Ian D. Eppler: J.D. Candidate, Class of 2019, Harvard Law School. Thanks are due to Professor Carol Steiker for her guidance and insight in drafting this article, and to the staff of the National Lawyers Guild Review for their thoughtful comments. All errors are my own.
The writ of habeas corpus—the so-called “great writ”1—has been described as the “the best and only sufficient defense of personal freedom.”2 Rooted in English common law, the writ allows a detainee or prisoner to petition a court to test the validity of his or her detention and order his or her release if there is no legal basis for continued detention.3 The writ’s role as a “vital instrument for the protection of individual liberty”4 was known to the Framers of the Constitution, who explicitly protected the writ by forbidding its suspension except “when in Cases of Rebellion or Invasion the public Safety may require it.”5 It is not an exaggeration to characterize the writ of habeas corpus as America’s oldest and most fundamental civil rights remedy.6
Under a series of Congressional statutory enactments7 and court decisions,8 the federal writ of habeas corpus has become a means of ensuring that state criminal proceedings comply with the Constitution. Federal habeas corpus review thus became an important tool to protect individual rights from oft-hostile state governments, from the lynching era9 to the civil rights era10 to the modern era of capital punishment and mass incarceration.11 At its high-water mark in the Warren Court era, habeas corpus was a mechanism that allowed for federal oversight of nearly all aspects of state criminal proceedings.12
However, decades of conservative courts and tough-on-crime legislatures have rendered federal habeas a shell of what it once was.13 In 1996, Congress sharply limited the scope of federal habeas by passing the Antiterrorism and Effective Death Penalty Act (AEDPA),14 which, among other provisions, required federal courts to defer to factual findings and legal conclusions of state courts in all but the most egregious cases of error.15 The Supreme Court has also limited access to federal habeas review in dozens of ways.16 For instance, the Supreme Court has precluded habeas petitioners from receiving the retroactive benefit of newly announced rules of criminal procedure in most circumstances,17 and insulated most Fourth Amendment claims against federal review.18 The effective result is that state prisoners, and the attorneys defending them, have lost an important tool to vindicate their rights.
Another significant way that the Supreme Court has cut back on the federal habeas remedy is through expansion of the procedural default doctrine. The doctrine of procedural default in federal habeas corpus review prevents a federal court from granting habeas relief to a petitioner who has “fail[ed] to raise a claim at the time or in the manner required by state procedures” during state appellate or post-conviction review.19 The procedural default doctrine has a particularly pronounced impact because of its intersection with other Supreme Court precedent that limits access to post-conviction counsel.20 Since criminal defendants are not entitled to counsel in post-conviction proceedings,21 and courts do not review the effectiveness of post-conviction counsel in instances where defendants are able to secure assistance,22 criminal defendants frequently procedurally default viable claims due to error by their attorneys or error that they themselves make while proceeding pro se.23 Procedural default therefore becomes the “‘principal escape route’ from federal habeas.”24 As a result, many criminal defendants are often unable to secure federal review of a host of important constitutional claims, such as ineffective assistance of trial counsel.25
Procedural default serves as a “particularly nefarious”26 barrier to federal review of state criminal convictions. Yet the news is not all bad. In recent years, the Supreme Court has taken tentative steps in the direction of easing strict procedural default rules and expanding the set of circumstances in which state procedural defaults may be excused, although the exact contours of the Court’s new doctrinal direction remain unclear. In Martinez v. Ryan27 and Trevino v. Thaler,28 the Supreme Court created a limited exception to the generally applicable procedural default rule, by which a subset of procedural defaults may be excused. This exception applies to a situation in which error during a petitioner’s first state collateral review proceeding results in the default of a substantial ineffective assistance of trial counsel claim, and the state collateral review proceeding had occurred in a state post-conviction review scheme that de jure or de facto requires claims of ineffective assistance of trial counsel to be brought in state collateral review proceedings. However, in its recent decision in Davila v. Davis,29 the Court cabined the Martinez/Trevino exception, holding it did not apply when error during a petitioner’s initial state post-conviction review proceeding results in the default of a claim of ineffective assistance of appellate counsel. After Davila, the scope of the nascent Martinez/Trevino doctrine—most notably, its applicability to other constitutional claims—remains unsettled, yet potentially fertile, terrain for criminal defense and civil rights attorneys looking to vindicate the rights of their clients in a federal forum. This Article provides an overview of the lay of the land. Part I summarizes the state of procedural default doctrine before Davila. Part II reviews and analyzes the Court’s opinion in Davila. Part III argues that Martinez, Trevino, and Davila collectively outline core principles that dictate when state procedural defaults may be excused and when they may not, and that those principles suggest that the Martinez/Trevino exception should be applied to at least one category of constitutional claims beyond ineffective assistance of trial counsel: namely claims arising under Brady v. Maryland that allege prosecutorial failure to produce material, exculpatory evidence.30
Procedural Default Doctrine Before Davila
The Supreme Court first addressed procedural default in Fay v. Noia.31 In Noia, the Court adopted a narrow rule of procedural default, “hold[ing] that the jurisdiction of the federal courts on habeas corpus is not affected by procedural defaults incurred by the applicant during the state court proceedings,”32 except when the federal habeas petitioner “understandingly and knowingly . . . deliberate[ly] by-pass[ed] . . . state procedures.”33 However, in the years following Noia, the Court expressed concern that Noia’s limited definition of procedural default did not give sufficient respect to “considerations of comity and concerns for the orderly administration of criminal justice.”34 In Wainwright v. Sykes,35 the Supreme Court replaced Noia’s “deliberate bypass standard” with a “cause and prejudice”36 standard, by which a federal habeas petitioner’s failure to raise a claim in state post-conviction proceedings in accordance with state procedural requirements would render the claim procedurally defaulted upon federal habeas review “absent a showing of cause for the noncompliance and some showing of actual prejudice resulting from the alleged constitutional violation.”37
The Supreme Court has repeatedly reaffirmed the Sykes “cause and prejudice” standard as the basic framework for procedural default,38 but Sykes itself left open several questions regarding the application of the standard.39 Among them was the question of whether attorney “error or negligence” in state proceedings, as opposed to the intentional circumvention of state proceedings contemplated in Noia, could qualify as the requisite cause for excusing a default.40 In Coleman v. Thompson,41 the Court ultimately ruled that attorney error, at least in post-conviction proceedings, could not constitute cause and prejudice. The Court relied on two lines of precedent in reaching its holding in Coleman. In Murray v. Carrier,42 the Court concluded that attorney error at trial or on direct appeal only constituted “cause” under Sykes if the attorney error amounted to Constitutionally ineffective assistance of counsel.43 In Pennsylvania v. Finley44 and its progeny,45 the Court held that there is no right to counsel in post-conviction proceedings. In Coleman, the Court merged these two lines of cases to hold that, since there is no constitutional right to counsel in post-conviction proceedings, post-conviction counsel cannot be unconstitutionally ineffective, and ineffective assistance of counsel on state collateral review can therefore never serve as cause for a procedural default.46
The strict rule of procedural default arising from Sykes and Coleman— under which a procedural default resulting from no counsel or ineffective assistance of counsel on state collateral review, would always preclude consideration of the defaulted claim on federal habeas review—prevailed until 2012, when the Supreme Court decided Martinez v. Ryan.47 In Martinez, the Court relied on equitable considerations such as the importance of effective assistance of trial counsel48 and the limited burden on state resources49 to establish what it defined as a “narrow”50 exception to the strict Sykes/Coleman regime: after Martinez, “[i]nadequate assistance of counsel at initial-review [state] collateral proceedings may establish cause for a prisoner’s procedural default of a claim of ineffective assistance at trial”51 when, under state law, the state post-conviction proceeding is “the first designated proceeding for a prisoner to raise a claim of ineffective assistance at trial,”52 and the claim of ineffectiveness is “substantial.”53 Subsequently, in Trevino v. Thaler,54 the Court expanded the newly announced Martinez doctrine to cases in which state law allows claims of ineffective assistance of trial counsel to be raised on direct appeal, but where the “structure and design of the [state] system in actual operation . . . make it ‘virtually impossible’ for an ineffective assistance claim to be presented on direct review.”55 Under the Martinez/Trevino doctrine, then, the following rule applies: generally, ineffective counsel or no counsel during state collateral review cannot excuse a state procedural default, except when the underlying claim is a “substantial” claim of ineffective assistance of trial counsel, the default occurred during a prisoner’s first state collateral review proceeding, and state law requires, either de jure or de facto, that claims of ineffective assistance of trial counsel be raised on post-conviction review.
