Brady in an Age of Innocence

pdfJennifer E. Laurin∞

I. Introduction. 505

II. Brady and the Changing Criminal Justice Zeitgeist. 507

A. Brady: A Child of and Ahead of Its Time. 507

B. An Age of Innocence. 510

III. Brady and the Innocence Effect. 511

A. The Innocence Effect’s Weakness 511

B. Why Is the Innocence Effect Less Pronounced in the Case of Discovery Reform? 515

IV. Lessons and Possibilities. 518




Just over fifty years ago, the Supreme Court reversed the death sentence of convicted murderer John Brady, making him the namesake of one of constitutional criminal procedure’s most bedrock guarantees: the defendant’s right to the disclosure of all of the state’s favorable evidence “material either to guilt or punishment.”[1] Decided against the backdrop of exceedingly restrictive discovery rights for criminal defendants in federal and state courts, Brady had all the makings of another Warren Court watershed.[2]

In the ensuing half century Brady has come to be both emblematic of and an outlier from the criminal procedure revolution initiated by the Warren Court. In the former respect, the decision in Brady exemplifies that Court’s focus on fair procedures in service of vindicating pure principles of justice—equality or fairness, for example—rather than in service of accurate outcomes. On the other hand, Brady’s mandate that favorable evidence be put in the hands of criminal defendants, especially as developed in subsequent decisions, is rooted not only in fairness per se but also accuracy. Brady, by its own terms, rooted the due process interest at stake in a concern for enabling refutation of the state’s case. Indeed, decisions following Brady have deepened its ties to accuracy concerns as by restricting the meaning of “material” favorable evidence to only that evidence that would have benefited the provably innocent.[3]

Proceeding from the premise that Brady is, perhaps uniquely, both reflective of the Warren Court’s emphasis on abstract procedural values and bound up in substantive concerns, this essay explores the question of how the decision’s legacy fares in a different era in criminal justice. Today’s era-defining jolt comes not from the Supreme Court but from the laboratory, in the form of DNA technology.[4] Long-simmering concerns that our system unwisely privileges procedure over accuracy have boiled to the surface, largely as a result of more than three hundred DNA exonerations over the past two and a half decades.[5] We are in an “age of innocence,”[6] in which securing accurate outcomes and avoiding the missteps catalogued in the growing list of exonerations is eclipsing procedural fairness as the primary focus of criminal justice advocates.[7]

In this age of innocence criminal justice advocates have leveraged mounting evidence of criminal justice error to achieve previously unseen reform successes. This essay identifies three key characteristics of that success and dubs them collectively the “innocence effect.” Those characteristics are (1) that reform is being won primarily in the sub-constitutional realm of state courts and legislatures; (2) that its arguments merge anecdotal evidence of the circumstances associated with wrongful convictions and social science research on law enforcement practices; and (3) that criminal justice insiders—police and prosecutors in particular—who traditionally have fought pro-defense reform measures have supported many innocence-based reforms. The innocence effect has propelled adoption of a range of accuracy-centered reforms—from recording of interrogations to shifts in eyewitness identification practices to regulation of police informants.[8]

Sub-constitutional rules governing disclosure of evidence in criminal cases—referred to here as “discovery doctrine”—would seem a likely candidate for expansion in the innocence era, given their accuracy-enhancing features. Discovery doctrine’s expansion would seem especially likely since Brady violations by prosecutors have increasingly become part of the public debate. In the notorious rape case against three Duke lacrosse players, for example, Brady violations played a key role in the prosecution of innocent people until the charges were dropped.[9]

However, as this essay will elucidate, discovery doctrine has not in fact reaped great benefit from the innocence effect. After placing Brady and discovery doctrine in historical and contemporary context in Part II, the essay develops its evaluative claim in Part III that reform of discovery doctrine has not exemplified the innocence effect as much as one might expect by examining the trajectory of discovery doctrine reform in the state and federal systems. The essay closes by assessing potential explanations for the surprisingly un-buoyed status of discovery reform in the age of innocenceexplanations that highlight some of the potential perils of innocence-centered reform approaches. The essay considers, in closing, strategies to better capture the innocence effect in discovery doctrine reform, as well as the possibility that innocence-centered discourse actually ill-serves Brady’s legacy and that re-seizing procedural fairness might better fortify prosecutorial professionalism and minimize error.



Brady and the Changing Criminal Justice Zeitgeist

  1. Brady: A Child of and Ahead of Its Time

Throughout the 1950s and ‘60s, the Warren Court aggressively distilled from the Due Process Clause of the Fourteenth Amendment what would amount to a constitutional “code of conduct” for police, prosecutors, and courts.[10] From strictures on police conduct in eyewitness identification procedures to the requirement of provision of counsel to efforts to thwart systemic discrimination in jury pools, the criminal procedure revolution touched upon nearly every phase of criminal investigation and prosecution.[11]

Critically, the code of conduct that emerged was of a distinctive character, in that it emphasized vindication of abstract values of the justice system at least as, if not more, prominently than accuracy.[12] In decision after decision, the Court articulated the rationale for extending constitutional protections as grounded in values such as fairness, equality, and autonomy, and the asserted primacy of such values over per se accurate outcomes.[13] Brady itself originally resonated in these terms. In its brief opinion, the Court said next to nothing about (obvious) harm to the innocent of prosecutorial withholding of favorable evidence; it did, however, extol the virtues of justice-seeking prosecution and elaborated on the imperative of fairly won victories.[14]

Yet Brady was soon understood and leveraged to advance reliability in the criminal justice system. In part, this relates to a rollback of the scope of Brady’s guarantee following the Warren Court era. Decisions elaborating on Brady’s undefined requirement that information subject to disclosure be “material” to guilt or punishment ultimately limited the due process right to evidence creating a reasonable probability of a different result at trial in the event of disclosure.[15] The upshot of that demanding standard is that only the likely innocent tend to prevail in Brady litigation. Even accepting the premise that the due process clause should protect only the innocent from prosecutorial nondisclosure there is reason to be skeptical that the doctrine is achieving that aim in application. Prosecutors make prospective disclosure determinations without information about the defense’s case and with cognitive biases that lead them to discount the materiality of favorable evidence.[16] Appellate courts in turn fashion restrictive Brady doctrine due to the hindsight biases generated by the fact of a defendant’s conviction.[17]

