’ ” of the asserted fundamental right. Washington v. Glucksberg, 521 U.S. 702, 720-21, 117 S.Ct. 2258, 138 L.Ed.2d 772 (1997) (quoting Reno v. Flores, 507 U.S. 292, 302, 113 S.Ct. 1439, 123 L.Ed.2d 1 (1993)). 58 When evaluating an executive action to determine whether it violates the asserted substantive due process right, a court must then undertake a threshold determination whether the conduct “shocks the conscience.” Sacramento, 523 U.S. at 846, 118 S.Ct. 1708. 59 If the government action meets this level of egregiousness, a court must then determine whether the right the government action infringes is fundamental. See id. at 846-47 & n. 8, 118 S.Ct. 1708 (noting that the question whether executive action shocks the conscience precedes the question whether the government infringes a fundamental right). To do this, it must determine whether the asserted fundamental right is “ ‘deeply rooted in this Nation’s history and tradition,’ ” Glucksberg, 521 U.S. at 721, 117 S.Ct. 2258 (quoting Moore v. City of E. Cleveland, Oh., 431 U.S. 494, 503, 97 S.Ct. 1932, 52 L.Ed.2d 531 (1977)), and “ ‘implicit in the concept of ordered liberty,’ such that ‘neither liberty nor justice would exist if [it] were sacrificed,’ ” Glucksberg, 521 U.S. at 721, 117 S.Ct. 2258 (quoting Palko v. Connecticut, 302 U.S. 319, 325-26, 58 S.Ct. 149, 82 L.Ed. 288 (1937), overruled on other grounds by Benton v. Maryland, 395 U.S. 784, 794, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969)). 60

There is some uncertainty as to the precise role of history and tradition in recognizing a fundamental right, but it appears that at least some historical support is essential. Glucksberg seemed to suggest a strong requirement, implying that a right implicit in the concept of ordered liberty would still not qualify as fundamental if it was not deeply rooted in the Nation’s history and tradition. See 521 U.S. at 721, 117 S.Ct. 2258 (“Our Nation’s history, legal traditions and practices thus provide the crucial guideposts for responsible decision-making that direct and restrain our exposition of the Due Process Clause.” (internal citation and quotation marks omitted)). 61 Sacramento subsequently reaffirmed Glucksberg’s contention that some showing of historical recognition was necessary for a right to qualify as fundamental, though it left open the required “level of specificity” of the showing. Sacramento, 523 U.S. at 847 n. 8, 118 S.Ct. 1708. However, the Supreme Court subsequently modified the need for a fundamental right to be based on history and tradition, noting that “ ‘[h]istory and tradition are the starting point but not in all cases the ending point of the substantive due process inquiry.’ ” Lawrence v. Texas, 539 U.S. 558, 572, 123 S.Ct. 2472, 156 L.Ed.2d 508 (2003) (quoting Sacramento, 523 U.S. at 857, 118 S.Ct. 1708 (Kennedy, J., concurring)). Though this statement might seem to suggest that there might be cases in which a right with no historical support at all was nonetheless fundamental, the statement followed, apparently by way of explanation, the Court’s emphasis of trends within the past 50 years supporting the fundamental right at issue in that case, Lawrence, 539 U.S. at 571-72, 123 S.Ct. 2472, and thus seems not to disturb the requirement of at least some historical support for an asserted fundamental right.

2. Definition of the Asserted Fundamental Right

I would find, if forced to decide the issue, that an individual has a fundamental right to avoid suffering criminal sanctions due to a prosecutor’s deliberate indifference as to whether or not the individual is innocent of the offense for which the sanctions are imposed. This asserted right is more carefully defined than those condemned by the court in Glucksberg. Cf. 521 U.S. at 722-23, 117 S.Ct. 2258 (criticizing the following descriptions of rights: “a liberty interest in determining the time and manner of one’s death”; “a right to die”; “a liberty to choose how to die”; “a right to control of one’s final days”; “the right to choose a humane, dignified death”; “the liberty to shape death”; “[the] interest in making decisions about how to confront an imminent death” (internal quotation marks omitted)). It is closer to those approved in that case and Flores, see Glucksberg, 521 U.S. at 723, 117 S.Ct. 2258 (approving the following descriptions of rights: “[a] right to refuse lifesaving hydration and nutrition”; “a right to commit suicide which itself includes a right to assistance in doing so”); Flores, 507 U.S. at 302, 113 S.Ct. 1439 (describing a right as “the alleged right of a child who has no available parent, close relative, or legal guardian, and for whom the government is responsible, to be placed in the custody of a willing-and-able private custodian rather than that of a government-operated or government-selected child-care institution”).

3. Whether Denial of Access to Physical Evidence Shocks the Conscience

I would find that under certain circumstances, denial of access to physical evidence for DNA testing does shock the conscience. The Supreme Court has “em phasized time and again that ‘[t]he touchstone of due process is protection of the individual against arbitrary action of government,’ ” Sacramento, 523 U.S. at 845, 118 S.Ct. 1708 (quoting Wolff, 418 U.S. at 558, 94 S.Ct. 2963), including protection against “the exercise of power without any reasonable justification in service of a legitimate governmental objective,” Sacramento, 523 U.S. at 846, 118 S.Ct. 1708.

Executive abuse of power is not cognizable unless it “shocks the conscience,” Sacramento, 523 U.S. at 846, 118 S.Ct. 1708. Negligently inflicted harm does not shock the conscience, while “conduct intended to injure in some way unjustifiable by any government interest” is most likely to shock the conscience. Id. at 849, 118 S.Ct. 1708. Deliberate indifference to the medical needs of pretrial detainees shocks the conscience, id. at 849-50, 118 S.Ct. 1708, but deliberate indifference to a risk of physical injury does not shock the conscience in fast-moving exigent circumstances such as a prison riot or a high-speed chase, id. at 852-54, 118 S.Ct. 1708. However, even in unhurried situations, defendants who are “ ‘subjected to the pull of competing obligations’ ” do not shock the conscience even when they intentionally lie. Matican v. City of N.Y., 524 F.3d 151, 159 (2d Cir.2008) (quoting Lombardi v. Whitman, 485 F.3d 73, 83 (2d Cir.2007)).

