Printer Friendly



Sandwiched between a state criminal trial and a federal habeas corpus proceeding is a lesser-known phase of criminal process called "state postconviction review" ("State PCR"). Whereas trials and federal habeas process have been lavished with centuries of legal attention, (1) State PCR is a younger phenomenon that has persisted in what one might call a state of malign neglect. There is no federal right to a state postconviction lawyer (2) because there is no federal right to state postconviction process at all. (3) Until recently, State PCR was a phase of criminal process that federal institutions (if not scholarship) (4) virtually ignored. Without federal intervention, it languished as an underfunded afterthought.

State PCR is a backwater no longer. The major structural changes lurking beneath the surface of American criminal punishment continue to undermine the premise that a meaningful "day in court" takes place in a single proceeding. Instead, a commitment to reliability increasingly entails effective collateral process, and the most potent doses of that process are available in state court. The growing need for postconviction review combines with restrictions on the federal habeas remedy to create hydraulic pressure on State PCR. Federal institutions have responded with "interventions" designed specifically to reinforce State PCR's burgeoning enforcement portfolio. My objectives in this Article are to model the effect of structural change on state postconviction remedies, to make sense of and evaluate the responsive federal intervention, and by extension to understand the role that State PCR will play in modern criminal process.

In Part I, I orient readers to the basic function and criticism of State PCR. For many years, State PCR was primarily a forum for relitigating constitutional challenges that were or could have been pressed in the "direct-review chain," by which I mean on appellate (rather than collateral) review of the conviction and sentence. (5) To the extent that collateral proceedings were necessary to vindicate constitutional rights, the federal habeas remedy did the heavy lifting. Federal courts were a robust backstop for dysfunctional State PCR process, which was vulnerable to criticism on two major grounds. First, states devoted inadequate resources to their State PCR. (6) Second, state collateral process could be hostile to Supreme Court decisions announcing new constitutional restrictions on criminal punishment. (7) Notwithstanding these serious problems, the robustness of the federal habeas remedy alleviated the need for federal institutions to intervene in State PCR. The constitutional rights of state inmates could be effectively enforced using a combination of direct appellate and federal habeas review.

In the last twenty years, however, the landscape has changed dramatically. In Part II, I provide a structural explanation for the growing institutional focus on State PCR: prominent constitutional defects may be discovered only after a conviction becomes final, (8) the Supreme Court increasingly announces new and retroactive decisions requiring a postconviction forum, (9) changes in forensic science might locate the most reliable evidence of innocence outside the trial record, (10) and adverse state postconviction outcomes have become virtually impossible to contest in federal habeas proceedings. (11) Because constitutional rules of newer vintage are less capable of being enforced effectively in the direct-review chain, a viable collateral forum is a growing imperative. At the same time, waves of federal habeas restrictions have all but eliminated federal courts from the collateral enforcement equation. The result is the emergence, by necessity, of State PCR as a crucial forum for enforcing federal law. In economic phrasing, demand for collateral process is rising at the same time that federal supply is falling.

Federal institutions have begun to use interventions to regulate State PCR, which is emerging as the only forum capable of enforcing a broad swath of constitutional law. In Part III, I provide a framework for understanding that intervention, which I subdivide into three forms. First, the Supreme Court has used "constitutional-law intervention" to constitutionalize small pockets of State PCR, thereby allowing itself to enforce those rules on appellate review of state postconviction dispositions. Second, federal institutions have increasingly attempted "habeas intervention" as a regulatory means of incentivizing the structural hygiene of State PCR--attention to the quality of representation and process in those proceedings. Third, federal institutions use "resource intervention" to siphon sources earmarked for federal habeas proceedings to state postconviction litigation.

In Part IV, I show that, as a descriptive matter, federal interventions primarily (but not always) conform to what one might call "Substantive-Constraint Exceptionalism"--an emphasis on substantive punishment restrictions, particularly Eighth Amendment proportionality rules. Substantive-Constraint Exceptionalism is the exclusive focus of constitutional-law intervention. The other forms of intervention are more likely to involve procedural constraints, but embody a phenomenon that I call "dispersed enforcement," whereby lower federal judges--rather than the Supreme Court--become the primary enforcement agents.

In Part V, I argue that, although such Substantive-Constraint Exceptionalism is the practiced reality of federal intervention, the Supreme Court's early attempts to justify it as something beyond an administrative necessity remain largely unpersuasive. Substantive constraints touch very few cases, and the per-case social cost of resolving such claims is law. Exceptionalism might be easy to administer, but it lacks an elegant conceptual justification. The Supreme Court's pro-Exceptionalism narrative appears in Montgomery v. Louisiana, (12) a 2016 case holding that the Federal Constitution requires states to apply federal retroactivity laws in their own postconviction proceedings. (13) Montgomery is a touch radical in this respect, and the Exceptionalist narrative it contains strongly suggests the difficult-to-defend proposition that a conviction and sentence remain "lawful" as long as the contaminating error is procedural.

My ultimate objectives are to diagnose, predict, and evaluate structural change in State PCR. Because claims and evidence necessary to enforce constitutional rights increasingly require a meaningful collateral forum, and because the federal collateral forum is so limited, State PCR is, for lack of a better term, the Last Man Standing. That status is not lost on the Supreme Court and lower federal judges, who are adapting available legal rules to try to improve the efficacy of collateral process in state court. And such adaptation does add to the bite of criminal-process rights, the underenforcement of which is perceived as a major blemish on American penal practice. What remains lacking, however, is a more satisfying theoretical explanation for the changes.


Notwithstanding how prominently it now figures in the administration of criminal punishment, State PCR largely operates below the broader legal community's awareness threshold. Most discourse instead centers on federal postconviction process--i.e., habeas corpus proceedings. (14) Before convicted inmates file their federal habeas petitions, however, they must exhaust remedies in State PCR. (15) Almost every inmate interested in collaterally challenging a state conviction will encounter the state postconviction apparatus.

State PCR is therefore a primary device by which the Federal Constitution is enforced. (16) Because of this function, federal institutions "intervene" in order to prevent underenforcement of specific constitutional guarantees. Left to develop on its own, however, State PCR fails to perform the enforcement function adequately, so searing criticism of State PCR is easy to find. (17) In Part I, I briefly familiarize readers with the historical function of State PCR and the chronic problems afflicting it.

A. The Paradigm Use of State PCR

At a criminal trial, a court adjudicates the guilt of a defendant, subject to various parameters imposed by law. Defendants can enforce those legal constraints at trial or on appellate review thereof--i.e., in the direct-review chain. As with any other type of judgment, legal regimes prefer that constitutional challenges to convictions and sentences be litigated in the direct-review chain rather than collaterally. After a defendant is convicted, however, collateral process does afford her a postconviction forum. Because of a federal exhaustion requirement, collateral litigation begins with State PCR and concludes with federal habeas corpus proceedings. (18) Figure 1 represents the sequence of criminal process visually.

A collateral allegation that a conviction (or sentence) violates federal law is called a "claim." At least in the minds of the legal institutions making postconviction law during the infancy of State PCR, most collateral claims represented attempts to litigate issues that the inmate either waived or failed to win in the direct-review chain. (19) For that reason, many states have laws barring their courts from collaterally entertaining claims that were or could have been resolved at some prior phase of the criminal process. (20) Challenges requiring a collateral forum for competent litigation were more exception than rule. Federal institutions then developed legal principles conforming to the familiar scenario--for example, that the Constitution does not require State PCR at all, (21) that there is no constitutional or statutory right to counsel for any State PCR process, (22) and that extreme deference was due to all State PCR determinations. (23) These one-size-fits-all rules followed naturally from the view that the federal interests at stake were small and undifferentiated. State PCR was regarded generally as a redundant forum for relitigating certain claims that could have been litigated in the direct-review chain or other claims that could be entertained on federal habeas review.

B. Resources and Counsel

State PCR is globally afflicted by resource scarcity, (24) which manifests in a spectacular failure to guarantee effective legal representation. (25) Some access-to-counsel issues are typical of any legal process dominated by indigent litigants, (26) but others flicker conspicuously as failures unique to State PCR. The bottom line is that there are extraordinary structural barriers to effective counsel during state postconviction proceedings. In some respects, each state has its own story, (27) but access to counsel is impaired by some phenomena common across jurisdictions.

Roughly eighty percent of prosecutions involve indigent defendants. (28) In state postconviction litigation involving capital inmates, every jurisdiction other than Alabama and Georgia requires that counsel be appointed. (29) In noncapital postconviction cases, however, a lawyer is appointed for somewhere in the neighborhood of ten to twenty-five percent of state inmates. (30) The remaining seventy-five to ninety percent litigate pro se.

The quality of legal representation, even for those inmates lucky enough to have an appointed lawyer, remains atrocious. (31) Most reflexively assume that State PCR is handled by the same public defender entities that handle trial proceedings in criminal cases. That assumption is incorrect. First, public defender entities managing a litigation portfolio across phases of criminal process will direct resources to the phases with the best returns: trials and appeals. (32) Second, even if a public defender office were inclined to maintain a state postconviction presence, their role in the trial phase frequently disqualifies them. Difficult ethical questions arise when a public defender office must assert a Sixth Amendment ineffective-assistance-of-counsel (IAC) claim against itself--thereby attacking its own attorneys--so it ordinarily will not provide postconviction representation when its own trial assistance is at issue. (33)

For these reasons, the only situation in which a public defender can effectively represent inmates in State PCR is when the office focuses exclusively on that phase of the proceedings. Some states have defender entities specializing in State PCR, but they almost always focus on capital cases. Texas, for example, has the Office of Capital and Forensic Writs (OCFW), but it has had a narrow mandate limited to the first round of State PCR for death-sentenced inmates. (34) Moreover, capital cases account for a tiny fraction of all felony cases, even in capitally active states. (35) In the grand scheme of State PCR, public defender organizations are actually capable of shouldering only a small fraction of the necessary litigation.

In the vast majority of cases where state courts appoint counsel that is not a public defender office, the appointee will be a private attorney, sometimes from a panel. (36) These lawyers are paid almost nothing, severely limiting the pool of attorneys even willing to accept appointments. (37) Miniscule flat-fee arrangements create perverse incentives dissuading the appointees from investing sufficient time in each case. (38) These appointed attorneys can be deeply unqualified for state postconviction representation. In some jurisdictions, judges make unsupervised appointments to newly minted lawyers, anticipating that they will learn by trial and error on the state postconviction docket. (39) In short, even in cases where state postconviction lawyers are appointed, massive structural impediments to quality representation remain.

Part of the problem involves how costs for postconviction representation are shared between counties and the state. County judges confronted with appointment-of-counsel decisions effectively administer a budget for which they are electorally accountable, and postconviction expenditures are never popular. (40) Nor do these local officials tend to suffer politically when a federal court orders a retrial, (41) so there is simply no internalized incentive for those officials to make unpopular expenditures to avoid that outcome. For those inclined to the law and economics explanation, the situation is a classic prisoner's dilemma.

Under-resourced state postconviction representation has dire consequences for enforcement of federal rights. Inadequate lawyers fail to perform investigation necessary to effectively litigate claims that require it. (42) They lack economic incentives to learn basic rules of state postconviction practice. They tend to be unfamiliar with available constitutional claims and the limits on litigation thereof. (43) The result is that many federal rights go unenforced, disappearing into State PCR without meaningful guarantees of adequate legal representation.

C. Hostility to Federal Rules of Decision

The resource-driven impediments to adequate state postconviction representation are the dominant problems for state inmates alleging violations of federal law. A secondary issue precipitating federal intervention is underlying hostility to federal rules of decision--that is, to rules of federal law that constrain criminal adjudication in the states. (44) Even though the distinction between faithful and subversive application can be in the eye of the beholder, hostility to federal decision rules is particularly evident in state decisions that pare back the effects of substantive rules. (It is present in the application of procedural rules, but the examples require too much doctrinal background for the space I have here.) In cases where the Supreme Court has declared a category of inmates ineligible for a particular punishment, some states have laws that consistently restrict the scope of the exemption. (45)

The first way hostility to federal law is expressed is through narrowing interpretations of the substantive rule itself. In jurisdictions that have been historically solicitous of the death penalty, for example, state institutions have interstitially defined capital punishment exemptions in very restrictive ways. Two examples of this phenomenon involve insanity and intellectual disability (mental retardation).

The aftermath of Ford v. Wainwright, (46) which barred the execution of incompetent (insane) offenders, (47) is an example of how some states tried to limit an Eighth Amendment exemption by stringently defining it. The Ford plurality opinion left "to the State the task of developing appropriate ways to enforce the constitutional restriction upon its execution of sentences." (48) Seizing on this language, which appears to contemplate procedure, several states claimed license to adopt narrow definitions of incompetence. (49) In Panetti v. Quarterman, (50) the Supreme Court ultimately rejected the attempts of several states--states that for two decades had cited the deferential language from Ford--to restrictively define incompetency. (51) Panetti also rejected state attempts, invoking the deferential Ford language, to exclude broader procedural protections specified in Justice Powell's Ford concurrence. (52)

After Atkins v. Virginia established that intellectual disability (ID, formerly "mental retardation") precluded a death sentence, (53) some states attempted to define that exemption narrowly. (54) Atkins contained the same deferential nod that Ford included: "As was our approach in Ford... 'we leave to the State [s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.'" (55) Invoking that deferential language, some states restrictively defined ID--by, for example, imposing stringent IQ cutoffs and adaptive functioning rules inconsistent with clinical ID criteria. (56) After a dozen years, in Hall v. Florida, (57) the Supreme Court rejected the IQ cutoff as overstepping the procedural delegation Atkins made to the states, (58) and it recently issued an even more full-throated endorsement of the idea that Atkins should adhere to clinical rules in Moore v. Texas. (59)

Some states resist federal rules of decision not by attempting to define the interstices of the substantive rule, but through state procedure. In Montgomery v. Louisiana, (60) for example, the state invoked its power to restrict remedies as a way around retroactive application of the Eighth Amendment prohibition against sentencing juveniles to life without parole (LWOP). (61) The use of procedure to resist unwelcome substantive law is not always detectable, and the Supreme Court may be in a position to correct it only when such resistance crystallizes in a particularly visible rule. For this reason, less visible procedural resistance is a particularly difficult problem to address through the federal intervention to which I turn shortly.

Given that it is afflicted by chronic resource shortages, and that it operates through voter-facing judges responsive to electorates that can be hostile to collaterally applied rules, it should be unsurprising that State PCR suboptimally enforced constitutional law in many jurisdictions. For reasons I explain in Part II, however, the downside of underenforcement was limited when the collaterally enforced rights were less consequential, and when there was a federal forum readily available to correct mistakes. In light of structural change, however, the comfort of such checks is rapidly disappearing. The stakes could not be higher--individual rights are only meaningful when they are capable of being enforced, and states are not all that good at ensuring that their PCR performs that function.


The problems afflicting State PCR are increasingly consequential, especially in light of two broad categories of structural change. First, disposition of crucial substantive questions increasingly requires reference to information outside the trial record, (62) thereby requiring a collateral proceeding. Second, the declining availability of federal habeas relief means that State PCR is usually the only collateral forum in which federal law is capable of being meaningfully enforced. (63) These two phenomena produce a hydraulic effect through which the increased emphasis on collateral litigation is routed from federal habeas proceedings to State PCR.

A. Substantive Changes

State PCR is becoming more and more important in part because of the explosion of substantive legal questions requiring reference to information necessarily outside the trial record. (64) Questions incapable of resolution in the direct-review chain require some sort of collateral forum, lest they assume the status of recognized-but-unenforced rights. Specifically, a claim requiring a collateral state forum might now belong to four different categories: a nontrial right accruing after a conviction becomes final, (65) a new trial right announced after a conviction becomes final, (66) an extant trial right nonetheless incapable of being enforced by reference to the trial record, (67) or new evidence of innocence. (68)

In the course of explaining one phenomenon elevating the importance of State PCR, then, Section II.A also represents a typology of sorts. The four claim categories specified above explain the growing need for a postconviction forum, but they also meaningfully differentiate the function that such a forum might need to perform. Claim categories might have similarities with others, but they differ in important ways--for example, in terms of the need for counsel, the resource intensiveness of litigation, or the likelihood of prior consideration.

1. Nontrial Rights Accruing After a Conviction Becomes Final

In the process of imposing criminal punishment, state and federal courts enforce a familiar panoply of trial rights: to a lawyer, (69) to confront witnesses, (70) to a speedy trial, (71) against the admission of involuntary confessions, (72) and so forth. Even if there are practical challenges to enforcement in the direct-review chain, when they are collaterally enforced they still remain trial rights--constraints on state activity during trial. Not every claim asserted during postconviction review, however, involves a trial right. Certain rights ripen only after direct review of a conviction concludes, and only then does the underlying constitutional error take place. These nontrial rights attach primarily in death penalty cases involving an inmate that becomes unsuitable for execution.

The only recognized claim in this category is a challenge to execution competency under Ford v. Wainwright (73) and Panetti v. Quarterman. (74) A so-called Ford claimant asserts that, despite a permissible death verdict, the state cannot execute him because he has dropped below a threshold of sanity. (75) The category could eventually include an as-yet-unrecognized "Lackey" claim--that the period between the death sentence and the execution is sufficiently large that an execution would violate the Eighth Amendment. (76) (Lackey theories lurk on the outskirts of auxiliary opinions in various Supreme Court cases, (77) although the Supreme Court has never taken the issue by the horns.) Whatever the theoretical composition of the nontrial right set, it remains small for now.

Nontrial rights differ from collaterally enforced trial rights in very important ways. For these nontrial rights, there is no theoretical possibility of direct-review enforcement because there is no right to enforce. Nor is there the possibility that an inmate would have had a prior right to counsel or resources to explore and factually develop the claim. Finally, the complicated retroactivity framework designed to restrict litigation over Supreme Court decisions postdating the conviction (discussed below) is an awful fit for any claim that necessarily arises after the conviction becomes final. (78)

2. Trial Rights Announced After a Conviction Becomes Final

The second claim category consists of trial rights--but trial rights that the Supreme Court recognizes after direct review of the conviction concludes. For such rights, the term "collateral review" can be a little misleading, because the claim was not recognized when litigation in the direct-review chain was pending. For example, the state may not sentence a juvenile offender to mandatory life without parole (LWOP), but that Eighth Amendment rule was not announced until 2012, in Miller v. Alabama. (79) If Miller is to run in favor of juvenile offenders receiving mandatory LWOP prior to 2012, then there must be a postconviction forum capable of retroactively enforcing the trial right.