But while accuracy concerns ultimately restricted constitutional discovery doctrine, they also animated an enlargement of criminal discovery secured by sub-constitutional reforms that were inspired, in part, by Brady itself. In contrast to civil practice, criminal litigation prior to Brady was less a match of adversaries and more trial by ambush.[18] Neither constitutional doctrine nor, in many jurisdictions, any rule-based regime entitled defendants to anything but the most minimal disclosure of evidence by prosecutors. In federal court prior to 1966, when revisions to Federal Rule 16 were adopted, defendants could demand that the state turn over no more than the defendants’ or other witnesses’ own papers or possessions—and even then only on a specific showing of need;[19] state regimes were equally if not more restrictive.[20] This was so notwithstanding an array of commentators decrying the regime’s deleterious impact on the truth-finding function of adjudication, as well as the unseemly advantage it gave to the comparatively well-resourced state.[21]

Brady’s requirement that favorable information material to guilt or punishment be disclosed by the state was in itself an exceedingly narrow grant of access. But Brady was catalytic in the sub-constitutional realm. Following Brady, nascent and isolated pushes to expand statutory discovery rights gained broader traction at the state and federal levels.[22] These changes opened discovery by comparison to the pre-Brady regime, but even these reforms contained critical omissions, including routine defense access to police reports and witness statements.[23] During this time, however, some called for more expansive discovery reform that would provide defendants with full access to police and prosecutors’ files in criminal cases, arguing that so-called “open file” discovery would promote more accurate outcomes, and also would, by neutralizing prosecutorial discretion in selecting evidence for disclosure, prophylactically ensure compliance with Brady’s constitutional dictates.[24] The call for “open file” discovery went unanswered almost without exception prior to the innocence era and, as discussed below, remains largely unanswered today, even in the age of innocence.[25]

  1. An Age of Innocence

The discourse of criminal justice reform today is markedly different from that of the Warren Court era. A wave of more than three hundred DNA exonerations in the last three decades has demonstrated that, contrary to prior supposition, our criminal justice system has a greater-than-negligible error rate.[26] At the same time, scholars and advocates have distilled patterns from these cases that now form a canonical list of causes of wrongful convictions.[27] Unreliable eyewitness evidence resulting from unduly suggestive identification procedures, false confessions resulting from either outright coercion or longstanding interrogation techniques, and flawed scientific evidence resulting from discredited or negligently administered forensic science methodologies top most such lists.[28] Together, these trends constitute what many have dubbed an “age of innocence.”[29]

One practical upshot of this new discourse of innocence has been a changed landscape of criminal justice reform, one characterized by new-found success for proposals aimed at curing the accuracy deficiencies revealed in the age of innocence. That success has come in a particular form, which this essay dubs the innocence effect. The effect is three-pronged. First, court-centered strategies have given way to legislative and other sub-constitutional advocacy aimed at securing more accurate criminal justice outcomes. Thus, while prior to the innocence era criminal justice reformers aimed primarily at pushing constitutional criminal procedure doctrine in their desired direction, now legislatures and state courts are key audiences.[30] Second, reformers have primarily argued not from abstract constitutional imperatives but instead from the stories of DNA exonerations, frequently enlisting social science to contend both that error is a not isolated and that systemic and remediable causes are driving it. From eyewitness identification procedures that conform with psychological research on memory and suggestion, to recommended reforms to long-standing interrogation methods, the literature has generated a litany of “best practices” to ensure accuracy in investigation and adjudication.[31] Finally, the innocence effect is characterized by an expansion of the political base of support for reform initiatives. While measures constraining the discretion of law enforcement and prosecutors and enlarging the arsenal of protections for defendants typically face uphill battles in legislatures and even courts,[32] criminal justice insiders have begun to buy into many of the proposals occupying the leading edge of innocence-driven reform.[33]


Brady and the Innocence Effect

  1. The Innocence Effect’s Weakness

One would intuitively expect the area of criminal justice discovery to be a likely site of the innocence effect, tied as it is to the full adversarial vetting of evidence, and live as the field has been with sub-constitutional debates over best practices.[34] And yet, the contemporary reality does not bear out this intuition.

Eyewitness identification is a helpful comparator. Over the course of more than three decades psychologists and other researchers have developed a sophisticated theoretical and practical understanding of how human perception and memory interact with forensic uses of eyewitness evidence, which has generated a battery of recommended best practices for police. These suggestions include double-blind administration of lineups, instructions to witnesses to counteract suggestibility, contemporaneous documentation of witness confidence, and perhaps showing suspects in sequence rather than simultaneously.[35] At least fifteen states over the last two decades have enacted standards for eyewitness identification that reflect at least some of these best practices or have initiated programs to study such reforms.[36] Moreover, the notion that law enforcement should follow identification procedures that are consistent with developed research findings and significantly exceed the requirements of due process now enjoys relatively broad consensus among police and prosecutors. The Department of Justice, for example, has long promulgated resources endorsing reform of eyewitness identification procedures to reflect the lessons of social science, and the Commission on Accreditation for Law Enforcement Agencies (“CALEA”), the leading national law enforcement accrediting body, has followed suit.[37] In New Jersey and Wisconsin, among other jurisdictions, the state’s chief prosecutor was the engine for eyewitness identification reform.[38]

A similar trend can be identified with regard to other investigative practices now under the innocence microscope. Changes to interrogation practices and forensic science oversight are both areas where political efforts have yielded gains and, crucially, support from insiders.[39] Indeed, the Department of Justice recently announced the adoption of interrogation recording, citing an emerging consensus among law enforcement as a reason for the change.[40]