In Martinez, the Supreme Court left open the question of whether its newly announced equitable rule applied outside of the context of claims of ineffective assistance of trial counsel, declaring that “[o]ur holding here addresses only the constitutional claims [ineffective assistance of trial counsel] presented in this case,”56 and the Court did not use Trevino to clarify whether the newly announced doctrine applied outside of the ineffective assistance of trial counsel context.57 Martinez and Trevino thus begged the question of whether the newly announced doctrine would apply to other constitutional claims commonly asserted in state post-conviction proceedings, such as claims of improper withholding of material exculpatory evidence in violation of Brady v. Maryland (“Brady claims”) 58 or claims of ineffective assistance of appellate counsel.59
In dissent in Martinez, Justice Scalia expressed concern “that the newly announced ‘equitable’ rule will [not] remain limited to ineffective-assistance-of-trial-counsel cases” because “[t]here is not a dime’s worth of difference in principle between those cases and many other cases in which initial state habeas will be the first opportunity for a particular claim to be raised,” citing Brady claims and claims of ineffective assistance of appellate counsel as examples of constitutional claims that could fall into Martinez’s newly announced exception.60 Perhaps inspired by Justice Scalia’s dissent, several commentators subsequently noted that the underlying logic of the Martinez doctrine suggests that it should be applied to a number of other constitutional claims beyond ineffective assistance of trial counsel: some suggested application to Brady claims,61 others suggested application to constitutional claims “impact[ing] fundamental fairness or the accuracy of the guilt/innocence determination,”62 and still others suggested application to all constitutional claims requiring introduction of extra-record evidence.63
In the wake of Martinez and Trevino, circuit courts also split on the application of the doctrine to constitutional claims beyond ineffective assistance of trial counsel. Several circuits addressed the applicability of Martinez and Trevino in the context of ineffective assistance of appellate counsel claims, with the Ninth Circuit concluding that the doctrine applied to claims of ineffective assistance of appellate counsel,64 while the other circuits that addressed the question concluded that it did not.65 The Ninth Circuit also rejected expansions of the Martinez doctrine to procedurally defaulted judicial bias claims66 and, over a dissent, procedurally defaulted Brady claims.67
Davila v. Davis
In Davila v. Davis, 68 the Supreme Court resolved some of the uncertainty regarding the application of the Martinez/Trevino doctrine outside of the ineffective assistance of trial counsel context. The petitioner, Erick Davila, was accused of shooting a woman and her granddaughter in Fort Worth, Texas in 2008.69 Davila was arrested shortly after the shooting and confessed, stating that “he ‘wasn’t aiming at the kids or the woman,’ but that he was trying to kill [the victim’s son] and the other ‘guys on the porch.’”70 Davila was charged with capital murder.71 At trial, the trial court sought to instruct the jury that it could convict Davila of capital murder on a theory of transferred intent: namely, it could convict Davila of capital murder in the deaths of the two victims, whom he did not intend to kill, based on his intent to kill the “guys on the porch.”72 Davila’s trial counsel objected to the proposed instruction, but the trial court overruled the objection and submitted the transferred instruction to the jury, and Davila was convicted of capital murder and sentenced to death.73
Davila subsequently appealed his conviction, but his appellate counsel did not challenge the trial court’s transferred intent instruction,74 and the Texas Court of Criminal Appeals affirmed his conviction.75 After the failure of his direct appeal, Davila sought post-conviction relief in Texas state court, but his counsel neither challenged the transferred intent instruction directly nor contended that his appellate counsel’s failure to challenge the transferred intent instruction on direct appeal constituted ineffective assistance of appellate counsel under Strickland v. Washington76 and its progeny.77 The state trial court denied Davila’s petition for post-conviction relief, and the Texas Court of Criminal Appeals affirmed.78
Davila then sought federal habeas relief, asserting among other claims that his counsel on direct appeal was unconstitutionally ineffective because of her failure to challenge the use of a transferred intent instruction at trial.79 Davila contended that, while his failure to raise an ineffective assistance of appellate counsel claim during state post-conviction proceedings would typically constitute a procedural default precluding consideration of the claim during federal habeas corpus proceedings under the regime set out in Sykes and Coleman, the Martinez/Trevino exception encompassed his ineffective assistance of appellate counsel claim, thus excusing his procedural default.80 The federal district court rejected Davila’s assertion that his claim of ineffective assistance of appellate counsel fell into the Martinez/ Trevino exception and denied habeas relief on the ineffective assistance of appellate counsel claim on the basis that it was procedurally defaulted.81 The 5th Circuit affirmed the district court’s denial of federal habeas relief,82 and the Supreme Court subsequently granted certiorari.83
In a 5-4 decision, the Supreme Court affirmed the 5th Circuit, declining to expand the Martinez/Trevino doctrine to encompass ineffective assistance of appellate counsel claims.84 Writing for the court, Justice Thomas defined the Martinez/Trevino doctrine as a “narrow . . . highly circumscribed, equitable exception”85 to the Sykes/Coleman regime and identified several reasons why claims of ineffective assistance of appellate counsel do not qualify for the Martinez/Trevino equitable exception. The Court’s primary basis for rejecting an expansion of the Martinez/ Trevino doctrine to the ineffective assistance of appellate counsel context was its conclusion that “the Court in Martinez was principally concerned about trial errors—in particular, claims of ineffective assistance of trial counsel.”86 The Court asserted that “[t]he criminal trial enjoys pride of place in our criminal justice system in a way that an appeal from that trial does not.”87 The Court went on to note several reasons for this “pride of place”, including that the Constitution guarantees the right to criminal trial but not appeal,88 the Court’s own extensive history of distinguishing the rights guaranteed at trial to the rights guaranteed post-conviction,89 and the fact that the “stakes for the defendant are highest”90 at trial because trial is where “a presumptively innocent defendant is judged guilty”91 and is “where the trial judge or jury makes factual findings that nearly always receive deference on appeal and collateral review.”92 Ultimately, the Court relied on these aforementioned differences between trial rights and appellate rights to conclude that there is “unique importance [in] protecting a defendant’s trial rights”93 that is not present in the appellate context.