Most new constitutional rules do not run in favor of inmates whose convictions are final, because collateral litigation is barred by a principle of "nonretroactivity"--the idea that the rules apply to prospective conduct. (80) A judicial decision is retroactive only if it applies to inmates whose convictions are final and for whom direct review is finished. (81)

The modern rule is one of nonretroactivity for any case not pending on direct review, with two exceptions: (1) some new rule of substantive law that prohibits criminal punishment of certain activity, or that prohibits a certain type of punishment for a certain type of offense; or (2) a "watershed" rule of criminal procedure. (82) There have been a number of decisions falling in the first category, (83) but the second is a null set. (84) (The Supreme Court has indicated that a felony defendant's right to a lawyer would have qualified, (85) but Gideon v. Wainwright (86) was decided before the arrival of the modern retroactivity framework. (87)) For claims asserting an exception, there must be some postconviction forum in which to seek relief.

The most familiar types of retroactive rules involve the death penalty and spring from the Eighth Amendment. Under Atkins v. Virginia, (88) the state may not capitally punish people with intellectual disability (mental retardation). (89) Under Roper v. Simmons, (90) the state may not capitally punish offenders that were juveniles at the time they committed the murder in question. (91) Under Kennedy v. Louisiana, (92) the state may not impose a death sentence for any crime against the person that does not result in death. (93) In each of these cases, the Court used the Eighth Amendment to announce a "proportionality rule" that barred capital punishment for certain categories of offenders or offenses. Even though proportionality jurisprudence has been around for over a century, (94) use of a more modern wholesale form--which produces rules fitting into the first retroactivity exception--is a relatively new phenomenon. (95)

3. Trial Rights Practically Incapable of Enforcement in Direct-Review Chain

Postconviction proceedings are required to process certain claims that, as a practical matter, are incapable of enforcement within the direct-review chain. Two examples should suffice to illustrate the scenario: one involving a Sixth Amendment ineffective assistance of counsel (IAC) claim that trial counsel was deficient and another involving a Fourteenth Amendment "Brady" claim that the prosecution withheld exculpatory evidence. (96)

An IAC claim asserts a violation of the Sixth Amendment right to counsel, (97) which entails a right to effective assistance of counsel. (98) Formally, the elements of the claim are (1) deficient performance and (2) prejudice to the trial outcome. (99) In most situations where trial counsel is inadequate, the very nature of the inadequacy prevents the construction of an appellate record disclosing deficient performance and harm. (100) Even in the limited situations where the appellate record fully captures the inadequacy, there is another barrier to IAC litigation: lawyers do not assert IAC claims against themselves. (101) Unless a state has an elaborate procedural mechanism to allow a new appellate lawyer to return a case to a trial court in order to consider and build a record documenting the trial lawyer's performance, there is virtually no way to enforce the Sixth Amendment right in the direct review chain.

To make the Sixth Amendment example more concrete, consider a common IAC scenario from a death penalty case. Under Wiggins v. Smith, (102) the Sixth Amendment entitles a capital defendant to an attorney that performs an adequate "mitigation investigation." (103) Mitigating evidence demonstrates reduced culpability and therefore bears on the appropriateness of a capital sentence. If trial and appellate counsel are the same person, the problem is obvious. That lawyer is not going to attack her own performance. The problem persists even if the defendant can get a new lawyer on appeal. In that scenario, the appellate record will not disclose what investigatory leads trial counsel ignored (the deficiency prong) or what an adequate investigation would have disclosed (the prejudice prong). Because a Wiggins claim necessitates additional factual development, it is ordinarily incapable of litigation within the direct-review chain and must be raised in postconviction proceedings. (104) The postconviction lawyer will have to identify information trial counsel possessed but mishandled, and must conduct the omitted investigation.

The second example involves a Fourteenth Amendment Brady violation, which gets its name from Brady v. Maryland. (105) A Brady claim asserts that the prosecution suppressed evidence that was exculpatory and material. (106) For reasons that are perhaps more intuitive than in the IAC scenario, "Brady material" is absent from the record on appeal. Brady material is often discovered years (even decades) after trial and direct review conclude. (107) A collateral forum is necessary to introduce the suppressed evidence and to explore the suppression's effect on the ultimate outcome of the case.

Whatever the doctrinal intricacies, IAC and Brady claims do not involve exotic factual scenarios. The most thorough study of federal postconviction cases to date reports that IAC claims were lodged by eighty-one percent of capital inmates and fifty percent of noncapital inmates. (108) Allegations that the prosecution lost, failed to disclose, or presented false evidence were lodged by forty-three percent of capital inmates and thirteen percent of noncapital inmates. (109) Of course, only a fraction of these claims are meritorious, but they also make up the lion's share of claims in which postconviction relief is actually granted. Moreover, both types of claims implicate the basic legitimacy of the criminal justice system. IAC claims challenge the "bedrock" guarantee of a competent lawyer, (110) and Brady claims are a judicial check against overzealous prosecution tactics that are increasingly understood as a widespread structural problem. (111) The guarantee of competent lawyers and honest prosecutors are, along with impartial judges and representative juries, the elemental normative commitments of American criminal process.

4. New Evidence of Innocence

The fourth phenomenon requiring a postconviction forum is not a category of legal claims per se, but a category of evidence. For much of human history, evidence necessary to evaluate guilt degraded over time. Eyewitness memories faded, documentary evidence disappeared, and the exculpatory value of changes in forensic science was limited. (112) In cases where an offender confessed, that confession was regarded as a virtually unimpeachable source of a guilt inference. (113)

Over the last twenty-five years, however, advances in forensic science have radically altered how legal institutions use certain types of evidence. Studies of exonerations in particular have exposed major flaws in the evidence that juries traditionally used to infer guilt against defendants. (114) False confessions, (115) erroneous eyewitness identification, (116) unreliable informants, (117) and junk science are the evidentiary categories most responsible for wrongful convictions. (118)

If the finality of criminal convictions is sacrosanct in part because of the reliability of evidence produced in a trial forum, (119) then at least two aspects of the forensic-science revolution favor collateral litigation. First, new DNA information might be evidence of historical fact that is superior to anything introduced at trial. DNA evidence is capable of categorically excluding a person as a source of biological material. (120) In situations where the conviction was secured before DNA evidence could be introduced at trial or where the DNA evidence is for some other reason newly available, that exclusion can be legally established only in a postconviction forum.

Second, as human knowledge evolves, categories of evidence that were historically treated as strong indicators of guilt are exposed as less reliable. Aggregated data from DNA exonerations, for example, performs a crucial diagnostic function--it shows what types of evidence tended to produce wrongful convictions. (121) The empirical data has shown, for example, that cross-racial eyewitness identifications are particularly unreliable. (122) The data has exposed similar overconfidence in confessions and jailhouse informants. (123)

Legal regimes ordinarily disfavor relitigation of judgments--even criminal convictions--because of the interest in finality. (124) Those regimes, however, do not promote that interest at all costs. One interest particularly likely to inspire maximal caution is the interest of human freedom at issue when the contested judgment is a criminal conviction. (125) If there is new DNA evidence or if advances in human knowledge diagnose as unreliable categories of evidence accounting for the conviction, then legal institutions will naturally want to honor the truth-finding function of criminal process by facilitating collateral review.

Collateral litigation belonging to the innocence category differs significantly from claims belonging to the others I specified previously. Whereas the impetus for providing collateral process in the other scenarios usually involves the inability to make an argument at some prior point in the course of capital litigation, innocence will have been litigated extensively within the direct-review chain. (126) Collateral litigation centered only on innocence does not so neatly implicate constitutional interests denominated as federal. Instead, such relitigation requires reconsideration of guilt, which is at least nominally considered at the core of state competence.

There are other features of innocence litigation that distinguish the collateral process it requires from the process afforded to litigate claims in other categories. The underlying error is perhaps the most profound--a court has criminally sentenced an offender that does not deserve punishment because she did not commit the offense. (127) But there is also a different ratio of claim frequency to merit; innocence is almost certainly substantially overclaimed. In light of the altered calculus, federal institutions might want to encourage a different balance of represented litigation, access to courts, fact development, and finality than they do in the context of litigation involving other categories.

B. Changes to Federal Habeas Law

The four structural phenomena discussed in Section II.A create demand for postconviction process suited to slightly different needs, depending on the claim category. The phenomenon discussed in Section II.B, by contrast, is about which postconviction forum fulfills that demand. State PCR and federal habeas proceedings are in some respects substitutes; (128) they represent distinct collateral forums in which an inmate may challenge a state conviction. When one forum becomes less efficacious, pressure mounts on the other. For that reason, new restrictions on the federal habeas remedy are a source of hydraulic pressure redirecting the critical collateral activity to state courts.

That pressure takes both substantive and procedural form. First, 28 U.S.C. [section] 2254(d) imposes the substantive limit on relief, restricting federal merits review only to cases in which no fairminded jurist could endorse a state disposition against the claim. (129) Second, federal lawmakers have created an interlocking web of procedural barriers to relief, which have the cumulative effect of making federal merits review largely inaccessible for state inmates. The cumulative effect of the substantive and procedural restrictions on the federal habeas remedy--which some prominent scholars now call a "pipe dream" (130)--is to transform State PCR into the pivotal postconviction forum.

1. Substantive Limits: 28 U.S.C. [section] 2254(d)

There is a long running dispute about how deferential a court exercising federal habeas jurisdiction should be towards a state disposition on a federal claim--whether the claim was decided at trial, on appeal, or in State PCR. In 1953, Brown v. Allen (131) declared--or reaffirmed, depending on your point of view--that the federal habeas statute permitted state inmates to obtain de novo federal review of claims that had been decided on the merits in state court. (132) For several decades, state inmates could simply relitigate claims lodged in federal habeas proceedings, even if such review was a second bite at the apple. (133) Speaking very generally, Brown embodied skepticism about the capacity of state institutions to enforce federal rights, (134) and conservatives--who were generally more solicitous of state enforcement--sought to roll Brown back. (135) Those seeking to restrict habeas relief won considerable victories from the Burger and Rehnquist Courts, (136) and the insurgent response eventually triumphed in the 104th Congress.

The Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), (137) passed in response to the bombing of the Murrah Federal Building in Oklahoma City, was loaded with statutory restrictions on federal habeas review. (138) The central AEDPA restriction is the substantive limit on relief, which is now codified at 28 U.S.C. [section] 2254(d). (139) Its significance transcends habeas law, as it touches more generally on how state and federal institutions share responsibility for enforcing constitutional obligations, the role of lower federal courts in developing constitutional law, and how legal institutions may limit remedies for constitutional violations. (140)

Analytically, [section] 2254(d) operates as a modified relitigation bar. If there is a state decision "on the merits" of a constitutional claim, then the relitigation bar kicks in, and it can only be overcome if an inmate shows that the state decision was either legally or factually unreasonable. (141) Even before AEDPA, courts exercising habeas jurisdiction had to defer to state fact-finding, (142) so [section] 2254(d) (1)'s rule involving legal unreasonability was by far the more dramatic change from existing practice.

The first landmark [section] 2254(d)(1) case was Williams v. Taylor, (143) in which the Supreme Court rejected the idea that [section] 2254(d) permitted federal courts to continue to decide constitutional claims simpliciter. Instead, the Court held that [section] 2254(d)(1) permitted federal merits consideration only if there was a sufficiendy grave legal defect in the state decision. (144) A defect in the state decision was only sufficient to disable the [section] 2254(d) relitigation bar, in turn, if it was "objectively unreasonable." (145) The opinion did not define "objective unreasonableness," (146) but it rejected the argument that a state decision was objectively unreasonable only if "all reasonable jurists" would consider it erroneous. (147)

Over the course of the last decade, the Supreme Court has ratcheted up the standard considerably, thereby amplifying the hydraulic effect on State PCR. Specifically, it has adopted by another name the "no reasonable jurist" standard that it rejected in Williams. (148) Section 2254(d)(1) decisions now routinely recite the rule that a state decision is not objectively unreasonable unless every "fair-minded jurist" would disagree with it. (149) The no-fair-minded-jurist standard is a formidable obstacle to a state inmate seeking relief, as a state respondent need only show that a reasonable jurist was capable of agreeing with the contested legal position. (150)

Because of the no-fairminded-jurist standard, federal habeas relief is largely unavailable for claims previously decided on the merits in state court. (151) The hydraulic effect of [section] 2254(d) is unmistakable: an inmate obtaining a state merits disposition does not get federal habeas review and the state postconviction proceeding is, as they say, the ballgame.

2. Procedural Barriers to Relief

In addition to [section] 2254(d), there are many procedural barriers to federal relief that turn on events in State PCR (152): an exhaustion rule, (153) a "procedural default" doctrine that bars consideration of claims forfeited in state court, (154) a statute of limitations, (155) and severe constraints on the availability of evidentiary hearings. (156) First, [section] 2254(b) requires that state inmates exhaust all state remedies before seeking federal habeas relief--effectively forcing State PCR to the front of the postconviction sequence. (Federal courts can deny unmeritorious claims without waiting for exhaustion, but cannot grant meritorious claims before exhaustion is complete. (157)) The exhaustion provision is only disabled if "there is an absence of available State corrective process" or if "circumstances exist that render such process ineffective to protect the rights of the [inmate]," (158) but those exceptions are almost never satisfied. (159)

Second, the "procedural default" doctrine generally bars federal habeas relief on a claim that an inmate forfeited in state proceedings. (160) If a state court rules against an inmate on an adequate and independent state procedural ground, then the federal courts are to give effect to the forfeiture and are not to reach the merits of the claim unless the forfeiture is excused. (161) Procedural default rules generally ensure that a state inmate must litigate a claim in State PCR or lose it forever.

Third, AEDPA enacted a one-year statute of limitations that is statutorily tolled only during the pendency of State PCR. (162) There was no limitations period before AEDPA, (163) although some courts used laches to limit untimely filing. (164) The limitations period creates a temporal incentive to initiate State PCR as soon as possible. The pendency of State PCR is the only thing that statutorily tolls the limitations period, (165) so a failure to quickly seek state remedies can create a disqualifying defect in a federal habeas claim.

Finally, there are major restrictions on fact development in federal court. An inmate seeking to avoid [section] 2254(d) may only use the state record--there is no fact development available in federal habeas proceedings. (166) Fact development becomes available only after an inmate has cleared the relitigation bar. (167) Even in scenarios where federal fact development is permitted, new evidence and hearings are subject to substantial limitation. Under [section] 2254(e)(1), state factual determinations are presumed correct and can be reversed only upon clear and convincing evidence to the contrary. Under [section] 2254(e)(2), federal hearings are unavailable for inmates that "failed to develop" claims in state court, unless they meet either a new-law or a new-fact-and-prejudice gateway. (168) The upshot is that, even in cases where limited federal relitigation is permitted, inmates are basically restricted to the factual record developed in State PCR.

These federal procedural rules reinforce the primacy of State PCR in the postconviction sequence. Exhaustion and procedural default rules combine to ensure that states have the first crack at any issue capable of being decided on the merits, the statute of limitations ensures that state inmates avail themselves of state remedies expeditiously, and an inability to supplement the record in federal litigation means that most major fact development must happen during state postconviction process.


The American legal experiment is one in which state courts must enforce federal law, (169) and state courts enforce federal laws that correspond to federal interests. One would therefore expect federal intervention in state proceedings to follow logically from affected federal interests. Indeed, when State PCR was largely a forum for redundant legal challenge, or when federal habeas process effectively secured the interests embodied in federal law, federal intervention in state PCR remained minimal. Part II explained how those conditions changed.

Part III provides a framework for mapping the corresponding adjustment in federal intervention. Different federal interests require different federal responses, and three modes of intervention dominate: (1) constitutional law constraining State PCR (constitutional-law intervention); (2) use of federal habeas rules as incentives for states to improve representation and other procedural features of State PCR (habeas intervention); and (3) supplementation of resources for state postconviction litigation (resource intervention).

One clarification is in order. I exclude from scrutinized intervention the scenario in which the Supreme Court directly superintends State PCR by reviewing the underlying merits of a claim of trial right. It is in some sense true that, whenever the Court uses its power to hear and adjust the scope of such a right--like the rights to effective counsel (170) or to exculpatory evidence in the prosecution's possession (171)--it necessarily (if indirectly) dictates the emphasis of state postconviction remedies. I nevertheless exclude such activity from the analysis because it regulates State PCR only incidentally, as it adjusts enforcement of the underlying trial right across all phases of criminal and postconviction litigation. In short, the object of the inquiry is not to regulate the State PCR practices. The interventions discussed below, by contrast, are procedural rules specifically targeting enforcement during State PCR, and are therefore the more appropriate subjects for inquiry into the federal enforcement priorities for that phase of criminal process.

A. Constitutional-Law Intervention

The first mode of federal intervention is the most intuitive: rules imposed by the Supreme Court when it undertakes appellate review of state postconviction adjudication. By "constitutional-law intervention," I do not mean the Supreme Court's direct review of whether State PCR correctly answered a question of constitutional error at trial, but intervention in which the Supreme Court declares that a rule of constitutional law directly constrains State PCR itself.

Until recently, constitutional-law intervention was virtually nonexistent. Only recently has the Supreme Court retreated from the basic premise of its abstention: that there exists no right to any State PCR, so there cannot be ancillary rights to things like lawyers and expert services. Still, use of constitutional-law intervention remains sparing, and its deployment signals an unusual level of institutional interest.

1. Constitutional Right to State PCR Itself

The foundational assumption behind constitutional-law abstention is that the Federal Constitution requires no State PCR at all. (172) The logic says that, if the Due Process Clause of the Fourteenth Amendment does not establish a right of direct appeal, then, a fortiori, it cannot establish a right to state postconviction process. (173) Nor does the federal habeas privilege, guaranteed against suspension in Article I, Section 9, (174) require a state postconviction forum. (175) To the extent that constitutional law applies to State PCR, it usually constrains the operation of State PCR after a state has made a decision to provide that process. (176) That the Federal Constitution does not require State PCR recurs as an express premise of greater-includes-the-lesser holdings denying other constitutional rights asserted during state collateral process. (177)

The only existing exception to the categorical rule that the Federal Constitution requires no State PCR involves an inmate's competency to be executed. (178) The Supreme Court first declared the Eighth Amendment proportionality rule against executing insane offenders in Ford v. Wainwright. (179) The (constitutional) rules for "execution competency" constrain states based on an offender's functioning at the time he is to be killed. Execution-competency rules are distinct from rules about when an inmate is not guilty by reason of insanity (180) and is not competent to stand trial. (181) Because a death-sentenced inmate's competency for execution is necessarily assessed after trial (and usually long after trial), it can only take place in a postconviction proceeding.