The story is different for discovery doctrine. While the benefits of expanded discovery are undoubtedly contested, full open file discovery is the central reform proposal advanced in the innocence era—the discovery equivalent of double-blind line-ups.[41] But while a small handful of states have enacted rules allowing full defense access to investigative files,[42] at least two-thirds of states and the federal government continue to restrict defense access to police reports and disclosure of witnesses known to the state but not covered by Brady.[43] Also revealing of the more limited innocence effect for discovery reform is that few states have moved toward open file discovery since the year 2000—the year when mass commutations on Illinois’s death row, followed by the publication of Actual Innocence, launched wrongful convictions into the national spotlight.[44] Furthermore, of the twenty states with the greatest number of exonerations, only six have enlarged discovery in this innocence era.[45]

This relatively halting trajectory of reform is at least partially due to significant political opposition to discovery reform from prosecutors and other criminal justice stakeholders who have largely supported other innocence-driven reforms. At the federal level, the Department of Justice’s positions on discovery and other innocence-driven reforms stand in stark contrast to its positions on other innocence-related reforms. Despite several recent high profile revelations of the suppression of evidence by federal prosecutors, including perhaps most notably, in the prosecution of the late Senator Ted Stevens,[46] the Department has publicly insisted that nondisclosure of evidence is a de minimis rather than systemic problem, and opposed reforms to expand the now-restrictive federal discovery regime.[47]

At the state level, the case for discovery reform, even when supported by evidence of error in the criminal justice system, falters when compared to other areas. Consider, for example, Illinois, where legislative reforms followed in the wake of a series of death row exonerations with documented evidentiary disclosure deficiencies.[48] The reforms included significant attention to eyewitness identification, recording of interrogations, and a number of discovery recommendations.[49] But the most consequential discovery proposals, unlike many of the interrogation and eyewitness reforms, were not adopted by the legislature.[50] Or consider Virginia, a state that, in the wake of some twenty-seven exonerations, has created an innocence commission, an independent and state-monitored crime laboratory, and standards for eyewitness identification, but which has faltered in discovery reform in the face of prosecutors’ ardent objections.[51]

To be sure, a number of jurisdictions have enacted significantly broadening discovery reform, in some cases under circumstances in which reform advocates were able to take full advantage of the innocence effect. Robert Mosteller has documented this dynamic in North Carolina, which in 2004 became the first state to mandate open file discovery in the wake of multiple and notorious instances of prosecutorial misconduct specifically linked to wrongful convictions and prosecutions.[52] In Texas, which in 2013 significantly broadened its existing regime of highly restricted criminal discovery, the reform legislation was precipitated by the exoneration of Michael Morton, which uncovered egregious Brady violations committed by Ken Anderson, a former district attorney and then-sitting judge.[53] Louisiana’s 2013 move to enact open file discovery followed not only the Connick v. Thompson litigation, which until the Supreme Court’s vacatur of judgment had led to a $14 million jury verdict in an exonerated man’s civil suit for Brady violations, but also the Supreme Court’s unusually overt frustration with Louisiana prosecutors who refused to concede error in the subsequent Brady case, Smith v. Cain.[54] However, these are all highly unusual instances where anecdotal circumstances suggested a causal link between discovery violations and wrongful convictions, and where greater openness of prosecutors’ files was deemed a remedy for those errors. For reasons explored in detail below, this linkage is uniquely difficult to demonstrate.[55]

  1. Why Is the Innocence Effect Less Pronounced

in the Case of Discovery Reform?

Why has the innocence effect not greased the wheels of discovery reform as thoroughly as reform in other arenas? Some would argue that the answer is clear: non-compliance with Brady is not a significant problem, existing discovery regimes are adequate, and in any case the downsides of expanded discovery—possible risks to witnesses, the potential for tailored and fabricated defenses—outweigh negligible advantages that open discovery may provide.[56] To be sure, the case for open file discovery is not incontestable, and this essay does not take on the task of justifying that proposal on its merits. Instead, the point to be made here is that there may well be factors inhibiting an innocence effect in the context of discovery doctrine reform that have less to do with the merits of reform proposals than with the limited ability of the innocence effect to make traction in the this unique realm. Here I offer two such factors.

The first concerns the role of empirical evidence in driving the innocence effect. Again, by way of contrast, consider eyewitness identification. A DNA exoneration in a case premised on eyewitness testimony or other directly inculpatory evidence negates, by its own force, the accuracy of that evidence; the defendant was not there; the witness was wrong; the evidence was unreliable.[57] Moreover, the anecdotal evidence suggesting a correlation between certain types of errors and wrongful convictions dovetails with decades of independent academic research preceding the innocence era to make an intuitively compelling (if not airtight) claim about how flawed eyewitness identification procedures produce error.[58] Armed with facts to argue both that the status quo features accuracy deficits with predictable sources, and that a particular reform might prevent getting “the wrong guy” (thereby missing the “right” one), advocates can claim at least a portion of the crime fighting mantle traditionally held only by opponents of reform initiatives that advantage defendants.[59]

By contrast, the fact of a DNA exoneration itself reveals little if anything about whether evidentiary disclosure was adequate in the case. One sees necessarily only that some evidence was wrong, not that more favorable evidence was available or that the defense would have found the problem. And while erroneous eyewitness evidence readily facilitates the argument that better identification procedures might prevent error, reversal of a conviction based on violation of Brady confounds the ability to determine how such error might have been avoided.[60] First, because Brady is transgressed by sheer nondisclosure even where prosecutors did not know of the evidence—because of, say, withholding by law enforcement, Brady violations do not by their own force reveal their source: If police are responsible for withholding evidence, discovery doctrine requiring greater openness of prosecutorial files would not have prevented error.[61] Second, the requirement under Brady that evidence be not only favorable but also material to guilt or punishment means courts frequently find incidents of undisclosed evidence that did not raise the probability of a different outcome.[62] This means that problems of nondisclosure may well be underrepresented by the number of adjudicated Brady violations; many instances of nondisclosure are essentially deemed harmless. These circumstances may go far to explain the relative dearth of research on causes of or solutions for evidentiary nondisclosure, as compared to a robust body of evidence concerning other categories of criminal justice error.[63]