The Court also rejected the argument that expanding the Martinez/Trevino doctrine to ineffective assistance of appellate counsel claims is necessary to protect the aforementioned “unique[ly] import[ant]” trial rights. The court reasoned that, unlike ineffective assistance of trial counsel claims, an exception to the Sykes/Coleman procedural default rule for claims of ineffective assistance of trial counsel is “not required to ensure that meritorious claims of trial error receive review by at least one . . . court—the chief concern identified by this Court in Martinez.”94 The Court noted that trial courts cannot address issues of alleged ineffective assistance of trial counsel “[b] ecause it is difficult to assess a trial attorney’s performance until the trial has ended,”95 and that in a regime in which appellants are de jure or de facto precluded from raising ineffective assistance of trial counsel claims on direct appeal, a procedural default of an ineffective assistance of trial counsel claim means that the underlying claim will never be reviewed by any court.96 By contrast, ineffective assistance of appellate counsel claims are fundamentally premised on errors that initially occurred in the trial court—in every instance in which a petitioner contends ineffective assistance of appellate counsel, the petitioner is doing so because appellate counsel failed to properly appeal an alleged error that occurred in the trial court.97 Thus, one of three circumstances is always true when a petitioner claims ineffective assistance of appellate counsel. Either appellate counsel failed to properly appeal an alleged trial error that was preserved at trial, in which case the trial court was able to contemporaneously rule on the alleged error by means of trial counsel’s objection;98 or appellate counsel failed to appeal an unpreserved error, which is “not deficient performance unless that claim was plainly stronger than those actually presented to the appellate court;”99 or, in the rare case in which the “unpreserved trial error was so obvious that appellate counsel was constitutionally required to raise it on appeal . . . trial counsel likely provided ineffective assistance by failing to object to it in the first instance,”100 thus qualifying the petitioner for the Martinez/Trevino exception based on ineffective assistance of trial counsel.101 Therefore, the court concluded, the concerns about unreviewability underlying Martinez/Trevino are not present in the ineffective assistance of appellate counsel context, implying that a further expansion is not warranted in this new context.
In its equitable analysis, the Court also relied on a line of reasoning, stemming from Martinez, regarding the equitable consequences of the state’s institutional design choices. In Martinez, the court reasoned that, when the state structured its post-conviction review system in a manner that forced ineffective assistance of trial counsel claims into post-conviction review, it “significantly diminishe[d] prisoners’ ability to file such claims” by moving them into a phase of review where counsel is not constitutionally guaranteed.102 Thus, it would be “inequitable to refuse to hear a defaulted claim of ineffective assistance of trial counsel when the State had channeled that claim to a forum where the prisoner might lack the assistance of counsel in raising it.”103 In Trevino, the Court expanded this equitable conclusion to state post-conviction regimes “pursuant to which collateral review was, “as a practical matter, the onl[y] method for raising an ineffective-assistance-of-trial-counsel claim.’”104 By contrast, the Court reasoned, in the ineffective assistance of appellate counsel context, state post-conviction review is inherently the first opportunity to review the claim, because review cannot take place until appeal is complete.105 Therefore, it is not inequitable to charge petitioners with the consequences of a default in this context, because the state is not responsible for structuring its scheme in a manner that makes a default likely.
Finally, the Court weighed the systemic costs and benefits of expanding the Martinez/Trevino exception to claims of ineffective assistance of appellate counsel, and concluded that the systemic costs outweighed the benefits.106 The Court reasoned that the systemic costs of expansion would be far more substantial than in Martinez for two reasons. The Court first noted that, while Martinez/Trevino only apply to states that de jure or de facto channel ineffective assistance of trial counsel claims to state post-conviction review, expanding Martinez/Trevino to ineffective assistance of appellate counsel claims would subject every state to the rule, because ineffective assistance of appellate counsel claims by their nature may not be raised before state post-conviction review.107 The Court also noted that expanding Martinez/Trevino to the ineffective assistance of appellate counsel context would “produce a domino effect”108 by which petitioners could “use those newly reviewable appellate ineffectiveness claims as cause to excuse the default of their underlying claims of trial error”109 and thus “knock down the procedural barriers to federal habeas review of nearly any defaulted claim of trial error.”110 The Court concluded that expansion would therefore “not only impose significant costs on the federal courts,” 111 but also aggravate the “intru[sion] on state sovereignty”112 caused by federal habeas review, “frustrate both the States’ sovereign power to punish offenders and their good-faith attempts to honor constitutional rights,”113 and “undermine the doctrine of procedural default and the values it serves.”114 By contrast, the Court concluded, “the benefit [of expansion] would—as a systemic matter—be small”115 because the “number of meritorious [ineffective assistance of appellate counsel] cases is ‘infinitesimally small.’”116 Therefore, the Court reasoned that expansion of the Martinez/Trevino exception to claims of ineffective assistance of appellate counsel would do little to protect genuine violations of constitutional rights while simultaneously imposing a significant burden on the federal courts and infringing on state sovereignty, suggesting that the equitable analysis in Martinez/Trevino weighed against expansion to this context.
On behalf of four justices, Justice Breyer dissented. As an initial matter, the dissenters highlighted the potentially arbitrary outcomes that could result from a regime that allows an exception to procedural default for claims of ineffective assistance of trial counsel but not claims of ineffective assistance of appellate counsel, noting that certain underlying constitutional claims such as improper jury instructions or prosecutorial misconduct are amenable to being framed on federal habeas review as either ineffective assistance of appellate counsel (because appellate counsel failed to raise the errors on direct appeal) or ineffective assistance of trial counsel (because trial counsel failed to raise them at trial), but that only the trial claim would be reviewable post-Davila if defaulted on state post-conviction review.117 The dissenters went on to critique the Court’s assertion regarding the “unique importance”118 of trial rights, noting that the Constitution guarantees both effective assistance of trial counsel119 and, should a state authorize appellate review, effective assistance of appellate counsel.120 The dissenters also contended that “the Court . . . misses the point”121 when it asserts that claims of ineffective assistance of appellate counsel, unlike claims of ineffective assistance of trial counsel, are distinguishable because they always involve underlying claims that have been reviewed by the initial trial court: the dissenters argued that the very purpose of appellate counsel is to ensure review of potentially erroneous trial court decisions, meaning that trial court review ex ante is no substitute for effective appellate assistance.122 The dissent then disputes the Court’s assertions regarding the systemic costs and benefits of an expansion of the Martinez/Trevino doctrine to claims of ineffective assistance of appellate counsel. The dissent notes that empirical evidence post-Martinez indicates that federal courts have not been overwhelmed by new habeas petitions alleging previously defaulted claims of ineffective assistance of trial counsel,123 and that the Court’s concerns about ineffective assistance of appellate counsel serving as a gateway to a broader range of underlying claims than ineffective assistance of trial counsel are misplaced, since both claims “could serve as the gateway to federal review of a host of trial errors.”124 Finally, the dissenters reiterate that the Court’s distinction between appeal and trial in Davila is arbitrary, and violates “the basic legal principle . . . that requires courts to treat like cases alike” that “should determine the outcome of this case.”125
The Implications of Davila for Brady Claims
In rejecting an extension of the Martinez/Trevino doctrine to claims of ineffective assistance of appellate counsel, the Court clarified the scope of the doctrine and attempted to ensure that the doctrine would remain a “narrow exception.”126 Perhaps counterintuitively, however, Davila nonetheless strengthens the argument that the Martinez/Trevino doctrine should apply to other classes of constitutional claims beyond ineffective assistance of trial counsel. In clarifying the basis for the Martinez/Trevino doctrine, the Supreme Court identified the characteristics by which courts can determine whether a category of claim qualifies for the Martinez/Trevino treatment. These characteristics include claims that implicate the fundamental fairness or accuracy of the trial, are likely to evade review without access to the federal forum via the Martinez/Trevino doctrine, were not addressed earlier in the adjudicative process because of the state’s own policy decisions, are regularly successful, and impose minimal systemic costs. In outlining this list of characteristics, the Court (perhaps unintentionally) suggested that Brady claims127 should be eligible for the Martinez/Trevino exception.
The Davila court’s rejection of the petitioner’s claims is primarily premised on its interpretation of Martinez and Trevino as announcing a limited doctrine rooted in the “unique importance of protecting a defendant’s trial rights.”128 Martinez, in turn, discusses the systemic purpose of the Constitutionally guaranteed trial rights in the context of explaining the importance of effective assistance of trial counsel: according to Martinez, “[d]efense counsel tests the prosecution’s case to ensure that the proceedings serve the function of adjudicating guilt or innocence, while protecting the rights of the person charged.”129 The purpose of the Martinez/Trevino doctrine, then, is not to protect core trial rights for the sake of protecting core trial rights. The purpose of the doctrine is to defend the systemic benefits that core trial rights, such as the right to effective assistance of counsel, are designed to protect: ensuring accurate adjudicative outcomes and promoting fundamental fairness.