Under Justice Powell's Ford framework, (182) advanced in a median concurrence but formally declared "'clearly established' law" in Panetti v. Quarterman, (183) an inmate making a "substantial [ ] showing" of incompetency is constitutionally entitled to a "fair hearing" and an "opportunity to be heard" on the claim. (184) Ford's Eighth Amendment rule against executing insane offenders is clear enough, but the auxiliary procedural rights--springing from the due process clauses of the Fifth and Fourteenth Amendments--are more general. States have enacted implementing statutes and, (185) where necessary, federal courts have particularized the rule in individual cases. (186)

The reason for the unique treatment of Ford claims is intuitive. Ford claims do not ripen until a state sets an execution date. If the Federal Constitution is to be the source of any rules for Ford claims at all, then those rules must regulate process afforded at the postconviction phase of the capital litigation.

2. Constitutional Right to Counsel During State PCR

The Sixth Amendment entitles a criminal defendant to "the assistance of counsel for his defence." (187) The Supreme Court gradually expanded the reach of the Sixth Amendment right to an appointed lawyer--first to capital defendants too impaired to defend themselves, (188) then to all federal defendants, (189) then to any capital defendant irrespective of impairment, (190) and, in Gideon v. Wainwright, (191) to any felony defendant in state or federal court. (192) The Sixth Amendment right extends to any critical phase of the adversary proceeding in the trial court, including the plea bargaining process. (193) On appeal, a convicted defendant has a due process and equal protection right to an appointed lawyer in "first-tier review"--that is, in the first appellate forum available for reviewing a particular determination. (194) In State PCR, however, the Supreme Court has not wavered in its refusal to recognize any constitutional right to counsel. (195) The most direct rejection of the possibility was articulated in Pennsylvania v. Finley. (196)

Finley rejected both due process and equal protection theories for a right to counsel. On the due process question, the Supreme Court analogized to Ross v. Moffitt, (197) which held that fundamental fairness did not require provision of counsel to use "as a sword to upset the prior determination of guilt." (198) Moffitt had explained that because a state need not provide appellate process at all, it need not provide a lawyer for whatever discretionary appellate process it electively maintains. (199) Using a familiar greater-includes-lesser rationale, Finley explained that "[p]ostconviction relief is even further removed from the criminal trial than is discretionary direct review," (200) and so the Court concluded that due process could not require a right to counsel in State PCR. (201)

Finley also rejected a Fourteenth Amendment equal protection theory that indigent inmates needed a right to counsel to have the same access to courts as inmates of means, (202) although such a rejection was a heavier lift. In Douglas v. California (203) the Supreme Court had used the Equal Protection Clause to declare a constitutional right to appointed counsel in the first as-of-right appeal. (204) (The distinction between the as-of-right and discretionary appeals explains the difference between Douglas and Moffitt.) In Finley, however, the Court explained that withholding appointed counsel for discretionary appeals or State PCR did not violate equal protection because the inmate's "access to the trial record and the appellate briefs and opinions provided sufficient tools for the pro se litigant to gain meaningful access to courts that possess a discretionary power of review." (205)

The reasons Finley gave for categorically rejecting a postconviction right to counsel no longer describes the category very well. First, the premise that no postconviction process is constitutionally required is now false. The Constitution requires that states provide process to consider content described in subsection II.A.l: nontrial rights that accrue after a conviction becomes final. Second, the premise that all unrepresented inmates can put themselves on equal footing with inmates of means because they have access to the trial and appellate records is also inaccurate. For each of the four different functions specified in Section II.A, inmates require extensive access to content outside those records in order to enforce constitutional rights.

Notwithstanding the deteriorating accuracy of its assumptions, Finley has been affirmed many times. (206) The major attempt to subvert Finley involved not the type of claim at issue, but the capital versus noncapital status of the claimant. In Murray v. Giarratano (207) however, the Supreme Court rejected the proposition that a capitally sentenced inmate possessed a unique right to state postconviction counsel. (208) Giarratano failed to produce a majority opinion, and the median vote came from Justice Kennedy, who was willing only to say that there was no right-to-counsel violation on the facts before the Court. (209) As Finley and Giarratano are currendy interpreted--and however one might parse the Justice alignment in the latter--there is no constitutional right to counsel in State PCR, without respect to the function that State PCR performs in a given case. (210)

3. The Applicability of Retroactivity Rules in State PCR

The newest intervention involves the Supreme Court's 2016 decision to apply federal retroactivity rules, by force of the Federal Constitution, during State PCR. (211) Understanding the significance of this particular constitutional-law intervention requires some additional discussion of retroactivity and basic familiarity with the potentially retroactive substantive rule. As for the rule: In Miller v. Alabama, (212) the Court held that the Eighth and Fourteenth Amendments precluded juveniles from receiving a mandatory sentence of life without parole (LWOP). (213) Miller naturally created a retroactivity question: Should juvenile offenders serving a mandatory LWOP when Miller was decided be able to invoke it to obtain a resentencing?

The framework from Teague v. Lane (214) governs retroactivity questions in federal habeas litigation. (215) Supreme Court cases decided after a trial concludes are classified as "old law" or "new law." "Old law" is treated as an explanation of existing precedent and postconviction claimants may invoke the later-in-time decision as a ground for relief. (216) "New law" is treated as a departure from precedent and is generally unavailable to inmates whose convictions are final, (217) except in two situations: (1) when it announces a new substantive rule that bars a finding of guilt or punishment for a particular offense; and (2) when it is a "watershed" rule of procedure. (218) The pressing question was whether Miller fit within the first exception.

The Supreme Court took up Miller's retroactivity in Montgomery v. Louisiana. (219) Even if Miller qualified as a new substantive rule for Teague purposes, there was a lurking jurisdictional question about whether state courts had to use the Teague framework at all. On one view, in refusing to give retroactive effect to Miller, Louisiana was simply applying a state limit on a nonmandatory state remedy; the federal retroactivity framework was arguably irrelevant. (220) Phrased in the black-letter jurisdictional language of federal-courts jurisprudence, the state ground for refusing relief was independent of any federal question (221)--because there was no federal law requiring Louisiana to provide a retroactive remedy. (222) Montgomery, however, declared that states had to retroactively apply new substantive rules during their PCR. (223) The Court, however, refused to extend the rule beyond the new-substantive-law category; it expressly reserved the question as to whether states had to give retroactive effect to watershed rules and it did not address the need to give retroactive effect to new precedent applying "old rules." (224) Montgomery is a classic split-the-baby case that yields more questions than answers, but is to date the Court's most robust attempt to explain the substantive-law focus of its State PCR intervention.

B. Habeas Intervention

Federal institutions use habeas intervention to encourage attention to the quality of representation and process during State PCR. Insofar as federal institutions can adjust the availability of federal habeas relief to reflect the quality of State PCR, they can incentivize preferred state practices.

Indeed, because federal habeas incentives link the availability of a state's federal habeas defenses to the efficacy of judicial remedies it administers, incentives create some dignitary and economic interest in good state postconviction hygiene. The dignitary interest arises from the state's preference to avoid a finding that its postconviction outcome was in some way defective. The economic interest arises from the state's preference to avoid both merits litigation in federal court (as a litigant) and any additional state proceedings necessitated by federal relief (as an adjudicator). As I explain below, however, these incentives are very muted and therefore represent the lowest-level federal intervention.

Federal habeas relief was not always linked to state postconviction process; for many years, federal judges reviewed constitutional claims de novo, no matter what happened in state court. (225) Eventually, starting in 1977, the Burger Court began to modernize decisions linking state outcomes with the availability of federal review. The first major point on the modern trend line was Wainwright v. Sykes, (226) when the Court held that even blameless inmates could not recover claims forfeited in state proceedings. (227) Sykes was the prototypical incentive--if state action caused the forfeiture, only then could the state inmate excuse it.

The concept of habeas incentives got a boost with the "Powell Committee Report," the work product of a committee charged by then-Chief Justice Rehnquist to study delays in the imposition of capital sentences. (228) The Committee ultimately concluded that a major cause of delay was the need for capital inmates, not always represented by competent counsel, to bounce back and forth between federal and state habeas proceedings to exhaust claims. (229) A considerable amount of responsibility for that state of affairs, the Powell Committee determined, was attributable to inadequate representation during State PCR. (230)

The Powell Committee therefore recommended a legislative quid pro quo for litigation initiated by capitally sentenced inmates: states guaranteeing adequate State PCR would benefit from enhanced procedural defenses during federal habeas proceedings. (231) The Committee's so-called "opt-in" recommendations found their way into federal legislation, but in an altered form that mooted the habeas incentives. When Congress passed AEDPA, it incorporated the Committee recommendations alongside a set of more general procedural restrictions applicable to both capital and noncapital cases--and applied those restrictions without respect to adequate representation during State PCR. (232) Because the default restrictions were so favorable to the states, none bothered to make changes necessary to avail themselves of the opt-in mechanism. The opt-in experience is emblematic of habeas incentives generally, insofar as such incentives have done little to improve the underlying resource and representation problems in State PCR.

The Supreme Court ultimately got into the habeas incentives game when it remade procedural default doctrine to account for defects in state postconviction representation. Recall that in Sykes, (253) the Supreme Court held that an inmate could not excuse state forfeiture without a showing of cause and prejudice. (234) The Court thereafter settled on a general rule that inadequate representation during State PCR was not sufficient to establish the excuse. (235) The rule was expressly built on the idea of agency--the professional negligence of a state postconviction lawyer was charged to the inmate, not to the government. (236) The Court carved out a narrow exception to that proposition when it decided Martinez v. Ryan in 2012. (237) In Martinez, the Court held that a federal court could entertain an otherwise defaulted IAC claim if the default was the result of inadequate state postconviction representation. (238)

Martinez is a window into the Supreme Court's strategic deployment of habeas incentives. For an IAC claim--which must usually be raised for the first time in a postconviction proceeding--the Court was willing to link state representation during State PCR to the availability of the habeas remedy. The incentive is straightforward: By promoting adequate state postconviction representation, states avoid having to litigate claims on federal habeas review. By failing to guarantee adequate state representation, they expose themselves to incremental litigation risk.

The Supreme Court has created similar habeas incentives for other procedural defenses. In Holland v. Florida, (239) for example, it devised an incentive related to the federal statute of limitations. (240) Specifically, if a state inmate files an otherwise untimely federal habeas petition because of serious defects in state postconviction representation--albeit defects that are more than mere negligence--the limitations period is equitably tolled and the claim is not time-barred. (241) Holland's incentive structure is also straightforward. In order to guarantee the availability of the federal limitations defense, the state must provide counsel that does not commit professional misconduct.

Across procedural doctrine--procedural default and the statute of limitations--federal institutions have formulated habeas incentives as a means of encouraging state focus on State PCR process and representation. The one area in which the Supreme Court has moved away from an incentives-based approach is with respect to 28 U.S.C. [section] 2254(d), the rule for state merits dispositions, discussed initially in subsection II.B.1. Section 2254(d) requires that when a state court decides the merits of a claim, federal habeas review begins with the state decision and not with an assessment of the underlying claim. If the state decision is legally or factually unreasonable, only then may the federal court consider the merits of the claim. On its face, [section] 2254(d) appears to draw a powerful link between reasonableness of state decisions and the availability of federal relief.

The Supreme Court's interpretation of [section] 2254(d), however, has suppressed that linkage. First, the Supreme Court has watered down the concept of what it means to be "reasonable." As explained above, a state decision is now considered legally unreasonable only if every "fair-minded jurist" would reject it. (242) In light of the no-fairminded-jurist standard, inmates face almost insurmountable hurdles in showing that state merits dispositions are legally unreasonable. Second, in Harrington v. Richter, (243) the Supreme Court mooted what might have been an incentive for desirable state postconviction process when it held that, for cases in which the state court produced no reasoned opinion in support of the decision, a federal court could not reach the merits of the underlying claim unless every conceivable line of reasoning sustaining the decision was unreasonable. (244) Third, in Cullen v. Pinholster, (245) the Court determined that inmates subject to adverse state merits determinations could not use evidence outside the state record to satisfy [section] 2254(d). (246) Pinholster effectively locks inmates into the record produced during State PCR, thereby entrenching any representational deficiency tainting that process. The decision is therefore of a piece with Richter and the no-fairminded-jurist standard. The Roberts Court has largely refused to use [section] 2254(d) as a means of promoting transparent state postconviction process, notwithstanding that one might defensibly interpret the plain language of the provision to provide habeas incentives for writing opinions, using strong reasoning, and facilitating fact development.

Habeas incentives also differ somewhat from other modes of intervention in terms of orientation, at least insofar as the most recent innovations appear preoccupied with the Sixth Amendment right to counsel rather than with substantive constraints. For reasons I ultimately set forth in Part IV, however, that deviation is less than meets the eye. Habeas incentives are, in effect, a lower order intervention, and they are appearing largely in addition to less apparent incentives already centered on Eighth Amendment substantive constraints.

C. Resource Intervention

Federal institutions also intervene in State PCR, albeit indirectly, through a process of statutory spillovers. In such spillover intervention, resources earmarked for federal habeas proceedings are used to support state postconviction litigation. (247) Because resource intervention is keyed to the appointment of a federal habeas lawyer for an indigent federal habeas claimant, the benefits are realized only in the small subset of state postconviction proceedings that follow the federal appointment. Even then, resources may be available only for discrete types of state postconviction litigation.

Resource intervention happens because some postconviction claimants are indigent and may obtain, by statute, federal counsel and fact-development resources--i.e., lawyers, investigators, and experts. (248) The appointment rules in State PCR vary by jurisdiction, but no state appoints lawyers for every state postconviction litigant. Even for inmates on death row, not every state provides legal representation. (249) Federal courts make appointments at the beginning of federal habeas litigation. Even though these appointments are made in federal court, they may cover state representation occurring after that appointment. States are loath to appoint counsel in most cases, reasoning that the vast majority of state postconviction claims are nonmeritorious.

The federal resource statute for noncapital inmates provides that an inmate with a federally appointed lawyer "shall be represented at every stage of the proceedings... including ancillary matters appropriate to the proceedings." (250) 18 U.S.C. [section] 3599(e), the provision specifically available to capital inmates, requires that the representation persist throughout "every subsequent stage of available judicial proceedings," for "all available post-conviction process," and for "other appropriate motions and procedures." (251) Moreover, [section] 3599(e) contains multiple references to proceedings that, for state inmates, must be outside the federal process, including "competency proceedings" and "executive or other clemency." (252)

The most prominent Supreme Court decision interpreting the federal appointment provisions for capital inmates is Harbison v. Bell, (253) in which the Court determined that "executive or other clemency" included state clemency provisions. (254) The Court refused to read "subsequent stage [ ] of available judicial proceedings" as statutory authorization for appointed counsel to represent inmates in state postconviction proceedings, reasoning that, by virtue of the exhaustion requirement, state postconviction proceedings were not a "subsequent" stage of process within the meaning of the statute. (255) In footnote seven, however, the Court held that, pursuant to [section] 3599(e)'s reference to representation for "other appropriate motions and procedures," federally appointed habeas counsel may be required to exhaust claims subsequent to the federal appointment on a "case-by-case" basis. (256) After Harbison, [section] 3599 appears to provide capitally sentenced state inmates with federally appointed lawyers to litigate Ford claims, to seek clemency, and, in certain cases, to exhaust claims.

The most important anomaly of resource intervention is therefore the most basic: inmates having undertaken out-of-order litigation may have the benefit of federal lawyers and investigative services when they return to state court to exhaust a claim. Inmates who appropriately exhaust claims, thereby encountering State PCR before they make any contact with a federal court, will not.

Resource intervention tracks almost nothing of functional significance. It siphons resources to whatever state postconviction litigation happens to follow the federal appointment. By way of one example, the federal appointment covers state Ford litigation that follows the appointment, but not litigation over other death penalty exemptions, such as intellectual disability. (257) By way of another, there is a provision permitting courts to make appointment of counsel for inmates serving noncapital sentences retroactive--and therefore a means of compensating lawyers for state postconviction activity--but not a provision doing the same for death-sentenced inmates. (258)

The arbitrariness of resource intervention persists in large part because courts are caught between, on the one hand, the impulse to use the federal statute to improve representation in State PCR and, on the other, the unforgiving language in the federal statutory provisions that govern appointment of counsel in federal habeas cases. Because resource intervention affects only state process commencing after the federal appointment, the spillover effect is limited largely to successive state postconviction litigation. (259) If one were to redirect federal resources from federal to state proceedings, a utility-maximizing designer would siphon resources to State PCR that effectively promote federal interests. A regime that redirects federal resources to successive state postconviction litigation is moving those resources to perhaps their lowest valued use. (260)

States often operate on the same principle that constrains federal habeas proceedings--that inmates should have one bite at the apple. (261) As a result, successive state postconviction litigation is severely restricted. (262) Relitigating claims decided in some prior state postconviction proceeding is next to impossible, and inmates generally have to meet strict exceptions to litigate new claims. (263) The cost to an inmate of enforcing rights in successive state postconviction litigation, as opposed to initial state postconviction litigation, is enormous.

Resource intervention therefore protects most effectively (if inadvertently) the narrow category of claims that I analyze in Part IV: substantive constraints incapable of being enforced at trial. The best example is the special treatment reserved for Ford litigation over whether an inmate is competent for execution (264)--litigation that requires a successive state forum. Enforcement of other restrictions on punishment that are new, substantive, and retroactive will also require a successive state proceeding, (265) because such claims can usually be asserted only after the first round of State PCR concludes.

Federal intervention occurs largely because State PCR underenforces federal law. Even though State PCR performs other functions, it is the application of federal rules of decision--almost always constitutional law--that is the primary concern of federal institutions. Federal intervention takes the three general forms that I describe here and, in light of the structural changes explored in Part II, that intervention is accelerating. The question that interests me, and that I turn to now, is whether that intervention conforms to a meaningful pattern.


In Part IV, I argue that federal intervention is best understood as a form of Substantive-Constraint Exceptionalism--a paradigm under which a small category of "substantive" limits on conviction and punishment are singled out for special treatment. (I use the term "substantive-constraint" rather than "Eighth Amendment" to modify Exceptionalism because the category is slightly broader than a pure set of Eighth Amendment rules.) Exceptionalism has the imperfections of any explanatory theory, but it rather robustly predicts the constitutional inflection of federal intervention. The Exceptionalism is "transmodal" in that it operates across intervention categories, but is most visible in the Supreme Court's constitutional-law intervention. The Exceptionalism embedded in the other modes of intervention has softer edges in that its priorities can include some procedural constraints.