I would conclude that if a prosecutor refuses a prisoner’s specific request for access to physical evidence for DNA testing, in circumstances where the testing could be performed at negligible cost to the state and the results of the testing, if exculpatory, would prove beyond a reasonable doubt that the prisoner did not commit the crime for which she is incarcerated, the prosecutor exhibits deliberate indifference to the possibility that the prisoner is actually innocent. Given the legal finding of guilt that attaches upon a verdict of conviction, see, e.g., Herrera, 506 U.S. at 399, 113 S.Ct. 853 (“Once a defendant has been afforded a fair trial and convicted of the offense for which he was charged, the presumption of innocence disappears.”), only indifference to evidence that could potentially prove innocence beyond a reasonable doubt would qualify as deliberate indifference to the possibility of innocence. However, given the negligible cost of actually performing the testing in those circumstances, the most charitable explanation for refusing to allow the testing is deliberate indifference to whether or not the prisoner is actually innocent. 62

I would further conclude that, in light of the special responsibility the state has toward those in its custody, this deliberate indifference shocks the conscience. See, e.g., Sacramento, 523 U.S. at 851-52, 118 S.Ct. 1708 (noting that deliberate indifference to a prisoner’s physical well-being shocks the conscience); Pena v. De-Prisco, 432 F.3d 98, 113 n. 22 (2d Cir.2005) (“Lewis also indicated that less culpable mental states may more easily shock the conscience when the victim is in state custody.”). The same factors indicating that deliberate indifference shocks the conscience with respect to medical attention apply here with similar force: While it is not “obligatory” for prosecutors to continue to ponder the innocence of their convicted defendants in the way it is obligatory for prison wardens to plan for the welfare of their inmates, prosecu tors have still caused the loss of the prisoner’s liberty, and it is far more “feasible” and less costly to allow access to the evidence under these circumstances than to provide even the most modest constitutionally-required medical attention. Cf. Sacramento, 523 U.S. at 851, 118 S.Ct. 1708 (“[I]n the custodial situation of a prison, forethought about an inmate’s welfare is not only feasible but obligatory under a regime that incapacitates a prisoner to exercise ordinary responsibility for his own welfare.”). And in the circumstances I describe, providing access to such evidence does not “ ‘clash with other equally important governmental responsibilities,’ ” id. at 852, 118 S.Ct. 1708 (quoting Whitley v. Albers, 475 U.S. 312, 320, 106 S.Ct. 1078, 89 L.Ed.2d 251 (1986)). Where the evidence, if exculpatory, would not merely undermine confidence in the outcome of trial but would prove innocence beyond a reasonable doubt, the impact on finality is even less than described in Section D.3.iii, supra. Indeed, the possibility that evidence exists which could prove innocence beyond a reasonable doubt itself is so unsettling that testing the evidence actually advances the cause of finality.

4. Whether the Asserted Right is Fundamental

I would find that this right is fundamental, though the fact that it turns on the subjective state of government officials severely circumscribes the remedies that are available to vindicate it. Wdiile the right is unlike those described in Glucksberg or Flores in including a subjective component, our history and tradition have continually emphasized a prosecutor’s duty to have particular psychological states, presumably due to the remarkable discretion our justice system vests in prosecutors.

As described in Section D.2.ii.b, supra, the Supreme Court has explicitly recognized the duty of the prosecutor not merely to disclose evidence before trial but to actively aim at just results. Brady was decided almost half a century ago, cf. Lawrence, 539 U.S. at 571-72, 123 S.Ct. 2472 (“In all events we think that our law and traditions in the past half century are of most relevance here.”), and drew on a line of older cases finding bad-faith deception and suppression of evidence to be unconstitutional. See Napue v. Illinois, 360 U.S. 264, 269, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959) (holding deliberate failure to correct an unsolicited false statement unconstitutional, and finding “[t]he principle that a State may not knowingly use false evidence, including false testimony, to obtain a tainted conviction” to be “implicit in any concept of ordered liberty”); Pyle v. Kansas, 317 U.S. 213, 215-16, 63 S.Ct. 177, 87 L.Ed. 214 (1942) (“Petitioner’s papers are inexpertly drawn, but they do set forth allegations that his imprisonment resulted from perjured testimony, knowingly used by the State authorities to obtain his conviction, and from the deliberate suppression by those same authorities of evidence favorable to him.”); Mooney v. Holohan, 294 U.S. 103, 112, 55 S.Ct. 340, 79 L.Ed. 791 (1935) (finding a prosecutor’s presentation of testimony known to be perjured to be “inconsistent with the rudimentary demands of justice”). The conviction that prosecutors have a duty to prosecute only the guilty and avoid charging the innocent, of course, dates back considerably farther than these cases. See generally Bruce A. Green, Why Should Prosecutors “Seek Justice”?, 26 Fordham Urb. L.J. 607, 612-14 (1999) (noting 19th-century recognition of the prosecutor’s duty to avoid prosecuting the innocent).

I am aware of no authority disputing the claim that a prosecutor has a duty to attempt to avoid convicting innocent defen dants. Even the Supreme Court’s recognition of absolute immunity for prosecutors was based on common-law principles intended to protect, not dissolve, the prosecutor’s duty to seek justice. See Imbler, 424 U.S. at 423, 96 S.Ct. 984 (noting prose-cutorial immunity at common law was intended in part to guard against the possibility that a prosecutor might “shade his decisions instead of exercising the independence of judgment required by his public trust”); id. at 424, 96 S.Ct. 984 (finding reasons underlying common-law immunity justified implying immunity under § 1983). In holding that prosecutors are entitled to absolute immunity, the Supreme Court took pains to note that while it removed a civil remedy for prosecutorial misconduct, it did not license prosecutors to commit misconduct. Imbler, 424 U.S. at 427-29, 96 S.Ct. 984. And as discussed in Section D.2.ii.b, supra, the Court also expressly acknowledged that a prosecutor’s ethical obligations toward his defendants did not end at the end of a case, noting that “after a conviction the prosecutor also is bound by the ethics of his office to inform the appropriate authority of after-acquired or other information that casts doubt upon the correctness of the conviction.” Id. at 427 n. 25, 96 S.Ct. 984.

I would find the right not to suffer criminal sanctions due to a prosecutor’s deliberate indifference to the possibility of one’s innocence not only deeply rooted in our history and tradition but also “implicit in the concept of ordered liberty.” See Glucksberg, 521 U.S. at 721, 117 S.Ct. 2258 (internal quotation marks omitted). As noted in Section D.2.i, supra, our government, constitution, and justice system are preoccupied with individualized determinations of guilt and innocence, and the concept of constitutionally-guaranteed due process is incompatible with the government as a whole being deliberately indifferent to an individual’s potential innocence. In light of the tremendous power prosecutors wield in our system, cf., e.g., Morrison v. Olson, 487 U.S. 654, 727-29, 108 S.Ct. 2597, 101 L.Ed.2d 569 (1988) (Scalia, J., dissenting) (noting the power prosecutors wield in selecting cases, quoting Attorney General Robert Jackson’s canonical address to federal prosecutors), a prosecutor’s deliberate indifference to the potential innocence of a defendant is tantamount to the government’s deliberate indifference, and incompatible with the very concept of justice, which necessarily entails attention to the distinction between guilt and innocence. See Herrera, 506 U.S. at 398, 113 S.Ct. 853 (“[T]he central purpose of any system of criminal justice is to convict the guilty and free the innocent.”).

While our legal system may be unable to provide a perfect remedy for violations of the right not to suffer criminal sanctions due to a prosecutor’s deliberate indifference toward the possibility of an individual’s innocence, it has given some legal effect to the prosecutor’s duty to seek justice, in Brady and its progeny. If it were necessary to decide the issue, I would conclude that the right is fundamental.