The Brady right is similar to the right to effective assistance of trial counsel in that it serves the systemic accuracy-promoting and fairness-promoting goals identified by the Court in Martinez and affirmed in Davila. In announcing the Brady rule, the Brady court (somewhat desultorily) identified systemic fairness as the rule’s fundamental purpose, contending that “[s] ociety wins not only when the guilty are convicted but when criminal trials are fair; our system of the administration of justice suffers when any accused is treated unfairly.”130 In subsequent cases, the Supreme Court emphasized the role of the Brady rule as an accuracy-promoting rule, noting that “[i]ts purpose is . . . to ensure that a miscarriage of justice does not occur”131 and construed the materiality element of the Brady rule announced in Brady itself to emphasize accuracy, limiting Brady to cases where “the nondisclosure was so serious that there is a reasonable probability that the suppressed evidence would have produced a different verdict.”132
The dual accuracy and fairness roles of the Brady rule—with a heavy focus, through a materiality requirement, on accuracy—mirrors the rules governing ineffective assistance of trial counsel. In Strickland v. Washington,133 the Supreme Court, just as it did in Brady, reasoned from conceptions of fundamental fairness reasoning to identify a trial right.134 However, just as it did in Brady, the Court used a materiality element to narrow the rule to cases where the accuracy of the trial outcome was called into question by the Constitutional violation.135 The similarities between the purpose and operation of the Brady doctrine—where the Court has not yet recognized a procedural default exception—and purpose and operation of the ineffective assistance of trial counsel doctrine—where the Court recognized an exception to general procedural default rules in Martinez and Trevino—thus weigh strongly in favor of expanding the Martinez/Trevino doctrine to Brady claims. The Court has identified both rules as core trial rights, with significant implications for fairness and accuracy. The Court has relied on similar reasoning to apply them narrowly to cases where trial outcome accuracy is called into question. It would be arbitrary to grant a procedural default exception to one of these very similar rules, but not the other.
In dismissing an expansion of the Martinez/Trevino doctrine to ineffective assistance of appellate counsel, the Court also relies heavily on assertions regarding the availability of an alternate forum to hear underlying claims. Under the Court’s reasoning, an expansion to ineffective assistance of appellate counsel is unnecessary because an appeal (whether effective or ineffective) always involves underlying conduct that occurred in the trial court, and any eventual habeas petitioner could have been heard on this allegedly objectionable underlying conduct in the trial court itself by objecting at trial.136 While the dissenters ably point out the flaws in that reasoning, noting that it misses the very point of appeal,137 the Court’s dubious arguments regarding an alternative forum are even less convincing in the context of Brady claims than they are in the context of ineffective assistance of appellate counsel claims. By their very nature, Brady claims involve evidence that is unavailable to the trial court.138 Brady claims inherently involve evidence that is withheld from the trial court: the prosecution has no incentive to submit material exculpatory evidence at trial, and the defense has no access to it because the prosecution withheld it in violation of its Brady obligations. There is thus no opportunity for the trial court to rule in the first instance on either the Brady claim or other constitutional issues (such as, for instance, other forms of prosecutorial misconduct) that may become apparent as a result of the Brady evidence. The possibility of ex ante trial court review—which the Davila court relies on so heavily in demonstrating that the concerns about unreviewability animating the Martinez/Trevino courts are not present in the ineffective assistance of appellate counsel context139—is not available in the Brady context. It is either post-conviction review or nothing, since no state has a regime by which Brady evidence may be reviewed outside of the post-conviction review system, except in very rare circumstances that are unlikely to occur in practice.140 If the Court is concerned about procedural default resulting in unreviewability in the Brady context, then an expansion of Martinez/ Trevino to cover Brady is necessary.
In Davila, the Court also distinguished ineffective assistance of appellate counsel claims from ineffective assistance of trial counsel claims by highlighting equitable concerns arising from a state’s institutional design choice to channel ineffective assistance of trial counsel claims into post-conviction proceedings, either de jure141 or de facto142, where there is no right to assistance of counsel, thereby increasing the risk that a petitioner, operating without the assistance of counsel, procedurally defaults such a claim on post-conviction review.143 Brady claims are similarly forced into state post-conviction review by virtue of a confluence of state institutional design choices and the behavior of state actors, and it would be similarly inequitable to charge a petitioner with the consequences of procedurally defaulting those claims on post-conviction review.
Brady claims inherently involve material, exculpatory evidence that, by its very nature, likely would have been introduced at trial but for the intentional or inadvertent144 acts of state officials, chargeable against the state,145 that kept Brady evidence from the defense. As a result of state action,146 then, evidence that otherwise would have been introduced at trial is excluded from the trial record. Many states have adopted procedural regimes in which evidence not found in the trial record—including Brady material—cannot be introduced on direct review.147 Others states have adopted procedural regimes that only allow a defendant to supplement the trial record—regardless of whether the supplemental evidence is Brady evidence—in very limited circumstances, such as by filing a motion for a new trial under filing deadlines ranging from five to thirty days post-verdict. 148 Therefore, the confluence of misconduct by state actors and state institutional design choices shunt Brady claims to state post-conviction review, where the risk of procedural default is high because of the lack of guaranteed counsel.149 As a result, the same equitable concerns—the risk that a petitioner may procedurally default and bear the consequences of the default as a result of the state’s choices—that animated the Court’s equitable reasoning in Martinez and Trevino arise when Brady claims are at issue.
The Davila court, noting the equitable nature of the Martinez/Trevino doctrine, weighed the systemic costs and benefits of the proposed expansion, concluded that the systemic costs would be high and the benefits small, and ultimately used the cost-benefit analysis as a basis for rejecting an expansion to cover ineffective assistance of appellate counsel claims. However, the Court’s cost-benefit reasoning in Davila is inherently context-specific, and in the Brady context, the cost-benefit calculus is different and weighs in favor of an expansion.
In analyzing the systemic costs and benefits of the proposed expansion, the Court came to the conclusion that the benefits of the proposed expansion were minimal because meritorious ineffective assistance of appellate counsel claims are, in the Court’s view, extremely rare.150 By contrast, meritorious Brady claims are ubiquitous and highly consequential. One study identified potential Brady violations in 29 of the first 250 DNA exonerations,151 and other studies have identified even higher rates of wrongful conviction involving Brady violations.152 Brady v iolations a re c ommon n ot o nly i n “run of the mill” criminal cases, but also in cases in which heightened reliability is expected, such as capital cases153 and politically charged high profile cases.154 The federal courts have also repeatedly recognized the problem of Brady noncompliance.155 Given the overwhelming evidence of outcome-determinative Brady noncompliance throughout the country, the systemic benefit—which benefits both individuals who are convicted as a result of Brady violations, and the system as a whole due to the accuracy-promoting function of Brady—of ensuring that Brady claims can be heard during federal habeas review, even in the face of a state procedural default, is far more significant than the rarely meritorious ineffective assistance of appellate counsel claims at issue in Davila. However, the Davila court was concerned not only with the systemic benefit of an expansion of the Martinez/Trevino doctrine, but also with the systemic costs of an expansion, particularly with respect to the potential burden on federal court dockets156 and the potential harm to federal-state comity that could result from a significant expansion of federal habeas jurisdiction.157 The significant systemic benefit of expansion in the Brady context may seem to imply significant systemic costs, as well: if there are significant numbers of meritorious Brady claims, and the Court adopts a rule that expands the jurisdiction of federal courts to hear them by limiting the circumstances in which they can be procedurally defaulted, then the resulting systemic costs are also likely to be significant, in the form of burdened federal habeas dockets and increased federal court involvement in state criminal cases.
Yet there is reason to believe that the significant systemic costs predicted by this syllogism will not materialize, as the number of cases in which a Brady claim is procedurally defaulted in state post-conviction proceedings, but revived in federal habeas proceedings, is likely to be small. For one, the number of habeas petitioners who allege Brady violations in the first instance is simply smaller than the number of habeas petitioners who allege ineffective assistance of trial or appellate counsel. An analysis of a random sample of federal habeas petitions found that non-capital federal habeas petitioners are almost four times more likely to allege ineffective assistance of counsel than “false, lost, or undisclosed evidence” (a category broader than Brady claims), and capital federal habeas petitioners are twice as likely to allege ineffective assistance of counsel than “false, lost, or undisclosed evidence.”158 There are several reasons why this is likely to be the case.
Unlike potential cases of ineffectiveness, which are readily apparent to potential habeas petitioners,159 Brady violations are difficult to identify post-conviction and the number of defendants who are able to identify them represent a small fraction of all defendants.160 While prosecutors are under a continuing obligation to comply with Brady, and that obligation continues post-conviction,161 in practice, the incentive structure confronted by prosecutors dissuades them from revisiting Brady issues sua sponte post-conviction.162 Discovery of Brady violations is then limited to the rare cases in which a defendant is able to identify Brady evidence using tools such as the Freedom of Information Act, an unrelated evidentiary hearing, or the discovery of previously unknown witnesses.163 In some cases, the discovery of Brady violations will be effectively miraculous: for instance, Brady violations have been discovered as a result of freak events such as inadvertent mailing of exculpatory recordings to defense counsel164 or the theft and release of exculpatory evidence from a prosecutor’s office by an individual in a romantic relationship with a staff member.165 The discovery of Brady violations will be “serendipitous”,166 rather than systemic. The task of identifying a Brady violation post-conviction—difficult even with the aid of counsel—is made Herculean for the vast majority of defendants, who have no right to counsel in post-conviction proceedings.167 The class of defendants able to enlist counsel and identify Brady claims post-conviction is likely to disproportionately consist of capital defendants,168 a context in which the Supreme Court has created exceptions to general procedural requirements due to a Constitutional requirement of heightened reliability,169 and non-capital defendants with especially meritorious claims, such as those who are able to demonstrate a colorable case of actual innocence and secure the aid of counsel from an Innocence Project. The difficulty of identifying Brady violations post-conviction, especially without counsel, is therefore likely to serve as a significant check on the flood of defaulted Brady claims into federal habeas proceedings, and those that remain are likely to be especially meritorious or occur in circumstances meriting heightened reliability, because cases exhibiting those circumstances are likely to be the ones in which the defendant is able to enlist counsel and identify a Brady violation.
The class of petitioners who cannot secure post-conviction relief on a Brady claim now, but would be able to do so if the Martinez/Trevino doctrine were expanded to cover Brady claims, make up an even smaller subset of habeas petitioners than the number of potential petitioners who can identify Brady violations post-conviction in the first instance. If a petitioner is able to identify a Brady violation prior to filing for state post-conviction review (or, depending on their state’s post-conviction review scheme, potentially after a ruling on an initial state post-conviction review petition), that petitioner could simply bring a Brady claim in state court: state post-conviction courts have proven willing to enforce Brady’s requirements in meritorious cases.170 In other cases of defaulted Brady claims—such as when the defendant fails to bring a Brady claim during state post-conviction review because the evidence remains withheld by the state and the state represents that it has made a complete disclosure—no expansion of the Martinez/Trevino doctrine is necessary: the Supreme Court held in Strickler v. Greene that, in certain cases, such as when the state claimed to have turned over all evidence under an “open file” policy but continued to suppress certain Brady evidence, an ongoing Brady violation can serve as the requisite cause and prejudice excusing procedural default in its own right.171
In the small subset of cases that would rely on an expansion of Martinez/ Trevino to cover Brady, the task of the reviewing federal habeas courts would also be relatively limited, at least relative to the burden associated with reviewing claims of ineffective assistance of counsel. Brady claims are discrete. A petitioner raising a Brady claim must identify a piece of Brady evidence, demonstrate that the government withheld the evidence, and demonstrate prejudice.172 In reviewing a Brady claim, the reviewing court need only address those three issues with respect to discrete pieces of evidence. By contrast, review of the ineffective assistance of counsel claims at issue in Martinez, Trevino, and Davila is far more extensive. The Supreme Court has set out only “general” standards for effective assistance of counsel,173 holding that a court reviewing a claim of ineffective assistance of counsel “must judge the reasonableness of counsel’s challenged conduct on the facts of the particular case, viewed as of the time of counsel’s conduct . . . in light of all the circumstances . . . .”174 Many aspects of counsel’s performance, at all stages of the adjudicative process, are potentially subject to federal review under Martinez/Trevino.175 This sort of sweeping, context-specific, holistic review required by the Supreme Court’s ineffectiveness precedents is likely to burden a reviewing court in a manner that review of discrete Brady claims does not. The narrower scope of Brady review as compared to ineffectiveness review thus provides further indication that the systemic costs that concerned the Supreme Court in Davila are less likely to materialize if the Martinez/Trevino doctrine is extended to Brady claims.
Finally, the Martinez/Trevino doctrine’s requirement that a claim be “substantial”176 in order to excuse a procedural default creates yet another limit on the scope of an expansion of the doctrine to cover Brady claims. Under Martinez, the standard for determining whether a habeas petitioner has made a sufficiently “substantial” claim tracks the standard for the issuance of a certificate of appealability of a federal habeas petition under the federal habeas statute177 and the Supreme Court’s interpretation of that statute in Miller-El v. Cockrell.178 This standard requires a petitioner to show that “reasonable jurists could debate whether . . . the petition should have been resolved in a different manner.”179 This standard is not particularly burdensome: while it requires a petitioner to “prove ‘something more than the absence of frivolity,’”180 it does not require a petitioner to show that he or she is likely to prevail on the merits.181 Nevertheless, the requirement that a claim that relies on the Martinez/Trevino doctrine to excuse a default must be “substantial” gives district courts another way to screen claims and prevents burdensome review on the merits when it is unwarranted, further reducing the risk that high systemic costs will result from the new rule.
Given the availability of Brady relief in state court and the availability of the Strickler rule as a means of excusing procedural default of a Brady claim when the state’s conduct is particularly egregious or deceptive, an expansion of Martinez/Trevino only reaches a small subset of the small number of Brady claims that are identified post-conviction. The expansion would reach only those cases that are substantial, but defaulted as a result of nonexistent or ineffective post-conviction counsel. The actual federal review of those cases would also be narrow in scope: unlike claims of ineffective assistance of counsel, which inherently involve sweeping review of counsel’s performance, review of alleged Brady claims is limited to discrete incidents. An expansion covering such a small subset of cases, and involving such limited review, is unlikely to implicate the federalism, comity, and docket-burdening concerns identified by the Court in Davila. Rather, an expansion to cover these cases is more properly understood as the sort of “guard against extreme malfunctions in the state criminal justice systems”182 that federal habeas corpus review is intended to provide.
An extension of Martinez and Trevino to cover Brady claims procedurally defaulted due to attorney error has potentially significant implications: both for the small, but likely tangible, number of prisoners with defaulted and otherwise unreviewable Brady claims who could get relief as a result of an extension, and for the coherent development of this convoluted area of law.
Attorneys who serve clients seeking post-conviction relief should scrutinize how the reasoning of Davila may actually serve to expand access to federal habeas review, even if it creates only a narrow opening for procedurally defaulted Brady claims. They should carefully raise these claims on behalf of their clients, and courts should take the opportunity to extend Martinez and Trevino to serve these worthy purposes. With hindsight, what appears now to be a narrow procedural development may actually mark the revival of the once-great Great Writ.
1 See, e.g., Louis H. Pollak, Proposals to Curtail Federal Habeas Corpus for State Prisoners: Collateral Attack on the Great Writ, 66 Yale L.J. 50 (1956).
2 Ex parte Yerger, 75 U.S. (8 Wall.) 85, 95 (1868).
3 Boumediene v. Bush, 553 U.S. 723, 740-42 (2008) (discussing the history of habeas corpus in English law). See also Paul D. Halliday & G. Edward White, The Suspension Clause: English Text, Imperial Context, and American Implications, 94 Va. L. Rev. 575 (2008).
4 Boumediene, 553 U.S. at 743.
5 U.S. Const. art. I, § 9, cl. 2. While the Suspension Clause does not explicitly state who is authorized to suspend the writ of habeas corpus, courts have interpreted the Clause as vesting the power to suspend habeas corpus in Congress, as opposed to the president. See Ex parte Merryman, 17 F. Cas. 144 (C.C.D. Md. 1861) (No. 9487) (Taney, J.).
6 See, e.g., Erwin Chemerinsky, Thinking About Habeas Corpus, 37 Case W. L. Rev. 748, 749 (1987) (“[Habeas] truly is one of the most, if not the single most, important part of the Constitution which protects individual rights.”); Noam Biale, Beyond a Reasonable Disagreement: Judging Habeas Corpus, 83 U. Cin. L. Rev. 1337, 1344
7 See, e.g., Act of February 5, 1867 (“Habeas Corpus Act of 1867”), Chap. 28, 39th Cong., 14 Stat. 385, 385-86 (1867) (authorizing federal courts to grant a writ of habeas corpus to anyone “restrained of his or her liberty in violation of the Constitution, or of any treaty or law of the United States”).
8 See, e.g., Brown v. Allen, 344 U.S. 443 (1953).
9 See, e.g., United States v. Shipp, 214 U.S. 386, 403-04 (1909) (discussing the use of federal habeas to protect an African-American man’s Constitutional rights from a “legal lynching” in the early 20th century South).
10 See, e.g., Robert J. Glennon, The Jurisdictional Legacy of the Civil Rights Movement, 61 Tenn. L. Rev. 869, 905-918 (1994) (reviewing how the Warren Court expanded the scope of federal habeas review in part to protect defendants prosecuted for civil rights demonstrations in the Jim Crow South from invasions of their federal rights).
11 See, e.g., Bryan A. Stevenson, Confronting Mass Imprisonment and Restoring Fairness to Collateral Review of Criminal Cases, 41 Harv. C.R.-C.L. L. Rev 339 (2006).
12 See, e.g., John H. Blume et al, In Defense of Noncapital Habeas: A Response to Hoffmann and King, 96 Cornell L. Rev. 435, 4 40 (2011) (describing t he b readth o f Warren Court-era habeas doctrine).
13 See, e.g., Stephen R. Reinhardt, The Demise of Habeas Corpus and the Rise of Qualified Immunity: The Court’s Ever Increasing Limitations on the Development and Enforcement of Constitutional Rights and Some Particularly Unfortunate Consequences, 113 U. Mich. L. Rev. 1219 (2015); Stephen B. Bright, The Richard J. Childress Memorial Lecture 2016 Keynote: The Continuing Denial of Counsel and Assembly-Line Processing of Poor People Accused of Crimes, 61 St. Louis U. L.J. 605, 608 (2016) (“A person convicted of a crime in the state courts can go to the federal court and file what used to be called the Great Writ of Habeas Corpus. It is now the shabby, barely existent writ of habeas corpus.”).
14 Pub. L. No. 104-132, 110 Stat. 1214.
15 Blume, supra note 12, at 441-42 (discussing the effects of AEDPA).
16 Blume, supra note 12, at 441 (describing the ways in which the Burger, Rehnquist, and Roberts Courts have curtailed the availability of federal habeas relief); see also Reinhardt, supra note 13 (providing a federal circuit judge’s perspective on this curtailment).
17 Teague v. Lane, 489 U.S. 288 (1989) (plurality opinion).
18 Stone v. Powell, 428 U.S. 465 (1976).
19 John C. Jeffries, Jr. and William Stuntz, Ineffective Assistance and Procedural Default in Federal Habeas Corpus, 57 U. Chi. L. Rev. 679, 681 (1990). See also 17B Charles Alan Wright and Arthur R. Miller, Federal Practice and Procedure § 4266 (3d ed. 2008).
20 See Pennsylvania v. Finley, 481 U.S. 551 (1987) (declining to extend the Sixth Amendment right to effective assistance of counsel to post-conviction proceedings).
22 Coleman v. Thompson, 501 U.S. 722, 752 (1991) (citing Wainwright v. Torna, 455 U.S. 586 (1982)) (“[A] petitioner cannot claim constitutionally ineffective assistance of counsel in [state post-conviction] proceedings.”).
23 Eve Brensike Primus, Effective Trial Counsel After Martinez v. Ryan: Focusing on the Adequacy of State Procedures, 122 Yale L. J. 2604, 2609 (2013).
24 Martinez v. Ryan, 566 U.S. 1, 22 (2012) (Scalia, J., dissenting) (citing statistics showing that procedural default is the most common basis for the dismissal of federal habeas corpus petitions).
25 Primus, supra note 20, at 2609–10.
26 Primus, supra note 20, at 2608.
27 566 U.S. 1 (2012).
28 569 U.S. 413 (2013).
29 137 S.Ct. 2058 (2017).
30 373 U.S. 83 (1963).
31 372 U.S. 391 (1962).
32 Ibid. at 438.
33 Ibid. at 439.
34 Francis v. Henderson, 425 U.S. 536, 539 (1976).
35 433 U.S. 72 (1977).
36 Ibid. at 87.
37 Ibid. at 84.
38 See, e.g. Davila v. Davis, 137 S.Ct. 2058, 2065 (2017).
39 433 U.S. at 87 (“We leave open for resolution in future decisions the precise definition of the ‘cause’-and-‘prejudice’ standard . . . .”).
40 See, e.g., Sykes, 433 U.S. at 100 (Brennan, J., dissenting) (“[L]eft unanswered is the thorny question that must be recognized . . . [h]ow should the federal habeas court treat a procedural default in a state court that is attributable purely and simply to the error or negligence of a defendant’s . . . counsel?”).
41 501 U.S. 722 (1991).
42 477 U.S. 478 (1986).
43 Ibid. at 488 (citing Strickland v. Washington, 466 U.S. 668 (1984)).
44 481 U.S. 551 (1987).
45 Murray v. Giarratano, 492 U.S. 1 (1989) (applying the Finley rule in death penalty cases).
46 Coleman, 501 U.S. at 752.
47 566 U.S. 1 (2012).
48 Ibid. at 12.
49 Ibid. at 15.
50 Ibid. at 10.
51 Ibid. at 9.
52 Ibid. at 11.
53 Ibid. at 13.
54 569 U.S. 413 (2013).
55 Ibid. at 417.
56 Martinez, 566 U.S. at 17.
57 569 U.S. at 430.
58 373 U.S. 83 (1963).
59 See, e.g., Keith A. Findley, Innocence Protection in the Appellate Process, 93 Marq. L. Rev. 591, 600 (2009) (“[T]ogether, ineffective assistance and Brady claims constitute the largest proportion of postconviction challenges to convictions.”).
60 Martinez, 566 U.S. at 19 (Scalia, J., dissenting).
61 See, e.g., Aziz Huq, Habeas and the Roberts Court, 81 U. Chi. L. Rev. 519, 592 (2014) (“[T] he overall fault-oriented structure of habeas jurisprudence suggests that Brady claims are an even stronger candidate for exculpating cause than [ineffective assistance of trial counsel] claims. . . .”); Megan Raker, Comment, State Prisoners with Federal Claims in Federal Court: When Can a State Prisoner Overcome Procedural Default?, 73 Md. L. ev. 1173, 1175 (2014) (concluding “that the nature of Brady claims—in how they are raised on collateral review and the constitutional rights they protect—are such that the Martinez exception can, and should, apply to Brady claims as well.”)
62 Emily Garcia Uhrig, Why Only Gideon?: Martinez v. Ryan and the “Equitable” Right to Counsel in Habeas Corpus, 80 Mo. L. Rev. 771, 806–807 (2015).
63 Kirsty Davis, Comment, Justice Scalia Was Right: “No One Really Believes that [Martinez] Will Remain Limited to Ineffective-Assistance-of-Trial-Counsel-Cases”, 54 Hous. L. Rev. 1349, 1374 (2017).
64 See, e.g., Nguyen v. Curry, 736 F.3d 1287, 1292–96 (9th Cir. 2013); see also Micah Horwitz, Note, An Appealing Extension: Extending Martinez v. Ryan to Claims of Ineffective Assistance of Appellate Counsel, 116 Colum. L. ev. 1207, 1223–24 (2016) (analyzing the Ninth Circuit’s reasoning in Nguyen and contending that it represents the correct interpretation of Martinez).
65 See, e.g., Long v. Butler, 809 F. 3d 299, 314–15 (7th Cir. 2015), aff’d sub nom Long v. Pfister, 874 F.3d 544 (7th Cir. 2017) (en banc); Reed v. Stephens, 739 F.3d 753, 778 n. 16 (5th Cir. 2014); Hodges v. Colson, 727 F.3d 517, 531 (6th Cir. 2013); Banks v. Workman, 692 F.3d 1133, 1147–48 (10th Cir. 2013); Dansby v. Norris, 682 F.3d 711, 729 (8th Cir. 2012), rev’d on other grounds sub nom Dansby v. Hobbs, 569 U.S. 1015 (2013). See also Horwitz, supra note 64, at 1221–23, 1224 (summarizing the conclusions of the 5th, 6th, 7th, 8th, and 10th Circuits regarding the application of Martinez to ineffective assistance of appellate counsel claims and arguing that their reasoning does not comport with Martinez).
66 Pizzuto v. Ramirez, 783 F.3d 1171, 1176–77 (9th Cir. 2015).
67 Hunton v. Sinclair, 732 F.3d 1124, 1126–27 (9th Cir. 2013).
68 137 S.Ct. 2058 (2017).
69 Ibid. at 2063.
75 Davila v. State, No. AP-76, 105, 2011 Tex. Crim. App. Unpub. LEXIS 43* (Tex. Crim. App., Jan. 26, 2011), cert. denied 565 U.S. 885 (2011).
76 466 U.S. 668 (1984).
77 See, e.g., Schwander v. Blackburn, 750 F.2d 494, 501–02 (5th Cir. 1985) (adopting the Strickland standard for claims of ineffective assistance of appellate counsel).
78 Ex parte Davila, Nos. WR-75, 356-01, 2013 Tex. Crim. App. Unpub. LEXIS 486 (Tex. Crim. App., Apr. 17, 2013), cert denied 134 S. Ct. 784 (2013).
79 Davila v. Stephens, No. 4:13-CV-506-O, 2015 U.S. Dist. LEXIS 51854, at *52-54 (N.D.Tex. Apr. 21, 2015).
80 Ibid. at *53.
81 Ibid. at *53-54.
82 Davila v. Davis, 650 Fed. Appx. 850, 867–68 (5th Cir. 2015).
83 Davila v. Davis, 137 S. Ct. 810 (2017).
84 137 S. Ct. at 2064.
85 Ibid. at 2065–66.
86 Ibid. at 2066.
88 Ibid (citing Halbert v. Michigan, 545 U.S. 605, 610 (2005)).
89 Ibid (quoting McFarland v. Scott, 512 U.S. 849, 859 (1994) (“The trial ‘is the main event at which a defendant’s rights are to be determined . . .’”)).
91 Ibid (citing Ross v. Moffitt, 417 U.S. 600, 610 (1974)).
92 Ibid. (citing Jackson v. Virginia, 443 U.S. 307, 318–19 (1979)).
94 Ibid. at 2067 (citing Martinez, 566 U.S. at 10, 12).
99 Ibid (citing Smith v. Robbins, 528 U.S. 259, 288 (2000)).
100 Ibid. at 2067–68.
102 566 U.S. at 13.
103 Davila, 137 S. Ct. at 2068.
104 Ibid (quoting Trevino v. Thaler, 569 U.S. 413, 427 (2013)).
106 Ibid. at 2070.
107 Ibid. at 2068–69.
108 Ibid. at 2069.
112 Ibid. at 2070 (quoting Harrington v. Richter, 562 U.S. 86, 103 (2011)).
113 Ibid. (quoting Calderon v. Thompson, 523 U.S. 538, 555–556 (1998)).
117 Ibid. at 2071–72.
118 Ibid. at 2066.
119 Ibid. at 2072 (citing Strickland v. Washington, 466 U.S. 668, 686 (1984)).
120 Ibid (citing Evitts v. Lucey, 469 U.S. 387, 396 (1985)).
121 Ibid. at 2073.
123 Ibid. at 2074 (citing “an increase of ‘dozens’ of cases out of 7,500 [habeas petitions] in total” filed in the 9th Circuit post-Martinez).
124 Ibid. (quoting Davila, 137 S.Ct. at 2069).
125 Ibid. at 2075.
126 Ibid. at 2062.
127 In Brady v. Maryland, 373 U.S. 83 (1963), the Supreme Court held that “the suppression by the prosecution of evidence favorable to an accused upon request violates [the Due Process Clause of the Fifth and Fourteenth Amendments] where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution.” Id. at 87. In subsequent cases, the Court clarified that “[t]here are three components of a true Brady violation: The evidence at issue must be favorable to the accused, either because it is exculpatory, or because it is impeaching; that evidence must have been suppressed by the State, either willfully or inadvertently; and prejudice must have ensued.” Strickler v. Greene, 527 U.S. 263, 282–83 (1999). A full discussion of contemporary Brady doctrine is beyond the scope of this paper. For a history of the Brady rule and a summary of contemporary doctrine, see, e.g., Bennett L. Gershman, Reflections on Brady v. Maryland, 47 S. Tex. L. Rev. 685 (2006).
128 Ibid. at 2067.
129 566 U.S. 1, 12 (2012).
130 373 U.S. 83, 87 (1963).
131 United States v. Bagley, 473 U.S. 667, 675 (1985).
132 See Strickler v. Greene, 527 U.S. 263, 282 (1999).
133 466 U.S. 688 (1985).
134 Ibid. at 685 (“An accused is entitled to be assisted by an attorney . . . who plays the role necessary to ensure that the trial is fair”).
135 Ibid. at 687 (“[T]he defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel’s errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.”).
136 Davila, 137 S.Ct. at 2067.
137 Ibid. at 2073.
138 See, e.g., Imbler v. Pachtman, 424 U.S. 409, 443 (1976) (White, J., concurring) (“[T]he judicial process has no way to prevent or correct the constitutional violation of suppressing evidence.”).
139 Ibid. at 2067.
140 Eve Brensike Primus, Structural Reform in Criminal Defense: Relocating Ineffective Assistance of Counsel Claims, 92 Cornell L. Rev. 679, 689–90 (2007) (noting that most jurisdictions do not allow defendants to supplement the trial record post-verdict, and those that do impose strict deadlines (between 5 and 30 days post-verdict) that almost no defendant who discovers Brady evidence post-trial will be able to meet).
141 Martinez v. Ryan, 566 U.S. 1, 13 (2012).
142 Trevino v. Thaler, 569 U.S. 413, 427 (2013).
143 137 S.Ct. at 2068.
144 Strickler v. Greene, 527 U.S. 263, 281–82 (1999) (“There are three components of a true Brady violation . . . [the] evidence must have been suppressed by the State, either willfully or inadvertently . . . .”).
145 Mooney v. Holohan, 294 U.S. 103, 112–13 (1935) (“[T]he action of prosecuting officers on behalf of the State . . . may constitute state action within the purview of the Fourteenth Amendment.”).
146 Even if the actions of individual prosecutors were not attributable to the state, it is arguable that Brady violations are the foreseeable result of the state’s choice to rely on prosecutorial discretion in determining what qualifies as Brady material, as opposed to proposed alternatives such as “open file” discovery,” in the design of its criminal discovery regime, just as the Court reasoned in Martinez/Trevino that procedurally defaulted ineffective assistance of trial counsel claims were the foreseeable result of the state’s choice to shift ineffective assistance of trial counsel claims out of the direct appeal process. See, e.g., Brian Gregory, Brady is the Problem: Wrongful Convictions and the Case for “Open File” Criminal Discovery, 46 U.S.F.L.Rev. 819 (2012).
147 Primus, supra note 140, at 689 (“[T]he vast majority of jurisdictions do not allow defendants to open or supplement the trial court record . . . .”).
148 Primus, supra note 140, at 689–90.
149 See Martinez, 566 U.S. at 13.
150 137 S.Ct. at 2070.
151 BrandonGarrett, Convictingthe Innocent 203 (2011).
152 See, e.g., Jon B. Gould et al., Predicting Erroneous Convictions, 99 Iowa L. Rev. 471, 491 (2014) (identifying prosecutorial withholding of evidence in violation of Brady as a statistically significant predictor of wrongful conviction based on a large data set of wrongful convictions and “near miss” cases); The National Registry ofExonerations, Exonerations in2016 6-7 (Mar. 7, 2017), https://www.law.umich. edu/special/exoneration/Documents/Exonerations_in_2016.pdf (identifying official misconduct, most commonly Brady violations, as a factor in 42% of exonerations, and over two-thirds of homicide exonerations, in 2016).
153 James S. Liebmanet al., ABrokenSystem, Part II: Why There Is So Much Error inCapital Cases, and What CanBe Done About It 41 (Feb. 11, 2002), http:// www2.law.columbia.edu/brokensystem2/report.pdf (documenting Brady violations as the basis for nearly 20% of reversals of capital convictions and/or sentences on post-conviction review).
154 Cynthia E. Jones, A Reason to Doubt: The Suppression of Evidence and the Inference of Innocence, 100 J. Crim. L. & Criminology 415, 416–19 (2010) (describing Brady violations during the corruption prosecution of former Sen. Ted Stevens).
155 See, e.g., United States v. Olsen, 737 F.3d 625, 626 (9th Cir. 2013) (Kozinski, C.J., dissenting) (“There is an epidemic of Brady violations abroad in the land. Only judges can put a stop to it.”); Emmet Sullivan, Enforcing Compliance with Constitutionally-Required Disclosures: A Proposed Rule, 2016 Cardozo L. Rev. de novo 138 (documenting a federal district judge’s efforts to ensure Brady compliance in criminal proceedings in his court).
156 137 S. Ct. at 2069.
157 Ibid. at 2070.
158 Nancy J. King, Fred L. Cheesman II & Brian J. Ostrom, Habeas Litigation in U.S. District Courts: An Empirical Study of Habeas Corpus Cases Filed by State Prisoners Under the Antiterrorism and Effective Death Penalty Act of 1996 5 (Vanderbilt Univ. Pub. Law & Legal Theory Working Paper Group, Working Paper No. 07-21, 2007), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1009640.
159 Tom Zimpleman, The Ineffective Assistance of Counsel Era, 63 S.C.L.Rev. 425, 445 (2011) (citing Diggs v. Welch, 148 F.2d 667, 670 (D.C. Cir. 1945)); Cf. Martinez v. Ryan, 566 U.S. 1, 20 (2012) (Scalia, J., dissenting) (questioning, rhetorically and pejoratively, whether “a duly convicted defendant [has] ever been effectively represented?”).
160 See, e.g., Connick v. Thompson, 563 U.S. 51, 80 (2011) (Ginsburg, J., dissenting) (“Brady violations . . . are not easily detected.”); Jones, supra note 136, at 433 (“[A]bsent extraordinary circumstances, it is very likely that the defense will never learn of the existence of favorable evidence”); Bennett L. Gershman, Litigating Brady v. Maryland: Games Prosecutors Play, 57 Case W. Res. L. Rev. 531, 536 n. 27 (2007) (summarizing materials on the difficulty of identifying Brady violations post-conviction).
161 Gershman, supra note 160, at 535.
162 See, e.g., Stephen A. Saltzberg, Perjury and False Testimony: Should the Difference Matter So Much?, 68 Fordham L. Rev. 1537, 1578–79 (2000) (“In many instances, withheld evidence will never see the light of day.”)
163 Gershman, supra note 160, at 537.
164 Jones, supra note 154, at 433 n. 81 (citing United States v. Arnold, 117 F.3d 1308 (11th Cir. 1997)).
165 Id. (citing Ken Armstrong and Maurice Possley, The Verdict: Dishonor, Chi. Tribune, Jan. 11, 1999, at A1).
166 See Connick v. Thompson, 563 U.S. 51, 86 (2011) (Ginsburg, J., dissenting) (describing the discovery of Brady evidence as “serendipitous” where, in a death penalty case, Brady evidence was discovered in an “eleventh-hour” search of crime lab archives days prior to an execution).
167 See, e.g., Pennsylvania v. Finley, 481 U.S. 551, 555 (1987) (“We have never held that prisoners have a constitutional right to counsel when mounting collateral attacks upon their convictions . . . and we decline to so hold today.”).
168 See, e.g., Daniel Givelber, The Right to Counsel in Collateral, Post-Conviction Proceedings, 58 Md. L. Rev. 1393, 1409 (1999) (“[M]ost states provide for the appointment of [capital post-conviction] counsel either by statute or judicial decision . . . .”).
169 See, e.g., Lockett v. Ohio, 438 U.S. 586, 604 (1978) (opinion of Burger, J.) (“[T]his qualitative difference between death and other penalties calls for a greater degree of reliability when the death sentence is imposed.”).
170 Liebman, supra note 153, at 41 (indicating a nearly equal rate of capital conviction reversals based on Brady in state and federal courts).
171 527 U.S. 263, 282–83 (1999) (citing Murray v. Carrier, 477 U.S. 478, 488 (1986)).
172 Strickler v. Greene, 527 U.S. 263, 282–83 (1999).
173 Bobby v. Van Hook, 558 U.S. 4, 7 (2009).
174 Strickland v. Washington, 466 U.S. 668, 690 (1984).
175 See Zimpleman, supra note 159, at 439–45 (describing the variety of circumstances in which post-conviction courts have identified ineffective assistance of counsel).
176 Martinez v. Ryan, 566 U.S. 1, 14 (2012) (citing Miller-El v. Cockrell, 537 U.S. 322 (2003)).
177 28 U.S.C. § 2253.
178 537 U.S. 322 (2003).
179 Ibid. at 336 (quoting Slack v. McDaniel, 529 U.S. 473, 484 (2002)).
180 Ibid. at 338 (quoting Barefoot v. Estelle, 463 U.S. 880, 893 (1983)).
181 Ibid. at 336.
182 Harrington v. Richter, 562 U.S. 86, 102 (2011) (quoting Jackson v. Virginia, 443 U.S. 307, 332, n.5 (1979) (Stevens, J., concurring in judgment)).