In short, discovery errors are difficult to see, and even when they are visible, they often do not suggest clear conclusions about what would have prevented them. As such, exonerations do not de facto make the case for discovery reform, and criminal justice insiders can hide behind this uncertainty. The nebulousness of the link between exonerations and undisclosed evidence at least partially explains the relative de-emphasis of Brady error on the “canonical list” of causes or risk factors underlying wrongful convictions.[64] While it is oft-repeated by courts and commenters that, for example, “[e]yewitness misidentification is the single greatest cause of wrongful convictions,”[65] lack of defense access to evidence in violation of Brady is a circumstance featured far less prominently in the “lessons learned” from wrongful convictions.[66] The relative silence on this issue has the consequence of limiting the ability of Brady-related reform work to benefit from the innocence effect.

A second force, little explored to date, is the complicated legal and political dynamic that mediates the relationship between prosecutors and the actors driving the innocence-based reform agenda. That relationship typically begins in the course of individual exonerations, the achievement of which hinges significantly on obtaining the cooperation of police and prosecutors in a number of respects—securing access to evidence, negotiating release, and so forth.[67] But the relationship continues in the broader policy work that the innocence organizations now pursue, as those groups often become strategic partners with prosecutors in their advocacy.[68] Because shifting the traditional political alignment against reform is frequently critical to success,[69] it stands to reason that it might be more difficult for reformers to draw attention in individual cases to questionable prosecutorial disclosure practices, or to prioritize those issues such as discovery reform that are directly anathema to prosecutors’ core interests. Moreover, to the extent that exonerations are litigated at least partially with an eye to subsequent civil claims for the former defendants, there is also an additional, even more concrete, incentive not to point fingers at prosecutors: in light of broad civil immunity for prosecutors’ Brady violations, identifying prosecutorial disclosure errors as the cause of a wrongful conviction could preclude an exoneree’s subsequent civil claim.[70] On the other hand, emphasizing other systemic concerns permits triangulation. Justice-seeking prosecutors can point the finger at other criminal justice entities for blemishes in eyewitness identification, forensics, and confessions.[71] In sum, there is reason to think that the complicated dynamics of navigating both the legal and political terrain of innocence work might, at least at times, suppress the impetus of reformers to tie other known causes of wrongful convictions to prosecutors and discovery, rather than (primarily) other actors and factors.


Lessons and Possibilities

If contemporary efforts to vindicate the promise of Brady have gained only limited traction from the innocence effect, two possible responses might improve the situation. One is to counter this trend and strengthen the link between discovery reform and innocence. The other is to embrace the disjunction and develop a distinctive rhetorical and political space for disclosure outside of an innocence-focused reform strategy.

As to the first strategy, one key step toward better deploying an innocence-based approach centers on data gathering. In contrast to the robust academic analysis of eyewitness identification, interrogation, and forensic science,[72] disclosure has received little systematic attention beyond limited surveys of actual practices. I suggest three areas of concentration for social scientists and other empirical researchers.

The first concerns ascertaining the frequency of Brady violations. Part II presented the difficulty of extrapolating the incidence and effect of Brady violations from appellate and post-conviction incidents of adjudicated Brady claims. A more promising tack would be to take up a suggestion made by Barry Scheck among others, and turn attention to “near misses”: documentation of instances prior to termination of a criminal case where the state would have failed to disclose evidence but for the intervention of supervisory or judicial oversight.[73] In order to accomplish this research task, jurisdictions would first have to establish discovery protocols along with rigorous documentation or supervision of compliance, and also engage in real-time observation and documentation of missteps.[74] While this approach requires working directly with prosecutors’ offices, an invitation that might meet resistance, if presented correctly it offers the substantial upside of stakeholder buy-in.

A second area ripe for study is the analysis of causal connections between discovery violations and erroneous outcomes in criminal cases, particularly through use of sophisticated empirical methodologies that move beyond mere counting of disclosure violations in exoneration cases. Exemplary is an analysis recently done by a team led by Jon Gould, which, employing multiple methods including bivariate and logistical regression as well as qualitative analysis, identified evidentiary non-disclosure as a significant predictor of error when false convictions were compared with similar cases resulting in acquittals.[75] As Gould and Richard Leo have explained, such “matched comparison” analysis permits “scholars to more accurately determine what factors are uniquely present in wrongful conviction cases, as well as to statistically test hypotheses about what factors may be causally related to or predict wrongful conviction.”[76]

Third, more research into prosecutorial behavior could helpfully contribute to an understanding of whether measures like open file policies are, as many have asserted, necessary to ensure Brady compliance.[77] Rigorous study of the presence and effect of cognitive biases among prosecutors, for example, is one promising route.[78] Conversely, more can be learned about best practices by close quantitative and qualitative study of outcomes in those jurisdictions and offices that have taken more expansive approaches to disclosure.[79] A combination of the time- and resource-intensive nature of such work, the difficulty of obtaining publicly available information about prosecutorial practices, and the reluctance of many offices to permit such examination makes engaging in such research burdensome and expensive. Scholars and grant funders should place greater priority on overcoming these challenges to better illuminate and learn from current practices.

None of these suggestions fully meets the difficulties posed by identifying and understanding the significance of disclosure problems. But the results of such inquiries might partially fill the evidence vacuum that still exists for discovery reform in an age of innocence, and will likely arm advocates for more open discovery with more favorable empirics than what is yielded by simple counting and correlation of nondisclosure and exonerations.