The presence of those softer edges itself reveals a familiar pattern. When intervention requires more active judicial supervision, the Supreme Court shifts that enforcement role to lower federal judges. The Court reserves constitutional-law intervention for a narrow sliver of substantive-law constraints, but enlists lower federal courts to enforce interventions that touch more broadly applicable procedural constraints. Put differently, interventions are more Exceptionalist when they require the Supreme Court to do more of the enforcement work. The dynamic by which lower federal courts are effectively deputized to enforce nonsubstantive constraints is reminiscent of the Supreme Court's enforcement approach for trial rights announced during the Warren Court's criminal procedure revolution. (266)

A. The Concept of Substantive Constraint

The precise contours of what I mean by "substantive constraint" are not self-evident, but what I mean by the term is this: a constraint is substantive if, notwithstanding perfect procedure, it bars conviction or the imposition of punishment based on attributes of the offender or offending conduct. New substantive constraints include new constitutional rules against conviction for certain offenses, or new decisions excluding certain offense conduct from the punitive coverage of a statute. They also include new constitutional rules against certain penalty-offense-offender combinations, or decisions that a statute no longer permits such a combination that it previously permitted.

Because federal courts do not narrow state criminal statutes responsible for state convictions and sentences, the category of "substantive constraints" that are the subject of State PCR interventions are substantive constraints imposed by the Federal Constitution. (267) If a court invalidates a statute under the First or Eighth Amendment, or as unconstitutionally vague, then state inmates convicted and serving a voidable sentence will seek to invoke the constraint in State PCR. Two features make these new substantive constraints intuitive candidates for exceptional treatment by the federal institutions that design interventions: (1) the claims are necessarily available only to the limited set of state inmates subject to the voided punishment and (2) the timing of new substantive constraints necessitates clear rules for postconviction enforcement.

Notwithstanding the theoretical scope of the term, the substantive constraints that federal institutions care most about are Eighth Amendment "proportionality" rules. Although I have referenced proportionality challenges throughout this Article, it is now worth pausing to explain their features in a bit more detail. As most lawyers know, the Eighth Amendment contains the so-called Cruel and Unusual Punishments Clause: "Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted." (268) The Clause is, among other things, the source of categorical rules against certain penalty-offense-offender combinations. By "proportionality" rules, I mean Supreme Court decisions declaring that the Eighth Amendment forbids conviction or punishment for a particular crime based on characteristics of the offense or the offender. (269)

The first major reference to "proportionality" in Eighth Amendment law was in Weems v. United States. (270) Weems was decided in 1910, when the Supreme Court declared "it is a precept of justice that punishment for crime should be graduated and proportioned to [the] offense." (271) Weems also foreshadowed the fluidity that would become a basic feature of proportionality doctrine, declaring that the content of Eighth Amendment rules was "not fastened to the obsolete but may acquire meaning as public opinion becomes enlightened by a humane justice." (272) The Court cemented its commitment to that fluidity in 1952, holding that a punishment was not proportional to a crime if it violated "the evolving standards of decency that mark the progress of a maturing society." (273)

Even after the Supreme Court incorporated the Eighth Amendment and its evolving-standards content against the states in 1962, (274) proportionality rules were spare. (275) In Coker v. Georgia, (276) the Court issued a proportionality rule against executing an offender who raped (but did not murder) an adult woman. (277) The first three decades of post-Coker precedent centered primarily on death penalty cases, (278) which naturally limited the set of offenders to which the Eighth Amendment rules applied. (279) For example, the number of inmates facing the death penalty and with viable claims that they are intellectually disabled under Atkins v. Virginia (280) or insane under Ford v. Wainwright (281) is only a subset of capitally sentenced offenders (282)--a subset that is itself only a fraction of all offenders. To the extent that the Court's important proportionality jurisprudence ventures beyond the death penalty, it involves LWOP for juveniles, (283) a sentence category that also entails natural limits on the size of any potential claimant class. (284) The important point is that, as a percentage of convicted inmates, the number touched by Eighth Amendment proportionality rules is small.

Another essential quality of proportionality claims--that after some period of penal experimentation, a punishment drifts outside the community's punishment norms--means that the Eighth Amendment rule will be announced while some inmates are actually subject to the newly forbidden penalty-offense-offender combination. There were intellectually disabled offenders on death row when the Supreme Court decided Atkins (285) there were juvenile offenders serving mandatory LWOP when it decided Miller (286) and so on and so forth. As a result, any proportionality rule will present pressing retroactivity questions for convicted inmates belonging to the verboten category.

B. Transmodality

Substantive-Constraint Exceptionalism operates transmodally--across constitutional-law intervention, habeas intervention, and resource intervention. The transmodality of the intervention is significant insofar as it reveals a consistent emphasis on enforcing substantive constraints. Moreover, because the interventions are mutually reinforcing, they combine to form an environment in which such constraints, and particularly Eighth Amendment claims, enjoy a uniquely privileged litigation status.

I do not mean to suggest that each mode of intervention conforms perfectly to an Exceptionalism model. Constitutional-law intervention does focus, for all intents and purposes, exclusively on substantive constraints. The other modes of intervention, however, have softer edges--which is to say they do sometimes co-prioritize procedural constraints on convicting and sentencing.

1. Exceptionalism in Constitutional-Law Intervention

The easiest-to-understand Exceptionalism is that which dominates constitutional-law intervention. Since the advent of State PCR, the Supreme Court has repeatedly intoned the premise that the Federal Constitution does not require the process at all. (287) The constitutional law of State PCR has been built, bolt by decisional bolt, around that foundational principle. In decisions turning back other constitutional rights, (288) the Court regularly relies upon the greater-includes-the-lesser proposition that, if the Federal Constitution requires no State PCR, then it can require no other rights to attach there. The two major exceptions to this general constitutional abstention strongly conform to the Exceptionalism model.

First, the only substantive constitutional rights that directly constrain State PCR are built around an Eighth Amendment proportionality rule. Ford v. Wainwright was the 1986 decision in which the Supreme Court announced the rule against executing insane (incompetent) offenders. (289) Panetti v. Quarterman was the 2007 decision announcing a more precise definition of competency (290) and affirming that an inmate with a "substantial" Eighth Amendment claim is entitled to a modicum of State PCR bootstrapped to a determination that the underlying Ford claim has some merit. (291) The only clear candidate to be added to the list of constitutional rules constraining state postconviction process is the so-called Lackey claim--also an Eighth Amendment challenge--that the Constitution forbids states from executing offenders that have spent too many years on death row. (292)

Second, Substantive-Constraint Exceptionalism explains a more complex constitutional-law intervention: the 2016 rule from Montgomery v. Louisiana, (293) in which the Supreme Court held that the Federal Constitution requires that states apply a decision retroactively if it announces a new rule of substantive law. (294) The vehicle for the decision was the Eighth Amendment proportionality rule against imposing mandatory LWOP against juvenile offenders, (295) and that vehicle is no accident.

Montgomery nicely illustrates Substantive-Constraint Exceptionalism not just because of what it is, but because of what it is not. The preexisting retroactivity framework required that federal courts apply new decisions retroactively in two other contexts--when the new decision is "old law" that is "dictated by precedent" and when it is "new law" that represents a "watershed" rule of procedure. (296) In Montgomery, however, the Court explicitly refused to require states to give retroactive effect to these other types of decisions. The result, then, is a special retroactivity carve-out applicable only to substantive constraints.

2. Habeas Intervention

Substantive-Constraint Exceptionalism has its softest edges with respect to habeas intervention, because habeas interventions consist of judge-made doctrine that is less likely to focus exclusively on substantive constraints. Insofar as Congress calibrates habeas incentives, it tends toward treatment that lumps substantive and procedural constraints together. The relitigation bar in 28 U.S.C. [section] 2254(d) does not operate differently depending on whether the state merits adjudication involved an Eighth Amendment proportionality claim or some other constitutional challenge. (297) Moreover, virtually none of the other federal claim-processing provisions pegged to state postconviction process make any facial distinctions based on the type of claim presented. (298)

To the extent that there are priorities for habeas intervention, they tend to find nonstatutory expression through the judge-made habeas law announced and refined by the Supreme Court. The priorities are most visible in the various ways for state inmates to revive claims that were procedurally defaulted during State PCR. One way that state inmates can revive otherwise forfeited claims is by showing that a failure to hear the claim would constitute a "miscarriage of justice." (299) The miscarriage-of-justice gateway is usually regarded as the gateway by which inmates revive claims associated with innocence, (300) but it is also used in capital cases to revive claims that a state inmate is "actually innocent of the death penalty." (301) Such "death ineligibility" claims, which invoke substantive constraints, usually take the familiar form of Eighth Amendment exemptions involving attributes of the offender or offense. (302)

If the miscarriage-of-justice gateway was the only mechanism for reviving otherwise defaulted claims, then habeas intervention would represent a harder-edged Exceptionalism. There is, however, another gateway through the otherwise-applicable bar to federal consideration of forfeited claims: the so-called "cause-and-prejudice" rule. (303) To satisfy the cause-and-prejudice gateway, an inmate must show (1) that some external phenomenon prevented state litigation of the claim (cause) and (2) some quantum of merit (prejudice). (304)

Until fairly recently, the cause-and-prejudice doctrine had no substantive valence because it did not privilege any particular type of constitutional challenge. The cause-and-prejudice gateway always worked the same way, regardless of what the underlying constitutional claim was. One of the most important rules was that, without respect to the underlying constitutional claim, the negligence of state postconviction counsel could not constitute "cause." (305) Practically speaking, such a rule meant that mistakes of postconviction lawyers were always attributed to clients and that they could never provide a basis for reviving claims.

Perhaps the single most significant development in recent federal habeas law involves a revision to this rule--the treatment of claims that were forfeited because state postconviction counsel was negligent. Martinez v. Ryan (306) and Trevino v. Thaler (307) represent attempts to ensure that state inmates get at least one bite at the apple on IAC claims incapable of being litigated during the direct-review chain or by an inadequate state postconviction lawyer. Doctrinally, Martinez and Trevino exempt certain IAC claims from the more general rule that a state inmate cannot use the negligence of state postconviction counsel to satisfy the cause-and-prejudice gateway. (308) The result of Martinez and Trevino is that habeas incentives target more than just substantive constraints and innocence claims. The Court-made habeas intervention is designed to be particularly solicitous of Sixth Amendment claims, largely on the ground that, absent meaningful state postconviction process, the procedural right would go unenforced. (309)

3. Resource Intervention

Substantive-Constraint Exceptionalism also suffuses resource intervention. Its influence on resource intervention does not appear to be the result of conscious institutional design in the same way it is for constitutional and habeas interventions. It is nevertheless noteworthy that even if the federal branches are not coordinating to implement a particular approach, the complex institutional interaction nonetheless produces a set of legal rules approximating more deliberate decisionmaking.

Some statutory background is necessary to understand how Exceptionalism affects resource intervention. The Criminal Justice Act (CJA) is the collection of federal statutory provisions, largely codified at 18 U.S.C. [section] 3006A, ensuring legal representation of federal postconviction litigants. (310) 18 U.S.C. [section] 3599, which Congress "sp[un] off from the CJA when it revived the federal death penalty in 1988, (311) is specific to death penalty cases, and requires federal habeas counsel to represent capitally sentenced inmates "in such competency proceedings... as may be available to the defendant." (312) The competency proceedings that take place after the appointment of federal habeas counsel, of course, are the means by which courts adjudicate Eighth Amendment proportionality claims under Ford and Panetti. (313) Eighth Amendment competency challenges happen after the federal appointment in death cases because, unlike many other challenges, a Ford challenge ripens only after the state sets an execution date, which is usually after federal habeas proceedings commence.

Resource intervention bears other substantial hallmarks of Substantive-Constraint Exceptionalism. Recall that the CJA provisions providing state inmates access to federal lawyers and resources generally apply only after an appointment is made in the initial federal proceeding. That sequence necessarily excludes the possibility that an inmate has such services and resources made available before the federal habeas proceeding begins--when State PCR usually takes place and considers procedural error. (314) Instead, federal lawyers and resources are available only for State PCR that happens to follow the federal appointment. (315)

The priorities embedded in resource intervention therefore reflect the types of claims that state inmates typically litigate in a successive posture, which fall roughly into two buckets: "new facts" and "new law." (316) If a new factual or legal basis of a claim materializes after a federal habeas appointment has been made, then the federal statute effectively permits the inmate to use federal lawyers and resources to support the state postconviction activity necessary to support the litigation. (317) "New facts" can be predicates for a variety of constitutional challenges, but "new law" refers generally to new substantive constraints--particularly Eighth Amendment proportionality decisions.

In short, the "new law" gateways for successive state postconviction litigation are largely gateways for litigating new substantive constraints, (318) and successive state postconviction litigation is the primary form of State PCR supported through federal resources. (319) When the Supreme Court recognizes new substantive constraints on punishment, there will always be some state enforcement activity that follows the federal habeas appointment. Even if the substantive constraint is capable of being applied prospectively by observing the rule in the direct-review chain, inmates whose convictions are final will still have to enforce the rule collaterally. (320)

C. Exceptionalism and Dispersed Enforcement

On one view, Martinez and Trevino appear to complicate an otherwise clear picture of Substantive-Constraint Exceptionalism. On a better view, softer-edged habeas intervention enriches it. Through a process that I call "dispersed enforcement," the Supreme Court carefully limits its own role in policing State PCR. In Martinez, the Supreme Court actually rejected an intervention that would have imposed a constitutional rule--enforceable by the Court itself on appeal--that states provide postconviction counsel to litigate viable Sixth Amendment claims. (321) In some ways, then, the Martinez rule was a way for the Court to avoid a more intrusive intervention, thereby reserving such interventions for substantive constraints.

When understood as an alternative to a more intrusive intervention, the rule of Martinez and Trevino is actually consistent with Substantive-Constraint Exceptionalism. Needing to enforce procedural constraints, the Justices opted for intervention that minimized their own enforcement role, instead dispersing it across the lower federal courts. Rather than imposing a constitutional rule that inmates are entitled to counsel during State PCR--a rule the Supreme Court itself would have to enforce on direct review of state postconviction dispositions--the Court formulated a rule that entails no direct interference with State PCR at all. After Martinez and Trevino, states never "have to" provide counsel in their postconviction proceedings; instead, they choose between ensuring adequate counsel for inmates with substantial IAC claims or permitting those claims to be considered de novo in a federal habeas proceeding.

Martinez and Trevino therefore mark a familiar path in which the Supreme Court, facing the prospect of a thick enforcement role, devised an alternative. In the 1950s and 1960s, when the Warren Court liberalized criminal procedure through a series of cases announcing constitutional rights (322) and incorporating them against the states, (323) it also balked at the prospect of direct appellate enforcement. Instead, it deputized lower federal courts to enforce the new suite of procedural constraints by activating the habeas remedy. (324) Martinez and Trevino represent a similar dispersion strategy. Constitutional-law intervention is thereby reserved only for the maximally manageable category of claims based on substantive-law constraints.


In Part V, I offer my assessment of Substantive-Constraint Exceptionalism, which is more capably defended as an administrative compromise than as a series of otherwise-principled distinctions between lawful and unlawful punishment. My position is not that there is no meaningful distinction between substantive and procedural constraints, but that the pro-Exceptionalism narrative has failed to justify why the particular laws in question turn on those distinctions.

The most self-conscious defense of Substantive-Constraint Exceptionalism is the Supreme Court's opinion in Montgomery v. Louisiana, (325) which held (among other things) that the substantive-law exception to the federal retroactivity framework applies in State PCR. (326) Whatever the administrative merit of Exceptionalism, Montgomery's bedrock premise--that convictions in violation of an Eighth Amendment proportionality rule are "unlawful" and convictions in violation of other constitutional provisions are not (327)--draws little support from penal theory or prior doctrine.

A. The Administrative Justification

Perhaps the strongest argument in favor of Substantive-Constraint Exceptionalism is the least inspirational: it is easily administered. Among the category of claims requiring a collateral forum for effective enforcement, substantive constraints are uniquely suited to relatively efficient litigation and disposition. This efficiency comes from two sources. First, the number of cases touched by substantive constraints is sufficiently small that exceptions do not affect much state postconviction litigation. (328) Second, the social resources necessary to litigate and decide substantive constraints in individual cases are less than those necessary to dispose of other constitutional challenges.

The first feature of Exceptionalism's administrative appeal involves the relatively small size of postconviction terrain it covers. The federal interventions involving retroactivity, the processing of Ford claims, and the availability of resources for state postconviction litigation following the appointment of a federal habeas lawyer all facilitate litigation of substantive constraints that are primarily Eighth Amendment proportionality claims.

The existing list of Eighth Amendment proportionality rules is small and touches few state convictions. Moreover, they generally involve the death penalty or, more recently, LWOP. As far as offenders are concerned, death sentences cannot be imposed on those who are insane or intellectually disabled, (329) or who were juveniles at the time they committed an offense. (330) Nor can juvenile offenders receive mandatory LWOP. (331) In terms of offenses, the death penalty can be imposed only for certain crimes against the state (such as treason) and for killings in which the offender actually murdered the victim or exhibited reckless indifference to human life or had substantial involvement in a felony that resulted in a killing. (332) When the substantive constraint arises from a federal constitutional provision other than the Eighth Amendment--something like a First Amendment decision invalidating a specific criminal provision as overbroad (333)--there is a similarly limited impact of such a holding in terms of the number of inmates affected.

For all of the existing substantive constraints, there is no pressing necessity for a postconviction forum, at least in the sense that those exemptions can be applied prospectively in the direct-review chain. Trial and appellate courts apply the substantive constraint by simply refusing certain penalty-offender-offense combinations. Collateral enforcement of these types of substantive constraints is largely an exercise in relitigating a direct-review determination. The only exception to this generalization is a Ford claim, which centers on a present state of mental health that is definitionally incapable of being adjudicated until an execution date approaches. (334)

Other than Ford claims, substantive constraints that require meaningful collateral enforcement are new constraints. In other words, State PCR enforcement is crucial when the Supreme Court announces a new substantive constraint that applies retroactively. In such situations, the constraint could not have been applied in the direct-review chain. For example, in the immediate aftermath of the Atkins decision, which announced the retroactive Eighth Amendment exemption for intellectually disabled offenders, (335) collateral proceedings were necessary to enforce the rights of inmates whose convictions were already final.