5. Potential Substantive Due Process Right of Access to Evidence for DNA Testing

For the reasons stated above, if I erred in my analysis in Section D, supra, I would find that an individual has a fundamental right not to suffer criminal sanctions due to a prosecutor’s deliberate indifference to the potential that the individual is innocent of the offense for which the sanctions are imposed. I would conclude that such deliberate indifference is expressed by a prosecutor’s refusal to allow access to physical evidence for the purpose of DNA testing, when the testing could be performed at negligible cost to the state and when exculpatory results would prove beyond a reasonable doubt that the prisoner did not commit the crime for which he is incarcerated. Of course, only those whose test results are exculpatory would actually suffer the criminal sanctions of continued incarceration due to this deliberate indifference. However, as this fundamental right would also constitute a liberty interest, for the same considerations of procedural due process discussed in Section D.3, supra, any prisoner would be entitled to access to physical evidence for DNA testing if it could be performed at negligible cost to the state and exculpatory results would prove beyond a reasonable doubt that the prisoner did not commit the crime of conviction. 63

F. Access to the Courts

Although McKithen raised this claim in his complaint, Compl. 11, the parties have not addressed it in their submissions. To the extent McKithen continues to advance this claim, as discussed in Section D.2.iii, supra, to the extent that a prisoner articulates a cause of action which cannot be brought without prior access to physical evidence for DNA testing, a prosecutor’s intentional denial of access would infringe on the prisoner’s right of meaningful access to the courts.

G. Eighth Amendment

McKithen raised this claim in his complaint, Compl. 11, but the parties have not addressed it in their submissions. At oral argument, McKithen made ambiguous statements about whether or not he continued to pursue it. To the extent McKi-then continues to pursue this claim, I need not decide whether the Eighth Amendment provides a basis on which to require post-conviction access to evidence for DNA testing. See Alley v. Key, 431 F.Supp.2d 790, 803-04 (W.D.Tenn.) (rejecting claims that the Eighth Amendment confers a right to post-conviction DNA testing), aff'd, — Fed.Appx. -, 2006 WL 1313364 (6th Cir.), cert. denied, 548 U.S. 921, 126 S.Ct. 2973, 165 L.Ed.2d 982 (2006). The only way the Eighth Amendment is possibly implicated is if it permits a freestanding claim of actual innocence. As discussed in Section D.2.i, supra, any evidence necessary to meet the extraordinary threshold possibly necessary to bring such a claim is evidence to which a prisoner is already entitled by virtue of her liberty interest in meaningful access to existing clemency procedures.

H. Confrontation Clause

McKithen raised this claim in his complaint, Compl. 11, but the parties have not addressed it in their submissions. McKi-then made ambiguous statements at oral argument about whether he continued to pursue the claim. To the extent McKithen continues to pursue this claim, I reject it. McKithen has not identified any witness he was prohibited from confronting, or any testimonial statements introduced against him without an opportunity for cross-examination. Cf. Davis v. Washington, 547 U.S. 813, 821, 126 S.Ct. 2266, 165 L.Ed.2d 224 (2006) (noting that only testimonial statements cause a declarant to be a witness within the meaning of the Confrontation Clause).

I. McKithen’s Claim

1. Issue Preclusion

“Under New York law, collateral estoppel will preclude a federal court from deciding an issue if (1) the issue in question was actually and necessarily decided in a prior proceeding, and (2) the party against whom the doctrine is asserted had a full and fair- opportunity to litigate the issue in the first proceeding.” McKithen VII, 481 F.3d at 105 (internal quotation marks omitted). Because the contours of the federal right are broader than those of the right guaranteed by state law, the issue in question was not necessarily decided in the prior proceeding, and collateral estoppel does not apply. 64

The state court deciding McKithen’s § 440.30(l-a) motion declined to order DNA testing, finding that there was “no reasonable probability that the results of such testing would have resulted in a verdict more favorable” to him. McKithen III, No. 3964/92, slip op. at 2. As discussed in Section D.3.Ü, supra, § 440.30(l-a) requires state courts to assess whether there is a reasonable probability that the DNA test results would be exculpatory and would have resulted in an outcome more favorable to the defendant. However, as stated in Section D.4, supra, the Due Process Clause of the Fourteenth Amendment requires a judicial determination of whether the results of DNA testing, assuming they were exculpatory, would be sufficiently relevant to undermine confidence in the outcome of trial (or, what amounts to the same thing, to raise a reasonable probability of a different outcome).

Accordingly, the state court did not necessarily decide the issue of McKi-then’s entitlement to DNA testing under the Due Process Clause. The state court did actually decide factual. issues that would determine McKithen’s entitlement to DNA testing of the blade for blood under the Due Process Clause, as it con-eluded that even exculpatory test results would have been immaterial to the verdict. See McKithen III, No. 3964/92, slip op. at 3 (“The presence or absence of her blood on the knife allegedly used in the stabbing would have little or no probative value in determining whether the wound was self-inflicted or inflicted by the defendant.”). However, McKithen now seeks to test the handle, not the blade of the knife. In any event, the state court’s determination was unnecessary to support its judgment, and thus has no preclusive effect here. See, e.g., NLRB v. Thalbo Corp., 171 F.3d 102, 109 (2d Cir.1999) (“Nor is litigation of an issue barred by collateral estoppel ... if its decision was not necessary to the judgment.”). Therefore, the issue ’ of McKi-then’s entitlement to DNA testing under the Due Process Clause is not precluded by the Queens Supreme Court’s determination of his § 440.30(l-a) motion.

2. Whether Exculpatory DNA Test Results Would Undermine Confidence in the Outcome of Trial

McKithen’s complaint suggests several possible exculpatory scenarios. In his complaint, McKithen appeared to seek to show that none of Rose McKithen’s blood was on the blade, but at oral argument he made clear that he sought only to test for the presence of DNA on the handle, which would most likely be present in the form of skin flakes. He argues that the DNA testing could reveal tissue of a person other than himself, his wife, or the officer who recovered the knife.

Even if McKithen has no constitutional right to subject his wife and the officer to nonconsensual DNA testing and neither of them consent to such testing, see supra note 53, McKithen could still establish that any skin flakes were shed by someone other than himself, his wife or his arresting officer if the skin flakes were shed by an individual with a DNA profile in the FBI’s Combined DNA Index System (CO-DIS), a database of DNA profiles collected for law enforcement purposes. Although collection of DNA for inclusion in CODIS is often nonconsensual under state and federal DNA database statutes, the database search itself does not subject any individual to nonconsensual DNA testing— the nonconsensual DNA collection and testing occurred by statute, not due to McKithen’s request for testing. See, e.g., United States v. Amerson, 483 F.3d 73, 76-77, 89 (2d Cir.2007) (describing federal DNA database statute and upholding it against Fourth Amendment challenge).