On the other hand, I want to sound a cautionary note about working to enhance the relationship between innocence and discovery doctrine. Attempting to strengthen the accuracy-based underpinnings of Brady risks exacerbating the doctrine’s Achilles heel: the judicial tendency to transmogrify Brady’s materiality prong into a harmlessness assessment that hinges on a judgment about the probable guilt of the defendant.[80] Reinforcing that understanding of Brady presents a real threat at a time when there are live doctrinal debates, especially in the realm of civil claims brought by former defendants aggrieved by evidentiary nondisclosure, over whether Brady promises a fair trial to the guilty as well as the innocent.[81] Making innocence the centerpiece of Brady’s guarantee might cabin sub-constitutional discovery doctrines as well. For example, closely linking discovery and innocence might diminish the impetus to meaningfully sanction non-compliance in cases of “harmless” disclosure error, or might lessen traction for pre-plea discovery—measures from which the guilty along with the innocent would benefit. Even if one elevates accuracy above all other values in criminal adjudication, this result might be viewed as problematic, since there is good reason to doubt that prosecutors ex ante or judges ex post will always reliably sort the guilty from the innocent when making judgments about whether disclosure is constitutionally required.[82]

In any event, we should consider whether at least as much good can come from a shift in the professional culture of prosecution as will ever come from judicially or politically enacted rules. Indeed, this shift was part of the initial attraction and promise of Brady, later to be cabined and subsumed, ironically, by the doctrine’s turn (through the materiality prong) to focus on the reliability of the result generated rather than the sheer unfairness of non-disclosure.[83] The potentially unwinnable empirical debates about the prevalence and consequences of nondisclosure, which pit prosecutors against discovery reform advocates in often hostile terms, divert energy from developing a robust and public prosecutorial self-conception centered on justice rather than adversarial victory.[84] Meanwhile, there are examples of justice-seeking prosecution practices, such as voluntarily enacted open file policies or the development of robust Conviction Integrity programs that, if conspicuously elevated and popularized, could shift the tide toward a public expectation that the prosecutorial role is to support rather than reject or evince skepticism toward greater transparency in criminal adjudication.[85]

Innocence-centered advocates have done some important work to tout such efforts[86], but it is fair to surmise that an agenda of documenting the link between prosecutorial practices and wrongful conviction might crowd out broader efforts at this more cooperative style of engagement. And yet, in a (still imagined) world where the pursuit of fairness as much as accuracy is a matter of political concern to elected prosecutors,[87] where prospective district attorneys could campaign on a platform of openness or face political risk by opposing reform, there would be hope that a fuller flowering of Brady’s conception of justice-seeking for its own sake might take hold.


∞ Assistant Professor, University of Texas School of Law.

[1]. See Brady v. Maryland, 373 U.S. 83, 87 (1963).

[2]. See infra Part II.A.

[3]. See infra Part II.A.

[4]. See, e.g., Brandon L. Garrett, DNA and Due Process, 78 Fordham L. Rev. 2919, 2921 (2010) (discussing structural impact of “DNA revolution”).

[5]. The Innocence Project, (last visited June 9, 2014) (listing 316 DNA exonerations as of June 9, 2014).

[6]. The phrase is not my own and has been used by a number of commenters. See, e.g., Marvin Zalman, An Integrated Justice Model of Wrongful Convictions, 74 Alb. L. Rev. 1465, 1499 (2011); Keith A. Findley, Defining Innocence, 74 Alb. L. Rev. 1157, 1204 (2011).

[7]. See, e.g., Kansas v. Marsh, 548 U.S. 163, 208 (2006) (Souter, J., dissenting) (arguing that maximizing death sentences is incompatible with a moral justice system in light of recent DNA-backed exonerations); Dan Simon, The Limited Diagnosticity of Criminal Trials, 64 Vand. L. Rev. 143, 214 (2011) (“The criminal justice system can no longer afford to skirt the issue of factual guilt.”); The Science and Standards of Forensics: Hearing Before the S. Comm. on Sci., Commerce, and Transp., 112th Cong. (2012).

[8]. See infra Part II.B. See also Findley, supra note 6, at 1176–77 (discussing systemic impact of “innocence movement”).

[9]. See Alafair S. Burke, Talking About Prosecutors, 31 Cardozo L. Rev. 2119, 2126–27 (2010) (discussing “high-profile examples”); Fred Klein, A View from Inside the Ropes: A Prosecutor’s Viewpoint on Disclosing Exculpatory Evidence, 38 Hofstra L. Rev. 867, 871 (2010) (discussing the Duke case as a recent prominent example of prosecutorial misconduct).

[10]. Pamela Karlan, The Paradoxical Structure of Constitutional Litigation, 75 Fordham L. Rev. 1913, 1915 (2007).

[11]. See, e.g., Stephen J. Schulhofer, The Constitution and the Police: Individual Rights and Law Enforcement, 66 Wash. U. L.Q. 11, 11–13 (1988).

[12]. See, e.g., William J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 Yale L.J. 1, 16–19 (1997) (contrasting Warren Court proceduralism with a hypothetical outcome-based approach).

[13].See Miranda v. Arizona, 384 U.S. 436, 460 (1966) (“[T]he constitutional foundation underlying the privilege [against self-incrimination] is the respect a government—state or federal—must accord to the dignity and integrity of its citizens.”); Gideon v. Wainwright, 372 U.S. 335, 344 (1963) (“From the very beginning, our state and national constitutions and laws have laid great emphasis on procedural and substantive safeguards designed to assure fair trials before impartial tribunals in which every defendant stands equal before the law. This noble ideal cannot be realized if the poor man charged with crime has to face his accusers without a lawyer to assist him.”); Mapp v. Ohio, 367 U.S. 643, 656-60 (1961) (“There are those who say, as did Justice (then Judge) Cardozo, that under our constitutional exclusionary doctrine ‘(t)he criminal is to go free because the constable has blundered.’ In some cases this will undoubtedly be the result. But, as was said in Elkins, ‘there is another consideration—the imperative of judicial integrity.’ The criminal goes free, if he must, but it is the law that sets him free.”) (internal citations and footnotes omitted).