The second reason why substantive constraints are especially suited to State PCR intervention involves the incremental cost of producing reliable outcomes in the incremental pieces of litigation. Virtually any constitutional challenge requires proof of some predicate fact--for example, an inmate with an Atkins claim must prove intellectual disability, (336) an inmate claiming an exemption from juvenile LWOP must prove her age, (337) and an inmate with a Sixth Amendment IAC claim must prove a prejudicial deficiency of the trial attorney. (338)

For at least some substantive constraints, the incremental cost of reliably enforcing the substantive constraint is far less than the cost of producing reliable outcomes in challenges involving procedural rules. The reason, in many cases, involves the incremental cost of litigating the factual predicate of the claim. For Eighth Amendment rules involving the qualities of the offense, there may be little more than a glance at the conviction and the sentence itself. After Kennedy v. Louisiana (339)--which held that states could not execute offenders convicted of crimes other than murder (340)--there was very little litigation over facts necessary to apply the exemption. An inmate was either sentenced to death for a murder or not; courts needed no experts to figure that information out. The same logic goes for convictions under a statute declared to be overbroad under the First Amendment. The major information that any court requires to apply the constraint is simply the judgment of conviction, which will disclose the offense for which an inmate has been punished.

For exemptions involving offender attributes, there are also situations where the incremental social cost of processing the claim is relatively low. For example, after Roper established that juveniles were ineligible for the death penalty, (341) litigation over the juvenile status of qualifying offenders was unlikely to impose a particularly taxing burden on the system. The same is true after Miller announced the Eighth Amendment prohibition on mandatory LWOP for juveniles. (342) For these offender-based exemptions and others like them, the incremental costs of processing the constitutional claim are negligible. In the same way that courts required no expert to tell them whether an inmate was convicted of murder or not, they required little help in reading a birth certificate.

Of course, it is true that there are certain substantive constraints for which per-claim adjudication costs are higher. For example, proving an inmate's entitlement to an Atkins exemption (intellectual disability) will often require extensive psychometric and behavioral testing, (343) as well as an investigation into family histories and conditions during the developmental period. (344) Moreover, much of the proof offered by the inmate--for example, an IQ score--will ordinarily be contested by the State. As a result, the social costs associated with developing, litigating, and resolving Atkins claims can be substantial. (345)

Generally speaking, however, Substantive-Constraint Exceptionalism is particularly appealing to federal institutions because it is particularly easy to administer. The lower case volume and per-case cost means that intervention consumes fewer state and federal resources from courts, state prosecutors, defense attorneys, and inmates.

B. The Montgomery Justification

Although the administrative justification likely exerts real influence on its Exceptionalism and its enforcement dispersion, the Supreme Court aspires to a more principled distinction for its treatment of substantive constraints. Its most energized articulation appears in Montgomery v. Louisiana, (346) where the Court determined that Miller v. Alabama's rule against imposing mandatory LWOP on juvenile offenders was a new but retroactive rule of substantive law. (347) I explore the Montgomery rationale below, but its gist is the problematic idea that a sentence violating a substantive constraint is "unlawful" and that a sentence violating some other constitutional provision is not. If the Supreme Court cannot use the concept of substantive constraints to draw a persuasive line between lawful and unlawful convictions, however, the pro-Exceptionalism narrative begins to fall apart.

Montgomery did not arise in the ordinary federal habeas posture in which the Supreme Court considers retroactivity questions. The Court heard the case on review of a state postconviction disposition (348) and the Louisiana Supreme Court had determined that state law withheld retroactive effect. (349) Such a holding appeared to be a state ground for deciding against the inmate, in which case the Court would lack appellate jurisdiction over the rest of the case. (350) The Court dealt with Montgomery's jurisdictionally disruptive retroactivity issue by declaring it to be a question of federal law, (351) thereby facilitating constitutional-law intervention.

In order to understand why Montgomery's pro-Exceptionalism narrative flounders, it is important to know a little bit more about retroactivity rules. Three types of Supreme Court decisions get retroactive effect under Teague v. Lane, "old law" that is dictated by precedent (which is not new law and is not covered by the Teague bar), "new" substantive law (the first Teague exception), and "watershed" procedure (the second). Montgomery changed two things about the existing understanding of retroactivity law. First, it posited some meaningful difference between new "substantive law" and Court decisions that might be retroactive for other reasons. Second, at least the substantive-law rule is of a constitutional order.

The Supreme Court explained: "Where state collateral review proceedings permit prisoners to challenge the lawfulness of their confinement, States cannot refuse to give retroactive effect to a substantive constitutional right that determines the outcome of that challenge." (352) The Court declared that any conviction inconsistent with a subsequently announced substantive rule is unlawful (353) and that the Constitution requires that states discharge any inmate incarcerated unlawfully, under that scenario. (354) The substantive rule permitting mandatory LWOP sentencing for juveniles was "unlawful," according to Montgomery, because "a State enforces a proscription or penalty barred by the Constitution." (355) The Court contrasted such sentences with those tainted by constitutional error that was procedural, which apparently did not render the sentence "unlawful" because the defendant was still eligible for the sentence. (356)

On the premise that only sentences inconsistent with new substantive rules are "unlawful," the Supreme Court announced that the Constitution requires states to apply only the corresponding part of the Teague retroactivity framework. (357) The Court did not decide whether the federal retroactivity rule had to be applied for claims invoking watershed procedural decisions, or new decisions on old laws. (358) Underneath the complex doctrinal discussion in Montgomery is the basic proposition that a penalty levied by way of constitutional error is "lawful" as long as the inmate was in some formal sense eligible for the sentence. This proposition is deeply problematic for at least two distinct reasons. First, the Court used the Eighth Amendment to fabricate a new and counterintuitive line between lawful and unlawful convictions. Second, the Court presented that line as one drawn by constitutional necessity.

First, the premise that sentences and convictions in violation of substantive constraints are "unlawful," whereas sentences and convictions inconsistent with procedural rules are not, flies in the face of some pretty basic ideas about the authority of the State to punish defendants. An offender can be unlawfully convicted if she is guilty, and may be unlawfully sentenced notwithstanding her formal eligibility for the penalty. If law enforcement forces a guilty defendant to confess at gunpoint and that confession is the dispositive evidence of guilt, for example, the conviction is still "unlawful." The touchstone of "lawfulness" is not innocence or eligibility. The question is whether the process for assigning guilt and fixing a penalty represents a legitimate exercise of state power. (359) An Eighth Amendment violation is sufficient for a conviction or sentence to be unlawful, but it is not necessary. The truth of the proposition that sentences inconsistent with substantive rules are not lawful does not establish that sentences consistent with them are.

Montgomery therefore conflates two distinct concepts: the set of unlawful punishments susceptible to collateral invalidation is not the same as the set of unlawful punishments. (360) It is true that only a limited set of constitutional infirmities permit a court to collaterally invalidate punishment, but that limitation does not reflect the view that only punishment inconsistent with substantive rules is unlawful. To suggest that only punishment in violation of a substantive rule is unlawful is to ignore a mountain of early-to-mid-twentieth century precedent about the relationship between the Due Process Clause and the viability of collateral challenges to state convictions. (361) There is nothing unusual about the proposition that only certain types of unlawful punishments are subject to collateral relief. Indeed, that proposition is at the very heart of Ex parte Siebold, (362) the crucial precedent recited in Montgomery itself. (363)

Second, Montgomery commits another serious mistake when it confuses a statutory remedy for constitutional violations with a constitutional rule that such a remedy be made available. The authority Montgomery discusses--especially Siebold itself--interprets the federal habeas statute. (364) The Judiciary Act of 1789 provided a statutory habeas remedy and the Habeas Corpus Act of 1867 made it available to state inmates confined in violation of the Constitution. (365) The question at issue in the operative precedent was whether a conviction under an unconstitutional statute entitled the aggrieved inmate to federal habeas relief under that statute. The proposition affirmed in the precedent--that such convictions were void and therefore triggered discharge under the statutory remedy (366)--is not equivalent to the proposition that the Constitution itself requires discharge. (367) Montgomery reads as though the Supreme Court is oblivious to Siebold's clear meaning. Instead of reporting Siebold as a case interpreting the statutory habeas remedy to reach particular types of defects, the Court reinterprets it as support for the idea that the Constitution requires the remedy.

There are more oddities about Montgomery's reliance on Siebold. Siebold is a single precedent in a long line of cases about what types of criminal judgments are "unlawful." (368) It is neither a starting nor end point on that line, so the selective emphasis on that case seems arbitrary. Subsequent cases--cases far more frequently cited as authoritative on the conditions for unlawful punishments--embrace the idea that unlawful punishment is a much broader category than the concept contemplated in Siebold. (369)

Why make mistakes that are so obvious? The answer might be that these were not mistakes. The Court wanted to hold that it always intervened but only intervened when custody was unlawful, and doing so required it to use the Eighth Amendment to redraw the line between lawful and unlawful convictions. Siebold's language about the lawfulness of the conviction, wrenched from its context, suggested that the Constitution--not any statute--required some form of retroactivity for attacks on certain types of punishments. (370) It also suggests a limitation on the idea of unlawful punishment, which excludes punishment violating procedural rules. Combining these two propositions is what allowed the Court to assume appellate jurisdiction in the case without, in the Court's view, imposing too extravagant an obligation on states. Louisiana's decision not to apply Miller's Eighth Amendment rule retroactively was positioned as a constitutional violation (and therefore a federal issue), but the constitutional rule reached no further than Teague's substantive-rule category. By constitutionalizing what amounts to a substantive-constraint exception to retroactivity, the Court simultaneously promoted the Eighth Amendment interest and preserved the administrability of its intervention.

Distorting precedent is the jaywalking of interpretive sins, but Montgomery's, collateral damage is significant. As explained above, Montgomery appears to embrace the radical proposition that a conviction is never unlawful unless substantive law bars a penalty for an offense. Punishment secured in violation of double jeopardy rules, by mob-dominated trials, and through prosecutorial misconduct are all declared by Montgomery to be in some unspecified sense "lawful." Of course, not every minor error of criminal procedure contaminates punishment such that it is open to collateral challenge, but there was already a device built into retroactivity law that performed the necessary sorting function--only watershed rules of criminal procedure were retroactively enforced. (371)

The other Montgomery revelation is the rule that the Teague framework--at least the substantive-rule exception--is a creature of constitutional law. If taken seriously, this piece of Montgomery would plunge much of modern postconviction law--both state and federal--into a state of profound uncertainty. If punishments in violation of substantive constraints are unlawful and the constitution requires government to permit collateral attacks in such cases, how can any statutory restrictions on such attacks survive? Montgomery appears to have had a blinkered focus on retroactivity rules in State PCR, but a constitutional rule is a constitutional rule, and it applies everywhere.

A rule that no court can enforce a punishment inconsistent with a substantive constraint would blow holes through a raft of federal restrictions on the habeas remedy. What, for example, is the status of the 28 U.S.C. [section] 2244(b)(2) restriction that limits successive habeas claims to those involving guilt questions, rather than eligibility for a punishment? (372) Could a federal court impose a time-bar on Eighth Amendment claims if the Constitution requires a discharge remedy? (373)

There are, of course, very meaningful differences between punishment that violates substantive constraints and punishment secured by way of a procedural violation. The problem is that the difference between substantive and procedural error has never mattered all that much to the law of judgments (when they involve property), (374) or to habeas law (when they involve liberty). (375) And although habeas law might have permitted enhanced collateral consideration of convictions in violation of substantive constraints, similarly enhanced consideration has always been available for punishment imposed by way of procedural defects. (376)

If the Supreme Court simply needed a response to the unique injustice of custody in violation of substantive constraints, it had much better tools in its toolbox. There were ways to ensure that qualifying juveniles got the benefit of Miller without the disruptive language the Supreme Court used to fashion the constitutional-law intervention. The Court, for example, could have held that the state nonretroactivity ground was "inadequate" to ensure protection of the substantive federal right, thereby permitting unfettered Supreme Court and federal habeas review of the state decision. (377) Or it could have accomplished that result by holding that the state nonretroactivity ground was insufficiently "independent" of federal law. (378) Or the Court could have applied the whole retroactivity framework to State PCR, without equating the concept of an unlawful sentence with the concept of an actually innocent defendant. The latter solution would still confound some precedent, but it would not require the Court to improvise new and disruptive theories about when a conviction is lawful and the source of a postconviction remedy. (379)

C. The Regulatory Efficacy of Dispersed Enforcement

A final (albeit secondary) criticism of the Exceptionalism model is that its dispersed-enforcement mechanism does little to improve what one might call states' postconviction hygiene. Specifically, Martinez and Trevino lack regulatory bite insofar as they are unlikely to alter the way states guarantee legal services to postconviction claimants. The doctrinal modifications treat the symptoms more than the disease; they deal with the effects of defective state postconviction representation by ensuring the availability of a federal forum, but the availability of that forum does not necessarily incentivize states to change much about the State PCR that they provide.

Before I discuss this particular problem, readers should understand that not every Justice voting for Supreme Court relief in Martinez and Trevino did so with the regulatory objective of reforming state postconviction practice. Nonetheless, the Justices were surely aware of the fact that federal habeas law might have such institutional effects. Indeed, the Martinez Justices were not making a binary choice between judgment for the inmate or the state; they were also choosing between whether to provide the equitable remedy that Martinez ultimately endorsed or to declare a narrow constitutional right to postconviction counsel that would have forced institutional change much more directly. (380) The point I want to make is that the theoretical justification for the equilibrium--whether that equilibrium was a conscious objective or a predictable byproduct of judicial decisionmaking--is lacking.

Understanding the empirical effect of Martinez and Trevino on administration of State PCR is difficult in light of available data. What little data there is, however, suggests that Martinez and Trevino are unlikely to substantially affect the volume of filing or hearing activity, or the provision of counsel. (381) The feedback loop, it seems, is just too attenuated. The institutional entities responsible for supplying representation during State PCR would have to conclude that, in light of the Martinez excuse, benefits exceed costs. Such a conclusion would require the entities--usually judges--to believe that a failure to appoint counsel would result in more federal decisions granting habeas relief, and that the costs of such unlikely relief (in the form of retrials) would exceed the cost of appointed counsel in State PCR. (382) Moreover, to the extent that state judges are concerned about the costs of federal habeas relief on Sixth Amendment claims, there are lower-cost responses than effectively enforcing a right to counsel in State PCR. (383)

All of this is to say that although Martinez and Trevino might be habeas interventions designed to redress Sixth Amendment violations, they do not necessarily function as a regulatory device that improves State PCR. States are unlikely to respond to the newly available federal forum by altering the institutional configuration of their postconviction processes. If the Supreme Court had the appetite to enforce it, the more effective regulatory tactic would be to include the Sixth Amendment in the suite of constitutional protections the Court was willing to protect using constitutional-law intervention. Instead, such interventions are reserved for substantive constraints.


The body of procedure that a community uses to punish lawbreakers will always have the criminal trial as its spine. Legal institutions nonetheless seem to recognize that the era of near-exclusive focus on the trial has ended. At the same time, and to borrow a basic economic concept, demand for postconviction process is increasing at the very moment that supply is falling--when federal habeas review becomes less and less viable as a means of enforcing federal rights. In light of these developments, federal institutions (primarily the judiciary) have devised intervention to ensure that State PCR more effectively secures the federal interests at stake in those proceedings.

That intervention increasingly discloses a Substantive-Constraint Exceptionalism that deprioritizes redress for other types of constitutional violations. Specifically, the most aggressive mode of intervention--direct review of state postconviction dispositions--is reserved exclusively for substantive constraints. The less aggressive modes of intervention, which do exhibit a growing focus on Sixth Amendment issues, are an exercise in what I call dispersed enforcement. The Supreme Court effectively offloads enforcement responsibility to the lower federal courts.

The justifications for such Exceptionalism are middling, and are less conceptual and doctrinal than they are administrative. There may be something unique about substantive constraints, but the Supreme Court's attempt to isolate them by suggesting that they define the universe of "unlawful punishment" is both unpersuasive and a little careless. The more intuitive but less emotionally stirring explanation is that review of punishment that violates substantive constraints is uniquely administrable.

Lee Kovarsky (*)

(*) Professor of Law, University of Maryland School of Law. For their insightful comments, I thank John Blevins, Zack Bray, Adam Gershowitz, Brandon Garrett, Nancy King, Leah Litman, and Eve Primus. I am also grateful for the opportunity to participate in the University of Maryland Legal Theory Workshop. Finally, I appreciate the excellent research assistance of Reagan Greenberg and Emily Levy. All errors are my own.

(1) See Vicki C. Jackson, Constitutional Law in an Age of Proportionality, 124 YALE L.J. 3094, 3107-08 (2015) (discussing the influence of Magna Carta and the English Bill of Rights on early American legal artifacts); Amanda L. Tyler, The Forgotten Core Meaning of the Suspension Clause, 125 HARV. L. REV. 901, 927 (2012) (explaining the influence of Blackstone and Coke on founding-era understanding of habeas corpus).

(2) Coleman v. Thompson, 501 U.S. 722, 752 (1991).

(3) See Pennsylvania v. Finley, 481 U.S. 551, 557 (1987); see also Dist. Attorney's Office v. Osborne, 557 U.S. 52, 69 (2009) (citing Finley for the proposition that the constraints on states providing postconviction relief are relaxed because they do not have to provide it); Lackawanna Cty. Dist. Attorney v. Coss, 532 U.S. 394, 402 (2001) (explaining that there is "no constitutional mandate" that states provide State PCR).

(4) I do not mean to suggest that nobody has paid attention to State PCR at all. There is, for example, a recent treatise on state postconviction remedies. See DONALD E. WILKES, STATE POSTCONVICTION REMEDIES AND RELIEF HANDBOOK (2013). Particularly after the Supreme Court decided Finley, more law review articles began to treat State PCR as a meaningful object of study. See, e.g., Eric M. Freedman, Giarratano Is a Scarecrow: The Right to Counsel in State Capital Postconviction Proceedings, 91 CORNELL L. REV. 1079, 1080-81 (2006) (disputing relevance of Supreme Court decision refusing a right to counsel in capital State PCR cases); Mary C. Hutton, Retroactivity in the States: The Impact of Teague v. Lane on State Postconviction Remedies, 44 ALA. L. REV. 421, 424 (1993) (exploring how State PCR ought to give effect to new Supreme Court rules). In the last several years, however, some scholars have begun to devote more attention to State PCR. See, e.g., Eric M. Freedman, State Post-Conviction Remedies in the Next Fifteen Years: How Synergy Between the State and Federal Governments Can Improve the Criminal Justice System Nationally, 24 FED. SENT'G REP. 298, 298 (2012) (arguing that several factors give state and federal governments a shared interest in procedurally sound State PCR); Nancy J. King, Enforcing Effective Assistance After Martinez, 122 YALK L.J. 2428, 2451-55 (2013) (considering how recent Supreme Court decisions about the availability of federal habeas relief will affect state postconviction remedies); Justin F. Marceau, Challenging the Habeas Process Rather than the Result, 69 WASH. & LEE L. REV. 85 (2012) (considering how 28 U.S.C. [section] 2254(d) and 42 U.S.C. [section] 1983 ought to be used to target serious defects in State PCR).