It does not seem particularly likely to me that skin flakes from any individual in CODIS would be found on the handle of the knife, but I must assume exculpatory results in assessing whether McKithen is entitled to testing. 65 Assuming that skin flakes from an individual in CODIS were found on the handle of the knife, it would not raise a probability that the result of trial would be different. That another individual in CODIS gripped the handle of a kitchen knife at some point is not necessarily inconsistent with McKithen wielding the knife and stabbing Rose McKithen, especially because other parties’ handling of the knife during its recovery and at trial, Tr. 391-92, may have removed any tissue McKithen deposited on it. McKi-then’s theory is that Rose McKithen’s unidentified boyfriend stabbed her and she falsely accused him instead. Assuming that the boyfriend is in CODIS and his DNA is found on the knife, his status as her boyfriend does not make it at all implausible for him to have gripped the kitchen knife and not stabbed her. 66 Given Rose McKithen’s testimony that McKithen stabbed her, Jones’s testimony that McKi-then entered the apartment, took the kitchen knife, and held Rose McKithen at knifepoint before forcing her out, and the evidence of McKithen’s inculpatory statement upon arrest, I cannot say that this evidence would “probably change the result” of a new trial, Tankleff, 848 N.Y.S.2d at 300 (internal quotation marks omitted). 67

However, the fact that an individual in CODIS held the knife with which Rose McKithen was stabbed, if that were the result of the testing, would undermine confidence in the outcome of trial. It is certainly possible that one of the guests at the McKithen house was profiled in CODIS and also grasped the kitchen knife at some point. However, it is also entirely possible that DNA from a CODIS profilee was on the knife handle because that individual grasped the knife and stabbed her. Such a test result would raise at least enough doubt to undermine confidence in the outcome of trial. I see no reason to suspect that DNA testing would reveal such a result, of course, but my suppositions are no substitute for forensic testing. The due process protections vindicating MeKithen’s right to meaningful access to New York’s clemency mechanisms give him the right to perform that testing.

CONCLUSION

Our legal system continues to grapple with the questions of which avenues of relief remain open to those advancing claims that they are wrongfully convicted. In some states, as in New York, there are statutory mechanisms to set aside a conviction based on newly discovered evidence. It is unclear whether there is a constitutional right to do so. The remaining resort for the innocent convicted is to avail themselves of the opportunity to petition for clemency in whatever form the state has authorized. States may debate the value of expanding or contracting any of these avenues; in light of the tremendous probative power of DNA evidence, it may be wise to strike a different balance between accuracy and finality in cases where it is available. The proper avenues for relief are matters for legislative judgment. The Petition Clause, however, secures a right of meaningful access to whatever avenues remain, and the Due Process Clause confers a procedural right of access to evidence for DNA testing, if the testing can be accomplished at little cost and exculpatory results would undermine confidence in the outcome of trial.

For the reasons stated above, McKi-then’s cross-motion is granted and Brown’s is denied. McKithen is entitled to an injunction directing Brown to provide him access to the knife for the purpose of DNA testing on the condition that McKithen abide by all necessary procedures to preserve its integrity as evidence and pay all costs associated with the testing.

So ordered.

1 . Brown styles his papers as an opposition to McKithen’s motion for summary judgment and a cross-motion to dismiss on statute of limitations grounds. Though his papers do not formally seek summary judgment, Brown does not assert that there are any genuine issues of material fact, and he argues throughout his papers that McKithen is not entitled to access the knife. Accordingly, and with his consent at oral argument, I construe Brown’s motion to be one for summary judgment. In light of this, for simplicity, I deem his cross-motion based on the statute of limitations to be incorporated into the cross-motion for summary judgment, arguing that no reasonable jury could conclude that his action is not time-barred.

2 . I use her first and last name to avoid confusion with the plaintiff.

3 . He has since changed his position, and now seeks to test the handle of the knife for DNA that would establish that another individual used the knife to stab Rose McKithen.

4 . McKithen also filed a habeas corpus petition on January 16, 2003, which I dismissed as time-barred on March 6, 2003. McKithen v. Walsh (McKithen IV), No. 03-CV-334 (JG) (E.D.N.Y. Mar. 6, 2003).

5 .This report and recommendation is the fifth cited written decision regarding McKithen's claims because it followed my March 6, 2003 order dismissing his habeas corpus petition. See supra note 4.

6 . The defense of claim preclusion may be raised in an answer. Fed.RXiv.P. 8(c)(1).

7 . McKithen does not argue otherwise.

8 . This conclusion is part of the Circuit’s holding because it was necessary to the reasoning leading the Second Circuit to find the Rooker-Feldman doctrine inapplicable. The doctrine prevents a federal court other than the Supreme Court from reviewing a state court judgment, and thus " 'federal plaintiffs are not subject to the Rooker-Feldman bar unless they complain of an injury caused by a state judgment.’ ” McKithen VI, 481 F.3d at 97 (emphasis omitted) (quoting Hoblock v. Albany County Bd. of Elections, 422 F.3d 77, 87 (2d Cir.2005)). The Second Circuit in this case concluded that "a party is not complaining of an injury ‘caused by’ a state-court judgment when the exact injury of which the party complains in federal court existed prior in time to the state-court proceedings, and so could not have been 'caused by’ those proceedings,” and found that to be "precisely the case” in this appeal, leading it to conclude that the bar was inapplicable. McKithen VII, 481 F.3d at 98. Although there might be other ways in which a court could find the doctrine inapplicable, see id. at 98 n. 9 ("By no means does this suggest that, in order to avoid the Rooker-Feldman doctrine, a party’s injury must have arisen prior to any state-court judgments.”), the finding that McKi-then’s injury arose prior to the state court proceeding was the way in which the Second Circuit found the doctrine inapplicable in this case, and thus was necessary to its holding.

9 . As described in Section D.4, infra, the constitutional violation was the deprivation of procedural due process to adjudicate whether his liberty interest in meaningful access to existing executive clemency mechanisms encompasses access to the knife introduced in evidence against him. McKithen argues that he is subject to continuing injury each time the government denies his requests. In light of my conclusion that McKithen's claim is not time-barred, I need not decide whether the statute of limitations is subject to this type of renewal.

10 . The redundant term is used to distinguish the requirements of procedural due process from the substantive rights the Due Process Clause protects as well. See generally Section E, infra.

11 . Parolees have a liberty interest in avoiding the revocation of parole, Morrissey v. Brewer, 408 U.S. 471, 482, 92 S.Ct. 2593, 33 L.Ed.2d 484 (1972), but this is distinct from the situation of inmates in custody, who do not have a liberty interest in being released on parole unless the distinct statutory scheme creates a conditional entitlement to parole, Greenholtz, 442 U.S. at 9-11, 99 S.Ct. 2100.

12 . New York’s parole statute does not confer any entitlement that gives rise to a liberty interest. Barna v. Travis, 239 F.3d 169, 171 (2d Cir.2001) ("The New York parole scheme is not one that creates in any prisoner a legitimate expectancy of release.”).

13 . "The term ‘clemency’ refers not only to full or conditional pardons, but also commutations, remissions of fines, and reprieves.” Herrera v. Collins, 506 U.S. 390, 411 n. 12, 113 S.Ct. 853, 122 L.Ed.2d 203 (1993). For convenience, I use it also to refer to parole decisions, which are analyzed under a similar constitutional framework. Dumschat, 452 U.S. at 464, 101 S.Ct. 2460 ("A commutation decision therefore shares some of the characteristics of a decision whether to grant parole.”).

14 .Chief Justice Rehnquist acknowledged that the capital prisoner retained "a residual life interest, e.g., in not being summarily executed by prison guards,” Woodard, 523 U.S. at 281, 118 S.Ct. 1244 (Rehnquist, C.J.), but found that any life interest the prisoner retained in "not being executed in accord with his sentence” was not cognizable as a matter of procedural due process, id. The residual life interest Chief Justice Rehnquist recognized in not being summarily executed seems loosely analogous to residual liberty interests the Court has found in conditions of confinement; a residual interest not to suffer in life or liberty beyond that duly authorized by the prisoner’s sentence.