[14]. See Brady v. Maryland, 373 U.S. 83, 86–88 (1963).

[15]. See United States v. Bagley, 473 U.S. 667, 675 (1985).

[16]. See Burke, supra note 9, at 2133–34.

[17]. See Keith A. Findley & Michael S. Scott, The Multiple Dimensions of Tunnel Vision in Criminal Cases, 2006 Wis. L. Rev. 291, 351–52.

[18]. William J. Brennan, The Criminal Prosecution: A Sporting Event Or a Quest for the Truth?, 1963 Wash. U. L.Q. 279, 279, 283–84.

[19]. Charles Alan Wright, Andrew D. Leipold, Peter J. Henning & Sarah N. Welling, 2 Fed. Prac. & Proc. Crim. § 251 (4th ed.).

[20]. See Robert L. Fletcher, Pretrial Discovery in State Criminal Cases, 12 Stan. L. Rev. 293, 321 (1960); Roger J. Traynor, Ground Lost and Found in Criminal Discovery, 39 N.Y.U. L. Rev. 228, 228–35 (1964).

[21]. See, e.g., Brennan, supra note 18, at 282–88; Traynor, supra note 20, at 228–29.

[22]. Traynor, supra note 20, at 228–35; Wright, Leipold, Henning & Welling, supra note 19, at § 252 nn.37–38 (noting that revisions to Rule 16 expanded in the year after Brady).

[23]. Wayne R. Lafave, Jerold H. Israel, Nancy J. King & Orin S. Kerr, 5 Crim. Proc. § 20.3(j) (3d ed. 2013) (describing widespread exclusion of police reports from discovery regimes); Bennett L. Gershman, Litigating Brady v. Maryland: Games Prosecutors Play, 57 Case W. Res. L. Rev. 531, 542–43 (2007) (listing documents “often viewed as critical to defense discovery” including “a list of the government’s witnesses, statements of those witnesses, summaries of statements made by witnesses, and relevant police reports”).

[24]. See Steve Williams, Implementing Brady v. Maryland: An Argument for a Pretrial Open File Policy, 43 U. Cin. L. Rev. 889, 904–07 (1974); Robert P. Mosteller, Exculpatory Evidence, Ethics, and the Road to the Disbarment of Mike Nifong: The Critical Importance of Full Open-File Discovery, 15 Geo. Mason L. Rev. 257, 262–63 (2008) (defining “open file”).

[25]. See infra notes 47–50, and accompanying text.

[26]. See Herrera v. Collins, 506 U.S. 390, 420 (1993) (O’Connor, J., concurring) (“Our society has a high degree of confidence in its criminal trials, in no small part because the Constitution offers unparalleled protections against convicting the innocent.”).

[27]. Findley, supra note 6, at 1158.

[28]. See, e.g., Brandon Garrett, Convicting the Innocent: Where Criminal Prosecutions Go Wrong passim (2010) (identifying these features chapter-by-chapter in analysis of first 250 DNA exonerations); Samuel R. Gross & Michael Shaffer, Exonerations in the United States 1989–2012: Report by the National Registry of Exonerations (June 2012), available at

1989_2012_ full_report.pdf.

[29]. See supra note 6 and accompanying text.

[30]. See, e.g., Katherine R. Kruse, Instituting Innocence Reform: Wisconsin’s New Governance Experiment, 2006 Wis. L. Rev. 645, 696–718 (describing priority of innocence movement on venues outside courts); Rebecca Brown & Stephen Saloom, The Imperative of Eyewitness Identification Reform and the Role of Police Leadership, 42 U. Balt. L. Rev. 535, 551 (2013) (describing concerted reform strategy outside the Supreme Court, especially in state courts).

[31]. James M. Doyle, Learning from Error in American Criminal Justice, 100 J. Crim. L. & Criminology 109, 115–18 (2010); Kruse, supra note 30, at 645.

[32]. See William J. Stuntz, The Pathological Politics of Criminal Law, 100 Mich. L. Rev. 505, 533–39 (2001).

[33]. See David Harris, Failed Evidence 16, 130–31, 136–38 (2012) (describing emergent law enforcement support for reforms notwithstanding pockets of opposition). Significantly, all of these areas of reform have been championed not simply by the usual suspects of defendant-aligned advocates, but also a new breed of reform-oriented stakeholders including a growing number of police and prosecutors loosely aligned under the banner of “smart on crime” strategies. Proponents of being “smart on crime,” have begun to embrace accuracy-focused criminal justice reforms in the name of reducing criminal justice error and reliably convicting the guilty. See Roger Fairfax, The “Smart on Crime” Prosecutor, 25 Geo. J. Legal Ethics 905 (2012); The Smart on Crime Coalition, Smart on Crime: Recommendations for the Administration and Congress 48–69 (2011), available at

[34]. See supra notes 30–31 and accompanying text.

[35]. See See Gary L. Wells, Eyewitness Identification: Systemic Reforms, Wis. L. Rev. 615, 622–31 (2006).

[36]. See Police Exec. Research Forum, A National Survey of Eyewitness Identification Procedures in Law Enforcement Agencies 23–27 (Mar. 8, 2013), available at (listing Maryland, Virginia, Texas, New Jersey, Wisconsin, North Carolina, West Virginia, Vermont, Florida, Connecticut, Rhode Island, and Illinois); Reforms by State, Innocence Project, (last visited Apr. 5, 2014) (listing also Georgia, Nevada, and Ohio).

[37]. See POLICE EXEC. RESEARCH FORUM, supra note 38, at 22; NAT’L INST. OF JUSTICE, U.S. DEP’T OF JUSTICE, EYEWITNESS EVIDENCE: A GUIDE FOR LAW ENFORCEMENT (Oct. 1999), available at Support among the rank and file as opposed to management and institutional actors is of course more varied. See, e.g., Harris, supra note 33, at 119–20 (describing police union opposition to recording of interrogations); Zoe Tillman, D.C. Council Weighs Eyewitness ID Reform Legislation, Blog of Legal Times (Mar. 21, 2013, 3:54 PM), (describing D.C. Police Department and union opposition to identification reform supported by representative of Police Executive Research Foundation, also mentioning D.A. opposition).