(6) See infra Section I.B.

(7) See infra Section I.C.

(8) See infra subsection II.A.1.

(9) See infra subsection II.A.2.

(10) See infra subsection II.A.4.

(11) See infra Section II.B.

(12) 136 S. Ct. 718 (2016).

(13) See id. at 732.

(14) But see supra note 4 (reciting recent academic articles focusing on State PCR).

(15) See 28 U.S.C. [section] 2254(b) (2012).

(16) State PCR is also, for example: (1) an important vehicle for enforcing features of state law and (2) a process by which states address local problems in criminal justice administration.

(17) See, e.g., King, supra note 4, at 2442-46 (noting the dearth of legal representation in State PCR); Eve Brensike Primus, Federal Review of State Criminal Convictions: A Structural Approach to Adequacy Doctrine, 116 MICH. L. REV. 75, 77 (2017) ("At the state level, modern postconviction review schemes are often so complicated and confusing that indigent criminal defendants have no realistic prospect of complying with the procedural rules.").

(18) The current version of the exhaustion requirement appears in 28 U.S.C. [section] 2254(b)(1)(A).

(19) See STEIKER & STEIKER, supra note 5, at 202-03.

(20) See, e.g., Parker v. Dugger, 550 So. 2d 459, 460 (Fla. 1989) (per curiam) (rejecting appellate paradigm for postconviction review); Leonard v. Commonwealth, 279 S.W.3d 151, 156 (Ky. 2009) (same); Carter v. Galetka, 44 P.3d 626, 630 (Utah 2001) (same).

(21) See supra note 3.

(22) Coleman v. Thompson, 501 U.S. 722, 752 (1991). But see infra Section III.B (explaining how some resources earmarked for federal habeas process might be used to support State PCR litigation).

(23) See 28 U.S.C. [section] 2254(d) (barring federal relitigation unless state decision is either legally or factually unreasonable); Harrington v. Richter, 562 U.S. 86, 101 (2011) (equating unreasonableness with a "no fairminded jurist" standard).

(24) See Bryan A. Stevenson, Confronting Mass Imprisonment and Restoring Fairness to Collateral Review of Criminal Cases, 41 HARV. C.R.-C.L. L. REV. 339, 342-45 (2006).

(25) See Eve Brensike Primus, Structural Reform in Criminal Defense: Relocating Ineffective Assistance of Counsel Claims, 92 CORNELL L. REV. 679 (2007).

(26) See, e.g., Paul T. Crane, Charging on the Margin, 57 WM. & MARY L. REV. 775, 825-27 (2016) (discussing effect of resource scarcity on indigent misdemeanor defendants).

(27) See King, supra note 4, at 2442.

(28) See Mary Sue Backus & Paul Marcus, The Right to Counsel in Criminal Cases, A National Crisis, 57 HASTINGS L.J. 1031, 1034 (2006).

(29) See AM. BAR ASS'N, EVALUATING FAIRNESS AND ACCURACY IN STATE DEATH PENALTY SYSTEMS: THE ALABAMA DEATH PENALTY ASSESSMENT REPORT 159 (2006),; AM. BAR ASS'N, State Standards for Appointment of Counsel in Capital Cases, ABA Death Penalty Representation Project (Feb. 2009); Am. Bar Ass'n, Evaluating Fairness and Accuracy in State Death Penalty Systems: The Georgia Death Penalty Assessment Report 155 (Jan. 2006),

(30) See King, supra note 4, at 2444.

(31) See John H. Blume, AEDPA: The "Hype" and the "Bite," 91 CORNELL L. REV. 259, 276 n.97 (2006).

(32) See King, supra note 4, at 2453-54.

(33) See Giovanna Shay & Christopher Lasch, Initiating a New Constitutional Dialogue: The Increased Importance Under AEDPA of Seeking Certiorari from Judgments of Stale Courts, 50 WM. & MARY L. REV. 211, 255-56 (2008).

(34) SeeTEX. GOV'T CODE ANN. [section] 78.054 (West 2017) (describing authority of the office).

(35) See, e.g., Thomas P. Sullivan, Efforts to Improve the Illinois Capital Punishment System: Worth the Cost?, 41 U. RICH. L. REV. 935, 968 (2007) (estimating two percent for Illinois, which has since discontinued capital sentencing).

(36) See Eve Brensike Primus, Culture as a Structural Problem in Indigent Defense, 100 MINN. L. REV. 1769, 1773 (2016).

(37) See, e.g., Donald J. Harris et al., Dispatch and Delay: Post Conviction Relief Act Litigation in Non-Capital Cases, 41 DUQ. L. REV. 467, 482 (2003) (explaining why more experienced lawyers avoid State PCR appointments); David Rudovsky, Gideon and the Effective Assistance of Counsel: The Rhetoric and the Reality, 32 LAW & INEQ. 371, 401 n.152 (2014) (describing the need for federal intervention "necessitated by the lack of qualified and properly compensated defense counsel in the [Pennsylvania] state appointment system").

(38) See Bruce A. Green, Criminal Neglect: Indigent Defense from a Legal Ethics Perspective, 52 EMORY L.J. 1169, 1179 (2003).

(39) See King, supra note 4, at 2443.

(40) See id. at 2453 n.91; Daniel S. Medwed, Up the River Without a Procedure: Innocent Prisoners and Newly Discovered Non-DNA Evidence in State Courts, 47 ARIZ. L. REV. 655, 706 (2005).

(41) See King, supra note 4, at 2452; see also Adam M. Gershowitz, Pay Now, Execute Later: Why Counties Should Be Required to Post a Bond to Seek the Death Penalty, 41 U. RICH. L. REV. 861, 866 (2007) (suggesting system in which counties have to internalize more costs of retrials).

(42) See Blume, supra note 31 (detailing examples for death-sentenced inmates).

(43) See Stephen B. Bright, Legal Representation for the Poor: Can Society Afford This Much Injustice?, 75 Mo. L. REV. 683, 690 (2010).

(44) See STEIKER & STEIKER, supra note 5, at 127.

(45) See infra notes 46-61 and accompanying text.

(46) 477 U.S. 399 (1986).

(47) Id. at 409-10.

(48) Id. at 416-17 (plurality opinion) (footnote omitted).

(49) See, e.g., Barnard v. Collins, 13 F.3d 871, 876 (5th Cir. 1994) (reciting "awareness" standard for competency that was subsequently overturned in Panetti).

(50) 551 U.S. 930 (2007).

(51) See id. at 959.

(52) See id. at 948-49.

(53) See Atkins v. Virginia, 536 U.S. 304, 321 (2002).

(54) See STEIKER & STEIKER, supra note 5, at 127-28 (discussing example in Texas).

(55) Atkins, 536 U.S. at 317 (second and third alterations in original) (internal citation omitted) (quoting Ford v. Wainwright, 477 U.S. 399, 416-17 (1986) (plurality opinion)).

(56) See Hall v. Florida, 134 S. Ct. 1986, 2004 nn.4-5 (2014) (Alito, J., dissenting).

(57) 134 S. Ct. 1986.

(58) Id. at 1990.

(59) 137 S. Ct. 1039, 1044 (2017).

(60) 136 S. Ct. 718 (2016).

(61) See id. at 727.

(62) See infra Section II.A.

(63) See infra Section II.B.

(64) This dynamic actually siphons litigation to postconviction process more generally.

(65) See infra subsection II.A. 1.

(66) See infra subsection II.A.2.

(67) See infra subsection II.A.3.

(68) See infra subsection II.A.4.

(69) See Strickland v. Washington, 466 U.S. 668, 687 (1984).

(70) See Crawford v. Washington, 541 U.S. 36, 68-69 (2004).

(71) See United States v. Ewell, 383 U.S. 116, 120 (1966).

(72) See Spano v. New York, 360 U.S. 315, 324 (1959).

(73) 477 U.S. 399 (1986).

(74) 551 U.S. 930 (2007).

(75) See id. at 949.

(76) The claim inherits its name from an opinion respecting the denial of certiorari in Lackey v. Texas, 514 U.S. 1045 (1995) (Stevens, J., memorandum respecting denial of certiorari).

(77) See, e.g., Boyer v. Davis, 136 S. Ct. 1446, 1447 (2016) (Breyer, J., dissenting from denial of certiorari); Glossip v. Gross, 135 S. Ct. 2726, 2756 (2015) (Breyer, J., dissenting); Valle v. Florida, 564 U.S. 1067, 1067 (2011) (Breyer, J., dissenting from denial of stay); Thompson v. McNeil, 556 U.S. 1114, 1114-15 (2009) (Stevens, J., statement respecting denial of certiorari); Baze v. Rees, 553 U.S. 35, 78-81 (2008) (Stevens,J., concurring in the judgment); Knight v. Florida, 528 U.S. 990, 993 (1999) (Breyer, J., dissenting from denial of certiorari).

(78) See infra notes 80-87 and accompanying text.

(79) 567 U.S. 460 (2012).

(80) Questions about retroactivity in criminal proceedings are merely a subcategory of long-standing questions involving retroactive application of any legal rule. See generally Richard H. Fallon, Jr. & Daniel J. Meltzer, New Law, Non-Retroactivity, and Constitutional Remedies, 104 HARV. L. REV. 1731 (1991) (arguing that the implementing doctrine for retroactivity principles should come from the law of constitutional remedies); Jill E. Fisch, Retroactivity and Legal Change: An Equilibrium Approach, 110 HARV. L. REV. 1055 (1997) (using destabilization model to suggest appropriate levels of retroactivity); Louis Kaplow, An Economic Analysis of Legal Transitions, 99 HARV. L. REV. 509, 511-12 (1986) (connecting retroactive application of judicial decisions with other contexts in which the scope of retroactive application must be determined).

(81) See Montgomery v. Louisiana, 136 S. Ct. 718, 728 (2016). Hypertechnically, a decision is retroactive if it applies to any convictions pending on direct appeal, but a new decision is always retroactive in that sense. See Teague v. Lane, 489 U.S. 288, 304 (1989) (plurality opinion).

(82) See Beard v. Banks, 542 U.S. 406, 416-17 (2004). The classic statement of the modern rule appears in Teague, 489 U.S. at 304 (plurality opinion).

(83) See infra notes 88-93 and accompanying text.

(84) See Dov Fox 8c Alex Stein, Constitutional Retroactivity in Criminal Procedure, 91 WASH. L. REV. 463, 466 n.15 (2016) (collecting cases rejecting membership in second Teague exception).

(85) See Whorton v. Bockting, 549 U.S. 406, 419 (2007) (collecting Court cases referencing Gideon as paradigm of second Teague exception).

(86) 372 U.S. 335 (1963).

(87) Gideon was decided in 1963, and Teague was decided in 1989.

(88) 536 U.S. 304 (2002).

(89) See id. at 321.

(90) 543 U.S. 551 (2005).

(91) See id. at 578.

(92) 554 U.S. 407 (2008).

(93) See id. at 413.

(94) Weems v. United States, 217 U.S. 349 (1910), is generally considered the first proportionality case. See Margaret Raymond, "No Fellow in American Legislation": Weems v. United States and the Doctrine of Proportionality, 30 VT. L. REV. 251, 252 (2006).

(95) See supra notes 88-93 and accompanying text.

(96) See generally Strickland v. Washington, 466 U.S. 668 (1984) (specifying modern IAC standard); Brady v. Maryland, 373 U.S. 83 (1963) (announcing rule against prosecutorial suppression of exculpatory and material evidence).

(97) See U.S. CONST, amend VI; see also Gideon v. Wainwright, 372 U.S. 335, 342 (1963) (incorporating an indigent criminal defendant's right to appointed counsel against the states).

(98) See Strickland, 466 U.S. at 685-86.

(99) Id. at 687; see also, e.g., Hinton v. Alabama, 134 S. Ct. 1081, 1087 (2014) (per curiam) (describing as "straightforward" the two-prong IAC rule specified by Strickland).

(100) See Primus, supra note 25, at 689.

(101) See Eve Brensike Primus, Effective Trial Counsel After Martinez v. Ryan: Focusing on the Adequacy of State Procedures, 122 YALE L.J. 2604, 2619 (2013).

(102) 539 U.S. 510 (2003).

(103) See id. at 523.

(104) See Samuel R. Wiseman, Habeas After Pinholster, 53 B.C. L. REV. 953, 989-90 (2012).

(105) 373 U.S. 83 (1963).

(106) See id. at 87.

(107) See Melanie D. Wilson, Anti-Justice, 81 TENN. L. REV. 699, 742 (2014).


(109) See id. at 30.

(110) See Strickland v. Washington, 466 U.S. 668, 687 (1984).

(111) See Angela J. Davis, The American Prosecutor: Independence, Power, and the Threat of Tyranny, 86 IOWA L. REV. 393, 432 (2001).

(112) See Larry W. Yackle, The Habeas Hagioscope, 66 S. CAL. L. REV. 2331, 2378 (1993).

(113) See Richard A. Leo & Richard J. Ofshe, The Consequences of False Confessions: Deprivations of Liberty and Miscarriages of Justice in the Age of Psychological Interrogation, 88 J. CRIM. L. & CRIMINOLOGY 429, 492 (1998).

(114) See, e.g., BARRY SCHECK ET AL., ACTUAL INNOCENCE: FIVE DAYS TO EXECUTION AND OTHER DISPATCHES FROM THE WRONGLY CONVICTED app. 2 (2000) (first 62 DNA exonerations); Brandon L. Garrett, Introduction: New England Law Review Symposium on "Convicting the Innocent," 46 NEW ENG. L. REV. 671 (2012) (analyzing first 250 DNA exonerees); Samuel R. Gross et al., Exonerations in the United States 1989 Through 2003, 95 J. CRIM. L. & CRIMINOLOGY 523, 524 (2005) (first 340 exonerations, including 144 DNA exonerations).

(115) See Brandon L. Garrett, Judging Innocence, 108 COLUM. L. REV. 55, 88-91 (2008).

(116) See id. at 78-81.

(117) See id. at 86-88.

(118) See id. at 81-86.

(119) See Brandon L. Garrett, DNA and Due Process, 78 FORDHAM L. REV. 2919, 2921 (2010).

(120) See generally Brandon L. Garrett, Claiming Innocence, 92 MINN. L. REV. 1629, 1652-99 (2008) (explaining role of DNA in criminal justice system).

(121) See Lee Kovarsky, Justice Scalia's Innocence Tetralogy, 101 MINN. L. REV. HEADNOTES 94, 103 (2016).

(122) See generally Sheri Lynn Johnson, Cross-Racial Identification Errors in Criminal Cases, 69 CORNELL L. REV. 934 (1984) (presenting overarching discussion of cross-witness identification issues in criminal cases).

(123) See Garrett, supra note 115, at 88-91 (false confessions); id. at 86-88 (informants).

(124) See Lee Kovarsky, Preclusion and Criminal Judgment, 92 NOTRE DAME L. REV. 637, 637 (2016).

(125) See Judith Resnik, Tiers, 57 S. CAL. L. REV. 837, 843 (1984).

(126) See supra notes 69-78 and accompanying text (claims of post-trial constitutional error); supra notes 79-95 and accompanying text (claims of trial-right violations based on decisions announced after conviction is final); supra notes 96-111 and accompanying text (claims of trial-right violations practically incapable of litigation in direct review chain).

(127) See Barry Friedman, Failed Enterprise: The Supreme Court's Habeas Reform, 83 CALIF. L. REV. 485, 503-12 (1995) (arguing that the Supreme Court's tepid embracement of federal "actual innocence" doctrine has been poorly executed); Robert J. Smith, Recalibrating Constitutional Innocence Protection, 87 WASH. L. REV. 139, 147-168 (2012) (arguing that innocence is the "transcendent" value in criminal process).

(128) They are also substitutes in a less colloquial, more economic sense--as the cost for one rises, so too does demand for the other.

(129) See 28 U.S.C. [section] 2254(d) (2012). I strongly dislike this term, but I do not dwell on my terminological point here because the dispute is not central to this discussion. See Kovarsky, supra note 124, at 668-69.

(130) Joseph L. Hoffmann & Nancy J. King, Essay, Rethinking the Federal Role in State Criminal Justice, 84 N.Y.U. L. REV. 791, 809 (2009).

(131) 344 U.S. 443 (1953).

(132) See id. at 497-513 (opinion of Frankfurter, J.). Justice Frankfurter's opinion, which was technically for the Court in the companion case of Daniels v. Allen, is generally considered controlling on the question presented in Brown as to the preclusive effect of a state disposition. See BRANDON L. GARRETT & LEE KOVARSKY, FEDERAL HABEAS CORPUS: EXECUTIVE DETENTION AND POST-CONVICTION LITIGATION 129 (2013); see also Lee Kovarsky, A Constitutional Theory of Habeas Power, 99 VA. L. REV. 753, 800-02 (2013) (discussing conflicting positions on precedent for Brown).

(133) See Jordan Steiker, Restructuring Post-Conviction Review of Federal Constitutional Claims Raised by State Prisoners: Confronting the New Face of Excessive Proceduralism, 1998 U. CHI. LEGAL F. 315, 323.

(134) See John C. Jeffries, Jr. & William J. Stuntz, Ineffective Assistance and Procedural Default in Federal Habeas Corpus, 57 U. CHI. L. REV. 679, 698 (1990).

(135) Lee Kovarsky, AEDPA's Wrecks: Comity, Finality, and Federalism, 82 TUL. L. REV. 443, 459-80 (2007).

(136) See John H. Blume et al., In Defense of Noncapital Habeas: A Response to Hoffmann and King, 96 CORNELL L. REV. 435, 440-41 (2011); see also, e.g., McCleskey v. Zant, 499 U.S. 467, 493 (1991) (adopting "cause-and-prejudice" standard for any claim not included in a prior federal habeas petition); Wainwright v. Sykes, 433 U.S. 72, 83, 87 (1977) (abandoning "deliberate bypass" rule for excusing claims forfeited in state court, in favor of a more difficult to satisfy standard of "cause and prejudice"); Stone v. Powell, 428 U.S. 465, 481-82 (1976) (holding that Fourth Amendment claims were not cognizable on federal habeas review).

(137) Pub. L. No. 104-132, [section][section] 101-108, 110 Stat. 1214, 1217-26 (1996) (codified in relevant part at 28 U.S.C. [section][section] 2244-2266 (2012)).