15 .In his opinion concurring in the Fourth Circuit’s denial of rehearing en banc in Harvey TV, Judge J. Michael Luttig suggested that Justice O’Connor may have intended the minimal procedural safeguards to apply to non-capital clemency proceedings as well. 285 F.3d at 314 (Luttig, J., respecting the denial of rehearing en banc). While it is true that Justice O’Connor did not distinguish between capital and noncapital clemency proceedings in describing the safeguards she envisioned, the context of Justice O'Connor's statement indicates that she intended them to extend only to capital clemency proceedings. See Woodard, 523 U.S. at 289, 118 S.Ct. 1244 (O’Connor, J.) ("When a person has been fairly convicted and sentenced, his liberty interest, in being free from such confinement, has been extinguished. But it is incorrect, as Justice Stevens’ dissent notes, to say that a prisoner has been deprived of all interest in his life before his execution. Thus, although it is true that pardon and commutation decisions have not traditionally been the business of courts ..., I believe that the Court of Appeals correctly concluded that some minimal procedural safeguards apply to clemency proceedings.” (citation omitted, first emphasis added)).

16 . Although Justice O’Connor’s plurality opinion does not spell out her reasoning in detail, she apparently implicitly balanced the procedural due process interests differently from Justice Stevens in order to arrive at the different procedural requirements for capital clemency proceedings.

17 . Judge Luttig construes Woodard slightly differently on this point, arguing that “the 'life interest’ of the capital prisoner would seem to continue (or not) to the same extent as does the 'liberty interest' of the noncapital prisoner,” Harvey IV, 285 F.3d at 314 (Luttig, J.), and that combining this conclusion with the minimal procedural safeguards Justice O’Connor contemplated for capital proceedings produces the result that Justice O’Connor and the justices who signed her opinion would hold, if the issue were "squarely put before” them, that noncapital prisoners have an ongoing liberty interest in release as well, id. Even if this is true, however, until the issue is squarely presented, Dumschat and Greenholtz control and compel the conclusion that noncapital prisoners have no liberty interest in release before the expiration of a valid sentence.

18 . For this reason, I also need not consider the precise conditions of the entitlement to life recognized in Woodard; it would be relevant to a capital prisoner’s attempt to access evidence only insofar as the life interest includes an interest in not being executed if innocent, the precise question the Supreme Court has avoided deciding.

19 . Clearly an innocent prisoner has a tremendous and justifiable interest, in any nontechnical sense of the term, in release from prison. See Schlup, 513 U.S. at 321, 115 S.Ct. 851 ("[A] prisoner retains an overriding interest in obtaining his release from custody if he is innocent of the charge for which he was incarcerated.” (internal quotation marks omitted)). However, a "liberty interest” for the purposes of procedural due process is a technical term which refers only to a conditional entitlement. See Dumschat, 452 U.S. at 465, 101 S.Ct. 2460 (finding that a prisoner has no liberty interest in receiving a pardon or commutation, "simply a unilateral hope”).

20 . In explaining Morello’s holding, Franco cited not only to Bounds but also to Washington v. James, 782 F.2d 1134 (2d Cir.1986), which itself cited Bounds for the proposition that the state has affirmative obligations to ensure the right of access to the courts, James, 782 F.2d at 1138. See Franco, 854 F.2d at 588 (citing James).

21 . As noted in Section D.2.g, infra, this analysis does not imply that a prisoner has any constitutional right to change the clemency mechanisms themselves to provide a better chance of clemency. Instead, it merely recognizes that a policy that drastically impacts a prisoner’s ability to obtain clemency through any existing clemency mechanism affects whether the prisoner’s access to that mechanism is meaningful or merely a " 'meaningless ritual,' ” Ross v. Moffitt, 417 U.S. 600, 612, 94 S.Ct. 2437, 41 L.Ed.2d 341 (1974) (quoting Douglas v. California, 372 U.S. 353, 358, 83 S.Ct. 814, 9 L.Ed.2d 811 (1963)).

22 .While the cases concerning access to records of court proceedings and access to appellate counsel typically discuss the Due Process and Equal Protection Clauses instead of the Petition Clause, Sliker, 751 F.2d at 490, the Supreme Court has made clear that they involve the right of meaningful access to courts, e.g., Johnson, 393 U.S. at 485-86, 89 S.Ct. 747; Bounds, 430 U.S. at 821-23 & n. 8, 97 S.Ct. 1491, which the Supreme Court has found to be grounded in those clauses as well as the Petition Clause, Christopher, 536 U.S. at 415 n. 12, 122 S.Ct. 2179. While it has noted that equal protection and due process inquiries are slightly different in theory, Ross, 417 U.S. at 608-09, 94 S.Ct. 2437, the Supreme Court has analyzed access claims similarly under the Due Process Clause and the Equal Protection Clause, see Evitts v. Lucey, 469 U.S. 387, 403-05, 105 S.Ct. 830, 83 L.Ed.2d 821 (1984) (rejecting, at length, the contention that access cases did not apply when due process alone was implicated), and has noted, in reviewing such cases, that its "case law reveals that, as a practical matter, the two Clauses largely converge,” Smith v. Robbins, 528 U.S. 259, 276, 120 S.Ct. 746, 145 L.Ed.2d 756 (2000).

23 . One can also simply petition a branch of government directly, bypassing any formal mechanisms set up for hearing petitions.

24 . It may also apply to evidence of unknown exculpatory value, if the defendant can establish that there is a reasonable likelihood that the evidence could have affected the outcome of the proceeding. See United States v. Valenzuela-Bernal, 458 U.S. 858, 873-74, 102 S.Ct. 3440, 73 L.Ed.2d 1193 (1982) (noting, in Compulsory Process Clause case borrowing due process principles, that sanctions would be appropriate for deportation of witnesses who never gave statements if the defendant could establish a reasonable probability that their testimony would have affected the trier of fact).

25 . It is "unsettled” in most circuits whether the Brady disclosure obligation continues during post-conviction proceedings. Fred C. Za-charias, The Role of Prosecutors in Serving Justice After Convictions, 58 Vand. L.Rev. 171, 189-91 (2005); but see Thomas v. Goldsmith, 979 F.2d 746, 749-50 (9th Cir.1992) (finding that Brady applies during post-conviction proceedings).

26 . The Court in Imbler was not merely referring to a duty extending during the pendency of an appeal. It referred to the actions of the prosecutor in that case, who wrote a letter to the governor regarding newly discovered evidence after the defendant’s conviction had been affirmed, id. at 412-13, 96 S.Ct. 984, as a fulfillment of this duty. Id. at 427 n. 25, 96 S.Ct. 984.

27 . Arguably during the pendency of a direct appeal, withholding of evidence could violate the right to meaningful access to the courts for the purposes of pursuing that appeal, but unless the appellate court can simply assimilate newly-discovered evidence into the record on appeal, it is hard to see why this would be. Generally, when new evidence comes to light during the pendency of a direct appeal, it is only relevant insofar as it bears on the defendant's ability to move for a new trial.

28 . The rationale for requiring disclosure even in the absence of a request was that in the context of trial preparation, the absence of a request is functionally similar to a general request for “all Brady material.’’ Agurs, 427 U.S. at 106-07, 96 S.Ct. 2392. This is, of course, not the case with DNA evidence, which can become available through scientific advances without anyone in the prosecutor's office having in mind its potential applicability to various pieces of evidence in past cases.

29 . The Second Circuit uses a slightly different formulation of the materiality test, but the Supreme Court has made clear that it is the same standard. The language used in the Second Circuit’s cases applying this standard is drawn from Agurs’s articulation of the materiality standard. See, e.g., Justice v. Hoke, 90 F.3d 43, 47 (2d Cir.1996) (finding that whether erroneous exclusion of evidence deprives the defendant of a fair trial "depends upon whether 'the omitted evidence [evaluated in the context of the entire record] creates a reasonable doubt that did not otherwise exist.'” (quoting Agurs, 427 U.S. at 112, 96 S.Ct. 2392)); Hawkins v. Costello, 460 F.3d 238, 244 (2d Cir.2006) (same, citing Justice). However, Strickland, Bagley, Kyles v. Whitley, 514 U.S. 419, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995), and Agurs itself all make clear that Agurs's formulation of the materiality standard is essentially the same as the later Bagley/Strickland “reasonable probability” formulation. See Strickland, 466 U.S. at 694, 104 S.Ct. 2052 (noting that "the appropriate test for prejudice finds its roots in the test for materiality of exculpatory information not disclosed to the defense by the prosecution” and citing Agurs); Bagley, 473 U.S. at 681—82, 105 S.Ct. 3375 (plurality opinion) (noting that the Court "relied on and reformulated the Agurs standard for the materiality of undisclosed evidence” in Strickland, and applying "the Strickland formulation of the Agurs test for materiality” to all Brady cases regardless of whether the defense had specifically requested the evidence); Kyles, 514 U.S. at 436, 115 S.Ct. 1555 (“Agurs thus opted for its formulation of materiality, later adopted as the test for prejudice in Strickland ....” (emphasis added)); compare Agurs, 427 U.S. at 104, 96 S.Ct. 2392 (“[IJmplicit in the requirement of Brady is a concern that the suppressed evidence might have affected the outcome at trial.”), with Bagley, 473 U.S. at 682, 105 S.Ct. 3375 (plurality opinion) ("The evidence is material only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.”).

30 . After reviewing, at least cursorily, all cases citing Brady which are binding on me, I have not found any that conflict with my interpretation. The case that comes closest to doing so is Leka, which noted that “Brady requires disclosure of information that the prosecution acquires during the trial itself, or even afterward.” 257 F.3d at 100. The only arguable conflict here is the fact that suppression of evidence discovered during the pendency of a direct appeal would, on my theory, not violate the right to a fair trial but the right to meaningful access to the courts for the purposes of filing a new trial motion, and the duty would be only the duty to disclose evidence which would establish a claim for a new trial. See supra note 27. Leka, on the other hand, seems to suggest that evidence discovered during the pendency of an appeal would be analyzed under Brady's materiality standard, which is more generous to a defendant. In fact, however, Lelca supports my interpretation more than it conflicts with it. First, Leka recognizes a continuing disclosure obligation of some kind. Second, it cites Imbler for this proposition, which distinguishes between trial-stage disclosure, which is enforced by due process, and post-trial disclosure, mandated by the ethics of the prosecutor's office. See Leka, 257 F.3d at 100 (citing Imbler, 424 U.S. at 427 n. 25, 96 S.Ct. 984). To the extent that Leka relies on Imbler as authority for a constitutional obligation to disclose evidence discovered after trial, it supports my theory that the prosecutor's ethical duties are enforceable through constitutional provisions other than the Due Process Clause. Finally, it is not even clear that Leka requires the same quantum of evidence to be disclosed post-trial as during the proceedings, but to the extent it does, it provides for broader post-trial disclosure than my interpretation, and thus supports my ultimate conclusion.

31 . The Sixth Amendment’s right to counsel did not apply; it was described as a ''trial-level” right and distinguished from the right to appellate counsel. Evitts, 469 U.S. at 393, 105 S.Ct. 830; see also id. at 408, 105 S.Ct. 830 (Rehnquist, J., dissenting) ("[T]he Sixth Amendment right to counsel applies only to trial level proceedings.”).

32 . It is noteworthy also that the right of meaningful access to courts attaches to appeals even in absence of a constitutional right to an appeal. Evitts, 469 U.S. at 400-01, 105 S.Ct. 830. Thus, the fact that clemency mechanisms are not constitutionally required, Herrera, 506 U.S. at 414, 113 S.Ct. 853, does not create a disanalogy between the right of meaningful access to existing clemency mechanisms and the right of meaningful access to the courts.

33 . It is noteworthy that, when it comes in response to a defendant's specific request for DNA testing, a prosecutor’s withholding is intentional and not merely inadvertent.

34 . It is true that a greater range of information is relevant to clemency proceedings, McKithen VI, 481 F.3d at 106, but the significance of any one piece of information is smaller because it confers no legal entitlement at such a proceeding.

35 . See Dumschat, 452 U.S. at 464, 101 S.Ct. 2460 ("A decision whether to commute a long-term sentence generally depends not simply on objective factfinding, but also on purely subjective evaluations and on predictions of future behavior by those entrusted with the decision.”).

36 . If anything, the argument that the state has an obligation to provide evidence of innocence on request is even stronger than in the case of a transcript. When prisoners have no other means of acquiring a transcript due to indigence, the state may bear little responsibility for their indigence. Prisoners, however, have no other means of acquiring evidence of innocence in the state’s possession due entirely to the state's decision to keep the evidence from them. In those circumstances, the obligation to allow access to the evidence is less in the nature of an affirmative duty to provide and closer to a duty not to withhold access.

37 . Additionally, I have found no indication that judges in this circuit applying § 753(f) look to determine whether a transcript is strictly necessary, such as inquiring whether an account of the proceedings could be ascertained by soliciting affidavits from those who were present. Indeed, in the rare situation in which even the most relevant transcript is entirely lost, a persuasive showing through affidavits would likely allow a court to decide the issues the transcript could have resolved. Thus, if § 753(f) authorized the provision of transcripts only when they were strictly necessary — such as when all of the participants in the court proceeding, including the petitioner, had no recollection of what occurred or were unavailable to provide affidavits — it would authorize them exceedingly rarely.

38 . Bounds is less directly analogous than the transcript cases because an adequate law library is a single resource not specific to any individual’s case, whereas transcripts are case-specific materials. It is not, however, clear that this makes any constitutional difference; I have found no case where a court discusses whether or not a resource is case-specific in deciding whether the right of meaningful access to the courts requires the state to provide that resource.

39 . The most persuasive distinction I can perceive between transcripts of court proceedings and evidence of innocence is that transcripts of court proceedings are available to the general public, whereas evidence of innocence is denied to all. However, as discussed in Section D.2.Ü.C, supra, the Supreme Court does not employ a merely comparative analysis. Given that a comparison to the access that others enjoy is not the entirety of the Court’s analysis or even a major part of it, the Brady line of cases and the cases involving denial of access to the courts by obstructive conduct suggest that the prosecutor's duty to seek justice provides at least as strong a presumptive entitlement to the resource as does the fact that other people have the ability to purchase court transcripts.

40 . It may also be significant that judicial recognition of newly-discovered evidence creates a risk that associates of a defendant will pressure witnesses to recant long after the defendant's conviction. Access to evidence for DNA testing does not, of course, give rise to such risks.

41 . Though a defendant who pleads guilty does not forfeit the right to petition for clemency generally, I leave aside the question whether a voluntary plea of guilty forfeits the right to plead for clemency on the grounds of actual innocence. Cf. The Innocence Project, Access to Post-Conviction DNA Testing, http://www. innocenceproject.org/Content/304.php (last visited July 21, 2008) (noting that 11 of the first 218 DNA exonerees pled guilty).

42 . Thus, to undermine confidence in the outcome of trial, evidence need not be so powerful that it is clear by a preponderance of the evidence that the outcome would have been different. Kyles, 514 U.S. at 434, 115 S.Ct. 1555.

43 . This is the reason I differ, with utmost respect, from the conclusion set forth in Judge Luttig’s penetrating and insightful opinion in Harvey IV. See 285 F.3d at 317 (Luttig, J.) (restricting the right to evidence that "concededly could prove beyond any doubt that the defendant did not commit the crime for which he was convicted”). While I follow him in recognizing the liberty interest in meaningful access to existing executive clemency mechanisms, I respectfully disagree with his assessment of the materiality evidence must possess in order to be essential for meaningful access. Judge Luttig limits the right to evidence so probative that it would prove beyond any doubt that the defendant was innocent. Id. However, I do not understand why he picks this quantum of proof rather than any other, besides his conception of "elemental fairness.” Id. Attempting to draw the closest parallel I can under existing precedent and finding the logic underpinning the Supreme Court’s cases regarding access to transcripts to be applicable here, I apply the standard that provides the closest analogy to the degree of importance a transcript must have in order to be necessary for meaningful access to the courts.

44 . As discussed in Section D.3.iii, this case does not implicate the question of the government’s duty to preserve evidence, only to disclose evidence presently in its possession.

45 . Because I conclude, as discussed in Section 1.2, infra, that McKithen would not be entitled to relief under § 440.10(1)(g)'s 'probability” standard, I may not confine my analysis to the narrower right of meaningful access to the courts and leave undecided the question whether the right of meaningful access to existing clemency mechanisms encompasses a right to access evidence for the purpose of DNA testing.

46 . This might seem to be such a case, though McKithen stated at oral argument that he was not alleging any error at trial.

47 . If, as the Supreme Court has hinted, there is in fact no freestanding actual innocence claim, in a jurisdiction without a statute such as § 440.10(l)(g) the only prisoners who would have an entitlement to DNA testing secured by the right of access to the courts would be those who had procedurally defaulted claims. This might at first glance seem bizarre; those who have defaulted on their claims are entitled to DNA testing under the circumstances I outline in Section D.4, infra, while those who have pursued their rights with diligence are not (not by the right of access to the courts, at least). However, the reason it seems bizarre simply underscores the fundamental importance of clemency in our system-without the possibility of clemency or a freestanding claim of innocence, DNA testing, even if exculpatory, would simply not be very helpful even to the actually innocent. Denial of DNA testing would not cause much harm to the wrongfully convicted, because proof of their innocence would entitle them to no relief. In the real world, however, the existence of clemency mechanisms allows the wrongfully convicted a residual opportunity to plead their case. This residual prospect of exoneration — through clemency even if not as of right — accounts for our intuition that DNA testing is of value to the wrongfully convicted. And the fact that the interest in meaningful access to clemency proceedings allows access to DNA testing as described in Section D.4, infra, vindicates the rights of all prisoners, whether or not their legal claims are defaulted, to plead their case of innocence.

48 . Given the Supreme Court's expressed reluctance to decide the issue, see Herrera, 506 U.S. at 417, 113 S.Ct. 853; House, 547 U.S. at 555, 126 S.Ct. 2064, I do not base any of my analysis on the proposition that a freestanding claim of actual innocence can give rise to a claim for habeas relief.

49 . The right of access to the courts, however, is significantly stronger in that judicial proceedings lack the discretionary character of clemency proceedings. Though I need not decide this, I note that in a case where a prisoner could not afford to conduct the DNA testing himself, the private interest in meaningful access to clemency proceedings might not outweigh the cost to the state (unless that cost diminishes significantly), but the private interest in meaningful access to the courts might, at least for reasonably inexpensive DNA testing. In New York, this would establish a right to state-funded DNA testing in situations where exculpatory results would establish a probability of a different outcome at trial, § 440.10(l)(g), and a right of access to evidence for privately-funded DNA testing in situations where exculpatory results would merely undermine confidence in the outcome of trial but not establish a probability of a different outcome. That question, however, is not presently before me.

50 .It is difficult to calculate the probability of a random match with precision, because the sequence of bases along a DNA strand is not random, but it has been conservatively estimated that the chance that two different DNA sequences have identical STRs at each of the 13 loci is on the order of one in 100 billion. Harvey IV, 285 F.3d at 305 n. 1 (Luttig, J.).

51 . Brown himself appears to agree that this is the correct interpretation of § 440.30(l-a). McKithen proposes that I find that a prisoner has a right to access evidence for DNA testing if there is a reasonable possibility that if a DNA test had been conducted in connection with the prosecution, and assuming exculpatory results, the' outcome would have been more favorable. PL’s Mem. 17. Brown’s incredulous reaction to what he describes as "an artificial 'can’t miss’ legal standard” created “out of whole cloth,” Def.’s Mem. 34, and his lengthy argument that the assumption of exculpatory results is inappropriate, id. at 35-36, both indicate that he interprets § 440.30(l-a) not to require a state court to assume exculpatory results.

52 . In addition to parole, New York State has a clemency mechanism in the form of the New York State Division of Parole’s Executive Clemency Bureau. N.Y. State Div. of Parole, Executive Clemency, http://parole.state.ny.us/ Clemency.asp (last visited July 21, 2008).

53 . I note that this is not a case in which the only way in which the tests could be performed effectively would be to subject individuals to unconsented DNA testing. Given that individuals’ privacy interests in being free from nonconsensual DNA testing have at least some constitutional protection, see, e.g., United States v. Amerson, 483 F.3d 73, 89 (2d Cir.) (upholding the federal DNA database statute under a Fourth Amendment balancing test), cert. denied sub nom. Amerson v. United States, - U.S. -, 128 S.Ct. 646, 169 L.Ed.2d 515 (2007); Nicholas v. Goord, 430 F.3d 652, 671 (2d Cir.2005) (upholding the New York DNA database statute under a Fourth Amendment balancing test); see gener ally Paul M. Monteleoni, Note, DNA Databases, Universality, and the Fourth Amendment, 82 N.Y.U. L.Rev. 247 (2007) (discussing Fourth Amendment limitations on DNA database statutes), these privacy interests might affect the outcome of the Mathews balancing test in such a scenario.

54 . As mentioned in notes 49 & 53, supra, I need not decide whether the absence of these conditions would alter the outcome of the balancing test.

55 . For reasons similar to those set forth in Section D.3, supra, the right of meaningful access to the courts would allow access to evidence for DNA testing where the evidence is in the government's possession; the testing is nonduplicative; and assuming exculpatory results, the testing would establish an element of a cause of action. In light of the heightened interest in access to a nondiscretionary proceeding, it is possible that testing imposing a greater burden on the government would still be constitutionally required in circumstances where exculpatory results would establish an element of a cause of action. See supra note 49 (discussing this possibility).

56 .As John Hart Ely has remarked, the term is logically similar to "green pastel redness.” John Hart Ely, Democracy and Distrust 18 (1980). It might be more natural to refer to this collection of substantive rights as privileges and immunities. But see SlaughterHouse Cases, 16 Wall. 36, 83 U.S. 36, 79-80, 21 L.Ed. 394 (1872) (construing the Privileges or Immunities Clause of the Fourteenth Amendment to protect only a narrow set of rights arguably already protected by the Supremacy Clause). In any case, under existing precedent, "[i]t can no longer be controverted that due process has a substantive component as well,” County of Sacramento v. Lewis, 523 U.S. 833, 856, 118 S.Ct. 1708, 140 L.Ed.2d 1043 (1998) (Kennedy, J., concurring), regardless how semantically awkward this may be.

57 . Even if I am correct in my conclusions in Section D.4, supra, that the procedural due process protections for the First Amendment right of meaningful access to existing clemency mechanisms grants a right of access to physical evidence for purpose of DNA testing, substantive due process analysis would be required in a jurisdiction without any existing clemency mechanisms. However, it is hard to see how any prosecutorial withholding of evidence could shock the conscience unless there was some legally protected use for that evidence. If there are no clemency proceedings, then the only other apparent legally protected use for this evidence (apart from overcoming a procedural default, if a prisoner has a procedurally defaulted claim) would be in an action challenging the conviction on the grounds of innocence. If the law permits no such action and there is no possibility of clemency, then it would not appear that withholding of legally useless evidence, no matter how malicious, could shock the conscience in the constitutional sense. (Of course, such a system would betray an appalling indifference to individualized guilt, at least by contemporary standards, but the shameful character of that system of justice would not create a legal right.) Thus, whether there is a substantive due process right of access to evidence in a system with no clemency mechanisms and no applicable statutory provisions for new trials on the basis of newly discovered evidence would turn on whether a freestanding claim of actual innocence is a cognizable habeas claim. Since -New York (along with every other state, Criminal Justice Policy Foundation, State Clemency Resources, http://www. cjpf.org/clemency/clemencystates.html (last visited July 21, 2008)) has at least some executive clemency mechanism, I need not reach this difficult question, and I decline to do so.

58 . Fundamental rights are often referred to as "fundamental liberty interests," e.g., Glucksberg, 521 U.S. at 721, 117 S.Ct. 2302 (internal quotation marks omitted), but I refer to them here as "fundamental rights” to avoid confusion with my use of the term "liberty interests” in my procedural due process analysis.

59 . This requirement is not present when evaluating statutes. Id. at 847 n. 8, 117 S.Ct. 2302 ("[E]xecutive action challenges raise a particular need to preserve the constitutional proportions of constitutional claims, lest the Constitution be demoted to what we have called a font of tort law.”). One might argue that this "threshold question,” id., precedes the careful description of the right to be asserted, but the assessment of whether an action shocks the conscience seems to require an understanding of the type of rights it infringes in cases where those rights are not readily apparent. See Flores, 507 U.S. at 302, 113 S.Ct. 1439 (" 'Substantive due process’ analysis must begin with a careful description of the asserted right....”).

60 .Generally, if the right is fundamental, the court applies strict scrutiny, prohibiting the government from infringing the right " 'unless the infringement is narrowly tailored to serve a compelling state interest.' " Glucksberg, 521 U.S. at 721, 117 S.Ct. 2302 (quoting Flores, 507 U.S. at 302, 113 S.Ct. 1439). This does not appear to be an independent step in evaluating executive actions, which only reach this stage of analysis if they are so egregious as to shock the conscience. The reasons an action shocks the conscience would obviously preclude a finding that the action is narrowly tailored to serve a compelling interest, so to the extent that strict scrutiny applies at all, it necessarily invalidates the action. See Pena v. DePrisco, 432 F.3d 98, 114 (2d Cir.2005) (finding that plaintiffs alleged a constitutional violation without considering whether the conscience-shocking executive action was narrowly tailored to serve a compelling state interest). Thus, I will not consider strict scrutiny as a separate step in my analysis.

61 . At least in the context of criminal prohibitions, the Second Circuit agreed in Quinones, holding that "a criminal law violates the constitutional command of the Due Process Clause only if it offends some principle of justice 'so rooted in the traditions and conscience of our people as to be ranked as fundamental.’ ” 313 F.3d at 62 (quoting Rochin, 342 U.S. at 169, 72 S.Ct. 205).

62 . A less charitable explanation is the possibility that the prosecutor actually suspects the potential innocence of the prisoner and wishes to avoid the embarrassment of the revelation of the fact of the prisoner’s innocence. This, of course, would amount to "conduct intended to injure in some way unjustifiable by any government interest,” Sac ramento, 523 U.S. at 849, 118 S.Ct. 1708, and would undoubtedly shock the conscience.

63 . It is possible that the potential importance of such evidence is so great that the state would be required to bear the costs of reasonably practicable DNA testing if the prisoner could not afford it. Although denial of a costly test is not fully arbitrary, it still might betray deliberate indifference if there is a real possibility of a wrongful conviction and the cost to the state is not overwhelming. However, that is a question I need not decide.

64 . Brown argues that there is no such right, or that if it exists, it is no broader than the right conferred by § 440.30(l-a), but he conceded at oral argument that collateral estop-pel would not apply if there was a federal right broader than the statutory right.

65 . There is a possibility that evidence exists which would require state-of-the-art low copy number ("LCN”) testing, which only the New York Office of the Chief Medical Examiner is certified to perform in criminal cases, and thus that McKithen might not have the means to perform the testing. However, there is also the possibility that he might be able to find DNA on the handle without the use of LCN testing.

66 . A CODIS hit not consistent with this theory — that is, a finding that the knife handle bears DNA from a CODIS profilee who is not Rose McKithen’s boyfriend — would not be more helpful to McKithen. Absent an intimate relationship with her assailant, it is hard to see why Rose McKithen would allow her assailant to go free and falsely accuse her husband instead.

67 .For similar reasons, if substantive due process analysis was appropriate, I would find that McKithen was not entitled to access to the knife. I concluded in Section E.5, supra, that if substantive due process analysis applied, it would only entitle a prisoner to evidence for the purpose of DNA testing if the results of the testing would prove beyond a reasonable doubt that the prisoner was innocent of the crime of conviction. Even exculpatory results of DNA testing conducted on the knife handle would fall far short of proving beyond a reasonable doubt that McKithen did not commit the crime.