[38]. See Wells, supra note 35 at 641–42.

[39]. See, e.g., Harris, supra note 33, at 136–38 (describing interrogation consensus); Resolution in Support of Efforts to Strengthen Forensic Science in the United States, Nat’l Dist. Attorneys Ass’n (Apr. 10, 2010), strengthen forensic science

resolution_4_10.pdf (expressing qualified support for forensic science oversight).

[40]. Michael S. Schmidt, In Policy Change, Justice Dept. to Require Recording of Interrogations, N.Y. Times, May 22, 2014,


[41]. See, e.g., Brandon L. Garrett, Aggregation in Criminal Law, 95 Calif. L. Rev. 383, 439 (2007) (listing open file discovery as common recommendation of innocence commissions).

[42]. I have grouped criminal discovery reform statutes into an unpublished chart, on file with me and with the N.Y.U. Rev. L. & Soc. Change. In my judgment, Minnesota, New Jersey, North Carolina, Ohio, and Texas qualify as “open file” jurisdictions based on their requirement of access to police reports and information concerning all known witnesses. The provisions on which this judgment is based are Minn. R. Crim. P. 9.01 subd. 1; N.J. R. Crim. P. 3:313-3; N.C. Gen. Stat. Ann. § 15A-903; Ohio R. Crim. P. 16; Tex. Crim. Proc. Code Ann. art. 39.14(a).

[43]. See LaFave, Israel, King & Kerr, supra note 23, §§ 20.2(b), 20.3(k) (describing limits on disclosure of reports in most jurisdictions). In my own survey of state statutes I classified regimes following the model of Federal Rule 16 or the original ABA standards for discovery as most restrictive, based on their failure to guarantee defendants access either to police reports or to information concerning witnesses known to the state who will not be called at trial. By that measure, thirty-three states were restrictive. See Discovery Statute Reform Chart (on file with author).

[44]. See Susan A. Bandes, Framing Wrongful Convictions, 2008 Utah L. Rev. 5, 16 (2008) (describing Illinois exonerations as important framing moment for innocence movement). See also Discovery Statute Reform Chart (on file with author) (showing eight such states). See generally Barry Scheck, Peter Neufeld & Jim Dwyer, Actual Innocence: Five Days to Execution and Other Dispatches from the Wrongly Convicted (2000). Two states, Florida and New Jersey, had long had open file discovery in place, but enacted recent statutory discovery amendments to specifically enumerate categories of disclosure arguably already covered but highly relevant to error concern—informant and identification information, respectively. See In re: Amendments to Florida Rule of Criminal Procedure 3.220, 2014 Florida Court Order 0018, No. SC13-1541, at 2 (May 29, 2014). Compare N.J. R. Crim. P. 3:313-3 (2012) with N.J. R. Crim. P. 3:313-3 (2011).

[45]. The top twenty states are, in order from the most exonerations to the fewest: California (119), Texas (114), Illinois (112), New York (104), Michigan (40) Florida (38), Louisiana (38), Pennsylvania (32), Massachusetts (31), Ohio (31), Wisconsin (27), Virginia (27), Washington (26), North Carolina (25), Missouri (21), Alabama (17), Oklahoma (16), Georgia (16), Indiana (14), Mississippi (13). See Exonerations by County 1989–2012, National Registry Of Exonerations, (last visited August 1, 2014). The six to have enlarged discovery are Texas, Florida, Louisiana, Massachusetts, Ohio, and North Carolina. See Discovery Statute Reform Chart (on file with author)

[46]. See, e.g., Ian Thoms, DOJ Stands Firm on Discovery Policy After Stevens Debacle, Law 360 (May 9, 2012, 1:47 PM),

[47]. See id. (discussing Rule 16 restrictions); Bruce A. Green, Federal Criminal Discovery Reform: A Legislative Approach, 64 Mercer L. Rev. 639, 640–41 (2013) (discussing the impact of the Stevens case).

[48]. See Steve Mills & John Biemer, Ford Heights 4 Inquiry Clears Cops, Prosecutors, Chi. Trib. Aug. 22, 2003, at 1.

[49]. Governor’s Comm’n on Capital Punishment, Report of Governor’s Commission on Capital Punishment 32–48 (Apr. 2002), available at

[50]. Susan S. Kuo & C.W. Taylor, In Prosecutors We Trust: U.K. Lessons for Illinois Disclosure, 38 Loy. U. Chi. L.J. 695, 710 (2007); Thomas P. Sullivan, Efforts to Improve the Illinois Capital Punishment System: Worth the Cost?, 41 U. Rich. L. Rev. 935, 955–56 (2007).

[51]. Peter Vieth, Virginia Prosecutors: Start Over on Criminal Discovery, Va. Law. Wkly., Jul. 12, 2013, available at 2013 WLNR 17587745 (reporting prosecutor view that “proposed . . . changes would tip the balance of fairness in favor of criminal defendants,” making the rule “unfair to society”).

[52]. See Mosteller, supra note 24, at 260.

[53]. Brandi Grissom, Perry Signs Michael Morton Act, Tex. Trib. (May 16, 2013),

[54]. See Ellen Yaroshefsky, New Orleans Prosecutorial Disclosure in Practice After Connick v. Thompson, 25 Geo. J. Legal Ethics 913, 913–15 (2013).

[55]. See infra notes 61–64 and accompanying text.

[56]. See, e.g., Ensuring That Federal Prosecutors Meet Discovery Obligations: Hearing on S. 2197 Before the S. Comm. on the Judiciary, 112th Cong. (2012) (Testimony of James M. Cole, Deputy Attorney General), available at (citing all these concerns in opposition to expansion of federal discovery).

[57]. See, e.g., Brandon L. Garrett, Innocence, Harmless Error, and Federal Wrongful Conviction Law, 2006 Wis. L. Rev. 35, 85–87 (“Once the person is exonerated, the court knows the witness’s identification was false, if not unreliable.”).

[58]. See Richard A. Leo & Jon B. Gould, Studying Wrongful Convictions: Learning from Social Science, 7 Ohio St. J. Crim. L. 7, 14–17 (2009) (comparing anecdotal causation claims with social science methodologies).

[59]. See Keith A. Findley, Toward A New Paradigm of Criminal Justice: How the Innocence Movement Merges Crime Control and Due Process, 41 Tex. Tech L. Rev. 133, 146–47 (2008).

[60]. See Emily M. West, Court Findings of Prosecutorial Misconduct Claims in Post-Conviction Appeals and Civil Suits in the First 255 DNA Exoneration Cases, Innocence Project (Aug. 2010),

[61]. See Kyles v. Whitley, 514 U.S. 419, 421–22 (1995) (holding due process violated whether police or prosecutor responsible for withholding evidence).

[62]. See Yaroshefsky, supra note 54, at 917–18.

[63]. Though for reasons discussed supra in Part III far more research is possible.

[64]. See Findley, supra note 6, at 1158.

[65]. State v. Henderson, 27 A.3d 872, 885 (N.J. 2011) (construing State v. Delgado, 902 A.2d 888 (N.J. 2006)). See also Perry v. New Hampshire, 132 S. Ct. 716, 739 (2012) (Sotomayor, J., dissenting). Invariably also making the list of wrongful conviction “causes” are the use of flawed forensic science and the procurement of false confessions. See Garrett, supra note 28.

[66]. The issue takes a conspicuous back seat to eyewitness identification, interrogation techniques, and failed science among other factors in Professor Garrett’s seminal work in the field. See Garrett, supra note 28. The Innocence Project’s website—a go-to resource for wrongful conviction analysis—includes “Government Misconduct” as a “cause[]” of wrongful convictions, but mentions failure to disclose exculpatory evidence as one of dozens of possible types of misconduct performed by prosecutors or police. Government Misconduct, Innocence Project,

[67]. See Steven A. Krieger, Why Our Justice System Convicts Innocent People, and the Challenges Faced by Innocence Projects Trying to Exonerate Them, 14 New Crim. L. Rev. 333, 353–54, 386–87 (2011).

[68]. See, e.g., Symposium, Voices from the Field: An Inter-Professional Approach to Managing Critical Information, 31 Cardozo L. Rev. 2037, 2069–74 (2010) (presentation of Terri Moore); Kruse, supra note 30, at 705.

[69]. Cf. Brown & Saloom, supra note 30, at 548–50 (describing work with law enforcement).

[70]. See Imbler v. Pachtman, 424 U.S. 409 (1976) (holding prosecutors absolutely immune from suit for withholding of evidence at or in preparation for trial). Developing evidence of prosecutorial knowledge of wrongdoing leading to a wrongful conviction can have the effect of precluding more than just Brady claims. See, e.g., Wray v. City of New York, 490 F.3d 189, 194–95 (2d Cir. 2007) (holding Section 1983 claim for suggestive identification procedure by police precluded in absence of evidence that police withheld facts of suggestiveness from prosecutors, and discussing related cases).

[71]. See, e.g., Tex. Dist. and Cnty. Att’ys Ass’n, Setting the Record Straight on Prosecutorial Misconduct 21–22 (Aug. 31, 2012),


[72]. See, e.g., Leo & Gould, supra note 58, at 29.

[73]. Barry Scheck, Professional and Conviction Integrity Programs: Why We Need Them, Why They Will Work, and Models for Creating Them, 31 Cardozo L. Rev. 2215, 2243 (2010).

[74]. Id. at 2239–44.

[75] Jon B. Gould, Julia Carrano, Richard Leo & Joseph Young, Predicting Erroneous Convictions: A Social Science Approach to Miscarriages of Justice ii—iii (Dec. 2012), available at

[76]. Leo & Gould, supra note 58, at 21.

[77]. See, e.g., Mosteller, supra note 24, at 307–08.

[78]. A rare example of work to date on this score is Barbara O’Brien, A Recipe for Bias: An Empirical Look at the Interplay Between Institutional Incentives and Bounded Rationality in Prosecutorial Decision Making, 74 Mo. L. Rev. 999 (2009).

[79]. While not focused on Brady issues or disclosure per se, the qualitative empirical work of Ron Wright and Marc Miller is exemplary in this regard. See, e.g., Ronald Wright & Marc Miller, The Screening/Bargaining Tradeoff, 55 Stan. L. Rev. 29, 30 (2002).

[80]. See, e.g., Strickler v. Greene, 527 U.S. 263, 297–302 (1999) (Souter, J., dissenting) (cautioning that materiality should not require “more likely than not” proof of acquittal).

[81]. Compare, e.g., Poventud v. City of New York, 750 F.3d 121, 140 (2d Cir. 2014) (Lynch, J., concurring) (emphasizing that Brady guarantees fair not accurate result), with id. at 155 (Jacobs, J., dissenting) (arguing civil Brady claim inconsistent with actual guilt).

[82]. See supra note 20 and accompanying text.

[83]. See supra Part II.A.

[84]. See, e.g., Tex. Dist. & Cty. Atty’s Assoc., supra note 71, at 7–13 (contesting claims by Innocence Project and others that evidentiary non-disclosure was a documented systemic problem in Texas).

[85]. See, e.g., Symposium, supra note 68, at 2070–72.

[86]. See Scheck supra note 73, at 2250 (highlighting work of Dallas County).

[87]. See Editorial, Duty a Strong Choice for Williamson, Austin Am. Statesman, Nov. 3, 2012, at A14 (highlighting Brady failures in district attorney endorsement).