(138) See generally Lee Kovarsky, Death Ineligibility and Habeas Corpus, 95 CORNELL L. REV. 329, 335-42 (2010) (discussing effects of AEDPA on procedural restrictions).

(139) See 28 U.S.C. [section] 2254(d).

(140) Cf. Vicki C. Jackson, Introduction: Congressional Control of Jurisdiction and the Future of the Federal Courts--Opposition, Agreement, and Hierarchy, 86 GEO. L.J. 2445, 2453-54 (1998) (noting important questions sounding in "both jurisprudence and constitutional law" (footnote omitted)).

(141) Specifically, [section] 2254(d)'s preambular text states as the general rule that: "An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless...." 28 U.S.C. [section] 2254(d) (emphasis added).

(142) 28 U.S.C. [section] 2254(d) required extensive deference to fact-finding before AEDPA. See Stephen B. Bright & Patrick J. Keenan, Judges and the Politics of Death: Deciding Between the Bill of Rights and the Next Election in Capital Cases, 75 B.U. L. REV. 759, 828 (1995) (discussing the pre-AEDPA regime).

(143) 529 U.S. 362 (2000).

(144) See id. at 409-10.

(145) See id. at 409-14.

(146) See id. at 409.

(147) See id.

(148) Compare Harrington v. Richter, 562 U.S. 86, 101 (2011) (equating unreasonableness with the "no fair-minded jurist" standard), with Williams, 529 U.S. at 410 (rejecting the "all reasonable jurists" standard).

(149) See, e.g., Woods v. Etherton, 136 S. Ct. 1149, 1151 (2016) (per curiam) (citing Richter formulation); Davis v. Ayala, 135 S. Ct. 2187, 2199 (2015) (same); White v. Woodall, 134 S. Ct. 1697, 1702 (2014) (same); Burt v. Titlow, 134 S. Ct. 10, 16 (2013) (same); Nevada v. Jackson, 569 U.S. 505, 1992 (2013) (per curiam) (same); Wetzel v. Lambert, 565 U.S. 520, 524 (2012) (per curiam) (same); Cullen v. Pinholster, 563 U.S. 170, 188 (2011) (same). Richter incorporated a passing reference to disagreement among "fairminded jurists" from Yarborough v. Alvarado, 541 U.S. 652, 664 (2004).

(150) Kovarsky, supra note 124, at 671.

(151) See supra note 150 and accompanying text.

(152) I do not discuss prominent procedural doctrines that do not turn on events occurring during State PCR, such as the rules for so-called successive petitions. See 28 U.S.C. [section] 2244(b) (2012).

(153) Id. [section] 2254(b).

(154) See Coleman v. Thompson, 501 U.S. 722, 735 (1991) (setting forth basic framework of default doctrine).

(155) 28 U.S.C. [section] 2244(d).

(156) Id. [section] 2254(e).

(157) See id. [section] 2254(b)(2).

(158) See id. [section] 2254(b)(1)(B).

(159) Cf. HERTZ & LIEBMAN, FEDERAL HABEAS CORPUS PRACTICE AND PROCEDURE [section] 23.4 (7th ed. 2016) (setting forth caselaw under the exceptions).

(160) See Johnson v. Lee, 136 S. Ct. 1802, 1804 (2016) (per curiam); see also generally HERTZ & LIEBMAN, supra note 159, [section] 26 (presenting the rules for procedural default and exceptions thereto).

(161) See Trevino v. Thaler, 133 S. Ct. 1911, 1917 (2013). State inmates can excuse defaults by showing either that failing to reach the claim is a miscarriage of justice or by showing cause and prejudice. See House v. Bell, 547 U.S. 518, 536 (2006) (miscarriage of justice); Edwards v. Carpenter, 529 U.S. 446, 451 (2000) (cause and prejudice).

(162) See 28 U.S.C. [section] 2244(d) (2).

(163) See Blume, supra note 31, at 288-89.

(164) See Day v. McDonough, 547 U.S. 198, 214-15 (2006) (Scalia, J., dissenting) (discussing the relationship of laches to the limitations statute).

(165) In Holland v. Florida, 560 U.S. 631 (2010), the Supreme Court declared equitable tolling available to state claimants, and held that deficient representation in State PCR could theoretically disable the limitations period. See id. at 645, 652.

(166) That an inmate must show [section] 2254(d)(2) factual unreasonableness by reference to the state record flows from the plain language of the provision, which requires that the reasonableness determination be made "in light of the evidence presented in the State court proceeding." 28 U.S.C. [section] 2254(d)(2). In Cullen v. Pinholsler, the Supreme Court determined that an inmate must also show [section] 2254(d)(1) legal unreasonableness by reference to the state record. 563 U.S. 170, 180-81 (2011).

(167) See Pinholster, 563 U.S. at 181-82.

(168) 28 U.S.C. [section] 2254(e)(2).

(169) The Madisonian Compromise rendered lower federal courts optional, and federal interests were secured through state adjudication that was subject to Supreme Court review. See Richard H. Fallon, Jr., Jurisdiction-Stripping Reconsidered, 96 VA. L. REV. 1043, 1060 n.72 (2010).

(170) See Strickland v. Washington, 466 U.S. 668, 687 (1984).

(171) See Brady v. Maryland, 373 U.S. 83, 87, 90-91 (1963).

(172) See sources cited supra note 3.

(173) See, e.g.. United States v. MacCollom, 426 U.S. 317, 323 (1976) (plurality opinion) ("The Due Process Clause of the Fifth Amendment does not establish any right to an appeal, and certainly does not establish any right to collaterally attack a final judgment of conviction." (citing Griffin v. Illinois, 351 U.S. 12, 18 (1956) (plurality opinion))).

(174) U.S. CONST, art. I, [section] 9, cl. 2.

(175) See Kovarsky, supra note 132, at 786-94 (documenting authority in support of the proposition that federal habeas power does not, by force of the Suspension Clause, require a state postconviction forum).

(176) See, e.g., Montgomery v. Louisiana, 136 S. Ct. 718, 731-32 (2016) (declaring that, if states provide postconviction process, the Supremacy Clause requires that they use the federal retroactivity framework).

(177) See, e.g., Coleman v. Thompson, 501 U.S. 722, 752 (1991) (denying a right to counsel in State PCR on the ground that the Constitution does not require State PCR at all).

(178) Of course, State PCR cannot be "fundamentally unjust" in the same way that any state action cannot be arbitrary. See Dist. Attorney's Office v. Osborne, 557 U.S. 52, 69 (2009).

(179) 477 U.S. 399, 410 (1986).

(180) See generally Stephen J. Morse, Excusing the Crazy: The Insanity Defense Reconsidered, 58 S. CAL. L. REV. 777 (1985) (arguing in favor of retaining the insanity defense).

(181) See Drope v. Missouri, 420 U.S. 162, 171 (1975) ("It has long been accepted that a person whose mental condition is such that he lacks the capacity to understand the nature and object of the proceedings against him, to consult with counsel, and to assist in preparing his defense may not be subjected to a trial.").

(182) A four-Justice plurality insisted that the Constitution required trial level safeguards, but Justice Powell concurred separately to urge more modest procedural protection. Ford, 477 U.S. at 410-18 (plurality opinion). Justice Powell's view of the required Ford procedure, id. at 418-31 (Powell, J., concurring in part and concurring in the judgment), was generally regarded as controlling pursuant to the principle that, when there is no majority opinion, the position of the median Justice can be treated as the opinion of the Court. See Marks v. United States, 430 U.S. 188, 193 (1977).

(183) 551 U.S. 930, 949 (2007).

(184) Id. (quoting Ford, 477 U.S. at 424-26 (Powell, J., concurring in part and concurring in the judgment)) (internal quotation marks omitted).

(185) See, e.g., TEX. CODE CRIM. PROG. ANN. art. 46.05 (West 2017) (specifying procedure for litigating Ford claims).

(186) See, e.g., Panetti v. Stephens, 727 F.3d 398, 410 (5th Cir. 2013) (articulating test as a "rational understanding" standard).

(187) U.S. CONST. amend. VI.

(188) See Powell v. Alabama, 287 U.S. 45, 73 (1932).

(189) See Johnson v. Zerbst, 304 U.S. 458, 463 (1938).

(190) See Hamilton v. Alabama, 368 U.S. 52, 54 (1961).

(191) 372 U.S. 335 (1963).

(192) See id. at 343-44.

(193) See Padilla v. Kentucky, 559 U.S. 356, 373 (2010).

(194) See Douglas v. California, 372 U.S. 353, 356-58 (1963).

(195) See infra notes 196-209.

(196) 481 U.S. 551, 555 (1987). The Court had previously suggested, in Johnson v. Avery, 393 U.S. 483 (1969), that no constitutional right to postconviction counsel existed. See id. at 488.

(197) 417 U.S. 600 (1974).

(198) See Finley, 481 U.S. at 555 (citing Moffitt, 417 U.S. at 610-11).

(199) See Moffitt, 417 U.S. at 602-11.

(200) Finley, 481 U.S. at 556-57.

(201) See id. at 556.

(202) See id. at 557.

(203) 372 U.S. 353 (1963).

(204) See id. at 357-58.

(205) Finley, 481 U.S. at 557.

(206) See, e.g., Williams v. Pennsylvania, 136 S. Ct. 1899, 1921 (2016) ("The postconviction petitioner has no constitutional right to counsel."); Coleman v. Thompson, 501 U.S. 722, 752 (1991) ("There is no constitutional right to an attorney in state postconviction proceedings.").

(207) 492 U.S. 1 (1989) (plurality opinion).

(208) See id. at 12-13.

(209) See id. at 14-15 (Kennedy, J., concurring in the judgment).

(210) See, e.g., Roe v. Flores-Ortega, 528 U.S. 470, 492-93 (2000) (Souter, J., concurring in part and dissenting in part) (explaining a predicament faced by state inmates because they have "no right to counsel in state postconviction proceedings").

(211) See Montgomery v. Louisiana, 136 S. Ct. 718, 729 (2016).

(212) 567 U.S. 460 (2012).

(213) See id. at 465.

(214) 489 U.S. 288 (1989).

(215) Part of Teague was only a plurality opinion, but it was subsequently adopted in toto. Penry v. Lynaugh, 492 U.S. 302, 313 (1989), abrogated on other grounds by Atkins v. Virginia, 536 U.S. 304 (2002).

(216) See Teague, 489 U.S. at 301 (plurality opinion).

(217) See id. at 310.

(218) See id. at 311.

(219) 136 S. Ct. 718 (2016).

(220) See id. at 727.

(221) See, e.g., Michigan v. Long, 463 U.S. 1032, 1040-41 (1983) (setting forth modern test and presumptions for independence inquiry).

(222) In Montgomery itself, the parties agreed that the Supreme Court had appellate jurisdiction to decide the case, so the argument that the state ground was independent was made by a Court-appointed amicus. See Montgomery, 136 S. Ct. at 727.

(223) See id. at 731-32.

(224) See id. at 729.

(225) See A. Christopher Bryant, Retroactive Application of "New Rules" and the Antiterrorism and Effective Death Penalty Act, 70 GEO. WASH. L. REV. 1, 6-7 (2002).

(226) 433 U.S. 72 (1977).

(227) See id. at 90-91.


(229) See id.

(230) See id. at 3240.

(231) See Mark Tushnet & Larry Yackle, Symbolic Statutes and Real Laws: The Pathologies of the Antiterrorism and Effective Death Penalty Act and the Prison Litigation Reform Act, 47 DUKE L.J. 1, 31 (1997).

(232) See Kovarsky, supra note 135, at 467.

(233) Wainwright v. Sykes, 433 U.S. 72 (1977).

(234) Id. at 87.

(235) See Coleman v. Thompson, 501 U.S. 722, 755 (1991).

(236) See Davila v. Davis, 137 S. Ct. 2058, 2065 (2017) (describing Martinez as an exception to the rule that "[a]ttorney error that does not violate the Constitution... is attributed to the prisoner").

(237) See Martinez v. Ryan, 132 S. Ct. 1309, 1315 (2012).

(238) Id. In Maples v. Thomas, the Court held that attorney abandonment could excuse forfeiture of any claim under the Sixth Amendment or under any other constitutional provision. 132 S. Ct. 912, 927 (2012).

(239) 560 U.S. 631 (2010).

(240) The federal limitations provision appears in 28 U.S.C. [section] 2244(d) (2012).

(241) 28 U.S.C. [section] 2244(d) (2) provides for the tolling of the limitations period only for the pendency of State PCR, but does not specify any condition under which failure to meet the limitations provisions should be excused.

(242) See supra note 149 and accompanying text.

(243) 562 U.S. 86 (2011).

(244) See id. at 98.

(245) 563 U.S. 170 (2011).

(246) See id. at 181-82.

(247) The spillover intervention is not limited to a subsidy for representation, but also includes a subsidy for fact development. For the small pocket of state postconviction litigation preceded by federal habeas litigation, there is a spillover from any investigation performed during the federal habeas phase. There is no way to "exclude" whatever information counsel uncovers from use in the subsequent proceeding. In economic parlance, as between the federal habeas and subsequent state postconviction phases, the information is nonexcludable. As a result, an inmate that develops a claim in federal court but is told to return to state court to exhaust it may have benefitted from resources to fact development. Indeed, such a subsidy appears in both the general and death-penalty provisions. The expert spillover is not particularly large, although it can be significant in situations where investigators uncover new claims while developing the federal record.

(248) Under 18 U.S.C. [section] 3599 (2012), any inmate sentenced to death is entitled to an appointed federal habeas lawyer. Subsection 3599(f) entitles those inmates to investigative resources that are "reasonably necessary" to the representation. Under 18 U.S.C. [section] 3006A, any inmate serving a noncapital sentence is entitled to CJA representation if a federal court "determines that the interests of justice so require."

(249) See, e.g., Gibson v. Turpin, 513 S.E.2d 186, 189 (Ga. 1999) (rejecting argument that capital inmate was entitled to state postconviction counsel); see also supra note 29 (documenting absence of right to counsel in Alabama and Georgia).

(250) 18 U.S.C. [section] 3006A(c).

(251) Id. [section] 3599(e).

(252) Id.

(253) 556 U.S. 180 (2009).

(254) Id. at 182-83.

(255) Id. at 189-90.

(256) Id. at 190 n.7.

(257) See supra notes 53-59 and accompanying text.

(258) Compare 18 U.S.C. [section] 3006(b) (2012) (noncapital inmate provision stating that the "appointment may be made retroactive to include any representation furnished pursuant to the plan prior to appointment"), with id. [section] 3599 (capital inmate provisions containing no parallel rule).

(259) Although reading the CJA statute to include such proceedings seems natural for the language, few jurisdictions have actually done so. Only the Fifth Circuit has unequivocally stated that the federal appointment includes subsequent state representation. See Wilkins v. Davis, 832 F.3d 547, 558 (5th Cir. 2016).

(260) There are rare scenarios where considerable state postconviction litigation follows the federal appointment. For example, if an inmate files a "mixed petition" containing exhausted claims, the federal proceeding will usually be stayed and held in abeyance to permit the inmate to take the unexhausted claims back to state court. See Rhines v. Weber, 544 U.S. 269, 275 (2005). Such stay-and-abeyance procedure increases the spillover effect.

(261) See, e.g., Jackson v. State, 983 So. 2d 562, 573 (Fla. 2008) (reciting the one-bite-at-the-apple principle); Ex parte Saenz, 491 S.W.3d 819, 824 (Tex. Crim. App. 2016) (same).

(262) See Andrew Hammel, Diabolical Federalism: A Functional Critique and Proposed Reconstruction of Death Penalty Federal Habeas, 39 AM. CRIM. L. REV. 1, 11 (2002).

(263) See id.

(264) See Stewart v. Martinez-Villareal, 523 U.S. 637, 643 (1998).

(265) By definition, a law is "new" if it is announced after a conviction, and a successive proceeding will be necessary unless the initial postconviction litigation remains pending.

(266) See infra note 322.

(267) Other types of substantive constraints come into play when federal interventions are targeted at federal custody because they touch retroactivity principles adjudicated under Teague v. Lane, 489 U.S. 288 (1989)--including decisions that, as a matter of statutory interpretation, narrow the applicability of a criminal statute. See Jason M. Zarrow & William H. Milliken, The Retroactivity of Substantive Rules to Cases on Collateral Review and the AEDPA, with a Special Focus on Miller v. Alabama, 48 IND. L. REV. 931, 955 (2015).

(268) U.S. CONST. amend. VIII (emphasis added).

(269) See generally John F. Stinneford, Rethinking Proportionality Under the Cruel and Unusual Punishments Clause, 97 VA. L. REV. 899 (2011) (arguing, based on original meaning, that proportionality is a constraint necessitated by retributivist limits on punishment).

(270) 217 U.S. 349 (1910).

(271) Id. at 367.

(272) Id. at 378.

(273) Trop v. Dulles, 356 U.S. 86, 101 (1958) (plurality opinion).

(274) See Robinson v. California, 370 U.S. 660, 667-68 (1962).

(275) See, e.g., Ewing v. California, 538 U.S. 11, 30-31 (2003) (plurality opinion) (rejecting proportionality challenge to "three-strikes" sentencing); Harmelin v. Michigan, 501 U.S. 957, 994-96 (1991) (rejecting proportionality challenge to mandatory LWOP possession of more than 650 grams of cocaine); Solem v. Helm, 463 U.S. 277, 279-84, 303 (1983) (sustaining proportionality challenge to LWOP for a defendant's seventh nonviolent felony); Rummel v. Estelle, 445 U.S. 263, 265 (1980) (rejecting proportionality challenge to "three-strikes" sentencing).

(276) 433 U.S. 584 (1977) (plurality opinion).

(277) See id. at 592.

(278) See supra notes 88-95 and accompanying text.

(279) See Rachel E. Barkow, The Court of Life and Death: The Two Tracks of Constitutional Sentencing Law and the Case for Uniformity. 107 MICH. L. REV. 1145, 1148 (2009).

(280) 536 U.S. 304 (2002).

(281) 477 U.S. 399 (1986).

(282) Estimates of the fraction of death-row inmates with intellectual disability range between four and twenty percent. See Peggy M. Tobolowsky, A Different Path Taken: Texas Capital Offenders' Post-Atkins Claims of Mental Retardation, 39 HASTINGS CONST. L.Q. 1, 143 (2011) (collecting authorities). Between five and ten percent of death-row inmates have severe mental illness sufficient to file a colorable Ford claim. See John H. Blume et al., Killing the Oblivious: An Empirical Study of Competency to Be Executed Litigation, 82 UMKC L. REV. 335, 356 (2014).

(283) See Miller v. Alabama, 567 U.S. 460, 465 (2012).

(284) See Graham v. Florida, 560 U.S. 48, 64 (2010) ("Thus, adding the individuals counted by the study to those we have been able to locate independently, there are 123 juvenile nonhomicide offenders serving life without parole sentences.").

(285) Atkins v. Virginia, 536 U.S. 304 (2002).

(286) 567 U.S. 460.

(287) See sources collected supra note 3.

(288) See Coleman v. Thompson, 501 U.S. 722, 752 (1991).

(289) 477 U.S. 399, 401 (1986).

(290) 551 U.S. 930, 959-60 (2007).

(291) See id. at 949-50.

(292) See supra notes 76-77 and accompanying text.

(293) 136 S. Ct. 718 (2016).

(294) See id. at 729.

(295) See id. at 732.

(296) See Teague v. Lane, 489 U.S. 288, 301 (1989) (plurality opinion) (old law); id. at 311 (watershed rules).

(297) Neither the general condition that the state decision be on the merits nor the sub-sectional exceptions apply differently depending on the underlying claim. See 28 U.S.C. [section] 2254(d) (2012).

(298) See, e.g., id. [section] 2254(b) (exhaustion rule); id. [section] 2254(d)(2) (statutory tolling). But see id. [section] 2254(e) (2) (A) (i) (pegging availability of federal hearings for nondiligent claimants to the announcement of a new and retroactive constitutional rule).

(299) See House v. Bell, 547 U.S. 518, 536 (2006).

(300) See, e.g., id. at 536-38 (explaining relationship between miscarriage-of-justice gateway and actual-innocence standard).

(301) See Sawyer v. Whitley, 505 U.S. 333, 340 (1992).

(302) Cf. Rocha v. Thaler, 626 F.3d 815, 826-27 (5th Cir. 2010) (specifying the different types of Eighth Amendment claims that fall in under the miscarriage-of-justice gateway). The Supreme Court has declined to answer whether the miscarriage-of-justice exception applies to claims that a sentence was erroneously enhanced--a logical equivalent of a death ineligibility claim--and has had no occasion to consider whether it extends to the noncapital eligibility rules. See Dretke v. Haley, 541 U.S. 386, 388-89 (2004) (avoiding question of how actual-innocence-of-death-penalty rule applied in noncapital scenario).

(303) See Coleman v. Thompson, 501 U.S. 722, 750 (1991).

(304) See id.

(305) See Edwards v. Carpenter, 529 U.S. 446, 451 (2000).

(306) 132 S. Ct. 1309 (2012).

(307) 133 S. Ct. 1911 (2013).

(308) See id. at 1918; Martinez, 132 S. Ct. at 1315.

(309) See Martinez, 132 S. Ct. at 1315.

(310) The CJA subdivides into a general provision for noncapital inmates, 18 U.S.C. [section] 3006A (2012), and a provision for capital inmates, id. [section] 3599.

(311) See Martel v. Clair, 565 U.S. 648, 659 (2012).

(312) 18 U.S.C. [section] 3599(e).

(313) See Panetti v. Quarterman, 551 U.S. 930 (2007); Ford v. Wainwright, 477 U.S. 399 (1986).

(314) See Harbison v. Bell, 556 U.S. 180, 189 (2009) ("State habeas is not a stage 'subsequent' to federal habeas. Just the opposite: Petitioners must exhaust their claims in state court before seeking federal habeas relief").

(315) See id. at 190 n.7 (explaining that federal resources are available for state postconviction litigation because they are within [section] 3599(e)'s reference to "other appropriate motions and procedures," rather than its reference to "subsequent post-conviction" proceedings.).

(316) See, e.g., TEX. CODE CRIM. PROC ANN. art. 11.07, [section] 4(a)(1) (West 2017) (providing for merits consideration for successive claims when "the factual or legal basis for the claim was unavailable on the date the applicant filed the previous application").

(317) See Harbison, 556 U.S. at 190 n.7; Wilkins v. Davis, 832 F.3d 547, 557-58 (5th Cir. 2016) (holding that [section] 3599(e) entitles inmates to lawyers and ancillary services for the purposes of state postconviction litigation that follows the federal appointment).

(318) See, e.g., Nishi Kumar, Note, Cruel, Unusual, and Completely Backwards: An Argument for Retroactive Application of the Eighth Amendment, 90 N.Y.U. L. REV. 1331 (2015).

(319) See, e.g., Sawyer v. Whitley, 505 U.S. 333 (1992).

(320) See Teague v. Lane, 489 U.S. 288, 308-09 (1989) (plurality opinion).

(321) Martinez v. Ryan, 132 S. Ct. 1309, 1315 (2012).

(322) See, e.g., Stovall v. Denno, 388 U.S. 293, 302 (1967) (acknowledging general rule barring use of out-of-court identification based on unnecessarily suggestive identification procedure, but allowing out-of-court identification in this case based on the totality of the circumstances); United States v. Wade, 388 U.S. 218, 228-39 (1967) (requiring that defendant have counsel at postindictment lineup); Miranda v. Arizona, 384 U.S. 436, 467 (1966) (holding that the Fifth Amendment requires that suspects be warned prior to custodial interrogation); Griffin v. California, 380 U.S. 609, 615 (1965) (announcing that the Fifth Amendment "forbids either comment by the prosecution on the accused's silence or instructions by the court that such silence is evidence of guilt"); Pointer v. Texas, 380 U.S. 400, 407-08 (1965) (recognizing a defendant's Sixth Amendment right to confront and cross-examine trial witnesses); Massiah v. United States, 377 U.S. 201, 207 (1964) (imposing a Sixth Amendment bar against using incriminating statements that law enforcement deliberately elicited after indictment and in counsel's absence); Brady v. Maryland, 373 U.S. 83, 87 (1963) (announcing prosecutors' duty to disclose exculpatory evidence); Douglas v. California, 372 U.S. 353, 357-58 (1963) (entitling an indigent defendant to appointed counsel during any mandatory state appeal); Griffin v. Illinois, 351 U.S. 12, 18-19 (1956) (plurality opinion) (giving indigent defendants rights to free trial transcripts to ensure adequate appellate consideration).

(323) See, e.g., Duncan v. Louisiana, 391 U.S. 145, 159-62 (1968) (Sixth Amendment jury trial right in nonpetty criminal cases); Pointer, 380 U.S. at 403 (Sixth Amendment confrontation clause right); Aguilar v. Texas, 378 U.S. 108, 111 (1964) (Fourth Amendment warrant requirement), abrogated on other grounds by Illinois v. Gates, 462 U.S. 213 (1983); Malloy v. Hogan, 378 U.S. 1, 6 (1964) (Fifth Amendment protection applicable to the states); Gideon v. Wainwright, 372 U.S. 335, 340, 342-44 (1963) (Sixth Amendment right to counsel in all felony cases); Robinson v. California, 370 U.S. 660, 667-68 (1962) (Eighth Amendment cruel and unusual punishments clause); Mapp v. Ohio, 367 U.S. 643, 653 (1961) (Fourth Amendment exclusionary rule).

(324) See Erwin Chemerinsky & Larry Kramer, Defining the Role of the Federal Courts, 1990 BYU L. REV. 67, 87 ("Less obviously, federal courts perform an essentially appellate function in reviewing petitions for writs of habeas corpus from state prisoners."); Barry Friedman, A Tale of Two Habeas, 73 MINN. L. REV. 247, 254 (1988) ("Accordingly, the federal habeas courts were to act as surrogates for the United States Supreme Court through habeas review, in effect exercising appellate jurisdiction over state criminal proceedings.").

(325) 136 S. Ct. 718 (2016).

(326) See id. at 727-32.

(327) See id. at 729-30.

(328) See, e.g., STEIKER & STEIKER, supra note 5, at 164.

(329) See Atkins v. Virginia, 536 U.S. 304, 321 (2002) (intellectual disability); Ford v. Wainwright, 477 U.S. 399, 401 (1986) (insanity).

(330) See Roper v. Simmons, 543 U.S. 551, 578-79 (2005).

(331) See Miller v. Alabama, 567 U.S. 460, 465 (2012).

(332) See Tison v. Arizona, 481 U.S. 137, 158 (1987).

(333) See generally Richard H. Fallon, Jr., Making Sense of Overbreadth, 100 YALE L.J. 853 (1991) (analyzing how overbreadth rulings interact with criminal punishment).

(334) See supra notes 46-52 and accompanying text.

(335) See Atkins v. Virginia, 536 U.S. 304, 321 (2002).

(336) See supra notes 53-59 and accompanying text.

(337) See supra notes 212-213 and accompanying text.

(338) See Strickland v. Washington, 466 U.S. 668, 687 (1984).

(339) 554 U.S. 407 (2008).

(340) See id. at 421.

(341) Roper v. Simmons, 543 U.S. 551, 578 (2005).

(342) Miller v. Alabama, 567 U.S. 460, 465 (2012).

(343) See, e.g., Hall v. Florida, 134 S. Ct. 1986, 1993-2001 (2014) (discussing treatment of IQ testing in Atkins litigation).

(344) See Richard J. Bonnie & Katherine Gustafson, The Challenge of Implementing Atkins v. Virginia: How Legislatures and Courts Can Promote Accurate Assessments and Adjudications of Mental Retardation in Death Penalty Cases, 41 U. RICH. L. REV. 811, 855 (2007).

(345) It is true that death penalty proceedings are substantially more expensive than postconviction claims involving noncapital sentences, and that many substantive constraints are restrictions on death sentencing. That the average postconviction cost of litigating a death sentence exceeds the average postconviction cost of litigating a noncapital sentence, however, does not change my observation about the incremental cost of enforcing a substantive constraint through a single claim.

(346) 136 S. Ct. 718 (2016).

(347) Id. at 734; see Miller, 567 U.S. 460.

(348) See Montgomery, 136 S. Ct. at 726.

(349) See id. at 727.

(350) See generally Kermit Roosevelt III, Essay, Light from Dead Stars: The Procedural Adequate and Independent State Ground Reconsidered, 103 COLUM. L. REV. 1888 (2003) (exploring the difference between "substantive" and "procedural" independence).

(351) See Montgomery, 136 S. Ct. at 729.

(352) Id. at 731-32.

(353) See id. at 729-30.

(354) See id. at 731.

(355) Id. at 729-30.

(356) Id. at 730.

(357) Id. at 729.

(358) Id.

(359) See Alice Ristroph, Regulation or Resistance? A Counter-Narrative of Constitutional Criminal Procedure, 95 B.U. L. REV. 1555, 1598 (2015); Nirej Sekhon, Willing Suspects and Docile Defendants: The Contradictory Role of Consent in Criminal Procedure, 46 HARV. C.R.-C.L. L. REV. 103, 121, 131 (2011).

(360) Montgomery explains that because a punishment tainted by procedural error might still be lawful, "a trial conducted under a procedure found to be unconstitutional in a later case does not, as a general matter, have the automatic consequence of invalidating a defendant's conviction or sentence." 136 S. Ct. at 730.

(361) See, e.g., Moore v. Dempsey, 261 U.S. 86, 89-90 (1923) (establishing that due process violations rendered convictions void). In Brown v. Allen, 344 U.S. 443 (1953), the Supreme Court eventually embraced the notion that convictions could be invalid because of due process violations--including violations occurring by way of constitutional provisions incorporated through the Fourteenth Amendment. See Daniels v. Allen, decided sub rum. Brown, 344 U.S. at 488-513 (1953) (opinion of Frankfurter, J.).

(362) 100 U.S. 371 (1879).

(363) See Montgomery, 136 S. Ct. at 730-32.

(364) Specifically, Siebold was an interpretation of the Judiciary Act of 1789, ch. 20, [section] 14, 1 Stat. 73, 81-82, which granted the Supreme Court appellate jurisdiction to issue original habeas writs to lower federal courts. The operative question in Siebold was, in light of the fact that there was an express grant of statutory jurisdiction, whether there was something inherent in the nature of the writ that precluded review of the inferior court judgment. See Paul M. Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 HARV. L. REV. 441, 468 n.59 (1963).

(365) See Act of Feb. 5, 1867, ch. 28, [section] 1, 14 Stat. 385 (establishing by statute the availability of habeas relief for state prisoners); Ex parte Dorr, 44 U.S. (3 How.) 103, 104-05 (1845) (explaining that the 1789 Act did not reach state custody).

(366) Just Scalia was explaining the distinction between a statutory and a constitutional rule when he wrote, "It is a decision about this Court's statutory power to grant the Original Writ, not about its constitutional obligation to do so. Nowhere in Siebold did this Court intimate that relief was constitutionally required...." Montgomery, 136 S. Ct. at 740 (Scalia, J., dissenting). Justice Scalia, however, fails to provide any explanation along the lines of that provided supra note 364, which more clearly conveys what Siebold was deciding.

(367) The Supreme Court declared: "A penalty imposed pursuant to an unconstitutional law is no less void because the prisoner's sentence became final before the law was held unconstitutional. There is no grandfather clause that permits States to enforce punishments the Constitution forbids." Montgomery, 136 S. Ct. at 731. Whatever that argument's abstract appeal, the entire theory of retroactivity jurisprudence is built on the idea that some convictions will be in violation of law but lack a remedy. See generally Fallon & Meltzer, supra note 80 (setting forth the canonical argument).

(368) Siebold was preceded by two other cases that began to expand the definition of "unlawful custody" beyond a conviction imposed by a court without jurisdiction. See Ex parte Parks, 93 U.S. 18, 21 (1876) (emphasizing that the court had no power to correct "mere error"); Ex parte Lange, 85 U.S. (18 Wall.) 163, 177-78 (1873) (in the context of a double-jeopardy violation, rejecting proposition that a judgment imposed by a jurisdictionally competent court is necessarily lawful). The Court subsequently held that a major due process violation could render a conviction unlawful, see Moore v. Dempsey, 261 U.S. 86 (1923), and the Court positioned Moore as a major reason for the result in Brown v. Allen, 344 U.S. 443 (1953), which is discussed supra note 361. No matter what side of the historical debate academics take, none appear to believe that Siebold represents an endpoint for the meaning of "lawful custody." Compare, e.g., Bator, supra note 364, at 463-99 (restrictive view of precedent), with Gary Peller, In Defense of Federal Habeas Corpus Relitigation, 16 HARV. C.R.-C.L. L. REV. 579, 604-63 (1982) (expansive view of precedent).

(369) See, e.g., Waley v. Johnston, 316 U.S. 101, 104 (1942) (per curiam) (permitting relief on a coerced confession claim because "petitioner's constitutional rights were infringed"); Johnson v. Zerbst, 304 U.S. 458, 467-68 (1938) (voiding a conviction on the violation of the right to counsel); Moore, 261 U.S. at 90-91 (holding that a due process violation voided a conviction).

(370) The proposition that retroactivity rules were something other than subconstitutional restraints on the federal habeas remedy was, notwithstanding Montgomery's lengthy suggestion to the contrary, completely novel. The Court sometimes positioned the retroactivity rule as something like federal common law, and other times positioned it as an interpretation of the federal habeas statute. See, e.g., Danforth v. Minnesota, 552 U.S. 264, 278 (2008) ("Teague's general rule of nonretroactivity was an exercise of this Court's power to interpret the federal habeas statute."); Zarrow & Milliken, supra note 267, at 955-64 (analyzing the Court's substantive-rule retroactivity doctrine as a common-law phenomenon). The Montgomery Court brushed aside prior cases operating under such an assumption as dicta--including foundational language from Teague--because Teague, it explained, "is best understood as resting upon constitutional premises." Montgomery, 136 S. Ct. at 729.

(371) See Teague v. Lane, 489 U.S. 288, 311 (1989) (plurality opinion).

(372) 28 U.S.C. [section] 2244(b) (2) (B) (2012) provides that new-fact petitions cannot proceed unless "no reasonable factfinder would have found the applicant guilty of the underlying offense."

(373) The federal limitations statute applicable to state-inmate claims appears in 28 U.S.C. [section] 2244(d).

(374) Indeed, civil judgments are voidable if process in the rendering jurisdiction does not satisfy constitutional standards. See Kovarsky, supra note 124, at 643-45.

(375) See, e.g., 28 U.S.C. [section] 2244(d)(1)(C) (describing scenario in which state inmates are foreclosed from federal habeas consideration in cases of a violated substantive constraint).

(376) See id. [section] 2254(a) (extending the habeas remedy to all "in custody in violation of the Constitution or laws or treaties of the United States").

(377) See generally Daniel J. Meltzer, State Court Forfeitures of Federal Rights, 99 HARV. L. REV. 1128, 1137-45 (1986) (describing four different categories of inadequate state procedural rules); Catherine T. Struve, Direct and Collateral Federal Court Review of the Adequacy of State Procedural Rules, 103 COLUM. L. REV. 243, 250-77 (2003) (developing a "Taxonomy of Adequacy").

(378) See generally PETER W. LOW ET AL., 2017 SUPPLEMENT TO FEDERAL COURTS AND THE LAW OF FEDERAL-STATE RELATIONS 5-7 (8th ed. 2017) (explaining decisional basis for such a holding).

(379) 1 am largely concerned with Montgomery's legal rule, but worth mentioning are the issues with Miller as a vehicle for declaring the constitutional requirement that substantive rules apply retroactively. Miller is not (as it is sometimes described) a rule against juvenile LWOP; it is a rule against mandatory juvenile LWOP. 567 U.S. 460, 465 (2012). Life without parole can still be imposed on juveniles; the sentencing must simply include individualized consideration. See id. Montgomery contemplates categorical rules against certain types of punishments for certain types of crime irrespective of procedure, and rules against mandatory imposition of otherwise permissible punishment do not fit comfortably within that framework.

(380) See Brief of Former State Supreme Court Justices as Amici Curiae in Support of Petitioner at 7, Martinez v. Ryan, 132 S. Ct. 1309 (2012) (No. 10-1001), 2011 WL 4427080.

(381) See generally King, supra note 4, at 2449-55 (explaining the limited effect that Martinez and Trevino will have on the state postconviction process).

(382) See id. at 2451.

(383) See id. at 2451-52 (collecting potential alternative responses).

[c] 2017 Lee Kovarsky. Individuals and nonprofit institutions may reproduce and distribute copies of this Article in any format at or below cost, for educational purposes, so long as each copy identifies the authors, provides a citation to the Notre Dame Law Review, and includes this provision in the copyright notice.
COPYRIGHT 2017 University of Notre Dame Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kovarsky, Lee
Publication:Notre Dame Law Review
Date:Dec 1, 2017

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters