UNIVERSITY OF UTAH
S.J. QUINNEY COLLEGE OF LAW

LEGAL STUDIES RESEARCH PAPER SERIES

Up the River without a Procedure: Innocent
Prisoners and Newly Discovered Non-DNA
Evidence in State Courts
Daniel S. Medwed

Arizona Law Review, Vol. 47, 2005
Research Paper No. 05-17
This paper can be downloaded without charge from the
Social Science Research Network Electronic Paper Collection:
http://ssrn.com/abstracts=833344

UP THE RIVER WITHOUT A PROCEDURE:
INNOCENT PRISONERS AND NEWLY
DISCOVERED NON-DNA EVIDENCE IN STATE
COURTS
Daniel S. Medwed∗
Introduction.......................................................................................................... 656
I. An Individual Injustice: People v. Schulz......................................................... 662
II. Overview: Newly Discovered Evidence in State Courts................................. 664
A. Newly Discovered Evidence Claims in Historical Context ........................ 666
1. Direct Remedies: Motion for New Trial .................................................. 666
2. Collateral Remedies: Coram Nobis and Habeas Corpus.......................... 669
B. Contemporary State Procedures Regarding Newly Discovered Evidence .. 675
1. New Trial Motions Today........................................................................ 675
2. Current Modes of Collateral Relief.......................................................... 681
III. Regime Change: Proposed Modifications to State Procedures ...................... 686
A. Decreasing the Risk of Procedural Default ................................................. 687
1. Statutes of Limitations ............................................................................. 690
2. Simplifying Post-Trial Procedures........................................................... 695
B. Full and Fair Hearings Before Unbiased Judges ......................................... 699
1. Selected Aspects of Behavioral Decisionmaking Theory ........................ 700
2. Implications of Behavioral Decisionmaking Theory for New Evidence
Claims .......................................................................................................... 703
C. Appellate Review ........................................................................................ 708
1. Theoretical Justifications for Divergent Standards of Review................. 711
2. Altering the Standard of Review Applicable to Summary Denials of Newly
Discovered Evidence Claims ....................................................................... 714
Conclusion ........................................................................................................... 715

∗
Associate Professor of Law, University of Utah—S.J. Quinney College of
Law. J.D., Harvard Law School, 1995; B.A., Yale College, 1991. I am particularly grateful
to the following people for their helpful comments regarding this project: Jensie Anderson,
Ursula Bentele, Dana Brakman Reiser, Stacy Caplow, Paul Cassell, Linda Feldman, Leslie
Francis, Daniel Greenwood, Will Hellerstein, Sharissa Jones, Claire Kelly, Erik Luna,
Steven Mulroy, Linda Smith, Marjorie Smith, George Thomas, and Manuel Utset. Special
thanks as well to the participants in a University of Utah faculty workshop at which I
presented a previous version of this paper. I would also like to acknowledge the able
research assistance conducted by Melissa Clark, Benjamin Rouse, Ryan Schriever, and
Nicole Skalla, and the financial support provided by the S.J. Quinney College of Law’s
Summer Stipend Program.

656

ARIZONA LAW REVIEW

[VOL. 47:655

INTRODUCTION
With all due respect to Edith Wharton’s literary masterpiece about the
golden era of New York high society, today may be the true “Age of Innocence.”1
Over the past fifteen years, the use of deoxyribonucleic acid (DNA) testing in
criminal cases has helped to expose the problem of wrongful convictions in the
United States, resulting in the post-conviction exoneration of 162 innocent
defendants.2 Indeed, dozens of state legislatures have recently enacted laws
implementing procedures through which prisoners may request and secure access
to post-conviction testing of biological evidence available in their cases.3 By
creating mechanisms that allow for the possibility that factually innocent prisoners
may have entrée to DNA testing and thereby gain their freedom, these statutes both
further the cause of justice at the individual level and bolster the institutional
credibility of the criminal justice system as a whole.4
Evidence suitable for DNA testing, however, exists only in a smattering
of criminal cases: an estimated 80–90% of cases do not have any biological
evidence.5 Even where biological evidence conducive to a DNA test is present at
the outset of a particular case, the evidence is often lost, destroyed, or degraded

1.
EDITH WHARTON, AGE OF INNOCENCE (1st ed. 1920).
2.
For a current tally of DNA exonerations, see Innocence Project Homepage,
http://www.innocenceproject.org (last visited Sept. 6, 2005). For other case studies of
exonerations, see EDWARD CONNORS ET AL., U.S. DEP’T OF JUSTICE, CONVICTED BY JURIES,
EXONERATED BY SCIENCE: CASE STUDIES IN THE USE OF DNA EVIDENCE TO ESTABLISH
INNOCENCE AFTER TRIAL (1996); Samuel R. Gross et al., Exonerations in the United States
1989 Through 2003, 95 J. CRIM. L. & CRIMINOLOGY 523, 524 (2005).
3.
See THE INNOCENCE PROJECT: POST-CONVICTION DNA TESTING STATUTES,
http://www.innocenceproject.org/docs/Model_DNA_Access_FactSheet.html (last visited
May 16, 2005) (noting that “thirty-eight states provide convicted persons access to DNA
testing”).
4.
Although post-conviction DNA testing legislation has generally received
extensive praise, some observers suggest that many details of the statutes could be
improved. See generally Kathy Swedlow, Don’t Believe Everything You Read: A Review of
Modern “Post-Conviction” DNA Testing Statutes, 38 CAL. W. L. REV. 355 (2002) (lauding
the enactment of these statutes, while also pinpointing some of their flaws).
5.
See Death Penalty Overhaul: Hearing Before the S. Comm. on the Judiciary,
107th Cong. (2002) (statement of Barry Scheck), available at 2002 WL 1335515 (“The vast
majority (probably 80%) of felony cases do not involve biological evidence that can be
subjected to DNA testing.”); Nina Martin, Innocence Lost, S.F. MAG., Nov. 2004, at 78, 105
(noting that “only about 10 percent of criminal cases have any biological evidence—blood,
semen, skin—to test”). Advancements in DNA technology might eventually produce results
in cases that are currently not subject to DNA tests. Seth Kreimer recently predicted a
“second wave” of DNA exonerations in the next decade, observing that “more sophisticated
and sensitive methods of DNA analysis are beginning to be tested that can engage in DNA
matching from ever smaller amounts of biological material. Laboratory reports have been
released of effective DNA matches from the small amounts of skin cells contained in
smudged latent finger or palm prints.” Seth F. Kreimer, Truth Machines and Consequences:
The Light and Dark Sides of ‘Accuracy’ in Criminal Justice, 60 N.Y.U. ANN. SURV. AM. L.
655, 658–59 (2005).

2005]

UP THE RIVER

657

over time.6 Post-conviction innocence claims based on DNA testing therefore
represent a small proportion of innocence claims generally,7 and this percentage is
bound to diminish in the future.8 That is, the growing availability of DNA
technology at the pre-trial stage should filter out a greater number of innocent
defendants up-front and, in so doing, decrease the volume of post-conviction
exonerations via DNA.9
Although post-conviction innocence claims hinging on DNA testing have
captured the attention of state legislators and the broader public, the far more
pervasive issue of innocence claims in cases that lack biological evidence has
largely escaped notice.10 The same problems that led to the wrongful convictions
of those innocent prisoners later freed through DNA—erroneous eyewitness
identifications, false confessions, witness perjury, ineffective assistance of counsel,
and the like—presumably appear in the scores of convictions procured without
biological evidence.11 In these non-DNA cases, prisoners must find alternative
6.
See Barry Scheck & Peter Neufeld, DNA and Innocence Scholarship, in
WRONGLY CONVICTED: PERSPECTIVES ON FAILED JUSTICE 241, 245 (Saundra D. Westervelt
& John A. Humphrey eds., 2001) (“In 75 percent of Innocence Project cases, matters in
which it has been established that a favorable DNA result would be sufficient to vacate the
inmate’s conviction, the relevant biological evidence has either been destroyed or lost.”).
7.
See, e.g., Steven J. Mulroy, The Safety Net: Applying Coram Nobis Law to
Prevent the Execution of the Innocent, 11 VA. J. SOC. POL’Y & L. 1, 7 n.35 (2003) (“Of the
102 cases since 1973 where persons were freed from death row based on evidence of
innocence, DNA only played a substantial role in 12 of those cases.”); Ronald J. Tabak,
Finality Without Fairness: Why We are Moving Toward Moratoria on Executions, and the
Potential Abolition of Capital Punishment, 33 CONN. L. REV. 733, 735 (2001)
(“Notwithstanding all the hoopla surrounding DNA testing, it can be performed only in a
small minority of situations in which significant biological evidence from the real culprit is
collected properly at the scene of the crime.”).
8.
See, e.g., BARRY SCHECK ET AL., ACTUAL INNOCENCE: WHEN JUSTICE GOES
WRONG AND HOW TO MAKE IT RIGHT 323 (2001) (“In a few years, the era of DNA
exonerations will come to an end. The population of prisoners who can be helped by DNA
testing is shrinking, because the technology has been widely used since the early 1990s,
clearing thousands of innocent suspects before trial.”).
9.
Id.
10.
See, e.g., Mulroy, supra note 7, at 1 (noting that, whereas “headline-making
DNA sleuthing can certainly help in remedying individual cases of injustice,” there should
be “adequate procedural mechanisms—and legal standards governing those mechanisms—
to address those situations where, after conviction, information comes to light raising
doubts” about the validity of a conviction).
11.
Scholarship analyzing wrongful convictions predates the DNA revolution by
decades, and a text by Edwin Borchard is often cited as one of the earliest accounts. EDWIN
M. BORCHARD, CONVICTING THE INNOCENT (1932). It is unclear whether the same error rate
applies to convictions with DNA evidence as to those lacking it. See, e.g., George C.
Thomas III et al., Is It Ever Too Late for Innocence? Finality, Efficiency, and Claims of
Innocence, 64 U. Pitt. L. Rev. 263, 271–73 (2003) (discussing what the DNA exonerations
signify regarding the total incidence of wrongful convictions). Even so, the same sources of
wrongful convictions appear in both DNA and non-DNA exonerations. See, e.g., SCHECK ET
AL., supra note 8, at 323 (“From Borchard’s review of cases stretching back to the dawn of
the American republic, all the way to the dawn of the twenty-first century, the causes of
wrongful convictions remain the same.”).

658

ARIZONA LAW REVIEW

[VOL. 47:655

means to support their innocence claims, frequently, “newly discovered evidence”
not susceptible to a test tube,12 such as confessions by the actual perpetrator,
statements by previously unknown witnesses, and/or recantations by trial
participants.13 Without a doubt, non-DNA cases are difficult for defendants to
overturn through state court proceedings given the subjectivity involved in
assessing most forms of new evidence and the absence of a method to prove
innocence to a scientific certainty.14
This inherent difficulty in litigating innocence claims predicated on newly
discovered non-DNA evidence is exacerbated by the structural design of most state
post-conviction regimes: in effect, the path to proving one’s innocence through

12.
This Article will focus principally on non-DNA, nonscientific forms of new
evidence given that states often provide special procedures for requesting scientific tests
that were technologically unavailable at trial. See, e.g., IDAHO CODE ANN. § 19-4902(b)
(2004) (“A petitioner may, at any time, file a petition before the trial court that entered the
judgment of conviction in his or her case for the performance of fingerprint or forensic
deoxyribonucleic acid (DNA) testing on evidence that was secured in relation to the trial
which resulted in his or her conviction but which was not subject to the testing that is now
requested because the technology for the testing was not available at the time of trial.”
(emphasis added)).
13.
The particular types of non-DNA, nonscientific evidence that have served as
bases for newly discovered evidence claims are often testimonial in nature. See, e.g.,
Michael J. Muskat, Note, Substantive Justice and State Interests in the Aftermath of Herrera
v. Collins: Finding an Adequate Process for the Resolution of Bare Innocence Claims
Through State Postconviction Remedies, 75 TEX. L. REV. 131, 132–33 (1996) (describing
the sources of newly discovered evidence, “including testimony by previously undiscovered
witnesses, recantation of testimony given by key prosecution witnesses, confession by the
real killer . . .” (internal footnotes omitted)). For a discussion of how state courts have
treated post-conviction claims hinging on third-party confessions, see Thomas R. Malia,
Annotation, Coram Nobis on Ground of Other’s Confession to Crime, 46 A.L.R.4th 468
(1986 & Supp. 2000). For a discussion of how courts have grappled with the issue of
whether to grant a motion for a new trial based on recanted testimony, see Keith A.
Mitchell, Note, Protecting Guiltless Guilty: Material Witness Recantation and Modern
Post-Conviction Remedies, 21 NEW ENG. L. REV. 429 (1985–86); Janice J. Repka,
Comment, Rethinking the Standard for New Trial Motions Based Upon Recantations as
Newly Discovered Evidence, 134 U. PA. L. REV. 1433 (1986); Daniel Wolf, Note, I Cannot
Tell a Lie: The Standard for New Trial in False Testimony Cases, 83 MICH. L. REV. 1925
(1985); Tim A. Thomas, Annotation, Standard for Granting or Denying New Trial in State
Criminal Case on Basis of Recanted Testimony—Modern Cases, 77 A.L.R.4th 1031 (1990
& Supp. 2005).
14.
See Edward K. Cheng, Reenvisioning Law through the DNA Lens, 60 N.Y.U.
ANN. SURV. AM. L. 649, 649 (2005) (“DNA is a special kind of evidence with few previous
analogs: It is powerful, physical evidence of identity that remains stable and available for
retesting long after trial . . . . [B]ecause the scientific community developed DNA typing,
DNA evidence comes pre-packaged with all the indicia of scientific reliability: population
statistics, pre-defined and pre-tested procedural standards, and known error rates.”); Scheck
& Neufeld, supra note 6, at 248–49 (explaining that DNA testing provides scientific
certainty to support innocence claims). In contrast, as Cheng notes, “[m]ost traditional
forms of evidence are fleeting—memories fade, eyewitnesses move away, and (written)
records are unwieldy to preserve and frequently lost.” Cheng, supra, at 650.

2005]

UP THE RIVER

659

new evidence has become virtually impassable due to procedural roadblocks.15 To
be sure, every state currently permits at least some form of post-trial relief on the
basis of newly discovered evidence.16 All too often, states even have multiple
potential remedies for a new evidence claim—including an ordinary motion for a
new trial and a collateral, post-conviction procedure17—yet navigating through
these procedural shoals can be a source of confusion and frustration for litigants.18
In general, these remedies are characterized by stringent statutes of limitations,19
limited access to discovery,20 and high legal and evidentiary thresholds.21 Merely
obtaining an evidentiary hearing pursuant to these procedures may be hard to
achieve.22 Even more, motions seeking relief on the grounds of new evidence are
often filed with the original trial judge,23 a person who may have a vested interest
15.
See Eli Paul Mazur, “I’m Innocent”: Addressing Freestanding Claims of
Actual Innocence in State and Federal Courts, 25 N.C. CENT. L.J. 197, 199 (2003) (“In
North Carolina state courts, inmates are required to present newly discovered evidence
suggesting actual innocence in a Motion for Appropriate Relief (“MAR”). This process is
plagued by unnecessary procedural hurdles, the nonsensical imputation of attorney
negligence upon indigent criminal defendants, and judicial conflicts of interests.”).
16.
See 1 DONALD E. WILKES, JR., STATE POSTCONVICTION REMEDIES AND
RELIEF: WITH FORMS, § 1-13, at 55–58 (2001) (noting that every state offers a direct remedy
in the form of a new trial motion based on newly discovered evidence, and many allow
newly discovered evidence as a ground for collateral, post-conviction relief).
17.
See infra notes 65–70 and accompanying text.
18.
1 WILKES, supra note 16, § 1-13, at 53–111 (discussing the potential
remedies and obstacles facing an inmate seeking to present newly discovered evidence of
innocence); see also infra note 274 and accompanying text (mentioning how many postconviction motions are filed by inmates without the assistance of an assigned attorney).
19.
1 WILKES, supra note 16, § 1-13, at 56 (commenting that many states have
brief statutes of limitations periods).
20.
Cynthia Bryant, When One Man’s DNA is Another Man’s Exonerating
Evidence: Compelling Consensual Sexual Partners of Rape Victims to Provide DNA
Samples to Postconviction Petitioners, 33 COLUM. J.L. & SOC. PROBS. 113, 122–23 (2000)
(noting how procedures for newly discovered evidence claims typically presuppose that
prisoners already have the exonerating evidence and mentioning that courts have offered
mixed views on whether to grant discovery in cases where the pertinent statute is silent); see
also Gibson v. United States, 566 A.2d 473, 478–79 (D.C. 1989) (granting defendant
discovery to support motion for new trial).
21.
See In re Clark, 855 P.2d 729, 739 (Cal. 1993) (holding that a conviction
may only be attacked collaterally on grounds of newly discovered evidence if the new
evidence creates fundamental doubt about the reliability and accuracy of the proceedings);
Mulroy, supra note 7 (analyzing the evidentiary standards used in Tennessee regarding
newly discovered evidence claims); Muskat, supra note 13 (discussing the need for more
permissive legal and evidentiary rules in the area of “bare innocence claims”); Margaret
Raymond, The Problem with Innocence, 49 CLEV. ST. L. REV. 449, 452 (2001) (“An
exoneration based on innocence requires, in the ordinary course, overwhelming proof,
which is necessary to overcome the substantial procedural barriers to relitigation of
outcomes in criminal cases.”).
22.
See Swedlow, supra note 4, at 356 (stating that “‘traditional’ post-conviction
remedies . . . are dominated by harsh procedural rules and . . . often function to deny
prisoners substantive review”).
23.
1 WILKES, supra note 16, § 1-13, at 55 (noting that new trial motions are
usually available in the convicting court).

660

ARIZONA LAW REVIEW

[VOL. 47:655

in the outcome,24 and that judge’s decision normally receives tremendous
deference on appeal.25 The hodge-podge of state newly discovered evidence
procedures must be critically examined in light of the realization, spawned by
DNA testing, that innocent defendants are convicted with alarming frequency, and
the reality that post-conviction innocence claims receive vastly different
procedural treatment within most states depending on whether the claim involves
DNA or non-DNA evidence.26
This Article aims to provide just such an examination: an analysis of the
state procedures that prisoners may employ after trial to litigate innocence claims
grounded on newly discovered non-DNA evidence.27 Ultimately, the result of this
examination is far from sanguine. Little-altered in decades beyond the trend
toward recognizing the benefits of DNA testing, the structure of most state
procedures means that a prisoner’s quest for justice may turn on the fortuity that a
biological sample was left at the crime scene and preserved over time.28 The fact
that DNA testing provides a modicum of certainty to an innocence claim does not
imply that claims lacking the possibility of such certainty are spurious; on the
contrary, DNA has unearthed holes in the criminal justice system, holes that are
likely also prevalent in cases without biological evidence.29
But, as a precondition to filling these holes and garnering relief in newly
discovered non-DNA evidence cases for potentially innocent defendants,30
24.
See infra notes 284–335 and accompanying text.
25.
The standard of review for the denial of a motion for a new trial on the
grounds of newly discovered evidence is generally abuse of discretion. See Penny J. White,
Newly Available, Not Newly Discovered, 2 J. APP. PRAC. & PROCESS 7, 13 (2000). For a
brief description of the abuse of discretion standard, see 5 AM. JUR. 2D Appellate Review
§ 695 (2004).
26.
See, e.g., Daniel F. Piar, Using Coram Nobis to Attack Wrongful
Convictions: A New Look at an Ancient Writ, 30 N. KY. L. REV. 505, 506–07 (2003)
(observing that coram nobis procedures bear a striking resemblance to their English
ancestors).
27.
Several scholars have studied the procedures in individual states. See, e.g.,
Josephine Linker Hart & Guilford M. Dudley, Available Post-Trial Relief After a State
Criminal Conviction When Newly Discovered Evidence Establishes “Actual Innocence”, 22
U. ARK. LITTLE ROCK L. REV. 629, 632 (2000) (determining that Arkansas’s “post-trial
procedures provide little opportunity for a prisoner to establish his or her actual innocence
through newly discovered evidence”); Mazur, supra note 15 (criticizing North Carolina’s
treatment of newly discovered evidence claims and advocating a series of reforms).
28.
See Mulroy, supra note 7, at 4 (commenting how, in many cases,
exonerations “came as the result of the ‘sheer accident’ that a biological DNA sample
happened to be available”).
29.
See, e.g., Richard A. Rosen, Innocence and Death, 82 N.C. L. REV. 61, 73
(2003) (observing “that for every defendant who is exonerated because of DNA evidence,
there have been certainly hundreds, maybe thousands, who have been convicted” on
comparable evidence yet whose cases lack physical evidence).
30.
In this Article, I will focus on claims of factual innocence as opposed to
wrongful convictions generally. See Paul G. Cassell, The Guilty and the “Innocent”: An
Examination of Alleged Causes of Wrongful Conviction from False Confessions, 22 HARV.
J.L. & PUB. POL’Y 523, 535–36 (1999) (arguing for a focus on factual innocence); Daniel
Givelber, Meaningless Acquittals, Meaningful Convictions: Do We Reliably Acquit the

2005]

UP THE RIVER

661

prisoners must first have adequate access to state courts—and such access is
deficient in most jurisdictions.31 To address this problem, states should: (1) refashion their procedures to minimize the chance newly discovered non-DNA
evidence claims will be rejected due solely to procedural default; (2) construct
each remedy so as to enhance the likelihood that viable claims will be heard in
open court in front of an unbiased judge; and (3) utilize a de novo standard of
review for appellate courts in assessing summary denials of motions for post-trial
relief based on newly discovered evidence, i.e., cases where the trial court declines
to hold an evidentiary hearing on the merits of an innocence claim prior to
rejecting it. Modifying state procedures in this manner, without adjusting the legal
and evidentiary thresholds that apply to innocence claims involving new
evidence,32 should allow more claims of potentially innocent prisoners to see the
light of day in a courtroom, yet not result in additional guilty inmates seeing the
light of day in the free world.
Part I of this Article offers a concrete example of how procedures in this
area are troubling by exploring a single case in which state remedies proved
inadequate to allow a prisoner to present his alleged new evidence in open court.
Next, Part II traces the historical development of state approaches to newly
discovered evidence claims, and then discusses the salient traits of contemporary
state procedures. Finally, Part III critiques these procedures and assesses
prospective reforms.33
Innocent?, 49 RUTGERS L. REV. 1317, 1346 n.92 (1997) (“Actual innocence means what it
says—the defendant did not commit the crime of which he has been convicted. Wrongfullyconvicted defendants may or may not be actually innocent; their defining characteristic is
that their convictions were secured as a result of a material legal error.”).
31.
See infra notes 152–215 and accompanying text.
32.
Some commentators have pressed for reforms in the legal and evidentiary
standards used in this area. See Mulroy, supra note 7, at 1–3 (contending that the law in
Tennessee, as in other states, is “not up to the task” of providing adequate post-conviction
procedures and urging that in Tennessee “defendants should be able to prevail on a coram
nobis petition if they can establish a ‘reasonable probability’ that the newly discovered
evidence would have changed the initial trial outcome”); Muskat, supra note 13, at 162,
176–86 (recommending that, when reviewing bare innocence claims in state post-conviction
proceedings, “[t]he reviewing judge should first assume the credibility of the new evidence
in the light most favorable to the petitioner, and then evaluate its strength by applying a
standard for relief which requires a judge wishing to grant relief to conclude that had the
new evidence been presented at trial, no rational juror would have found the petitioner
guilty beyond a reasonable doubt”). Muskat has also suggested that this standard “better
assures that potentially meritorious claims will receive an evidentiary hearing to evaluate
the credibility of the evidence.” Muskat, supra note 13, at 162, 176–86. Outside the sphere
of post-conviction remedies, scholars have also proposed the reform of trial and direct
appellate procedures in cases of factual innocence. See D. Michael Risinger, Unsafe
Verdicts: The Need for Reformed Standards for the Trial and Review of Factual Innocence
Claims, 41 HOUS. L. REV. 1281 (2004) (arguing for, among other things, the adoption of an
“unsafe verdict” standard of review on direct appeal).
33.
This Article will not explore the full gamut of state court mechanisms that
may be at an inmate’s disposal in raising a non-DNA innocence claim, including common
law and statutory remedies unrelated to new evidence that allow collateral attacks based on
alleged “off-the-record” constitutional violations. See 1 WILKES, supra note 16, § 1-13, at

662

ARIZONA LAW REVIEW

[VOL. 47:655

I. AN INDIVIDUAL INJUSTICE: PEOPLE V. SCHULZ
To demonstrate the problems that can be generated by state procedures
involving newly discovered evidence of innocence, I will draw upon my prior
experience representing inmates in post-conviction proceedings for three years as
assistant director of an innocence project at Brooklyn Law School.34 Among the
litany of claims that I investigated—and the few that I actually litigated—one
stands out in particular, People v. Schulz.35 The basic facts of the case are as
follows.36 A large white man entered a diner in a Long Island town on February 3,
1999.37 The man placed his food order with the diner’s cook, who then entered the
kitchen to prepare the meal.38 While the cook was in the kitchen, the man robbed a
waitress at knifepoint; she was staffing the restaurant’s cash register and was the
only other person present in the dining area at the time.39
The next day, both the cook and the waitress viewed a photo array and
identified Stephen Schulz, a tall white man who lived nearby, weighed
approximately 250 pounds, and had a prior criminal record, as the perpetrator.40
The cook later identified Schulz in a lineup—the waitress never participated in any
identification procedures beyond the initial photo array—and Schulz was charged
with first-degree robbery.41 Prior to trial, the prosecution made Schulz an offer to
plead guilty to a lesser crime and receive three years’ imprisonment.42 Schulz
rejected the deal, steadfastly claiming to be innocent.43

55 (citing “numerous cases where an innocent person obtained postconviction relief, not on
grounds of newly discovered evidence of innocence, but rather on grounds the person was
denied a constitutional right in the proceedings leading to the conviction”); Ursula Bentele,
Does the Death Penalty, by Risking Execution of the Innocent, Violate Substantive Due
Process?, 40 HOUS. L. REV. 1359 (2004) (discussing some of the constitutional arguments
against capital punishment in light of the possibility of executing innocent defendants). This
Article also declines to tackle the treatment of newly discovered evidence claims in federal
court. Moreover, questions related to inmates who plead guilty and later seek to overturn
their guilty pleas due to new evidence may implicate different state law remedies and, thus,
will not be discussed.
34.
For a description of the Second Look Program at Brooklyn Law School, see
Brooklyn Law School: Academic Program: Second Look Program, http://www.brooklaw.
edu/academic/courses/description/?course=116 (last visited Jan. 20, 2005).
35.
774 N.Y.S.2d 165 (App. Div. 2004).
36.
The facts of this case are far more complex than those presented in this
Article. For the sake of brevity, I have put forth only the details essential to the issues raised
in this Article, namely, the facts relevant to the defendant’s newly discovered non-DNA
evidence of innocence claim. There were also other issues presented in the direct appeal and
concerns regarding ineffective assistance of trial counsel.
37.
Brief for Defendant-Appellant at 11, 14, People v. Schulz, 774 N.Y.S.2d 165
(App. Div. 2004) (A.D. Nos. 2003–01596, 2000–09423, 1999–10592) [hereinafter
Appellant’s Brief].
38.
Id. at 11–12, 14–15.
39.
Id. at 12, 15.
40.
Id. at 10, 13, 16.
41.
Id.
42.
Id. at 22.
43.
Id.

2005]

UP THE RIVER

663

At trial, the cook maintained his belief in Schulz’s guilt, but his testimony
was replete with inconsistencies and reasons to question his credibility.44
Specifically, the cook denied he had received any benefit in exchange for his
testimony, a claim placed in doubt by the fact that his own lingering felony charge
for weapon possession, pending at the time of the robbery, had been favorably
resolved in the interim through a plea to the minor crime of disorderly conduct.45
Then the waitress (the actual victim of and sole eyewitness to the robbery itself)
took the stand. At a pivotal juncture in her testimony, when asked whether the man
who had robbed her was present in the courtroom, she answered “No” and
indicated the culprit was taller and heavier than the defendant.46 The jury,
nonetheless, found Schulz guilty of robbery in the first degree, and the court
eventually sentenced him to eleven years in prison.47
More than two years later, my students and I began to follow a lead that
Schulz’s trial attorney had not comprehensively pursued: the possibility that
another person had committed the crime. In particular, a white man named
Anthony Guilfoyle, who was both taller and heavier than Schulz, had pled guilty to
six factually analogous robberies that occurred in the vicinity of the diner from
January to March 1999.48 In 2002, we managed to locate the waitress and, upon
viewing a photograph of Guilfoyle, she identified him with 90% certainty as the
man who had robbed her;49 evidently, neither the police nor anyone else had ever
shown her Guilfoyle’s picture before.50 We also interviewed Schulz’s roommate at
the time of the robbery, a disabled mechanic with no criminal record, who declared
that he was home with Schulz watching television when the incident allegedly took
place.51
After obtaining affidavits from the waitress and roommate, as well as
gathering other materials,52 we filed a motion under the pertinent state postconviction remedy with the New York state judge before whom Schulz was tried
and convicted, requesting a new trial on the basis of newly discovered evidence or,
at the very least, an evidentiary hearing at which our witnesses could be heard to

44.
Id. at 11–14.
45.
Id. at 14.
46.
Id. at 15.
47.
Id. at 21–22.
48.
Id. at 17–19, 27; see also Barbara Durkin, The Description: Big; Robbery
Suspect Used Size to Bully, Cops Say, NEWSDAY, Mar. 10, 1999, at A31 (describing
Guilfoyle’s crime spree and plea bargain).
49.
Appellant’s Brief, supra note 37, at 24–25.
50.
The waitress was not shown Guilfoyle’s photograph during the trial, and
there is no indication, from the police reports or other sources, that law enforcement
officials ever showed her a picture of Guilfoyle. See generally id.
51.
Notably, Schulz’s roommate did not testify as an alibi witness at trial. Id. at
26.
52.
The other materials included a polygraph report demonstrating Schulz had
passed a lie detector test and an affidavit from Schulz’s trial attorney explaining why he
chose not to cross-examine the waitress and ask her about Guilfoyle, noting that he had not
had the opportunity to interview her before trial. Id. at 24–28.

664

ARIZONA LAW REVIEW

[VOL. 47:655

determine if a new trial was warranted.53 On January 6, 2003, the court heard oral
arguments to decide whether to hold an evidentiary hearing on the points raised in
the motion.54 As defense counsel, I urged the court to follow case law favoring the
use of post-conviction hearings “to promote justice if the issues raised are
sufficiently unusual and suggest investigation.”55 Unmoved, the trial judge stated
his recollection that the waitress had failed to identify Schulz at trial because of
“fear,”56 and one month later formally rejected our motion without a hearing.57
Turning to the appellate courts, we first had to seek leave to appeal the
summary disposition of the newly discovered evidence claim.58 Although we
gained authorization to appeal the decision, the state intermediate appellate court
unanimously affirmed the trial judge’s decision, concluding that the court had
“providently exercised its discretion” in denying the motion summarily.59
Thereafter, we petitioned for further review from the state Court of Appeals, the
court of last resort in New York, and received permission to appeal. In a six-to-one
decision, however, the Court of Appeals sustained the intermediate court’s holding
in May 2005.60 To date, the waitress has never testified in open court as to whether
Guilfoyle committed the crime, and Schulz’s alibi witness never will; he has since
died. People v. Schulz is an example of how state court procedures in the sphere of
newly discovered non-DNA evidence claims may not always achieve the result
that justice and common sense require: full and fair airings of claims of innocence.

II. OVERVIEW: NEWLY DISCOVERED EVIDENCE IN STATE COURTS
Post-conviction remedies are designed primarily to address egregious
legal errors: mistakes of either jurisdictional or constitutional dimensions.61 As a
result, newly discovered evidence claims do not seem at first blush to be likely
candidates for post-conviction relief given their fact-based orientation. Moreover,
state courts have traditionally viewed newly discovered evidence claims with
disdain,62 fearing the impact of such claims on the finality of judgments and the
53.
Id. at 24, 27–28; see also N.Y. CRIM. PROC. LAW § 440.10(1)(g) (McKinney
2004) (describing the standard for moving for a new trial on the basis of newly discovered
evidence).
54.
Appellant’s Brief, supra note 37, at 28.
55.
Id. at 28–29.
56.
Minutes of Argument at 22, People v. Schulz, No. 368-99 (N.Y. Sup. Ct.,
Jan. 6, 2003) (“There are no papers that can be put into words what [the waitress] was like
at the time of trial, Mr. Medwed. And I say this: It was fear written all over her face.”).
57.
Appellant’s Brief, supra note 37, at 29.
58.
Id. at 31; see also N.Y. CRIM. PROC. LAW §§ 450.15, 460.15 (McKinney
2004) (providing procedures through which a defendant may seek permission to appeal the
denial of a post-trial newly discovered evidence claim).
59.
People v. Schulz, 774 N.Y.S.2d 165, 167 (App. Div. 2004).
60.
People v. Schulz, 829 N.E.2d 1192, 1199 (N.Y. 2005).
61.
1 WILKES, supra note 16, § 1-5, at 17 (noting that “postconviction relief
usually is available only on grounds involving egregious legal error, i.e., errors that are
jurisdictional, constitutional, or otherwise fundamental”).
62.
See, e.g., People v Sutton, 15 P. 86, 88 (Cal. 1888) (mentioning that claims
of newly discovered evidence warranting a new trial “are to be regarded with distrust and
disfavor”); Sanders v. State, 370 N.E.2d 966, 968 (Ind. Ct. App. 1977) (“A motion for a
new trial on the basis of newly discovered evidence should be received with great caution,

2005]

UP THE RIVER

665

historic role of the jury as the true arbiter of fact,63 and harboring doubts about the
underlying validity of new evidence.64
Asserting newly discovered evidence as a ground for relief, though, has
become an integral part of the state court landscape for criminal defendants. First,
every state provides for a motion for a new trial on the basis of newly discovered
evidence.65 Such motions are viewed as direct, rather than collateral, remedies that
are filed in the court of original conviction66 and usually carry with them strict
statutes of limitations.67 Second, a number of states allow newly discovered
evidence to serve as grounds for collateral, post-conviction relief through
procedures in the nature of a writ of error coram nobis or, less often, habeas
corpus.68
Therefore, a host of procedures—including, on occasion, multiple
procedures within a particular state69—affect newly discovered evidence of
and the alleged new evidence should be carefully scrutinized.”); Hart & Dudley, supra note
27, at 632 (“Newly discovered evidence is the least-favored ground for the granting of a
new-trial motion [in Arkansas].”).
63.
See, e.g., Lee v. Moore, 213 So. 2d 197, 198 (Ala. 1968) (noting, in the
context of a civil case, that “trial courts do have the power to grant motions for new trials in
order to prevent irrevocable damage. Such power should be hesitantly exercised, because
the verdict of a jury results from one of the most precious rights in our system of
government, that is, the right of trial by jury”).
64.
Courts are often dubious of recantations or, for that matter, any evidence that
otherwise contains ample opportunities for perjury. See supra note 13 and accompanying
text. Likewise, courts are typically skeptical of third-party confessions. See, e.g., Brown v.
State, 955 S.W.2d 901, 902 (Ark. 1997) (“The mere fact that another person has confessed
to a crime cannot, alone, be grounds for relief for such confessions are not uncommon and
must be approached with some skepticism.”). Moreover, judicial skepticism toward new
evidence claims in general may be exacerbated by the overall volume of frivolous motions
filed by inmates. See, e.g., Daniel S. Medwed, The Zeal Deal: Prosecutorial Resistance to
Post-Conviction Claims of Innocence, 84 B.U. L. REV. 125, 148–49 (2004) (noting how the
large number of post-conviction filings, many of which lack merit, deters prosecutors from
viewing each one as potentially viable).
65.
Herrera v. Collins, 506 U.S. 390, 410–11 (1993); 1 WILKES, supra note 16, at
30 n.15.
66.
See 1 WILKES, supra note 16, §§ 1-3 to 1-5, at 13–30 (describing the general
difference between direct and post-conviction remedies).
67.
See id. A major distinction between new trial motions and collateral
remedies concerning newly discovered evidence relates to statutes of limitations. Id. § 1-13,
at 56 (“In those state jurisdictions which have affixed a statute of limitations to a particular
postconviction remedy, the statute of limitations is usually longer than the limitations period
for direct remedies, including the motion for new trial; and claims for relief based on newly
discovered evidence are often exempted from the limitations period applicable to a
particular state postconviction remedy.”); Piar, supra note 26, at 508 n.35 (“The principal
difference between a coram nobis petition and a motion for new trial is one of timing: a
motion for new trial must normally be sought within a limited time (a maximum of three
years), while traditional coram nobis relief could be sought at any time.”).
68.
See infra notes 190–92 and accompanying text.
69.
For example, there are two post-trial statutory methods for prisoners to put
forth newly discovered evidence claims in Tennessee. See Mulroy, supra note 7, at 12–25;
Piar, supra note 26, at 517–25. A defendant may move for a new trial on that ground within

666

ARIZONA LAW REVIEW

[VOL. 47:655

innocence raised by state prisoners after the completion of their trials. The motion
for a new trial, albeit technically a direct remedy, is one of the tools available to
convicted criminal defendants and must be analyzed in conjunction with existing
collateral remedies to assess the manner in which newly discovered non-DNA
evidence is treated in state courts. Also, on a conceptual level, new trial motions
and collateral remedies centering on newly discovered evidence are extremely
similar and ought to be viewed in tandem.70
A. Newly Discovered Evidence Claims in Historical Context
1. Direct Remedies: Motion for New Trial
England began to permit new trials in criminal cases in the late
seventeenth century.71 In the United States, the First Congress acknowledged this
remedy, authorizing new trials for the “reasons for which new trials have usually
been granted in courts of law,”72 and individual states soon followed suit.73 From
the outset, motions for a new trial were normally filed with the original trial judge
who presided over the case.74 The power of a trial judge to order a new trial stems
from the equitable concept that litigants and society as a whole deserve some
mechanism to mend fundamental miscarriages of justice.75 In accord with these
equitable underpinnings, newly discovered evidence became generally recognized
in federal and state jurisdictions as one of the grounds upon which new trial relief
thirty days of the judgment, TENN. R. CRIM. P. 33(b), and may seek collateral relief within
one year under the Tennessee Post-Conviction Relief Act, TENN. CODE ANN. § 40-30-102
(2004); see also infra notes 263–72 and accompanying text.
70.
See LARRY W. YACKLE, POSTCONVICTION REMEDIES § 8, at 31 (1981) (“To
begin, to the extent its objective is a new trial, the writ [of coram nobis] has something in
common with ordinary post-trial motions, particularly a motion for a new trial.”); Malia,
supra note 13, § 2(a), at 471 (“It has been said that a petition for a writ of coram nobis is in
the nature of a motion for a new trial, at least to the extent that the result sought, that of a
new trial, is the same.”); Mulroy, supra note 7, at 9 (“Coram nobis . . . alleges no error by
the original court or in any of the court’s findings; rather, it simply calls to that court’s
attention additional facts which were not known to the court at the time and which may
change the result. Of course, in this form, it does parallel a motion for new trial based on
newly discovered evidence.”). Yackle, however, notes several “important differences”
between new trial motions and the writ of coram nobis, including the fact that “[w]hile in
some jurisdictions coram nobis is considered a part of the original criminal case, in others
the writ initiates an independent civil lawsuit collaterally attacking the judgment.” YACKLE,
supra, at 31; see also In re Cruz, 129 Cal. Rptr. 2d 31, 38 (Ct. App. 2003) (“By analogy, a
petition for a writ of habeas corpus based on newly discovered evidence resembles a motion
for a new trial based on newly discovered evidence.”).
71.
Herrera v. Collins, 506 U.S. 390, 408 (1993).
72 .
Id. (citing Act of Sept. 24, 1789, ch. 20, § 17, 1 Stat. 83).
73.
Id. at 409–10.
74.
See William Renwick Riddell, New Trial in Present Practice, 27 YALE L.J.
353, 360 (1917) (describing how the colonies followed the English common law system
with respect to new trials, and noting how “the trial judge (at least in most cases) sat as the
court and not as a mere commissioner; and he it was to whom the application for a new trial
was made”). Riddell also observed that “[a]t the present time in practically every state of the
Union, the trial judge has power to grant a new trial.” Id. at 361.
75.
See 58 AM. JUR. 2D New Trial § 8 (2004).

2005]

UP THE RIVER

667

could be granted.76 Still, the notion that the discovery of new evidence could
theoretically provide the foundation for a new trial in early American courts belied
the reality that such claims were embraced by neither the judiciary nor the
legislature.
For instance, most jurisdictions had rigid statutes of limitations on filing
motions for new trials.77 Early federal cases even abided by the English common
law rule that a new trial could only be granted during the term of the court in
which the original judgment in the case was entered; if the claim were to be filed at
a subsequent term, the court merely had the power to correct mistakes in form or
clerical errors.78 Likewise, most states initially heeded the common law rule
regarding new trials—in jurisdictions that allowed new trial motions, those
motions usually had to be filed prior to the end of the court term during which the
original decision had been issued.79 In due course, states made new trial motions
available in a variety of situations, including claims of newly discovered
evidence.80 Although some states enlarged the time limit for filing beyond the
court term, many continued to impose a brief time limit.81
Even if a newly discovered evidence claim did not fall prey to a statute of
limitations problem, stern legal and evidentiary tests guided their resolution.82 In
1851, the Georgia Supreme Court first articulated the “probability” standard to
govern new trial motions based on newly discovered evidence.83 Berry v. Georgia
involved accusations that a white man, James Berry, had hatched a plan for two
African-American slaves to steal money from another white man’s house.84 The
state tried and convicted Berry of larceny,85 and thereafter Berry filed a request for
76.
Herrera, 506 U.S. at 408–10.
77.
Id.
78.
Id. at 408. In 1934, the United States Supreme Court retreated from the
common law rule and approved a sixty-day statute of limitations for the filing of new trial
motions premised on newly discovered evidence in federal court. Id. In referring to the
sixty-day limit for new trial motions, the Supreme Court opined in 1946 that:
[T]he extraordinary length of time within which this motion can be made
is designed to afford relief where, despite the fair conduct of the trial, it
later clearly appears to the trial judge that, because of facts unknown at
the time of trial, substantial justice was not done. It is obvious, however,
that this privilege might lend itself for use as a method of delaying
enforcement of just sentences.
United States v. Johnson, 327 U.S. 106, 112 (1946). In federal court, the time period for
filing a new trial motion based on newly discovered evidence is now three years. FED. R.
CRIM. P. 33(b).
79.
Herrera, 506 U.S. at 409.
80.
Id. at 409–10.
81.
Id.
82.
See, e.g., Berry v. Georgia, 10 Ga. 511, 527 (1851); see also State v.
Montgomery, 109 P. 815, 817 (Utah 1910) (holding that newly discovered evidence must be
so conclusive as to raise a reasonable presumption that the result of a second trial would
differ from the first).
83.
Berry, 10 Ga. at 527; see also Wolf, supra note 13, at 1925 n.5 (crediting
Berry with being the first case to implement the probability test).
84.
Berry, 10 Ga. at 513–16.
85.
Id. at 514.

668

ARIZONA LAW REVIEW

[VOL. 47:655

a new trial on, among other grounds, newly discovered evidence.86 In wrestling
with Berry’s claim, Georgia’s highest court began by commenting that new trial
motions based on newly discovered evidence “are not favored by the Courts,”87
and noted that in neighboring South Carolina new trials are not granted on this
ground alone.88 Citing “a pretty general concurrence of authority,”89 the court then
outlined a multi-factor test for handling new evidence claims:
It is incumbent on a party who asks for a new trial, on the ground of
newly discovered evidence, to satisfy the Court, 1st. That the
evidence has come to his knowledge since the trial. 2d. That it was
not owing to the want of due diligence that it did not come sooner.
3d. That it is so material that it would probably produce a different
verdict, if the new trial were granted. 4th. That it is not cumulative
only—viz.; speaking to facts, in relation to which there was
evidence on the trial. 5th. That the affidavit of the witness himself
should be produced, or its absence accounted for. And 6th, a new
trial will not be granted, if the only object of the testimony is to
impeach the character or credit of a witness.90

The Berry court, applying this test to the facts of the case, found the
defendant’s motion wanting and denied the request for a new trial.91 In the ensuing
decades, the test put forth in Berry became the model to which other states looked
in formulating their own responses to newly discovered evidence claims.92
A subsequent federal case, Larrison v. United States,93 also influenced the
treatment of newly discovered evidence claims where the claim hinged on witness
recantations.94 Larrison announced that a new trial motion should be granted
where: (1) the court is satisfied that the testimony of the material witness is false;
86.
Id. at 515–16.
87.
Id. at 527.
88.
Id.
89.
Id.
90.
Id.
91.
Id. at 528, 531.
92.
See, e.g., Salinas v. State, 373 P.2d 512, 514 (Alaska 1962) (“A motion for a
new trial based on the ground of newly discovered evidence has to meet the following
requirements: (1) it must appear from the motion that the evidence relied on is, in fact,
newly discovered, i.e., discovered after the trial; (2) the motion must allege facts from
which the court may infer diligence on the part of the movant; (3) the evidence relied on
must not be merely cumulative or impeaching; (4) must be material to the issues involved;
and (5) must be such as, on a new trial, would probably produce an acquittal.” (quoting Pitts
v. United States, 263 F.2d 808, 810 (9th Cir. 1959))); People v. Salemi, 128 N.E.2d 377,
381 (N.Y. 1955) (holding that, to be considered newly discovered in New York, the
evidence must meet the following six criteria: “(1) it must be such as will probably change
the result if a new trial is granted; (2) it must have been discovered since the trial; (3) it
must be such as could have not been discovered before the trial by the exercise of due
diligence; (4) it must be material to the issue; (5) it must not be cumulative to the former
issue; and (6) it must not be merely impeaching or contradicting the former evidence”
(quoting People v. Priori, 58 N.E. 668, 672 (N.Y. 1900))).
93.
24 F.2d 82 (7th Cir. 1928).
94.
See Mulroy, supra note 7, at 7 (noting that witness recantation evidence “is
the most common source of newly discovered evidence in criminal cases”).

2005]

UP THE RIVER

669

(2) the jury might have reached a different conclusion without the testimony; and
(3) the party seeking a new trial was surprised when the new testimony was given
and did not know of the original testimony’s falsity until after the trial.95 Several
jurisdictions later adopted the Larrison test as the proper standard in ruling on new
trial motions involving witness recantations.96 The evolution of the Berry and
Larrison tests, coupled with harsh statutes of limitations, reflected the judicial and
legislative cynicism toward newly discovered evidence claims in the nineteenth
and early twentieth centuries—these claims may have been cognizable through
motions for a new trial, but relief was seldom forthcoming. 97
2. Collateral Remedies: Coram Nobis and Habeas Corpus
In addition to the new trial motion remedy, the modern approach to newly
discovered evidence claims has its roots in the English common law writs of error
coram nobis98 and, to a much lesser extent, habeas corpus.99 The remedy of coram
nobis, which literally means “before us,”100 was available in the court of original
judgment in order to amend its own proceedings.101 Developed in the sixteenth
century, coram nobis served to correct significant errors of fact rather than law in
criminal cases in England; claims made collaterally under this writ asserted the
existence of facts unknown to the court at the time of judgment that bore upon the
soundness of a conviction.102 Classic functions of the writ included rectifying
clerical errors or mistakes concerning the process of notice and pleading.103 Most
notably, this writ contained no statute of limitations—it was cognizable “however
late discovered and alleged”104—and its trademark form of relief was to vacate the

95.
See Larrison, 24 F.2d at 87–88.
96.
See generally Thomas, supra note 13.
97.
See supra notes 82–96 and accompanying text.
98.
See 1 WILKES, supra note 16, § 1-5, at 16 (citing that “there are two basic
categories of principal postconviction remedies in the states. They are (1) the remedy of the
writ of habeas corpus, and (2) the remedies in the nature of the writ of error coram nobis”);
Piar, supra note 26, at 506–07.
99.
Historically, there were several substantive differences between the writs of
coram nobis and habeas corpus. See YACKLE, supra note 70, §§ 1-13, at 1–69; Piar, supra
note 26, at 508 n.39 (“The major practical difference between habeas corpus and coram
nobis in modern law is that habeas corpus will lie only where the petitioner is in custody,
whereas coram nobis can be used even after a sentence has been served and a petitioner
released. Further, coram nobis is an attack on the validity of a conviction because of facts
unknown at trial, while habeas corpus is an attack on the legality of detention for reasons
that may or may not have to do with the factual basis for the conviction.” (citation
omitted)).
100.
See Mulroy, supra note 7, at 9.
101.
See Piar, supra note 26, at 506.
102.
Id.
103.
See YACKLE, supra note 70, § 8, at 32.
104.
Piar, supra note 26, at 507 n.20 (quoting Blackstone). Nonetheless, postconviction remedies have traditionally been subject to some timing restrictions, if not
statutes of limitations. See, e.g., 1 WILKES, supra note 16, § 1-5, at 18 (observing that
historically post-conviction remedies could not be invoked until after direct remedy
proceedings, i.e., the direct appeal of the conviction).

670

ARIZONA LAW REVIEW

[VOL. 47:655

conviction with leave for the state to re-try the defendant.105 Despite the absence of
a statute of limitations, parties seeking to use the remedy of coram nobis were
required to prove they had proceeded with reasonable diligence.106
The United States Supreme Court recognized coram nobis as early as
1810,107 and individual states started to accept the writ’s legitimacy near the
beginning of the nation’s history.108 Together with acknowledging the concept of
coram nobis, American jurisdictions imported many of the writ’s mechanical traits
from England. Application for the writ, for example, typically occurred in the
court of conviction through motion or petition accompanied by affidavits.109 Early
observers deemed it preferable for the same judge who had rendered the initial
judgment to receive the application due to that jurist’s familiarity with the facts.110
In assimilating coram nobis into their legal frameworks, however,
American courts did not mirror the English model entirely. Daniel Piar suggests
that coram nobis jurisprudence in the United States evolved “along slightly more
liberal lines than it had in England,”111 as American judges showed less concern
for rigid adherence to common law forms than for the achievement of substantive
justice.112 Over time, coram nobis emerged as an extraordinary remedy to which
courts resorted sparingly and only to repair significant miscarriages of justice.113
Even so, not all of the writ’s potential uses were welcomed with open arms in the
United States, especially its availability in the realm of innocence claims founded
on newly discovered evidence.
Whereas the Berry and Larrison tests, or variations thereof,114 soon
governed most state courts’ treatment of new trial motions relating to newly
discovered evidence of innocence in the nineteenth and twentieth centuries,115
105.
See Piar, supra note 26, at 507.
106.
See Mulroy, supra note 7, at 11.
107.
See Piar, supra note 26, at 507 (citing Strode v. The Stafford Justices, 23 F.
Cas. 236 (Marshall, Circuit Justice, C.C.D. Va. 1810) (No. 13,537)).
108.
See Piar, supra note 26, at 507 n.23. Donald Wilkes identifies a Virginia
state court case as the first known coram nobis proceeding in the United States, Gordon v.
Frazier, 2 Wash. 130 (Va. 1795), and mentions that for many years coram nobis in state
courts was only available in regard to civil proceedings. 1 WILKES, supra note 16, § 2-3, at
149. Wilkes further observes that a decision of the Missouri Supreme Court, Ex parte
Toney, 11 Mo. 661 (1848), represented the initial use of coram nobis as a post-conviction
remedy in state court to attack a criminal conviction, and notes that “it was not until the
period from 1927 to 1931 that the majority of coram nobis cases reaching the state appellate
courts involved attacks on criminal rather than civil judgments.” Id. § 2-3, at 150.
109.
See YACKLE, supra note 70, § 8, at 34.
110.
Id.
111.
Piar, supra note 26, at 507.
112.
See id. at 507–08.
113.
Id. at 508 n.32; see also 1 WILKES, supra note 16, § 2–4, at 155 (“As late as
1916 there were only 30 reported state postconviction coram nobis cases . . . .”); Mulroy,
supra note 7, at 11 (“Originally, American courts did not often make use of the writ in
criminal cases, and usage declined well into the 20th century.”).
114.
See Thomas, supra note 13 (observing that some jurisdictions developed
standards independent of either test, at least in the area of recantations).
115.
See supra notes 92–96 and accompanying text.

2005]

UP THE RIVER

671

judges appeared to be even more grudging in accepting the idea that new evidence
alone might form the basis for a collateral remedy. In fact, the predominant rule in
nineteenth century American courts decreed that newly discovered evidence never
warranted coram nobis or habeas corpus relief.116 The policies behind this rule lay
primarily in concerns about preserving the finality of judgments and aversions to
the relitigation of guilt or innocence,117 mainly in situations involving witness
recantations.118 Given that petitions for coram nobis relief generally lacked statutes
of limitations, courts were also fearful that a defendant could bolster his chance of
success by purposely letting time pass before petitioning for the writ.119
Furthermore, possible injustices revealed through the discovery of new evidence
could supposedly be cured by an alternative remedy: the executive clemency
power.120
Notwithstanding the aforementioned reservations, beginning in the 1920s,
several state courts began to concede that common law post-conviction relief could
be granted on the basis of newly discovered evidence.121 The courts were frugal in
fashioning such remedies, though, affording relief typically only upon determining
that the newly discovered evidence would conclusively result in a different
verdict.122 By the 1940s, a handful of other state courts had held that, while
common law coram nobis could not be used to resolve innocence claims based on
newly discovered evidence, judges could still create a special post-conviction
remedy comparable to coram nobis to fix errors in cases where proof of innocence
was conclusive.123 Around the same time, the general use of common law coram
nobis experienced an upsurge in the wake of the Supreme Court’s decision in

116.
1 WILKES, supra note 16, § 1-13, at 56.
117.
Id. § 1-13, at 56–57; see also Humphreys v. State, 224 P. 937, 940 (Wa.
1924) (“It seems highly probable that in all cases of newly discovered evidence touching
exclusively the merits of the issue actually tried and determined there can be no relief under
any circumstances at the hands of the courts. To open the door to such inquiry would be to
create a condition wherein the judgments of courts would have no finality . . . .”).
118.
See supra note 13 and accompanying text.
119.
As the Kansas Supreme Court warned in 1901, “[w]hen time had removed or
so scattered the witnesses of the prosecution, or the memory of those who could be obtained
had grown dim, a writ of coram nobis would result in certain acquittal.” Dobbs v. State, 65
P. 658, 660 (Kan. 1901).
120.
See, e.g., Sharpe v. Commonwealth, 143 S.W.2d 857, 858 (Ky. 1940)
(holding that “an appeal for executive clemency is the remedy of a person convicted of a
crime which, after the judgment of the court convicting him has become final, it is shown
by subsequently discovered evidence he did not commit”); Humphreys, 224 P. at 940
(suggesting that a lack of finality for court judgments would “be fruitful of greater evil than
would flow from very rare cases of possible injustice, which would, however, not be beyond
all cure, for, if injustice results from any such condition, it is readily curable, as far as
human ingenuity can safely do, upon proper showing by a resort to the pardoning power”).
121.
1 WILKES, supra note 16, § 1-13, at 57. Wilkes pinpoints Davis v. State, 161
N.E. 375 (Ind. 1928), as the “first major state appellate decision holding newly discovered
evidence may be a basis for postconviction coram nobis relief.” Id. § 2-4, at 159 n.34.
122.
Id. § 1-13, at 57.
123.
Id.

672

ARIZONA LAW REVIEW

[VOL. 47:655

Mooney v. Holohan,124 which prescribed that state post-conviction procedures
must be broad enough to address deprivations of federal constitutional rights.125
Shortly thereafter, a trend arose whereby states enacted legislation
forming more comprehensive post-conviction regimes to supplant traditional
common law remedies.126 In a sense, the creation of statute- and rule-based
systems reflected an effort to grapple with the unruly nature of the common law
writs: to tame coram nobis, above all, a remedy that a Kentucky judge once termed
“the wild ass of the law which the courts cannot control.”127 By 1965, eighteen
states had promulgated a modern post-conviction remedy through statute or court
rule,128 a movement that has continued into the present day.129 The extent to which
coram nobis survived this development differs significantly across the country.130
In some states, coram nobis was officially incorporated into the legislative
framework by statute or endured as a secondary common law remedy; in other
states, it was abolished or superseded by new statutory forms of relief.131
Establishing statutory post-conviction regimes by and large altered the scope of
relief available in state courts, yielding a narrowing of potential relief in some
locales while other systems retained relief coextensive or possibly greater than that
afforded by common law coram nobis.132

124.
294 U.S. 103 (1935).
125.
Id.; see also Young v. Ragen, 337 U.S. 235, 238–39 (1949) (urging states to
augment their state post-conviction remedies); Mulroy, supra note 7, at 11–12 (noting how
Mooney triggered a revival of coram nobis).
126.
The Illinois Post-Conviction Hearing Act is credited as the first statute
establishing an entirely new, comprehensive post-conviction remedy. 1 WILKES, supra note
16, § 2-5, at 161 (“The Illinois Post-Conviction Hearing Act of 1949 created a modern
postconviction remedy—that is, a remedy which (1) authorizes relief on grounds of
violations of constitutional or other basic rights generally, (2) is unhampered by obsolete,
unjustified, and irksome procedural obstacles to relief, and (3) is designed to be used in lieu
of traditional, narrower habeas corpus and common law coram nobis remedies.”). The
remedy formed by the Illinois statute can be considered “in the nature of coram nobis, i.e., it
was available, like the common law coram nobis remedy, in the convicting court . . . .” Id.;
see also YACKLE, supra note 70, § 10, at 42–43 (noting that “the Act channeled all
prisoners’ postconviction claims to the sentencing court”).
127.
Anderson v. Buchanan, 168 S.W.2d 48, 55 (Ky. 1943) (Sims, J., dissenting).
128.
See 1 WILKES, supra note 16, § 3-2, at 188.
129.
See infra notes 184–87 and accompanying text. According to Wilkes, the
writ of habeas corpus is currently the principal post-conviction remedy in twelve states. 1
WILKES, supra note 16, § 3-3, at 206. In four of those states, “the remedy results principally
from liberal judicial interpretation of traditional habeas corpus statutory provisions,”
whereas “[i]n the remaining eight states the remedy is the result of either a modernizing
statute . . . or a modernizing judicially promulgated rule of court . . . .” Id. As for the other
thirty-eight states, Wilkes characterizes their principal post-conviction remedy as “a remedy
in the nature of coram nobis.” Id. § 3-4, at 213. In twenty-six of those states, “the remedy
was created by statutory enactment,” and in the other twelve it emerged through “a
judicially promulgated rule of court.” Id.
130.
See Mulroy, supra note 7, at 11–12.
131.
Id. at 12.
132.
See Piar, supra note 26, at 530.

2005]

UP THE RIVER

673

The codification of most state post-conviction practices may have
signaled the demise of the common law writ of error coram nobis as a principal
state law remedy,133 yet not of newly discovered evidence claims as grounds for
post-conviction relief in state courts. Indeed, some states explicitly cited newly
discovered evidence as a basis for relief when overhauling their post-conviction
regimes by statute or court rule,134 and others found such claims cognizable
through judicial decisions.135 Additionally, the 1966 and 1980 versions of the
Uniform Post-Conviction Procedure Act (UPCPA), a model statute implemented
to some degree in several jurisdictions,136 endorsed newly discovered evidence as
an acceptable ground for collateral relief.137 This feature of the UPCPA also met
with the approval of both the 1968 and 1978 editions of the American Bar
Association’s Post-Conviction Standards.138
As for legal and evidentiary thresholds, many jurisdictions that counted
newly discovered evidence among their post-conviction causes of action expressly
imposed heavy burdens on defendants seeking to prevail under that theory. By
statute or court rule, some states required that factors comparable to those
developed in the context of new trial motions (e.g., Berry’s probability standard)139
should govern new evidence claims raised collaterally as well.140 Even where the
pertinent post-conviction rule was silent in regard to the precise legal standards
controlling newly discovered evidence claims, case law in certain states adopted
criteria similar to those announced in Berry,141 and still other states kept the rule
133.
See 1 WILKES, supra note 16, § 3-5, at 232 (observing that common law
coram nobis, while a secondary post-conviction remedy in eighteen states, is no longer a
principal post-conviction remedy in any state).
134.
Id. § 1-13, at 57.
135.
Id. § 1-13, at 58.
136.
See id. app. A at 762 (describing the grounds for relief in Idaho’s postconviction statute as identical to those of 1966 UPCPA); Piar, supra note 26, at 530–31
(noting how a number of states hew closely to the UPCPA in providing post-conviction
relief that is liberal both as to timing and scope); IDAHO CODE ANN. § 19-4901(a) (2004).
137.
1 WILKES, supra note 16, § 1-13, at 58 (commenting that both versions of the
UPCPA “expressly provide for relief if there is evidence of material facts, not previously
presented or heard, that requires vacation of the conviction or sentence in the interest of
justice”).
138.
Id.; see also STANDARDS FOR POST-CONVICTION REMEDIES 22-2.1(a)(v)
(1978) [hereinafter ABA 1978 STANDARDS]; STANDARDS RELATING TO POST-CONVICTION
REMEDIES § 2.1(a)(v) (1968) [hereinafter ABA 1968 STANDARDS].
139.
See supra notes 82–92 and accompanying text.
140.
See, e.g., ARIZ. R. CRIM. PROC. 32.1(e) (including, as a ground for postconviction relief, a claim that “[n]ewly discovered material facts probably exist and such
facts probably would have changed the verdict or sentence. Newly discovered material facts
exist if: (1) The newly discovered material facts were discovered after the trial. (2) The
defendant exercised due diligence in securing the newly discovered material facts. (3) The
newly discovered material facts are not merely cumulative or used solely for impeachment,
unless the impeachment evidence substantially undermines testimony which was of critical
significance at trial such that the evidence probably would have changed the verdict or
sentence”).
141.
See, e.g., People v. Johnson, 793 N.E.2d 591, 598 (Ill. 2002) (“A claim of
actual innocence based on newly discovered evidence may be raised in a post-conviction
petition . . . . The supporting evidence must be new, material, noncumulative, and so

674

ARIZONA LAW REVIEW

[VOL. 47:655

that newly discovered evidence only merited collateral relief upon “conclusive”
proof of innocence.142
With respect to habeas corpus, the history of the “Great Writ”143 in state
courts is of little service to this Article’s emphasis on newly discovered evidence
claims because habeas corpus traditionally applied to prisoners challenging the
lawfulness of their detention and extended only to errors of law, not fact.144 Factual
questions of guilt or innocence therefore had no bearing on a court’s assessment of
a habeas corpus petition.145 With that history in mind, courts have been reticent to
address newly discovered evidence claims raised via habeas corpus; as the Iowa
Supreme Court observed in 1917 in rejecting a habeas corpus filing based on
alleged new evidence of innocence, “[W]e are not yet ready to make so radical a
venture.”146 Nevertheless, given that several states have since made that venture
and currently treat habeas corpus as a means by which prisoners may present
newly discovered evidence of innocence,147 a few of the writ’s distinctive aspects
conclusive that it would probably change the result on retrial.”); Harrington v. State, 659
N.W.2d 509, 516 (Iowa 2003) (holding that, with respect to a petition filed pursuant to
Iowa’s post-conviction procedures, “[t]o prevail on his newly discovered evidence claim
[the defendant] was required to show: (1) that the evidence was discovered after the verdict;
(2) that it could not have been discovered earlier in the exercise of due diligence; (3) that the
evidence is material to the issues in the case and not merely cumulative or impeaching; and
(4) that the evidence probably would have changed the result of the trial”).
142.
See, e.g., Richardson v. State, 546 So. 2d 1037, 1038 (Fla. 1989) (holding
that, in order to receive coram nobis relief, the alleged newly discovered evidence must be
of a character such that it would conclusively have prevented entry of the judgment).
143.
See, e.g., YACKLE, supra note 70, § 4, at 7–15 (discussing the history of
habeas corpus).
144.
See, e.g., id. §§ 3-6, at 4–29. Habeas corpus was often only available to
correct jurisdictional defects. See 1 WILKES, supra note 16, § 2-2, at 136 (commenting that,
from roughly the mid-nineteenth century through 1935, “[a]s a general rule, the state courts
refused to grant habeas corpus relief to persons imprisoned pursuant to an allegedly invalid
criminal conviction or sentence unless the conviction was void for lack of either personal or
subject matter jurisdiction, or the sentence was void for lack of jurisdiction”). In general,
courts began to utilize habeas corpus to redress constitutional errors in the 1930s, especially
after Johnson v. Zerbst, 304 U.S. 458 (1938), where the Supreme Court adopted the legal
fiction that, in a criminal case, constitutional violations deprive a trial court of jurisdiction.
YACKLE, supra note 70, § 5, at 15–16.
145.
YACKLE, supra note 70, § 3, at 5 (“The writ in theory has nothing to do with
the prisoner’s guilt or innocence but is concerned only with the process employed to justify
detention under attack.”); see also Ex parte Presnell, 49 P.2d 232, 234 (Okla. Crim. App.
1935) (“It has long since been established that the office of the writ of habeas corpus is not
to determine the guilt or innocence of the prisoner, and the only issue it presents is whether
or not the prisoner is restrained of his liberty by due process of law.”).
146.
Springstein v. Saunders, 164 N.W. 622, 623–24 (Iowa 1917); see also
Anderson v. Gladden, 383 P.2d 986, 991 (Or. 1963) (holding that “[a]s a general rule,
habeas corpus (or its statutory counterpart in post-conviction proceedings) does not provide
relief from a conviction resulting from a mistake of fact, where proof of the jury’s mistake
must depend on the credibility of newly discovered evidence”).
147.
See In re Weber, 523 P.2d 229, 243 (Cal. 1974) (holding that newly
discovered evidence fails to warrant habeas relief unless it thoroughly undermines the entire
structure of the prosecution’s case, and such evidence undermines the prosecution’s case

2005]

UP THE RIVER

675

should be mentioned. Unlike coram nobis applications, state habeas corpus
petitions were historically submitted to a court in the prisoner’s county of
confinement, and petitioners were required to be in custody.148 Moreover, the form
of relief upon a grant of a habeas corpus petition often consisted of absolute
release from custody with no provision for further proceedings against the
petitioner.149
Thus, by the late twentieth century, prisoners could try to utilize coram
nobis (and occasionally habeas corpus) to prove their innocence through newly
discovered evidence. Yet, in doing so, inmates had to overcome burdensome legal
and evidentiary requirements150 and, more abstractly, judicial opposition to the
very idea of employing collateral remedies to present newly discovered evidence
at all.151
B. Contemporary State Procedures Regarding Newly Discovered Evidence
In contemporary practice, newly discovered evidence claims filed by
defendants after trial surface in a mélange of direct and collateral remedies:
motions for a new trial, procedures created by statutes and court rules in the nature
of coram nobis, applications for common law coram nobis relief, and even habeas
corpus petitions. Through resort to either direct and/or collateral remedies,
prisoners convicted in each of the fifty states and the District of Columbia may
avail themselves of a post-trial procedure to put forth a claim of innocence based
on newly discovered non-DNA evidence.152 In practice, though, prisoners
encounter a daunting task considering that these remedies are marked by severe
procedural limitations.
1.

New Trial Motions Today

Each state currently permits a motion for a new trial on the basis
of newly discovered evidence, nominally providing any state prisoner who may be
factually innocent—and who has new evidence to substantiate that claim—with
recourse to a post-trial procedure in state court.153 Such motions, which Donald
only if it is conclusive and unerringly points to innocence); Ex parte Elizondo, 947 S.W.2d
202, 209 (Tex. Crim. App. 1996) (finding state habeas relief available in cases of actual
innocence where the petitioner shows “by clear and convincing evidence that no reasonable
juror would have convicted him in light of the new evidence” (emphasis added)); infra note
192 and accompanying text.
148.
See 1 WILKES, supra note 16, § 2-2, at 137.
149.
Id. § 1-8, at 38.
150.
See supra notes 139–42 and accompanying text.
151.
See supra notes 116–19 and accompanying text.
152.
See infra notes 153, 190–92 and accompanying text. Indeed, prisoners in
many jurisdictions have access to at least one post-trial procedure in order to present newly
discovered evidence of innocence that—be it a motion for a new trial or a post-conviction
remedy—has a relatively favorable statute of limitations or none whatsoever. See 1 WILKES,
supra note 16, § 1-12, at 51 (“There are in fact 38 states which now permit relief based on
newly discovered evidence of innocence, whether in a direct or a postconviction
proceeding, instituted more than three years after conviction.”).
153.
See 1 WILKES, supra note 16, § 1-13, at 55; Mitchell, supra note 13, at 464–
65.

676

ARIZONA LAW REVIEW

[VOL. 47:655

Wilkes dubs the “traditional and usual” means of presenting claims of innocence
conditioned on new evidence,154 are direct remedies that adhere to the historic
requirements mentioned above; they typically must be filed in the court of original
judgment155 within a short time frame.156 These features, then, dictate that
prisoners often must not only seek relief from the specific judge before whom they
were tried and convicted, but they must find evidence supporting their claim of
innocence soon after the curtain falls on their trial. And, in the event the original
trial judge denies a defendant’s request for a new trial based on newly discovered
evidence, state rules of appellate procedure tend to make it difficult for defendants
to challenge this decision on appeal, much less obtain a reversal.
First, many time limits governing motions for a new trial on the grounds
of newly discovered evidence are remarkably brief.157 As a result, these remedies
are of limited utility to the bulk of criminal defendants who, in the immediate
aftermath of their convictions, might not have the resources or the good fortune to
find new evidence. According to Wilkes, litigants in seventeen states must file
their motions within sixty days of judgment,158 and one state still follows the
common law rule that motions for a new trial can only be filed during the court
term in which the original judgment was entered.159 Seventeen other states and the
District of Columbia have time restrictions on new trial motions spanning from
one to three years.160 Of the remaining states, six allow new trial motions to be
filed beyond three years after conviction,161 with four of those jurisdictions
boasting waivable time limits of less than 120 days.162 A sparse number of
jurisdictions—nine—have no limitations period whatsoever.163 Also, many of the
154.
1 WILKES, supra note 16, § 1-13, at 55.
155.
See OR. REV. STAT. § 136.535 (West 2005) (making Rule 64(A) of the
Oregon Rules of Civil Procedure applicable to criminal actions); OR. R. CIV. P. 64(A) (“A
new trial is a re-examination of an issue of fact in the same court after judgment.”); supra
notes 74–75 and accompanying text; cf. GA. CODE ANN. § 5-5-43 (West 2004) (“A judge
who did not try the case may, if presented with a motion for new trial within 30 days from
the date of verdict or judgment sought to be set aside, allow the filing of, issue rule nisi
thereon, and decide the motion . . . .”).
156.
See supra notes 77–81 and accompanying text; infra notes 157–63 and
accompanying text.
157.
See, e.g., Thomas et al., supra note 11, at 279 (“Arkansas requires claims of
newly discovered evidence be filed within thirty days of sentencing. If that period expires,
the common law writ of coram nobis rule is available for four narrow categories of claims
that do not include newly discovered evidence.”).
158.
Herrera v. Collins, 506 U.S. 390, 410 (1993); 1 WILKES, supra note 16, § 113, at 56. Until recently, Virginia required defendants to present newly discovered evidence
within only twenty-one days of judgment. See Philip H. Yoon, Statute Note, Va. Code Ann.
§ 19.2-327.01, 15 CAP. DEF. J. 513, 513 (2003). In 2003, however, both the Virginia Senate
and the House of Delegates passed a bill extending the deadline to ninety days, and the law
became effective on July 1, 2004. Id.
159.
Herrera, 506 U.S. at 410; 1 WILKES, supra note 16, § 1-13, at 56.
160.
Herrera, 506 U.S. at 410; 1 WILKES, supra note 16, § 1-13, at 56.
161.
Herrera, 506 U.S. at 411; 1 WILKES, supra note 16, § 1-13, at 56.
162.
The remaining two of these six states have waivable time limits in excess of
120 days. Herrera, 506 U.S. at 411; 1 WILKES, supra note 16, § 1-13, at 56.
163.
Herrera, 506 U.S. at 411; 1 WILKES, supra note 16, § 1-13, at 56.

2005]

UP THE RIVER

677

states that impose time limits on new trial motions view those conditions as
jurisdictional in nature, barring trial courts from granting tardy motions even when
such motions are based on newly discovered evidence.164
Some states have carved out exceptions to general statutory time limits
when new trial motions involve newly discovered evidence.165 Similarly, if the
alleged new evidence could not have been discovered prior to the end of the
limitations period, several jurisdictions have held that the trial court has the
latitude to consider the motion.166 Where the purported new evidence is coupled
164.
See, e.g., Ex parte O’Leary, 438 So. 2d 1372, 1374 (Ala. 1983) (“Where a
motion for new trial filed within 30 days of a judgment of conviction does not contain a
ground relative to newly discovered evidence, a defendant is not in a position to make an
assertion regarding newly discovered evidence by motion for new trial after expiration of
the 30-day period, even though the new evidence could not have been discovered until after
that time period had elapsed.” (quoting O’Leary v. State, 417 So. 2d 214 (Ala. Crim. App.
1980))); Taylor v. United States, 759 A.2d 604, 609 (D.C. 2000) (“The time periods for
filing the new trial motion are jurisdictional; this court has no power to consider an untimely
new trial motion, even if the result seems harsh and it is difficult to see how the alleged
newly discovered evidence would have been available within the jurisdictional period.”);
Peterson v. State, 810 So. 2d 1095, 1100 (Fla. Dist. Ct. App. 2002) (holding that the trial
court lacked jurisdiction over a new trial motion grounded on newly discovered evidence
that was filed after the statutory deadline); State v. Reed, 712 So. 2d 572, 582–83 (La. Ct.
App. 1998) (deeming motion for new trial based on newly discovered evidence filed after
the one-year statute of limitations untimely and rejecting arguments that the time bar was
unconstitutional “as it is the legislative prerogative to set reasonable procedural formalities
and requirements such as the time limitations provided in [the pertinent state procedure]”);
cf. People v. Parker, 227 N.W.2d 775, 779 (Mich. 1975) (“A delayed motion for a new trial
can always be filed at any time on leave granted by the trial court.”).
165.
See, e.g., N.C. GEN. STAT. § 15A-1415(c) (2004) (providing that, irrespective
of the general time limits on a Motion for Appropriate Relief, such motions based on newly
discovered evidence need only be filed within a “reasonable time” after discovery of the
evidence); COLO. R. CRIM. P. 33(c) (“A motion for a new trial based upon newly discovered
evidence shall be filed as soon after entry of judgment as the facts supporting it become
known to the defendant, but if a review is pending the court may grant the motion only on
remand of the case. A motion for new trial other than on the ground of newly discovered
evidence shall be filed within fifteen days after verdict or finding of guilt or within such
additional time as the court may fix during the fifteen-day period.”); IDAHO CRIM. R. 34 (“A
motion for a new trial based upon the ground of newly discovered evidence may be made
only before or within two (2) years after final judgment. A motion for a new trial based on
any other ground may be made at any time within fourteen (14) days after verdict, finding
of guilt or imposition of sentence, or within such further time as the court may fix during
the fourteen (14) day period.”).
166.
See, e.g., Noffke v. State, 422 P.2d 102, 106–07 (Alaska 1967) (“There is
nothing in the record to show that appellant’s trial counsel had any knowledge, within the
time limits of Crim.R. 33 [rule governing new trial motions], of the fact that the trial judge
had given the jury this supplemental instruction. In such a situation we believe that it would
work an injustice to appellant to hold that he is now precluded from questioning the
propriety of the supplemental instruction.”); State v. Condon, 808 N.E.2d 912, 917 (Ohio
Ct. App. 2004) (granting leave to file a delayed motion for a new trial on the basis of newly
discovered evidence where the defendant presented clear and convincing proof that he was
unavoidably prevented from discovering the new evidence within 120 days of the jury’s
verdict).

678

ARIZONA LAW REVIEW

[VOL. 47:655

with a strong claim of innocence, some state courts gravitate toward entertaining
the motion even if untimely filed pursuant to procedural norms.167 A number of
states, however, have neglected to create exceptions to rigid time limits for new
evidence claims, and a few states openly distinguish between scientific and
nonscientific evidence in implementing statutes of limitations, treating the former
with greater leniency.168 Only at great cost, therefore, may state prisoners delay in
pursuing potential non-DNA evidentiary leads and submitting their innocence
claims to the trial court after conviction.
Second, rules requiring that new trial motions be filed in the court of
conviction owe their continued existence not merely to historical inertia wrought
from British custom, for the policy rationale behind them is an obvious one—
efficiency. After all, the trial judge presided over the case and ideally has retained
some memory of those prior proceedings.169 This knowledge, in turn, could inform
and assist the evaluation of any newly discovered evidence, particularly where
such evidence depends on witness credibility from the trial (namely, the
recantation context).170 Regardless of whether witness credibility lies at the crux of
a defendant’s claim, trial judges necessarily must review decisions from the
original proceedings when evaluating new trial motions based on newly discovered
evidence because the legal standard in most states for granting a new trial has
incorporated some offshoot of the Berry requirements: whether, among other
factors, the newly discovered evidence is noncumulative, does not simply impeach

167.
See Thomas et al., supra note 11, at 280 (“The trend is undoubtedly in the
direction of finding a basis to allow powerful claims of innocence to be heard even if filed
too late under the rules of procedure.”).
168.
For example, Connecticut, Illinois, Missouri, and Tennessee have time limits
on new trial motions involving nonscientific evidence, yet lack limits for motions based on
scientific testing that could not have been conducted at trial. Id. at 277–78 (describing time
limits for scientific and nonscientific evidence in Illinois, Missouri, and Tennessee); see
also CONN. GEN. STAT. ANN. § 52-582 (West 2005) (“No petition for a new trial in any civil
or criminal proceeding shall be brought but within three years next after the rendition of the
judgment or decree complained of, except that a petition based on DNA (deoxyribonucleic
acid) evidence that was not discoverable or available at the time of the original trial may be
brought at any time after the discovery or availability of such new evidence.”).
169.
See, e.g., Salinas v. State, 373 P.2d 512, 513 (Alaska 1962) (“[T]he trial
judge is in a better position to determine the possible effect and merit of the alleged newly
discovered evidence since he presided over the original trial and heard all the evidence
there.”); see also 58 AM. JUR. 2D New Trial § 369 (2004) (“A motion for a new trial is best
heard and decided by the judge who presided at the trial of the cause, since that judge’s
familiarity with the case better enables him or her to rule upon the questions which are
raised by the motion.”); YACKLE, supra note 70, § 8, at 34 (“Indeed, it was considered
preferable that the same judge who had rendered the judgment should receive the
application, since that judge was already familiar with the case and thus able to determine
quickly whether the facts alleged in the petition had been adjudicated at trial.”).
170.
See supra note 169 and accompanying text. Courts have historically frowned
upon recantation evidence. See supra note 13 and accompanying text; Yarborough v. State,
514 So. 2d 1215, 1220 (Miss. 1987) (“No form of proof is so unreliable as recanting
testimony . . . . Our skepticism does not translate into callousness, however.”).

2005]

UP THE RIVER

679

the prosecution’s witnesses, and would probably yield a different result in a new
trial.171
As will be discussed in more depth in Part III of this Article, there may be
good reason to reconsider the propriety of directing newly discovered evidence
claims to the original trial judge in spite of any potential inefficiency that might
arise.172 Having played a vital role in the initial decision, the trial judge arguably
has a vested interest in preserving that outcome173 and, as a matter of common
sense, human beings undoubtedly struggle to some extent when asked to review
their own work and examine it critically.174 Yet criminal defendants wary of filing
new trial motions with the original trial judge in their case have few available, let
alone viable, alternatives. If the trial judge is no longer in office, ordinarily her
successor is conferred with jurisdiction to handle the motion.175 Save those
circumstances, to avoid filing with the original trial judge the defendant is often
forced to assert the existence of a blatant conflict of interest or otherwise ask the
judge to recuse herself.176 This is no easy feat; the applicable rules and case law
generally demand parties seeking recusal to proffer specific factual bases for
challenging the judge’s ability to be fair and impartial, a requirement not often met
by bald allegations of a “vested interest in the outcome.”177

171.
See supra notes 82–92 and accompanying text.
172.
See infra notes 284–335 and accompanying text.
173.
See, e.g., Mazur, supra note 15, at 201, 207, 233 (urging North Carolina to
abandon the preference for allocating a Motion for Appropriate Relief—a catch-all
provision encompassing traditional new trial motions—based on newly discovered evidence
of innocence to the original trial judge).
174.
See Medwed, supra note 64, at 144 (analyzing the practice of trial
prosecutors being assigned post-conviction motions in cases they handled and positing that
“there is an inherent problem” in asking people to review their own work product); infra
notes 284–335 and accompanying text.
175.
See 58 AM. JUR. 2D New Trial § 369 (2004). As Eli Mazur points out, the
statutory preference for the original trial judge to handle post-trial motions can be quite
potent; in North Carolina, the relevant statute permits the original trial judge to entertain the
Motion for Appropriate Relief even after the expiration of her term. Mazur, supra note 15,
at 207, 230; see also N.C. GEN. STAT. § 15A-1413(b) (2004).
176.
See, e.g., UTAH R. CRIM. P. 29(c)(1)(A) (requiring that a motion to disqualify
a judge must be “accompanied by a certificate that the motion is filed in good faith and shall
be supported by an affidavit stating facts sufficient to show bias, prejudice or conflict of
interest”). While the terms “recusal” and “disqualification” originally had different
meanings, the concepts today are often used interchangeably. See Debra Lyn Bassett,
Judicial Disqualification in the Federal Appellate Courts, 87 IOWA L. REV. 1213, 1214 n.4
(2002).
177.
At common law, a direct financial interest was the only basis for invoking
recusal or disqualification. Bassett, supra note 176, at 1223; see also RICHARD E. FLAMM,
JUDICIAL DISQUALIFICATION: RECUSAL AND DISQUALIFICATION OF JUDGES 106 (1996) (“At
common law, however, a judge could not be disqualified because of bias, except insofar as
such bias might be inferred from the fact of his pecuniary interest in the cause.”). Over time,
the rules affecting recusal have expanded to address other forms of potential bias. See
MODEL CODE OF JUDICIAL CONDUCT Canon 3E(1)(a)–(b) (2000) (“A judge shall disqualify
himself or herself in a proceeding in which the judge’s impartiality might reasonably be
questioned, including but not limited to [situations where] the judge has a personal bias or

680

ARIZONA LAW REVIEW

[VOL. 47:655

Third, after a trial judge rejects a new trial motion based on newly
discovered evidence, state procedures limit the capacity of defendants to overturn
that decision on appeal. As an initial matter, denials of new trial motions are not
always appealable as of right; the defendant may have to petition the appellate
court for permission to appeal.178 Even if a state appellate court agrees to review
the denial of a new trial motion, the standard of review applied to that denial is
extraordinarily deferential—the defendant normally must prove the trial court
abused its discretion in rendering its decision179 or failed to exercise that
discretion.180 The deferential nature of the abuse of discretion standard reflects
(and reinforces) the belief that the trial judge is best able to weigh the new
prejudice [against] a party or a party’s lawyer, [has] personal knowledge of disputed
evidentiary facts[, or] has served as a lawyer [or] been a material witness [in the case].”). In
any event, states often require that parties seeking to disqualify a judge substantiate that
claim with ample facts. See, e.g., IND. R. POST-CONV. REM. 1(4)(b) (“Within ten [10] days
of filing a petition for post-conviction relief under this rule, the petitioner may request a
change of judge by filing an affidavit that the judge has a personal bias or prejudice against
the petitioner. The petitioner’s affidavit shall state the facts and the reasons for the belief
that such bias or prejudice exists, and shall be accompanied by a certificate from the
attorney of record that the attorney in good faith believes that the historical facts recited in
the affidavit are true. A change of judge shall be granted if the historical facts recited in the
affidavit support a rational inference of bias or prejudice.”). Trial judges also have a great
deal of autonomy in considering recusal requests—with any resulting decision rarely
overturned. See Leslie W. Abramson, Deciding Recusal Motions: Who Judges the Judges?,
28 VAL. U. L. REV. 543, 545 (1994) (“In the majority of states, the decision of whether to
grant or deny a motion to recuse is within the sound discretion of the challenged judge.”). In
some states, however, the recusal motion is forwarded to another judge for consideration.
See, e.g., UTAH R. CRIM. P. 29(c)(2) (“The judge against whom the motion and affidavit are
directed shall, without further hearing, enter an order granting the motion or certifying the
motion and affidavit to a reviewing judge.”). Moreover, the standard of review that applies
to a denial of a recusal motion is frequently quite deferential. See Abramson, supra note
177, at 556 (“Reversal of a trial judge’s decision not to recuse occurs if there is an abuse of
discretion.”).
178.
See, e.g., N.Y. CRIM. PROC. LAW §§ 450.15, 460.15 (McKinney 2004)
(providing procedures through which a defendant may seek permission to appeal the denial
of a post-trial newly discovered evidence claim).
179.
See Mills v. State, 786 So. 2d 547, 549 (Fla. 2001) (“Absent an abuse of
discretion, a trial court’s decision on a motion based on newly discovered evidence will not
be overturned on appeal.”); State v. Weaver, 554 N.W.2d 240, 244 (Iowa 1996) (observing
that the state Supreme Court reviews a trial court’s decision on a new trial motion on the
basis of newly discovered evidence under the abuse of discretion standard and finds abuse
“‘only when discretion is exercised on grounds clearly untenable or to an extent clearly
unreasonable’” (quoting Preferred Mktg. Assocs. Co. v. Hawkeye Nat’l Life Ins. Co., 452
N.W.2d 389, 393 (Iowa 1990) (internal citation omitted))); State v. Stukes, 571 S.E.2d 241,
244 (N.C. Ct. App. 2002) (“The decision of whether to grant a new trial in a criminal case
on the ground of newly discovered evidence is within the trial court’s discretion and is not
subject to review absent a showing of an abuse of discretion.”); DANIEL J. MEADOR ET AL.,
APPELLATE COURTS: STRUCTURES, FUNCTIONS, PROCESSES, AND PERSONNEL 207 (1994)
(noting that trial court rulings on new trial motions are treated with deference on appeal);
White, supra note 25, at 13 (stating that the abuse of discretion standard is the norm in
reviewing denials of new trial motions based on newly discovered evidence).
180.
See 58 AM. JUR. 2D New Trial § 429 (2004).

2005]

UP THE RIVER

681

evidence and thus deserves a large margin for error.181 Under this standard of
review, trial judges routinely enjoy vast leeway in determining not only whether to
grant a new trial, but also whether even to order an evidentiary hearing to explore
the character of the new evidence before issuing a final decision.182 What this
means for everyday courtroom combat is that new trial motions are often decided
on paper submissions alone, with the new evidence allegedly proving the inmate’s
innocence never developed in open court and therefore absent from the record on
appeal.183
2.

Current Modes of Collateral Relief

As previously noted, states have largely revamped their collateral, postconviction procedures in recent decades, codifying the available forms of relief by
statute or court rule.184 This shift from common law systems of state postconviction relief in favor of statute- and rule-based regimes is revealed by the fact
that only four states have yet to pass a statute or court rule creating a modern postconviction remedy in the nature of coram nobis or a modern version of habeas
corpus.185 Regarding the jurisdictions that have chosen a statute or court rule to
displace common law coram nobis or traditional habeas corpus as their chief form
of post-conviction relief, eight of those states have installed a remedy akin to
habeas corpus186 and thirty-eight have promulgated a remedy in the nature of
coram nobis.187 Though on the wane, the common law writ of coram nobis
remains as a secondary post-conviction remedy in eighteen jurisdictions.188 True to
181.
See, e.g., G. Fred Metos, Appellate Advocacy: Standards of Appellate
Review, CHAMPION, Dec. 1996, at 31, 32 (observing that the abuse of discretion standard “is
generally applicable to trial court decisions that involve balancing various interests or
factors. A trial judge is generally in a better position than an appellate court to make these
decisions.”).
182.
Some jurisdictions treat a trial court’s decision whether to hold an
evidentiary hearing on a new trial motion based on newly discovered evidence as a matter
of discretion. See, e.g., State v. Cossette, 856 A.2d 732, 738 (N.H. 2004) (“We hold that it
was well within the trial court’s discretion to deny the defendant’s request for a hearing on
the newly discovered evidence [alleged in a new trial motion].”); State v. Butler, No. 2003
CA 26, 2004 WL 869371, at *2 (Ohio Ct. App. Apr. 23, 2004) (stating that “the decision of
whether a hearing is warranted upon [a new trial motion grounded on newly discovered
evidence], also lies soundly within the discretion of the trial court”); see also 58 AM. JUR.
2D New Trial § 356 (2004) (noting that the decision to grant or deny a motion for a new trial
based on newly discovered evidence in a federal criminal case falls within the sound
discretion of the trial judge, as does the decision whether to hold an evidentiary hearing in
connection with the motion).
183.
See, e.g., 58 AM. JUR. 2D New Trial § 400 (2004) (noting that “a motion for a
new trial ordinarily may be decided upon affidavits without an evidentiary hearing”).
184.
See supra notes 126–32 and accompanying text.
185.
1 WILKES, supra note 16, § 2-5, at 163–64. According to Wilkes, in
California, Connecticut, New Hampshire, and Virginia, “the principal postconviction
remedy exists by authority of a traditional habeas corpus statute that has been modernized
principally by a process of liberal judicial construction.” Id. § 3-2, at 189.
186.
Id.
187.
Id.
188.
Id. § 3-5, at 232.

682

ARIZONA LAW REVIEW

[VOL. 47:655

the origins of coram nobis, state courts still refer to the writ as an extraordinary
remedy, a singular weapon in the judicial arsenal geared to tackle errors of fact
where no other form of redress exists.189
The degree to which claims of innocence hinging on newly discovered
evidence may be raised collaterally in state courts varies greatly at the moment,190
but on the whole, states seem more amenable to recognizing those claims than in
the past.191 Newly discovered evidence of innocence has emerged as an
appropriate basis for collateral relief in numerous jurisdictions, as evidenced by
Donald Wilkes’s conclusion that it represents a ground for relief through the
principal state post-conviction remedies in thirty-two states.192 Yet—as in the
189.
See, e.g., Cloird v. State, 76 S.W.3d 813, 815 (Ark. 2002) (“A writ of error
coram nobis is an extraordinarily rare remedy, more known for its denial than its approval.
The writ is allowed only under compelling circumstances to achieve justice and to address
errors of the most fundamental nature.” (internal citations omitted)); People v. Carty, 2 Cal.
Rptr. 3d 851, 854 (Cal. Ct. App. 2003) (“Importantly, the ‘purpose [of a petition] is to
secure relief, where no other remedy exists.’” (emphasis in original) (internal citations
omitted) (quoting People v. Adamson, 210 P.2d 13 (Cal. 1949))); Edwards v. State, 633
N.W.2d 623, 625 (S.D. 2001) (per curiam) (“The writ of coram nobis can only be used to
remedy a profound injustice where the petitioner has no other available remedy.”).
190.
See, e.g., Mulroy, supra note 7, at 12 (“Depending on the state, coram nobis
may be available under the common law, abolished by statute, supplanted by the statutory
provision of other remedies, or expressly adopted by statute.”). Some states permit newly
discovered evidence of innocence to be presented through state habeas corpus procedures.
See infra note 192 and accompanying text. There may even be other post-conviction
methods of presenting newly discovered evidence depending upon the jurisdiction. For
instance, Georgia allows newly discovered evidence claims to be raised collaterally through
an “extraordinary” motion for a new trial. See, e.g., Dick v. State, 287 S.E.2d 11, 13 (Ga.
1982) (noting that extraordinary motions are similar to ordinary motions for a new trial,
except that the former is filed after the deadline for submitting an ordinary motion and is
subject to stricter requirements when, for example, the motion is based on newly discovered
evidence).
191.
See supra notes 116–19 and accompanying text.
192.
See 1 WILKES, supra note 16, § 1-13, at 57–58. In twenty-four of those
states, the primary means for presenting newly discovered evidence collaterally is a
procedure in the nature of coram nobis. Id. § 1-13, at 57–58. The technique for putting forth
such a claim in the remaining eight states derives from a habeas corpus remedy. Id. at 57;
see also Summerville v. Warden, 641 A.2d 1356, 1369 (Conn. 1994) (noting that “a
substantial claim of actual innocence is cognizable by way of a petition for a writ of habeas
corpus, even in the absence of proof by the petitioner of an antecedent constitutional
violation that affected the result of his criminal trial”); Ryan Edward Shaw, Avoiding a
Manifest Injustice: Missouri Decides Not to Execute the “Actually Innocent”, 69 MO. L.
REV. 569, 577–80 (2004) (discussing state court decisions granting habeas corpus relief on
freestanding actual innocence claims); Laura Denvir Stith, A Contrast of State and Federal
Court Authority to Grant Habeas Relief, 38 VAL. U. L. REV. 421, 433–38 (2004) (same).
Many states retain the historic view that newly discovered evidence by itself is an
inappropriate basis for habeas corpus. See, e.g., Boyles v. Weber, 677 N.W.2d 531, 537
(S.D. 2004) (“Newly discovered evidence is generally an insufficient ground for habeas
relief when the evidence pertains to guilt rather than a deprivation of constitutional rights or
lack of jurisdiction.”). Furthermore, new evidence of innocence is cognizable through
common law coram nobis in seven of the eighteen states where that common law writ
lingers as a secondary post-conviction option. 1 WILKES, supra note 16, § 1-13, at 57.

2005]

UP THE RIVER

683

sphere of new trial motions193—the actual availability of newly discovered
evidence as a method to prove one’s innocence in state post-conviction
proceedings is tempered by the harsh reality inflicted by an array of procedural
obstacles.
Although post-conviction remedies have historically lacked statutes of
limitations,194 states have become increasingly willing to place time restrictions on
the use of collateral measures.195 This broad proposition is buttressed by the data;
in 1972, only three states (Illinois, New Jersey, and Wyoming) attached a time
limit to their principal post-conviction remedy whereas now thirty-one states have
such requirements.196 In terms of duration, the limitations periods operating in
those states run the spectrum from sixty days (Missouri) to ten years
(Maryland).197 Collectively, moreover, states that installed limitations periods as
part of their post-conviction regimes have displayed reluctance to lengthen the
time frames.198 Despite the fact that many states with statutes of limitations on
post-conviction relief provide exceptions for newly discovered evidence claims,
especially where the evidence could not have been found with due diligence prior
to the expiration of the period,199 not every state provides prisoners with such a
safety valve200 and some states only earmark DNA evidence for preferential
193.
See supra notes 153–83 and accompanying text.
194.
See supra note 104 and accompanying text.
195.
See infra notes 196–98 and accompanying text.
196.
1 WILKES, supra note 16, § 3-2, at 191.
197.
Id. § 3-2, at 192.
198.
Id. (“Since 1973 seven states with a statute of limitations applicable to the
principal postconviction remedy—Arkansas, Idaho, Illinois, Louisiana, Montana,
Tennessee, and Wyoming—have shortened the limitations period, whereas only two
states—Oregon and Utah—have lengthened it.”).
199.
See, e.g., id. §§ 1-13, 3-2, at 56, 193 (observing that “claims for relief based
on newly discovered evidence are often exempted from the limitations period applicable to
a particular state postconviction remedy”); Thomas et al., supra note 11, at 278–81
(mentioning a few states that afford litigants freedom from rigid time constraints in filing
post-conviction petitions based on newly discovered evidence); see also IOWA CODE ANN.
§ 822.3 (2004) (stating that the general three-year statute of limitations on post-conviction
petitions “does not apply to a ground of fact or law that could not have been raised within
the applicable time period”); FLA. R. CRIM. PROC. 3.850(b)(1) (providing for a two-year
statute of limitations on petitions for post-conviction relief unless, among other bases, “the
facts on which the claim is predicated were unknown to the movant or the movant’s
attorney and could not have been ascertained by the exercise of due diligence”). Also, some
states that impose time restrictions on newly discovered evidence allow the limitations
period to begin to run as of the date of discovery of the newfound facts. For example,
Utah’s Post-Conviction Remedies Act permits post-conviction challenges based on newly
discovered evidence to be filed within one year of the “date on which [the defendant] knew
or should have known, in the exercise of reasonable diligence, of evidentiary facts on which
the petition is based.” UTAH CODE ANN. § 78-35a-107(1), (2)(e) (2004); see also ALA. R.
CRIM. PROC. 32.2(c) (providing that, pursuant to the state post-conviction statute, a claim of
newly discovered evidence must be brought within six months after discovery of the
evidence); infra note 261 and accompanying text.
200.
See People v. Ambos, 51 P.3d 1070, 1073 (Colo. Ct. App. 2002) (noting that,
while newly discovered evidence is a basis for state post-conviction relief, it is subject to
the three-year statute of limitations); State v. Mixon, 983 S.W.2d 661, 670 (Tenn. 1999)

684

ARIZONA LAW REVIEW

[VOL. 47:655

treatment.201 The frequency with which states have fastened statutes of limitations
to their post-conviction procedures in recent decades also indicates the trend is
toward endorsing as opposed to abandoning time restrictions.202
In addition, prisoners seeking redress through state post-conviction
procedures may face forum-related dilemmas comparable to those in the area of
new trial motions, such as requirements that petitions go to the original trial
judge203 and onerous standards for obtaining recusal.204 As mentioned above,
historically the common law writ of error coram nobis was directed toward the
judge who presided over the case at trial.205 In converting from common law to
rule- or statute-based post-conviction regimes, many jurisdictions that crafted a
remedy in the nature of coram nobis as their principal mode of relief borrowed that

(enforcing the one-year time limit for newly discovered evidence claims under the state
post-conviction statute); Piar, supra note 26, at 632 (discussing Wyoming’s intractable fiveyear statute of limitations on post-conviction relief, and noting that he has found no cases
extending or excusing this limit); Thomas et al., supra note 11, at 279 (“A few states still
appear to have rigid deadlines for filing claims based on newly discovered evidence, with
either no exception for fairness or the interests of justice or one that is difficult to satisfy.”).
201.
Virginia, for instance, has strict time limits on filing state habeas petitions
but allows for issuance of a writ “of actual innocence” in cases involving biological or
scientific evidence, regardless of any deadlines. Thomas et al., supra note 11, at 278–79.
Florida also affords flexibility regarding its statute of limitations in post-conviction cases
involving DNA testing. Id. at 279 (“Though the Florida habeas statute does not seem to
have a loophole for good cause shown, the Florida Supreme Court has indicated that a
prisoner could request a DNA test even if the request is outside the two year statute of
limitations as long as the prisoner satisfied due diligence.”). Connecticut differentiates as
well between post-trial claims based on DNA and those based on other grounds; a threeyear time limit governs coram nobis relief and new trial motions in the state, except for
claims involving DNA evidence. See Piar, supra note 26, at 525.
202.
Daniel Piar has described the willingness of states such as Arkansas,
Connecticut, and Tennessee to put statutes of limitations on coram nobis relief as a
“disturbing trend.” See Piar, supra note 26, at 524–28.
203.
As with respect to new trial motions, this requirement usually stems from
statute or court rule. See, e.g., COLO. R. CRIM. P. 35(c)(3) (defendants seeking postconviction relief, among other grounds, on the basis of newly discovered evidence “may file
a motion in the court which imposed the sentence to vacate, set aside, or correct the
sentence”); DEL. SUPER. CT. CRIM. R. 61 (providing that a motion for post-conviction relief
is filed first in the office of the prothonotary in the county in which the conviction occurred
and, if the motion is facially sufficient, it shall be allocated to the superior court judge who
presided at the moving party’s trial); MASS. R. CRIM. P. 30(a) (directing post-conviction
relief motions to the “trial judge”).
204.
See, e.g., Patrick J. Quinn & John J. Hynes, Impact of Recent Decisions upon
Proceedings under the Post-Conviction Hearing Act, 34 LOY. U. CHI. L.J. 639, 642 (2003)
(noting how, regarding the principal post-conviction remedy in Illinois, “the same judge
who presided over the defendant’s trial should hear his post-conviction petition, unless it is
shown that the defendant would be substantially prejudiced. Thus, defendants do not have
an absolute right to substitution of a judge at a post-conviction proceeding, but rather they
must show that they will be substantially prejudiced if the motion for substitution is
denied”).
205.
See supra notes 100–01 and accompanying text.

2005]

UP THE RIVER

685

procedure, assigning petitions to the court of conviction.206 Even in the realm of
habeas corpus, some states have departed from the conventional practice of filing
requests for relief in the court having jurisdiction over the region in which the
prisoner is incarcerated.207 Specifically, in four of the eight states that permit
newly discovered evidence of innocence to be raised via habeas corpus—Nevada,
South Dakota, Texas, and Utah—the post-conviction petition is usually filed in the
convicting court.208
Finally, as for appellate review, denials of state post-conviction petitions
are often subject to standards analogous to those applied to new trial motions.209 At
common law, litigants were ordinarily not allowed to appeal rejections of habeas
corpus petitions,210 and appellate review of coram nobis applications was similarly
circumscribed.211 The creation of comprehensive state post-conviction regimes
modified the specific parameters surrounding the reviewability of denials of postconviction relief, but not the general tenet that lower courts warrant deference in
this domain of activity; although some states provide for appeals as of right from
denials of post-conviction petitions,212 others deem such review discretionary213
206.
See 1 WILKES, supra note 16, § 1-5, at 16–18; YACKLE, supra note 70, § 7, at
29–30. The fact that many states demand filing in the court of conviction or sentencing
court does not necessarily indicate that court clerks direct post-conviction petitions to the
original trial judge herself, although it appears that many do so. Notably, Vermont, whose
principal post-conviction remedy is in the nature of coram nobis, explicitly requires filing in
the superior court in the county of sentencing, but bars the original trial judge from hearing
the petition. 13 VT. STAT. ANN. § 7131 (2004); 2 WILKES, supra note 16, app. A at 684–85
(describing Vermont’s chief post-conviction remedy as a “remedy in the convicting court in
the nature of coram nobis”).
207.
See supra note 148 and accompanying text.
208.
See 1 WILKES, supra note 16, § 3-3, at 206–07. California permits state
petitions for writs of habeas corpus to be filed originally in a variety of courts: the superior
court, the state court of appeal, or the state supreme court. CAL. PENAL CODE § 1508(a)–(c)
(West 2005).
209.
See supra notes 178–83 and accompanying text.
210.
YACKLE, supra note 70, § 158, at 581 (“At common law, there was no right
of appeal from judgments in habeas corpus.”). Even today, Louisiana, for instance,
apparently does not permit appellate review of denials of state habeas corpus petitions. See
LA. CODE CRIM. PROC. ANN. art. 369 (West 2004).
211 .
YACKLE, supra note 70, § 8, at 35 (“In jurisdictions that considered coram
nobis proceedings to be independent of the original criminal case, trial court judgments
denying relief were generally reviewable on writ of error or appeal—though, again, the
scope of review was narrow. In states that considered applications for the writ to be part of
the criminal prosecution, appellate review was sometimes denied in the absence of a statute
expressly establishing the right.”).
212.
See, e.g., 22 OKLA. STAT. ANN. §§ 1051, 1087 (West 2005) (prescribing that
a defendant may appeal as of right any judgment against him and detailing the procedure for
appealing the judgment on a post-conviction petition to the Court of Criminal Appeals).
213.
See, e.g., CONN. GEN. STAT. ANN. § 52-470(b) (West 2005) (providing that a
denial of relief under the principal post-conviction remedy, habeas corpus, may only be
appealed if the losing party petitions the habeas judge or a judge of the state supreme court
or state appellate court to certify that the case involves a question worth reviewing, and the
judge certifies to that effect); ME. R. APP. P. 19(a) (mandating discretionary review of
denials of post-conviction petitions when criminal defendants are seeking to appeal); N.M.

686

ARIZONA LAW REVIEW

[VOL. 47:655

and still others seemingly balk at any appellate review.214 Significantly, where
denials of post-conviction relief are appealable, states tend to endorse the abuse of
discretion standard or a close facsimile of it in reviewing petitions grounded on
newly discovered evidence, sometimes even when the petition is dismissed
without an evidentiary hearing.215

III. REGIME CHANGE: PROPOSED MODIFICATIONS TO STATE
PROCEDURES
As discussed in Part II, state procedures in the area of newly discovered
non-DNA evidence bear similar traits to those of their British ancestors and, more
notably, contain strict procedural requirements. Prominent among the procedural
barriers facing inmates are statutes of limitations, restrictions on the forum in
which the motion must be filed, and rules regarding appellate review. Whereas the
restrictions vary from procedure to procedure, and from jurisdiction to jurisdiction,
the end result remains largely the same: after trial, convicted criminal defendants
seeking to prove their innocence through newly discovered non-DNA evidence
have trouble obtaining access to full-fledged evidentiary hearings in state courts.
In this Part of the Article, I urge states to revise their procedures to afford
prisoners the access needed to allow courts to make thorough evaluations of the
validity of post-conviction innocence claims. Simply put, it is not that states should
unlock the gates of their prisons, but rather that states should more readily open the
doors to their courthouses in the context of newly discovered non-DNA evidence
given the lessons learned by the DNA revolution. On a fundamental level,
waylaying prisoners’ attempts to receive evidentiary hearings on alleged newly
R. CRIM. P. 5-802(G)(2) (specifying that a final judgment denying relief under the principal
state post-conviction remedy is reviewable in the discretion of the New Mexico Supreme
Court upon timely filing of a writ of certiorari).
214.
See, e.g., In re Reed, 663 P.2d 216, 217 n.2 (Cal. 1983) (noting that statutory
law in California provides no appeal from the denial of a habeas corpus petition by a
superior court, and that the proper remedy would be to file a new habeas petition in the
court of appeal).
215.
Although some jurisdictions use different standards of review depending on
whether an evidentiary hearing occurred, see infra note 339 and accompanying text, the
abuse of discretion standard has been applied in assorted jurisdictions to summary denials of
newly discovered evidence claims as part of post-conviction petitions. See, e.g., People v.
Schulz, 774 N.Y.S.2d 165, 167 (App. Div. 2004) (holding that the trial court “providently
exercised its discretion in denying the defendant’s motion pursuant to CPL 440.10(1)(g)
[newly discovered evidence prong of post-trial remedy statute] without a hearing”); State v.
Apanovitch, 667 N.E.2d 1041, 1042–52 (Ohio Ct. App. 1995) (holding that the lower court
did not abuse its discretion in summarily dismissing petition for post-conviction relief
allegedly based partially on newly discovered evidence). States frequently apply the abuse
of discretion standard to post-hearing denials of collateral relief petitions. See, e.g., Kirby v.
State, 652 So. 2d 797, 798 (Ala. Crim. App. 1994) (stating that, in the context of an appeal
of a denial of post-conviction relief after the completion of an evidentiary hearing, “[t]he
standard applied by this court when reviewing the denial of a Rule 32 petition that alleges
newly discovered evidence is whether the court abused its discretion in denying the
petition”); Callier v. Warden, 901 P.2d 619, 628 (Nev. 1995) (finding that the trial court did
not abuse its discretion in denying a habeas corpus petition based on newly discovered
evidence after conducting an evidentiary hearing).

2005]

UP THE RIVER

687

discovered evidence of innocence by means of procedural obstacles clashes with
core concepts of justice as well as constitutional safeguards,216 and does nothing to
inspire confidence in the underlying correctness of the decisions rendered through
the criminal justice system.217
State courts, though, have historically looked askance at newly
discovered evidence claims,218 signaling that this traditional animus—and the
policies behind it—must be taken into account in molding any prospective
reforms. To that end, states should weigh several proposed modifications that
might achieve a satisfactory balance between finality and efficiency, on the one
hand, and justice for the actually innocent on the other. First, in order to curb the
risk of procedural default, states should abolish the use of statutes of limitations
and simplify their mechanisms for presenting newly discovered evidence claims by
adopting a single procedure in lieu of multiple ones. Second, defendants ought to
be able to submit their innocence claims to a judge other than the original trial
judge without making a showing equal to that required in a recusal motion. Third,
appellate courts should have the authority to review summary dismissals of newly
discovered evidence claims de novo. Any one of these reforms by itself would
improve the adjudication of post-conviction innocence claims in the courts and, if
employed together, these changes could go beyond improving the system and truly
foster the exoneration of innocent prisoners.
A. Decreasing the Risk of Procedural Default
Presupposing that potentially innocent prisoners with newly discovered
evidence at their disposal deserve the chance to have that evidence presented in
state courts, the question then becomes how to provide such access without unduly
harming finality and judicial economy.219 Granted, omitting all procedural
restrictions from new trial motions and post-conviction petitions to facilitate the
analysis of new evidence claims would almost surely hurt the very people
(wrongfully convicted prisoners) such a shift would be designed to help; the
ensuing flood of claims would diminish the likelihood that courts could

216.
Thomas et al., supra note 11, at 292 (“Limits on powerful claims of
innocence are not only prudentially unjustifiable but also violate the fundamental premises
of the Supreme Court’s own procedural due process jurisprudence.”); see also Seth F.
Kreimer & David Rudovsky, Double Helix, Double Bind: Factual Innocence and
Postconviction DNA Testing, 151 U. PA. L. REV. 547 (2002) (arguing for a constitutional
right to the disclosure of biological evidence for DNA testing in cases of actual innocence).
217.
It has been argued that the ability of the American criminal justice system to
reliably ascertain “true” guilt or innocence is central to its legitimacy. See, e.g., Christopher
A. Bracey, Truth and Legitimacy in the American Criminal Process, 90 J. CRIM. L. &
CRIMINOLOGY 691, 693 (2000) (reviewing WILLIAM PIZZI, TRIALS WITHOUT TRUTH (1999))
(“The American trial system is fundamentally ‘weak,’ according to Professor Pizzi, because
it privileges fairness norms at the expense of ‘truth.’”).
218.
See, e.g., Berry v. Georgia, 10 Ga. 511, 527 (1851).
219.
See, e.g., Kreimer & Rudovsky, supra note 216, at 561 n.50 (citing sources
arguing for the importance of finality in the criminal justice system).

688

ARIZONA LAW REVIEW

[VOL. 47:655

successfully differentiate the wheat from the chaff.220 Moreover, crime victims
have an interest in gaining some element of closure to their cases, and policies that
encourage prisoners to submit frivolous claims—or at least fail to discourage such
filings—risk needlessly reopening old wounds.221
Along with promoting finality, procedural restrictions arguably enhance
systemic efficiency in other ways. For instance, in certain jurisdictions, a properly
filed new trial motion may serve as a basis for a litigant to move to stay any direct
appeal of the judgment of conviction pending the outcome of the motion, and thus
possibly avoid an unnecessary appeal.222 Valuing finality and efficiency as assets
in the world of post-trial litigation, however, does not mean they should trump
competing policy objectives, particularly the accuracy of criminal adjudication.223
Ultimately, how should states accommodate, or even synthesize, these potentially
conflicting policies in their procedures?
Lessening the threat of procedural default in presenting claims of
innocence based on newly discovered evidence, without significantly altering the
substantive standards used in adjudicating them,224 would appear to reach a
delicate equilibrium between finality of judgments and individualized justice for
220.
See Thomas et al., supra note 11, at 294 (“If there were no way at some point
to impose a sentence with finality, prisoners would endlessly search for scraps of new
evidence and bombard the courts with petitions to reopen their cases.”).
221.
See, e.g., Kreimer & Rudovsky, supra note 216, at 606 (noting that the
Supreme Court has consistently proclaimed that “finality is essential to both the retributive
and deterrent functions of the criminal law and to the interests of victims of crimes in
obtaining closure”); Medwed, supra note 64, at 146 (noting that “alerting victims to a postconviction motion in a case is not a task cherished by prosecutors and, accordingly, one not
readily undertaken for fear of evoking painful memories”). For a discussion of postconviction innocence cases from the perspective of a victims’ rights advocate, see Dan S.
Levey, Wrongfully Convicted: A No-Win Situation for the Victim, 52 DRAKE L. REV. 695
(2004).
222.
See, e.g., KY. R. CRIM. P. 10.06(2) (“After a motion for a new trial is filed
and if there is an appeal pending, either party may move the appellate court for a stay of the
proceedings in the appellate court, whereupon the clerk of the appellate court shall notify
the clerk of the trial court that the motion has been filed.”). Some state post-conviction
regimes explicitly aim to consolidate collateral challenges with the direct review of a
conviction. See Swedlow, supra note 4, at 360 n.20 (“It is undisputed here that California is
a unitary review State, which is a State that allows prisoners to raise collateral challenges in
the course of direct review of the judgment, such that all claims may be raised in a single
state appeal.” (citing Calderon v. Ashmus, 523 U.S. 740, 743 n.1 (1998))). Moreover, the
shorter the statute of limitations period on the new trial motion, the greater the prospect that
the direct appeal will not yet have been decided.
223.
See, e.g., Piar, supra note 26, at 533 (“While the state has a legitimate
interest in the finality of criminal judgments, that interest cannot overcome a person’s right
to be permitted to challenge his conviction absent some culpable procedural default.”).
224.
This Article focuses on the procedural rules relating to the litigation of nonDNA claims of innocence rather than the legal and evidentiary standards used in assessing
those claims in court. As a result, I will refrain from commenting much on those standards,
except to note that several observers have suggested that the legal and evidentiary standards
used today should perhaps be less onerous. See generally supra note 32 and accompanying
text.

2005]

UP THE RIVER

689

defendants. To be more precise, easing the procedural requirements surrounding
entry to state courts on newly discovered evidence claims after trial while
simultaneously maintaining relatively high legal and evidentiary standards in
considering those claims ought to enhance the possibility that (a) actually innocent
prisoners will receive evidentiary hearings in newly discovered evidence cases and
(b) only valid claims of innocence eventually obtain relief.
As for the legal and evidentiary standards pertaining to new trial motions
and post-conviction petitions, nearly every jurisdiction requires that the defendant
prove (at a minimum) that the purported new evidence could not have been
discovered with due diligence at the time of trial, is neither merely cumulative nor
impeachment evidence, and would probably result in a different verdict if it were
received at trial—by no means effortless hurdles to clear.225 What is more, the
defendant customarily bears the burden of proof in making these assertions,226
which is appropriate considering the presumption of innocence has long since
vanished. The enforcement of these legal and evidentiary standards should provide
the requisite safeguard to prevent guilty prisoners from slipping through the
opening created by removing some procedural roadblocks that confront new
evidence claims.227 That is, even if a factually guilty defendant were to have his
new evidence claim assessed during a post-trial evidentiary hearing due to a
relaxation of procedural barriers, he would still likely struggle to carry the burden
of proving that all of the requirements for relief are met.228 And, because the
trademark brand of relief sought under most state procedures is a new trial, a
further backstop is embedded within the process to ensure that few guilty prisoners
are released—the subsequent trial.229 The combination of rather mild procedural
rules and stringent legal and evidentiary thresholds in the realm of newly
discovered non-DNA evidence claims would also promote transparency and lend
credibility to the judicial decisionmaking process.230

225.
See supra notes 82–92, 139–42 and accompanying text.
226.
See, e.g., Berry v. Georgia, 10 Ga. 511, 527 (1851) (“It is incumbent on a
party who asks for a new trial, on the ground of newly discovered evidence, to satisfy the
Court . . . .”); N.Y. CRIM. PROC. LAW § 440.30(6) (McKinney 2004) (providing that, on a
motion filed pursuant to Article 440, “the defendant has the burden of proving by a
preponderance of the evidence every fact essential to support the motion”).
227.
For obvious reasons, the fear that guilty prisoners may benefit is often raised
as a challenge to any proposed reforms designed to enhance the ability of potentially
innocent prisoners to gain their freedom. See, e.g., C.C. Horton II, Utah’s DNA Actual
Innocence Bill, UTAH B.J., Dec. 2001, at 12 (“What we wanted to achieve [in creating a
post-conviction DNA testing statute in Utah] was a balanced approach that would give the
truly innocent every opportunity to be exonerated, while not creating a mechanism which
would be abused by the guilty.”).
228.
See supra notes 225–27 and accompanying text.
229.
Of course, as the Kansas Supreme Court cautioned, in certain cases the
passage of time could lead to fading memories and missing prosecution witnesses, possibly
resulting in acquittals for some not necessarily innocent defendants. See Dobbs v. State, 65
P. 658, 660 (Kan. 1901).
230.
Augmenting state review of newly discovered evidence claims also affirms
principles of state sovereignty by limiting the likelihood of federal habeas corpus review.
See Muskat, supra note 13, at 162–65.

ARIZONA LAW REVIEW

690

[VOL. 47:655

In this vein, I would recommend two specific procedural reforms to state
regimes. For one, discarding statutes of limitations on new evidence claims would
reduce the risk of procedural default caused solely by lapse of time and
concomitantly increase the chance that viable innocence claims will be heard in
open court. Additionally, unifying the disparate methods of presenting new
evidence claims within any particular jurisdiction into a single remedy—namely,
fusing the new trial motion and collateral relief options—might minimize the
complexity facing prisoners, thereby decreasing the threat of procedural default,
and possibly even lower the overall administrative burden on state courts by
channeling newly discovered non-DNA evidence claims into one straightforward
procedure.
1.

Statutes of Limitations

Admittedly, prisoners in many jurisdictions may resort to at least one
post-trial procedure to present newly discovered evidence of innocence that—be it
a motion for a new trial or a post-conviction remedy—has no statute of limitations,
a relatively benign time restriction, or an exception to the general limitations
period for such claims.231 But not every jurisdiction has such favorable rules
regarding timing,232 and the tendency of states to attach limitations periods to posttrial remedies in recent years suggests many prisoners must grapple with time bars
in filing their innocence claims or will confront these stumbling blocks in the
future.233 Thus, the advantages and disadvantages of applying statutes of
limitations to newly discovered evidence claims must be explored.234
As Oliver Wendell Holmes once asked, “[W]hat is the justification for
depriving a man of his rights, a pure evil as far as it goes, in consequence of the
lapse of time?”235 In seeking to answer that question, proponents of statutes of
limitations generally cite several distinct policy interests. First, time limits
arguably promote finality and the right to “repose” by pinpointing a date at which
the parties know that any exposure stemming from the relevant incident has

231.
1 WILKES, supra note 16, § 1-12, at 51 (“There are in fact 38 states which
now permit relief based on newly discovered evidence of innocence, whether in a direct or a
postconviction proceeding, instituted more than three years after conviction.”); see, e.g.,
D.C. CODE ANN. § 22-4135 (2004) (allowing a person convicted of a crime to move for a
new trial or have a sentence vacated on grounds of actual innocence at any time, and
providing the court shall vacate the conviction if, after reviewing the new evidence, the
court concludes the defendant is actually innocent by clear and convincing evidence); IND.
R. POST-CONV. REM. 1(6) (imposing no statute of limitations in seeking post-conviction
relief on, among other grounds, newly discovered evidence).
232.
See supra notes 157–68, 195–202 and accompanying text.
233.
See supra note 202 and accompanying text.
234.
For discussions of the purposes and policies allegedly served by statutes of
limitations, see generally Ehud Guttel & Michael T. Novick, A New Approach to Old
Cases: Reconsidering Statutes of Limitations, 54 U. TORONTO L.J. 129 (2004); Tyler T.
Ochoa & Andrew J. Wistrich, The Puzzling Purposes of Statutes of Limitations, 28 PAC. L.J.
453 (1997).
235.
Oliver W. Holmes, Jr., The Path of the Law, 10 HARV. L. REV. 457, 476
(1897).

2005]

UP THE RIVER

691

ceased.236 Among the benefits of finality offered by backers of time limits are
gaining peace of mind, avoiding disturbing settled expectations, and reducing
uncertainty and the costs associated with uncertainty.237 A second basic policy aim
is to prevent the deterioration of evidence: to encourage the filing of claims while
evidence is fresh and witnesses are available, which aids the accuracy of the factfinding mission.238 Third, statutes of limitations help adversaries by putting them
on notice about the existence of a potential claim and, thus, assist them in
mounting a defense.239 Fourth, temporal restrictions on filing actions reinforce the
cultural value of diligence by encouraging claimants to act promptly.240 Finally,
advocates often suggest that limitations periods function to decrease the toll on the
court system by shrinking the aggregate volume of litigation.241
Nevertheless, in the context of newly discovered evidence of innocence,
many of the policies favoring statutes of limitations wither in the face of critical
examination. At its core, the enforcement of statutes of limitations that begin to
run as of the date of conviction242 fundamentally impedes the presentation of
236.
See, e.g., Ochoa & Wistrich, supra note 234, at 460–71 (discussing the
policy of “repose”). The concept of repose in criminal cases should be distinguished from
“statutes of repose” that state legislatures frequently pass to govern a significant amount of
civil tort litigation. That is, in the tort context, legislators may differentiate statutes of repose
from those of limitations. See, e.g., Stephen J. Werber, The Constitutional Dimension of a
National Products Liability Statute of Repose, 40 VILL. L. REV. 985, 989 (1995) (“A statute
of repose is a form of statute of limitations that runs from a point in time which disregards
the date of injury . . . a true statute of limitations begins to run when a person discovers, or
reasonably should have discovered, the existence of a claim through violation of a legal
right whereas a statute of repose runs from a fixed date or event.”). In contrast, time
restrictions in the area of criminal cases are typically designated as statutes of limitations,
even if they often resemble civil statutes of repose by commencing at a date certain, and the
term “repose” usually refers to a right of a party (most often a potential criminal defendant)
to have an end to the uncertainty of possible litigation. See Yair Listokin, Efficient Time
Bars: A New Rationale for the Existence of Statutes of Limitations in Criminal Law, 31 J.
LEGAL STUD. 99, 99 n.3 (2002).
237.
Ochoa & Wistrich, supra note 234, at 460–71.
238.
Id. at 471–83; see also James Herbie DiFonzo, In Praise of Statutes of
Limitations in Sex Offense Cases, 41 HOUS. L. REV. 1205, 1210 (2004) (“Time fades
memories, witnesses die or disappear, and documentation is destroyed or irretrievably
misplaced.”).
239.
See Ochoa & Wistrich, supra note 234, at 483–88.
240.
Id. at 488–92.
241.
Id. at 495–500. Ochoa and Wistrich cite two additional policies often
espoused by supporters of statutes of limitations: the goal of encouraging the prompt
enforcement of the law, and the need to avoid retrospective application of contemporary
standards. Id. at 492–95.
242.
In all but a few states, time limits on post-conviction relief generally start to
run upon completion of direct appellate review or when the time for submitting a direct
appeal has expired. See 1 WILKES, supra note 16, § 3-2, at 192–93. The starting point of the
statute of limitations periods applicable to new trial motions varies; some commence at the
date of verdict or sentencing, whereas others begin as of the final judgment. See, e.g., ALA.
R. CRIM. PROC. 24.1(b) (“A motion for a new trial must be filed no later than thirty (30)
days after sentence is pronounced.”); DEL. SUPER. CT. CRIM. R. 33 (providing that a
“motion for a new trial based on the ground of newly discovered evidence may be made

692

ARIZONA LAW REVIEW

[VOL. 47:655

innocence claims given that the new evidence may not even be discoverable prior
to the expiration of the limitations period.243 Resolving a dispute on a basis other
than the substance of the claim leaves litigants dissatisfied and also weakens the
legitimacy of the legal system, the chief purpose of which is to adjudicate
controversies on their merits.244
In addition to inhibiting the chance for prisoners to present meritorious
innocence claims at all, time restrictions fail to advance many of the policies
described above. As for the notion of “repose” in criminal cases,245 this goal is
effectively turned on its head where an innocence claim is rejected solely on
procedural grounds; any peace of mind or sense of certainty acquired by the
decision might be undermined by the uneasy feeling that substantive justice was
denied and that an innocent person may remain incarcerated while the actual
culprit is at large, free to commit other crimes. Moreover, the costs saved by
evading litigation through procedural default are passed on to the state prison
system in the form of the expense of continuing to house and feed a potentially
innocent prisoner.246 With respect to the other justifications for statutes of
limitations, those aims could still be achieved by a less severe procedural rule, one
that adequately juxtaposes the desire for finality against the opportunity for
prisoners to put forth their innocence claims. Specifically, in place of statutes of
limitations, states should wholly rely on a more equitable method of upholding
finality and spurring prisoners to proceed with haste: the doctrine of due or
reasonable diligence.
To elaborate, even if states do not expressly affix time limits to newly
discovered evidence claims, they typically oblige litigants to exercise due
diligence in pursuing their allegations after the new evidence is discovered—as

only before or within two years after final judgment”); FLA. R. CRIM. P. 3.590(a) (requiring
that a motion for a new trial be made within ten days of rendition of the verdict). Some
states have created a “discovery rule” for newly discovered evidence claims that allows the
statute of limitations to commence as of the date the new evidence is or should have been
discovered. See infra note 261 and accompanying text.
243.
See generally Piar, supra note 26 (arguing that rigid time limits fail to
recognize that complexities may arise in pursuing or uncovering extraordinary grounds for
relief).
244.
See Ochoa & Wistrich, supra note 234, at 500–01.
245.
See supra note 236 and accompanying text.
246.
The annual costs of housing a prisoner are exorbitant, both in terms of gross
expenditures and the lost opportunity to allocate these funds to other endeavors. See, e.g.,
Paul G. Cassell, Too Severe?: A Defense of the Federal Sentencing Guidelines (and a
Critique of Federal Mandatory Minimums), 56 STAN. L. REV. 1017, 1031 (2004)
(mentioning that, in an address to the American Bar Association, “Justice Kennedy noted
the monetary expense of incarceration, explaining that in California the cost per prison
inmate per year was about $26,000”); John M. Darley, On the Unlikely Prospect of
Reducing Crime Rates by Increasing the Severity of Prison Sentences, 13 J.L. & POL’Y 189,
191 (2005) (“The cost of a year in prison has been accurately estimated to be quite close to
a year in college. The funds for prisons come mainly from discretionary state budgets and,
thus, they compete with other possible draws on these budgets, such as medical and
educational payments for citizens.”).

2005]

UP THE RIVER

693

was the case with coram nobis at common law.247 Under the doctrine of due
diligence, litigants must present their claims shortly after discovery of the
evidence;248 analogues include laches or good faith pleading requirements.249
Whether cast in the rhetoric of laches, good faith, due diligence, or a comparable
theory grounded in equity, these standards demonstrate respect for the finality of
judgments by signaling to defendants that dilatory filing will not be excused.250 At
the same time, the use of these equitable benchmarks as opposed to “hard and fast”
timing rules accommodates the need for individualized justice by accepting the
fact-specific nature of new evidence claims; in particular, identifying the date of
discovery as the trigger point of any timing analysis fails to penalize defendants
for the vagaries of their post-trial investigation, an adverse byproduct of some
regimes at present.251 This reform would also alleviate the procedural due process
concerns generated by current practices in jurisdictions that enforce firm post-trial
time limits on claims that allege newly discovered evidence of innocence.252
Revamping state post-conviction regimes in this manner would not
necessarily displease supporters of time limits. Innocent prisoners would still have
incentives to pursue their claims conscientiously—for fear that tardy filings will
face dismissal from the courts—and to defend against evidence decay.253 Likewise,
the sanctity of the trial process and the primary role of the jury in resolving
247.
See supra note 106 and accompanying text. Regardless of whether a
jurisdiction has adopted a statute of limitations, states may forbid defendants from raising
such claims if the evidence could have been found with due diligence at the time of trial.
See, e.g., N.Y. CRIM. PROC. LAW § 440.10(1)(g) (McKinney 2004).
248.
See supra note 247 and accompanying text.
249.
Although traditionally the doctrine of laches only extended to postconviction proceedings in a few states, in jurisdictions where it is applicable today “laches
may operate to prevent an applicant for postconviction relief from obtaining relief, even
though the claim is meritorious, if the applicant slumbered on his or her rights and the delay
in applying for postconviction relief has prejudiced the government.” 1 WILKES, supra note
16, § 1-5, at 18; see, e.g., Sanders v. State, 733 N.E.2d 928, 930 (Ind. 2000) (noting that,
while there is no statute of limitations on the application for post-conviction relief under
Rule 1 of the Indiana Rules of Post-Conviction Remedies, the state may plead laches and
claim the petitioner unreasonably delayed filing the petition and that such delay prejudiced
the state); Brewer v. State, 446 N.W.2d 803, 805 (Iowa 1989) (mentioning that laches is
available to the state as an affirmative defense to a post-conviction relief petition).
250.
See Piar, supra note 26, at 529 n.224 (observing that laches or bad faith
pleading rules can be used to deal with the problem of petitioners being deliberately
dilatory).
251.
As noted above, some states start their statutes of limitations for postconviction relief claims at, for instance, the date of final judgment. See supra note 242 and
accompanying text. States that have such rules and lack “discovery” rules for newly
discovered evidence claims, see id., risk potentially harming defendants who are unable,
whether by virtue of insufficient funds and/or bad luck, to find the evidence within the time
frame.
252.
See Thomas et al., supra note 11, at 301 (“Our argument is, ultimately, a
simple one. Both a cost-benefit Mathews calculus and principles of fairness conclude that it
is unjust to reject powerful claims of innocence because of rules about when those claims
can be made.”).
253.
It should be noted that defendants naturally would have no incentive to
proceed rapidly for fear that the prosecution’s evidence might be lost.

694

ARIZONA LAW REVIEW

[VOL. 47:655

questions of factual guilt or innocence, oft-mentioned justifications for imposing
time limits specifically on new evidence claims,254 would remain intact
considering the relief sought is almost universally a new trial.255 If a new trial is
granted in a particular case, then juries and trial judges would have another chance,
this time equipped with the newfound evidence, to determine whether a conviction
is deserved. So as to guard against the possibility that the government will be
disadvantaged, i.e., cases where its own evidence is no longer available at the time
a defendant discovers and wishes to present new evidence,256 courts could consider
any undue prejudice to the prosecution in deciding whether to order a new trial.
To be sure, supplanting statutes of limitations altogether with due
diligence-type standards in the area of innocence claims based on newly
discovered non-DNA evidence would insert some procedural uncertainty into the
post-conviction mix. Prosecutors and defendants would not know precisely when a
particular claim might be rejected as untimely, and forcing courts to engage in a
facts and circumstances assessment of due diligence for each case would strain
judicial resources and damage finality to a degree.257 On the whole, courts might
also obtain additional filings in the absence of statutes of limitations given that a
certain number of prisoners would not be dissuaded from submitting their claims
through threat of an automatic procedural bar.258
It is my contention, however, that any modest rise in inefficiency
spawned by embracing the due diligence standard is appropriate to bear in
exchange for reaping the rewards of ensuring that viable claims of innocence
receive substantive evaluation. Ridding state post-conviction regimes of
unforgiving time limits on newly discovered evidence claims not only accrues to
the benefit of potentially innocent defendants, but also reflects well upon the
criminal justice system itself, sending a message that courts are willing and able to
look anew at legitimate claims of innocence and not hide behind a veil of

254.
See, e.g., CONNORS ET AL., supra note 2, at 28–29 (“The reason for limiting
the time to file appeals based on new evidence is to ensure the integrity of the trial process
and jury verdicts.”).
255.
See generally supra notes 71–215 and accompanying text.
256.
See supra note 119 and accompanying text.
257.
In 1999, the Tennessee Supreme Court cited finality concerns in upholding
the one-year statute of limitations period for newly discovered evidence claims filed under
the state post-conviction statute and determining that the period begins to run as of the date
of judgment. State v. Mixon, 983 S.W.2d 661, 670–71 (Tenn. 1999). Moreover, the court
noted that, whereas in the past it was appropriate for due diligence to comprise the time
limit on the writ of error coram nobis, “criminal procedure has drastically changed in the
past thirty years” and defendants have a plethora of other methods for raising their claims.
Id.
258.
Nonetheless, abandoning strict time limits could partially diminish the
amount of frivolous motions predicated on newly discovered evidence by removing the
inducement for prisoners to file a motion—regardless of the actual content of the
evidentiary claim—prior to the expiration of the limitations period for fear of otherwise
being barred.

2005]

UP THE RIVER

695

procedural default.259 Furthermore, to offset the uncertainty created by the
departure of statutory time limits, state courts would likely establish approximate,
de facto guidelines as to how long litigants have after discovery of new evidence to
file their claims.260 The implementation of somewhat flexible timing standards of
this nature, which would start to run at the date of discovery and be determined
judicially on a case-by-case basis, would advance the airing of bona fide innocence
claims yet still largely sustain the systemic goal of finality. Short of this reform,
simply pegging the statute of limitations period for new evidence claims to the
time of discovery or the time when the evidence should have been discovered as
opposed to the date of conviction—essentially tolling the time period during the
search for new evidence—would be a step in the right direction.261
2.

Simplifying Post-Trial Procedures

Along with abandoning rigid time limits, a second proposed reform
geared toward decreasing procedural default in newly discovered non-DNA
evidence cases relates to harmonizing—in effect, simplifying—the procedures
through which prisoners must raise their claims. As described in Part II of this
Article, newly discovered evidence of innocence may generally be presented
259.
See ABA 1978 STANDARDS, supra note 138, 22-2.4(a) (“A specific time
period as a statute of limitations to bar post-conviction review of criminal convictions is
unsound.”). As Ochoa and Wistrich point out:
[Permitting] litigants their “day in court” promotes the dignitary value of
the legal process. It is frustrating and demeaning not to be allowed to be
heard when a person believes that he or she possesses a valid complaint.
Creating such feelings of frustration and powerlessness causes
disaffection with the legal system, and possibly with the political system
as well.
Ochoa & Wistrich, supra note 234, at 501–02.
260.
For instance, in New York, which has no statute of limitations on newly
discovered evidence claims, the courts have given litigants an indication of what constitutes
a lack of due diligence through decisions suggesting that filing more than one year after
discovering the evidence may constitute a failure to exercise due diligence. See, e.g., People
v. Stuart, 509 N.Y.S.2d 824, 829 (App. Div. 1986).
261.
A number of states have endorsed this option. See, e.g., 15 ME. REV. STAT.
ANN. § 2128(5) (West 2005) (“A one-year period of limitation applies to initiating a petition
for post-conviction review seeking relief from a criminal judgment . . . . The limitation
period runs from the latest of the following: . . . . C. The date on which the factual predicate
of the claim or claims presented could have been discovered through the exercise of due
diligence.”); 42 PA. CONS. STAT. ANN. § 9545(b)(2) (West 2004) (requiring a request for
post-conviction relief based on new evidence, among other grounds, to be “filed within 60
days of the date the claim could have been presented”); S.C. CODE ANN. § 17-27-45(C)
(2004) (mandating that post-conviction applications based on newly discovered evidence be
filed within one year of actual discovery or from the date when such facts should have been
discovered through reasonable diligence). North Dakota has a hybrid new trial motion time
limitation, requiring that the motion be filed within thirty days after the discovery of the
facts and within three years after the verdict or finding of guilt. N.D. R. CRIM. P. 33(b); see
also 22 OKLA. STAT. ANN. § 953 (West 2005) (“A motion for a new trial on the ground of
newly discovered evidence may be made within three (3) months after such evidence is
discovered but no such motion may be filed more than one (1) year after judgment is
rendered . . . .”); supra note 199 and accompanying text.

696

ARIZONA LAW REVIEW

[VOL. 47:655

through either a direct and/or collateral remedy.262 While the presence of multiple
remedies at the state court level may seem desirable or at least better than the
alternatives, a single option or no remedy at all, the interrelationship between these
devices within any given jurisdiction can be perplexing. Consider the situation in
Tennessee.
Tennessee offers two mechanisms for presenting newly discovered,
nonscientific evidence of innocence, each of which has a statute of limitations.
Tennessee Rule of Criminal Procedure 33(b) allows defendants to move for a new
trial on the basis of newly discovered evidence within thirty days of judgment,263
after which a defendant’s sole outlet may lie with the Tennessee Post-Conviction
Relief Act.264 Pursuant to that Act, defendants may seek collateral relief due to
new evidence within one year of judgment.265 The only exceptions to this deadline
arise in cases involving scientific proof of innocence266 or capital matters
concerning the discovery of exculpatory evidence.267
Viewed in isolation, these procedures would appear to interact quite
seamlessly—a defendant unable to meet the new trial motion deadline can
conceivably take advantage of the post-conviction remedy. Yet Tennessee courts
have added a mysterious judicial gloss to the superficial symmetry of these
procedures by utilizing different legal standards for each of the remedies.268 To
obtain a new trial under the post-conviction relief statute, a defendant must show
that she is without fault in neglecting to put forth the purported new evidence at
the original trial and that such evidence “may have resulted” in a different outcome
had it been presented at the earlier proceeding.269 This language, “may have
resulted,” conflicts with the standard applicable to Rule 33 new trial motions in
Tennessee, where the defendant must prove “that the evidence will likely change
the result of the trial.”270 This dissimilar treatment between collateral petitions and
new trial motions, which Steven Mulroy terms “somewhat incongruous,”271 could
produce unwanted filing incentives for defendants, namely, inducements to bypass
the thirty-day limit for a new trial motion to profit from a more lenient legal
standard and submit a post-conviction petition before the one-year period
expires.272 The availability of “incongruous” procedures in the sphere of newly
262.
See generally supra notes 71–215 and accompanying text.
263.
TENN. R. CRIM. P. 33(b).
264.
TENN. CODE ANN. § 40-30-102 (2004).
265.
Id. The Tennessee courts have strictly construed the one-year statute of
limitations. See, e.g., State v. Mixon, 983 S.W.2d 661, 670–71 (Tenn. 1999).
266.
TENN. CODE ANN. § 40-30-102(b)(2) (2004); Piar, supra note 26, at 519
(mentioning the exception from the one-year deadline in instances of actual innocence
based on scientific testing).
267.
See Mulroy, supra note 7, at 6 (noting that the Tennessee Supreme Court has
ruled that the deadline can be extended indefinitely in regard to capital convictions where
the late discovery of exculpatory evidence was not the defendant’s fault).
268.
Id. at 10–19.
269.
Id. at 13–16.
270.
TENN. R. CRIM. P. 33. According to Mulroy, this equates to a probability or
preponderance standard commonly found in other states. Mulroy, supra note 7, at 16.
271.
Mulroy, supra note 7, at 16.
272.
Id. at 16–18.

2005]

UP THE RIVER

697

discovered evidence claims also enhances the risk of divergent outcomes for
factually comparable claims depending on the specific procedure used by a
litigant, a danger that undermines the legitimacy of the state court system.273
Not to overstate the implications of Tennessee’s post-trial regime, but the
unevenness of its precise procedures may cause problems beyond creating
detrimental filing incentives and/or inconsistent results: the existence of
nonintegrated, possibly contradictory, and likely confusing remedies for newly
discovered evidence claims within a particular state heightens the risk of
procedural default. Despite the fact that many states offer some form of a right to
counsel for indigent prisoners in state post-conviction proceedings, newly
discovered evidence claims are often filed by prisoners pro se,274 adding to the
chance of excessive procedural default in jurisdictions with multiple—and rather
baffling—post-trial remedies.
Accordingly, states might consider taking a unitary approach to newly
discovered non-DNA evidence claims, constructing a single remedy that fits
squarely within neither the new trial motion nor the coram nobis camp. New York
provides just such a procedure.275 Historically, New York recognized a morass of
post-trial remedies, among them, common law coram nobis and post-judgment
motions for a new trial on the basis of newly discovered evidence.276 Article 440 of
the Criminal Procedure Law was enacted in 1970, in the view of one commentator,
“collectively to embrace all extant non-appellate post-judgment remedies and
273.
Assuming that the truth-seeking function of the criminal justice system is
indeed crucial to its legitimacy, see supra note 217 and accompanying text, then procedures
that might yield different results for factually similar claims may be problematic.
274.
There is no constitutional right to counsel for collateral, post-conviction
proceedings. See Daniel Givelber, The Right to Counsel in Collateral, Post-Conviction
Proceedings, 58 MD. L. REV. 1393, 1393 (1999) (“Hornbook constitutional law tells us that
the state has no obligation to provide counsel to a defendant beyond his first appeal as of
right. The Supreme Court has rejected arguments that either the Due Process Clause or the
Equal Protection Clause require that the right to counsel apply to collateral, post-conviction
proceedings.”); Jonathan G. Neal, Note, “Critical Stage”: Extending the Right to Counsel
to the Motion for New Trial Phase, 45 WM. & MARY L. REV. 783 (2003) (contending that a
right to counsel should apply to motions for a new trial). Nevertheless, many states have
passed legislation creating such a right. See ALA. CODE § 15-12-23 (2004) (providing for
appointment and payment of counsel to represent indigent defendants applying for state
post-conviction relief); N.J. STAT. ANN. § 2A:158A-5 (West 2004) (prescribing that public
defenders may be appointed to represent defendants in post-conviction proceedings); 1
WILKES, supra note 16, § 3-2, at 196 (noting that thirty-one states “generally guarantee,
whether by statute, rule of court, or caselaw, the right to appointed counsel to indigents
applying for the state’s principal postconviction remedy,” and observing that only seven
states “refuse to provide, as a matter of right, free counsel to any and all indigent persons
applying for the state’s principal postconviction remedy”). The right to counsel in state
post-conviction proceedings, however, is often discretionary with the court. See, e.g., State
v. Parmar, 544 N.W.2d 102, 105 (Neb. 1996) (noting that the district court has discretion to
appoint counsel in a post-conviction relief matter, and a failure to appoint counsel is not
error absent abuse of discretion).
275.
N.Y. CRIM. PROC. LAW § 440.10(1)(g) (McKinney 2004).
276.
See Peter Preiser, Practice Commentaries, N.Y. CRIM. PROC. LAW § 440.10,
at 246–49 (McKinney 2004).

698

ARIZONA LAW REVIEW

[VOL. 47:655

motions to challenge the validity of a judgment of conviction.”277 Section
440.10(1)(g) of that Article addressed the issue of newly discovered evidence and
selected a solitary remedy for all such claims:
1. At any time after the entry of a judgment, the court in which it
was entered may, upon motion of the defendant, vacate such
judgment upon the ground that . . . (g) [n]ew evidence has been
discovered since the entry of a judgment based upon a verdict of
guilty after trial, which could not have been produced by the
defendant at the trial even with due diligence on his part and which
is of such character as to create a probability that had such evidence
been received at the trial the verdict would have been more
favorable to the defendant; provided that a motion based upon such
ground must be made with due diligence after the discovery of such
alleged new evidence[.]278

Although the New York State Legislature amended Article 440 in 1994 to
incorporate a specific provision governing requests for post-conviction DNA
testing,279 and in the process became the first state to enact such a statute,280
section 440.10(1)(g) continues to steer the resolution of newly discovered nonDNA evidence claims in the state. By harmonizing new trial motion and collateral
petition relief, section 440.10(1)(g) provides a single procedure whose foremost
virtue, besides its relative clarity, is the omission of any statute of limitations on
newly discovered evidence claims and, instead, espousal of the due diligence
standard. Undoubtedly, the New York post-conviction regime is not without its
flaws, as reflected by the account of the Stephen Schulz case in Part I,281 but by
277.
Id. at 247–48 (attributing this description to Judge Denzer, the principal
draftsperson of the Criminal Procedure Law and author of the original Practice
Commentary); see also YACKLE, supra note 70, § 10, at 12 (Supp. 2004) (“Under New York
law, ancient common law writ of coram nobis has been largely superseded by the motion to
vacate a judgment, and remains available as remedy only in those situations not explicitly
covered by N.Y. McKinney’s CPL § 440.10 which governs motions to vacate.”). State
habeas corpus in New York is not covered by Article 440 of the Criminal Procedure Law,
but rather by a provision of the Civil Practice Law and Rules. See N.Y. C.P.L.R. art. 70
(McKinney 2004).
278.
N.Y. CRIM. PROC. LAW § 440.10(1)(g) (McKinney 2004). If the purported
newly discovered evidence is presented to the trial judge during the period between the jury
verdict and sentencing, the defendant technically must submit the claim pursuant to a
different yet substantively comparable provision. See id. § 330.30.
279.
Id. § 440.30(1-a).
280.
See INNOCENCE PROJECT, NYSA § 440.30, NY COURT OF CLAIMS ACT SEC.8B, HB BILL A07003, http://www.innocenceproject.org/legislation/display_description.
php?id=a09250&sort=jurisdiction&filterStatus=all&filterJurisdiction=New%20York&filter
Category=all (last visited Aug. 19, 2005).
281.
See YACKLE, supra note 70, § 11, at 50–51 (criticizing certain provisions of
section 440.10(2) that provide for the mandatory denial of requests for relief, which thereby
“deprives the courts of New York of one last opportunity to reach the merits of federal
claims and correct errors that may otherwise open the judgment to attack in federal habeas
corpus”); id. § 11, at 60–61 (observing that “[t]he effectiveness of New York’s scheme . . .
is difficult to measure” and highlighting many of the advantages and disadvantages of the
regime); supra notes 34–60 and accompanying text.

2005]

UP THE RIVER

699

offering a unitary approach to post-trial claims of innocence based on newly
discovered evidence that snubs harsh time limits, it is a model worthy of
consideration and even emulation by other states.
Not incidentally, simplifying a jurisdiction’s newly discovered evidence
remedies could lighten the overall administrative load borne by state courts. That
is, in lieu of multiple prospective filing options, a prisoner might be limited to a
single remedy—and, at that, a remedy somewhat impervious to exploitation
through repeated use. As it stands, some states currently ban the submission of
second or “successive” claims under a particular post-conviction remedy,282 or at
least impose additional filing burdens to deter inmates wishing to do so.283 With
the availability of only one potential remedial measure that is not subject to rigid
time limits, prisoners would be prompted to consider carefully the optimal moment
to file a newly discovered evidence claim, and make a decision unencumbered by
artificial time constraints. Even so, to ensure that inmates with legitimate
innocence claims are not harmed by this streamlined approach, states that adopt a
single remedy should refrain from barring successive petitions outright; rather,
prisoners should be allowed to file additional claims in certain cases, e.g., upon
finding compelling “new” newly discovered evidence of innocence after a
previous application has been rejected as insufficient.
B. Full and Fair Hearings Before Unbiased Judges
Throughout this Article, I have repeatedly mentioned that many new trial
motions and post-conviction petitions premised on newly discovered non-DNA
evidence are directed to the trial judge who handled the case originally.284 As part
of these references, I have intimated my displeasure with such arrangements,
speculating that it is only human nature for people to struggle when asked to
reexamine their own work.285 Nonetheless, in the post-conviction and myriad other
areas, the criminal justice system condones judges engaging in this type of
analysis. For instance, neither codes of judicial conduct nor appellate cases
generally prohibit judges from ruling on the validity of search warrants that they
coincidentally had authorized, or ruling on motions to suppress evidence seized
pursuant to those warrants.286 Implicit and at times explicit in this system, then, is
the idea that judges can put aside any potential biases stemming from their role in
282.
Along with limiting the time frame in which defendants may file their postconviction petitions, some states have opted to curtail the number of petitions that any
individual litigant may file; Arkansas, Maryland, and Missouri, for instance, have each
amended their principal post-conviction remedy so as to disallow the filing of any more
than one application for relief. 1 WILKES, supra note 16, § 3-2, at 194.
283.
See, e.g., IND. R. POST-CONV. REM. 1(12)(a) (“A petitioner may request a
second, or successive, Petition for Post-Conviction Relief by completing a properly and
legibly completed Successive Post-Conviction Relief Rule 1 Petition Form in substantial
compliance with the form appended to this Rule. Both the Successive Post-Conviction
Relief Rule 1 Petition Form and the proposed successive petition for post-conviction relief
shall be sent to the Clerk of the Indiana Supreme Court, Indiana Court of Appeals, and Tax
Court.”).
284.
See supra notes 169–77, 203–08 and accompanying text.
285.
See supra notes 173–74 and accompanying text.
286.
See Abramson, supra note 177, at 544 n.4.

700

ARIZONA LAW REVIEW

[VOL. 47:655

the original outcome of a case and maintain impartiality.287 But recent scholarship
has shed light on a topic previously shrouded in darkness: the prevalence of
cognitive biases that might influence judicial decisionmaking.288
1. Selected Aspects of Behavioral Decisionmaking Theory
Scholars have studied an array of cognitive biases that may apply to the
manner in which most state courts currently resolve newly discovered non-DNA
evidence claims.289 Even if a judge has no objective stake in the outcome of a
case,290 as Dan Simon notes, she always owns a “stake of professionalism” in the
matter, “a professional interest in the soundness and effectiveness of the decision
rendered.”291 This professional interest in a judicial decision can manifest itself in
287.
In the words of William Blackstone, “the law will not suppose a possibility
of bias or favor in a judge, who is already sworn to administer impartial justice, and whose
authority greatly depends on that presumption and idea.” WILLIAM BLACKSTONE, 3
COMMENTARIES *361. Some courts have acknowledged the inherent potential for bias or the
perception thereof in situations where judges are asked, through a habeas corpus petition, to
review cases in which they presided over the state court proceedings. See, e.g., Clemmons
v. Wolfe, 377 F.3d 322, 329 (3d Cir. 2004) (“[We] now . . . require that each federal district
court judge in this circuit recuse himself or herself from participating in a 28 U.S.C. § 2254
habeas corpus petition of a defendant raising any issue concerning the trial or conviction
over which that judge presided in his or her former capacity as a state court judge.”).
288.
See, e.g., Bassett, supra note 176, at 1222 (“Recent psychological studies
suggest that unconscious bias is far more prevalent than originally believed.”); see also
Chris Guthrie et al., Inside the Judicial Mind, 86 CORNELL L. REV. 777 (2002) (analyzing
the impact of cognitive biases on the judicial decisionmaking process); Dan Simon, A
Psychological Model of Judicial Decision Making, 30 RUTGERS L.J. 1, 27–29 (1998)
(describing the study of “judicial behaviorism” as a movement that explores “the
relationship between judges’ personal predispositions and the decisions they make. This
approach is based on the claim that judicial decisions are affected to a large degree by extralegal or quasi-legal factors, namely, the judge’s general attitudes, values, and other socially
determined behavioral traits”).
289.
In a series of lectures that were later published in a book, Benjamin Cardozo
attempted to describe the factors and impulses that enter into judicial decisionmaking,
mentioning several modes of analysis to which judges resort in resolving disputes, including
comparison to precedent, references to history and customs, and sociological concerns, yet
observing that “[m]ore subtle are the forces so far beneath the surface that they cannot
reasonably be classified as other than subconscious. It is often through these subconscious
forces that judges are kept consistent with themselves, and inconsistent with one another.”
See BENJAMIN N. CARDOZO, THE NATURE OF THE JUDICIAL PROCESS 11–12 (1921). Cardozo
also noted that:
I have spoken of the forces of which judges avowedly avail to shape the
form and content of their judgments. Even these forces are seldom fully
in consciousness. They lie so near the surface, however, that their
existence and influence are not likely to be disclaimed . . . . The great
tides and currents which engulf the rest of men do not turn aside in their
course and pass the judges by.
Id. at 167–68.
290.
A financial interest is an example of an objective stake in a case and
historically served as a basis for judicial disqualification. See supra note 177 and
accompanying text.
291.
Simon, supra note 288, at 40–41.

2005]

UP THE RIVER

701

a variety of ways for the individual jurist, including an aversion to the prospect of
reversal on appeal.292 The desire to preserve the status of a decision in the face of
appellate review may reflect fears about being viewed as incompetent and/or
unworthy of advancement within the judicial hierarchy.293
A number of other potential biases may surface when judges evaluate the
merits of post-conviction innocence claims relating to trials over which they
previously presided. Behavioral psychologists have chronicled, for example, how
people often act in ways that allow them to preserve a “positive self-image.”294
Specifically, individuals may be hesitant to acquire information that undermines
the affirmative image they carry of themselves and, when confronted with
information that could negate that positive self-identification, people are less likely
to deem it relevant or may simply interpret it away.295 Intertwined with the concept
of preserving a positive self-image is the cognitive heuristic known as the
“egocentric bias,” a term that describes the proclivity of individuals to develop a
positive—and perhaps over-inflated—vision of their own abilities.296 With regard
to judges, most members of the bench may perceive themselves as fair individuals
who, in the main, render or oversee correct decisions. The potential effect of the
egocentric bias on judges emerged in a recent survey of 167 federal magistrate
judges that asked the respondents to gauge their individual reversal rates relative to
their co-equals in the study;297 nearly 90% of the judges believed they had lower
reversal rates than at least half of their colleagues.298
Researchers have also identified a phenomenon known as the “status quo
bias.”299 Proponents of this theory contend that people often have difficulty
292.
See, e.g., id. at 13 n.68.
293.
See, e.g., Mazur, supra note 15, at 232 n.274 (“There is a natural
disinclination to admit wrong, [because] people are concerned they will . . . diminish their
stature . . . .”).
294.
See Manuel Utset, A Theory of Self-Control Problems and Incomplete
Contracting: The Case of Shareholder Contracts, 2003 UTAH L. REV. 1329, 1368 n.143 (“A
positive self-image is something that individuals value, and self-confidence and optimism
play an important role in preserving and bolstering that self-image.” (citing Roy F.
Baumeister, The Self, in 1 THE HANDBOOK OF SOCIAL PSYCHOLOGY 680, 688–92 (Daniel T.
Gilbert et al. eds., 4th ed. 1998))).
295.
See Baumeister, supra note 294, at 690 (“Given the powerful motivation to
think well of oneself, it is necessary to ask how people manage to maintain such selfflattering views in the face of mixed and contrary evidence.”); Utset, supra note 294, at
1368 n.144 (“Economic actors may engage in such ‘strategic ignorance’ aimed at preserving
their current levels of self-confidence.”).
296.
See Guthrie et al., supra note 288, at 784; Mazur, supra note 15, at 231–32.
297 .
See Guthrie et al., supra note 288, at 784.
298.
Id. at 814.
299.
See, e.g., Daniel Kahneman et al., Anomalies: The Endowment Effect, Loss
Aversion, and Status Quo Bias, 5 J. ECON. PERSP. 193, 197–98 (1991) (“One implication of
loss aversion is that individuals have a strong tendency to remain at the status quo, because
the disadvantages of leaving it loom larger than advantages.”); Russell Korobkin, The Status
Quo Bias and Contract Default Rules, 83 CORNELL L. REV. 608, 625 (1998) (mentioning
empirical findings indicating “that people systematically favor maintaining a state of affairs
that they perceive as being the status quo rather than switching to an alternative state, all
else being equal”). As Korobkin notes, the term “status quo bias” is often used

702

ARIZONA LAW REVIEW

[VOL. 47:655

deviating from a prior decision because that decision has become the reference
point to which they compare and contrast newfound information.300 In effect, they
ascribe significant and likely undue weight to that reference point and, in order to
compel the decisionmaker to spurn that earlier decision and embrace a different
solution, much more information (“evidence”) is required than had that previous
decision never been made.301 To put it more bluntly, even if the information
objectively suggests that Option B is the best route, the decisionmaker will be less
likely to choose that option if she had selected Option A before than if she had
never previously made a determination in the matter at all.302
The status quo bias has widespread implications for judicial
decisionmaking—and not solely for judges asked to encounter their decision a
second time. Studies suggest that, when given a chance to revisit a decision, a
person may be predisposed toward the outcome of the earlier decision in the matter
regardless of whether she was responsible for that decision in the first instance, a
concept that has been referred to as “conformity effects.”303 Pursuant to this
theory, individuals are profoundly influenced by others and may lean toward
acting in conformity with those whom they admire.304 In particular, people are
often swayed by the decisions of others deemed to have access to greater—and
possibly “better”—information or who are thought to enjoy such access.305
Research also indicates that conformity effects are amplified in situations where

interchangeably with a phenomenon known as the “endowment effect.” Id. at 626 n.58; cf.
Matthew Rabin, Psychology and Economics, 36 J. ECON. LIT. 11, 14 (1998) (treating the
endowment effect and the status quo bias as related yet distinguishable ideas, and describing
the former as “[o]nce a person comes to possess a good, she immediately values it more
than before she possessed it”).
300.
See, e.g., Amos Tversky & Daniel Kahneman, Loss Aversion in Riskless
Choice, in CHOICES, VALUES, AND FRAMES 146 (Daniel Kahneman & Amos Tversky eds.,
2000) (commenting that “a decision maker who is indifferent between x and y from t will
prefer x over y from x, and y over x from y,” and crediting William Samuelson and Richard
Zeckhauser with introducing “the term ‘status quo bias’ for this effect of reference
position”).
301.
See supra notes 299–300 and accompanying text.
302.
A study involving health plan options available to employees at Harvard
University is illustrative. Samuelson and Zeckhauser discerned that, after the school added a
number of new health care options, a larger proportion of faculty members who had been
hired prior to the creation of the new options retained their former plans and rejected the
new options in comparison with faculty members hired after the availability of the new
options. See William Samuelson & Richard Zeckhauser, Status Quo Bias in Decision
Making, 1 J. RISK & UNCERTAINTY 7, 26–31 (1988); see also Colin F. Camerer, Prospect
Theory in the Wild: Evidence from the Field, in CHOICES, VALUES, AND FRAMES, supra note
300, at 294–95 (describing the Harvard health plan study and other empirical data related to
the status quo bias).
303.
See, e.g., Susan Burkhardt, The Contours of Conformity: Behavioral
Decision Theory and the Pitfalls of the 2002 Reforms of Immigration Procedures, 19 GEO.
IMMIGR. L.J. 35, 54–56 (2004).
304.
Id. at 56.
305.
Id.

2005]

UP THE RIVER

703

the decisionmakers are colleagues or self-identify as members of the same
group.306
2. Implications of Behavioral Decisionmaking Theory for New Evidence
Claims
The above survey of behavioral decisionmaking theory, albeit brief,307 has
significance with respect to how judges approach post-trial claims of innocence
predicated on newly discovered evidence in cases over which they presided
originally. To begin with, professional interests in maintaining the result of the
initial trial may affect how judges react to the new trial motion or post-conviction
petition when it comes across the transom; judges may fear having “gotten it
wrong” and feel cowed by the ramifications of acknowledging that error publicly.
The result sought—a new trial—could further expose a judge’s evidentiary rulings
and other decisions at the original trial to scrutiny from both the public and
colleagues within the judiciary.
Similarly, a particular judge’s positive self-image could be threatened by
the presentation of newly discovered evidence that, in theory, signals the
possibility an innocent person was convicted at a trial held on that judge’s watch
and during which she issued evidentiary and legal rulings that bore upon the
result.308 In the face of this potential threat to one’s self-image, a judge might
unconsciously dismiss the alleged newfound information as irrelevant or otherwise
characterize it as not outcome-determinative. Indeed, the authors of the
aforementioned study of federal magistrate judges expressed grave concerns about
the impact of egocentric biases in criminal cases, suggesting that judges may fail to
set aside judgments as often as they should in cases they originally handled.309
Although the status quo bias has been explored most often in the area of
economic choices, the concept inevitably surfaces in post-conviction litigation
involving newly discovered evidence.310 The fact that there has already been a
306.
Id.
307.
Indeed, an exhaustive study of the literature on behavioral decisionmaking
theory and behavioral law and economics far exceeds the scope of this Article. For further
information about the field, see generally Christine Jolls et al., A Behavioral Approach to
Law and Economics, 50 STAN. L. REV. 1471 (1998); Rabin, supra note 299.
308.
See Mazur, supra note 15, at 232 (“The trial judge is psychologically
invested in the justness of the trial result; for instance, in a criminal case the trial judge
likely denied a motion to dismiss and a motion for a directed verdict based on the
insufficiency of the evidence.”).
309.
Guthrie et al., supra note 288, at 815 (noting that these biases “might lead the
judge to react too skeptically to the suggestion that the trial over which he presided
produced an erroneous result”).
310.
Not only has this bias typically been observed in an economic setting, but
much of the empirical work has been done in the area of consumer goods. See Korobkin,
supra note 299, at 629. Nevertheless, the status quo bias pervades many aspects of everyday
life. For instance, the National Football League’s instant replay system exemplifies this
bias. Shane Frederick, Automated Choice Heuristics, in HEURISTICS AND BIASES: THE
PSYCHOLOGY OF INTUITIVE JUDGMENT 555 (Thomas Gilovich et al. eds., 2002) (“When the
‘on-field’ call is challenged, it is reviewed by the replay official, who has access to multiple
angles and slow motion photography. However, despite these advantages, the on-field call is

704

ARIZONA LAW REVIEW

[VOL. 47:655

verdict—guilty—becomes the reference point to which a judge may look in
assessing the new evidence; it may require much greater evidence to prod the
judge to move away from that reference point than had no verdict ever been
rendered and, say, the judge were merely evaluating a motion to dismiss at the
close of the prosecution’s case at trial. Even more, the possible effect of the status
quo bias is magnified by the convention of assigning newly discovered evidence
claims to the actual judge who handled the case at trial. Not only has “guilty”
become the reference point, but the judge witnessed the process through which
guilt was determined, presumably making her more deeply attached to the status
quo than a judge lacking any prior connection to the case.311
What this discussion of behavioral decisionmaking theory means for state
post-conviction regimes is that there are legitimate and compelling reasons for
rethinking the propriety of directing newly discovered evidence claims of
innocence to the original trial judge. Frankly, allocating these claims to different
judges is a “second-best” solution. At some level, efficiency considerations
militate in favor of letting judges entertain such motions in cases they handled at
trial; they observed the trial witnesses’ testimony and demeanor, received the
evidence, and theoretically stand in the best position to assess whether the
purported new evidence warrants a new trial.312 Moreover, the concept of
conformity effects raises doubt as to whether assigning newly discovered evidence
claims to a different judge will truly improve the likelihood of a correct decision.
That is, the new judge may be prone toward acting in conformity with the original
trial judge—a colleague on the bench who had extensive access to the factual
background of the case during the initial proceedings.313 Yet the justifications for
permitting the original trial judge to receive newly discovered non-DNA evidence
claims rest on flawed assumptions.314 As indicated above, studies allude to the
possibility that biases infect the decisionmaking processes of many judges, biases
that may be accentuated when a judge reviews a case in which she previously
given preference in any ‘close calls’; the replay official is instructed to defer to the on-field
call unless he has ‘indisputable visual evidence.’ Although there are reasons to minimize
challenges of on-field calls (such as maintaining the flow of the game), once a ruling has
been challenged and is going to be reviewed anyway, it makes little sense to give more
weight to the first call when the second can be made with better information.”). One’s
assessment of a particular situation may also depend largely on that individual’s personal
experiences instead of any objective, absolute evaluation of the situation. See, e.g., Rabin,
supra note 299, at 13 (mentioning the common occurrence in which “the same temperature
that feels cold when we are adapted to hot temperatures may appear hot when we are
adapted to cold temperatures”).
311.
The status quo bias likely has profound implications for a trial judge looking
at a previous decision anew. Even so, a judge may be more willing to excuse an incorrect
trial outcome when the challenge is based on newly discovered evidence rather than other
grounds in the sense that a judge could always rationalize the previous decision as having
been based on incomplete information.
312.
The question of whether the original trial judge (or any trial judge for that
matter) is genuinely in the best position to assess a newly discovered evidence claim may
hinge on whether that judge orders an evidentiary hearing on the claim. See infra notes 336–
72 and accompanying text.
313.
See supra notes 303–06 and accompanying text.
314.
See supra notes 74, 110 and accompanying text.

2005]

UP THE RIVER

705

played a role.315 The presence of potential biases undercuts the notion that the
original trial judge can view newly discovered evidence claims with the requisite
detachment; as the Third Circuit recently proclaimed, “[T]he passage of time
cannot overcome a reasonable person’s doubts about a judge’s impartiality in
judging his or her own past works.”316
Furthermore, sending post-trial innocence claims to a judge different from
the one who heard the case originally may serve to mitigate any unwelcome
political pressures on the resolution of those claims in jurisdictions where judges
are elected by the public.317 In many states, judges ascend to the bench by dint of
popular elections,318 and for those judges, political considerations may inexorably
(and regrettably) enter into the post-conviction decisionmaking equation.319 The
impact of political pressure on judges may be particularly acute when such
pressure is “direct”: where the judge who handled the trial initially must address
the case once again in the form of a post-conviction innocence claim and thus

315.
See supra notes 288–302 and accompanying text.
316.
Clemmons v. Wolfe, 377 F.3d 322, 327–28 (3d Cir. 2004). The fact that a
judge in certain circumstances may have a limited recollection of a case—and thus be
“emotionally removed”—might minimize any potential bias and allow her to offer a
relatively fresh perspective. See Muskat, supra note 13, at 169.
317.
See generally Mazur, supra note 15, at 233 (noting the effect of electoral
pressure on judges in evaluating post-trial motions); Steven Zeidman, To Elect or Not to
Elect: A Case Study of Judicial Selection in New York City 1977-2002, 37 U. MICH. J.L.
REFORM 791 (2004) (discussing the pros and cons of electing judges).
318.
See, e.g., Michael R. Dimino, Pay No Attention to that Man behind the Robe:
Judicial Elections, the First Amendment, and Judges as Politicians, 21 YALE L. & POL’Y
REV. 301, 310 (2003) (“At present, a clear majority of states—thirty-nine—elect some or all
of their judges.”); see also Clive S. Thomas et al., Interest Groups and State Court
Elections: A New Era and Its Challenges, in JUDICIAL POLITICS: READINGS FROM
JUDICATURE 53, 53 (Elliot E. Slotnick ed., 3d ed. 2005) (stating that “judges seeking
election, reelection, or reconfirmation are subject to codes of conduct that restrict what they
can say in campaigns, but there are few restrictions on interest groups”); Adam Liptak,
Judicial Races in Several States Become Partisan Battlegrounds, N.Y. TIMES, Oct. 24,
2004, at 1 (describing the recent rise in campaign spending and television advertisements,
fueled by interest groups, in judicial races).
319.
See John Blume & Theodore Eisenberg, Judicial Politics, Death Penalty
Appeals, and Case Selection: An Empirical Study, 72 S. CAL. L. REV. 465 (1999)
(comparing the relative willingness of partisan-elected appellate judges with other judges to
uphold capital sentences); Stephen B. Bright, Elected Judges and the Death Penalty in
Texas: Why Full Habeas Corpus Review by Independent Federal Judges is Indispensable to
Protecting Constitutional Rights, 78 TEX. L. REV. 1805, 1826 (2000) (“Once in office, any
vote that might be perceived as ‘soft on crime’ or as delaying executions—no matter how
clear the law requiring it—carries with it the risk that the judge will be voted out of office in
the next election.”); Richard R.W. Brooks & Stephen Raphael, Life Terms or Death
Sentences: The Uneasy Relationship Between Judicial Elections and Capital Punishment,
92 J. CRIM. L. & CRIMINOLOGY 609, 638 (2002) (noting that defendants in Chicago from
1870–1930 were 15% more likely to be sentenced to death in a judicial election year); cf.
Michael Heise, Mercy by the Numbers: An Empirical Analysis of Clemency and its
Structure, 89 VA. L. REV. 239, 288–97 (2003) (suggesting that political considerations did
not surface as significant factors in an empirical study of clemency decisions).

706

ARIZONA LAW REVIEW

[VOL. 47:655

experience renewed public inspection about the case within her district.320 As a
result, in states where judges are elected rather than appointed,321 sending a posttrial motion to a judge outside the district—or at least a different judge within the
district—might obviate to some extent any direct political demands on the judge to
affirm the conviction.322
Given the possibility of cognitive biases and political pressures impairing
judicial decisionmaking, as well as the high stakes involved in resolving innocence
claims, states should contemplate allowing defendants to file post-trial motions
based on newly discovered evidence with a judge other than the one who
conducted the original trial. This prompts the crucial question of who should
properly field those motions, and two alternatives seem most promising upon
reflection. First, newly discovered evidence claims could be randomly assigned to
another trial judge in the county or district of conviction.323 Due to conformity
effects, a new judge may be loathe to order a new trial in a case over which a peer
within the same jurisdiction presided,324 but she likely is less susceptible to many
of the biases described above than the original trial judge. In particular, she would
own a limited professional stake in the outcome of the newly discovered evidence
claim considering she had no hand in the prior proceeding. Likewise, any positive
self-image to which she may feel obliged to cling would remain unaffected by her
decision on the post-trial motion, and the status quo bias would probably be less
pronounced than if the original trial judge were handling the claim. Another
possible virtue of this proposal is that it preserves a modicum of efficiency; the
office and even the attorneys who prosecuted the case in the specific county could
stay involved in the post-trial matter if that happens to be the custom within the
jurisdiction.325
320.
See Mazur, supra note 15, at 233.
321.
A thorough analysis of the relative merits of electing versus appointing
judges surpasses the scope of this Article. For a more detailed analysis of the processes
surrounding how judges rise to the bench, see Dimino, supra note 318; Deborah Goldberg,
Public Funding of Judicial Elections: The Roles of Judges and the Rules of Campaign
Finance, 64 OHIO ST. L.J. 95 (2003); Zeidman, supra note 317.
322.
See James S. Liebman et al., Capital Attrition: Error Rates in Capital Cases,
1973-1995, 78 TEX. L. REV. 1839 (2000) (summarizing the chief findings from a study on
errors and the death penalty); Mazur, supra note 15, at 233–34 (“Using a judge from a
separate judicial district [in North Carolina] will remove the ‘direct electoral pressure’ that
Liebman’s study revealed as a ‘root cause’ of increased judicial error rates.”).
323.
See, e.g., 13 VT. STAT. ANN. § 7131 (2004) (compelling filing in the superior
court in the county of sentencing but banning the original trial judge from hearing the
petition).
324.
See supra notes 303–06, 313 and accompanying text. The term “structural
bias” is occasionally used to characterize the potential bias exhibited by individuals in favor
of colleagues within an organization or group, such as how directors make decisions
affecting other members of a corporate board of directors. See, e.g., Biondi v. Scrushy, 820
A.2d 1148, 1164 n.40 (Del. Ch. 2003) (describing “the danger that the difficult-to-detect
influence of fellow-feeling among directors (i.e., so-called ‘structural bias’)” may impose
on special litigation committees within corporations).
325.
The involvement of the original trial prosecutors in post-conviction
innocence cases, however, raises a number of other issues in its own right. See Bruce A.
Green & Fred C. Zacharias, Prosecutor Neutrality, 2004 WIS. L. REV. 837 (describing the

2005]

UP THE RIVER

707

Second, as is the practice for many habeas corpus petitions, newly
discovered evidence claims could be filed in the county of confinement as opposed
to the county of conviction.326 The advantages of this change include not merely
removing the original trial judge from any adjudicatory role in the post-trial
proceeding, but also blunting any conformity effects that may derive from the
assignment of the motion to a judicial cohort within the same county or district.327
As a practical matter, though, it may be burdensome to compel prosecutors from
the county of conviction to prepare for and attend evidentiary hearings in what are
often far-flung locales and a likely alternative measure, allocating the case to
prosecutors in the county of confinement, would impose even greater demands on
efficiency by forcing the newly-assigned attorneys to learn the case from
scratch.328 More significantly, many state prisons are located in sparsely populated
counties where the courts are already inundated with litigation stemming from
intraprison crimes and disputes.329 Even if clustering post-conviction innocence
claims within particular courts could build up expertise among those judges in
evaluating newly discovered evidence, foisting another set of cases on these
counties would further strain their court systems.330 Finally, selecting the county of
confinement as the proper forum for newly discovered evidence claims would
factors and principles that tend to affect prosecutorial decisionmaking). See generally ABA
1968 STANDARDS, supra note 138, § 1.3(b), at 27–28 (“Because of the nature of postconviction claims and their probably pervasive importance to criminal process in any state,
it is preferable to charge the office of the attorney general or comparable official with basic
responsibility for representing the state in such cases. Such an office will develop a greater
expertise on the manifold phases of this type of litigation than can be expected in most local
prosecutors’ offices.” (citation omitted)); Medwed, supra note 64.
326.
Although state habeas corpus petitions are typically filed in the county of
incarceration, several states allow filing in the county of conviction. See supra notes 207–08
and accompanying text.
327.
See supra notes 303–06, 313, 324 and accompanying text.
328.
Assigning post-conviction motions to prosecutors in the state attorney
general’s office might be a worthwhile option. See ABA 1968 STANDARDS, supra note 138,
§ 1.3(b), at 27–28 (suggesting that lawyers in the state attorney general’s office or
comparable officials should handle these motions).
329.
See, e.g., Larry Yackle, The Misadventures of State Postconviction
Remedies, 16 N.Y.U. REV. L. & SOC. CHANGE 359, 365 n.35 (1987–88) (describing the
administrative tension in state habeas corpus jurisprudence in which “courts situated near
penal institutions could be swamped with prisoner petitions, while sentencing courts, with
more convenient access to records and witnesses, handled other business”).
330.
If the application were granted by a judge in the county of confinement and a
new trial ordered, presumably the case would be transferred back to the county of
conviction for all further proceedings. Otherwise, in addition to burdening the county of
confinement, the prospect of retrying the defendant in a region where the composition of the
jury pool may differ vastly from that of the place where the crime occurred raises genuine
fairness issues; the right of a defendant to a jury of her “peers” could possibly be affected by
a change of venue. See James Oldham, The History of the Special (Struck) Jury in the
United States and its Relation to Voir Dire Practices, the Reasonable Cross-Section
Requirement, and Peremptory Challenges, 6 WM. & MARY BILL RTS. J. 623, 626 (1998). At
its core, this concept derives from the belief that a jury should contain “a representative
number of people that share the defendant's cultural, linguistic, ethnic, or possibly socioeconomic circumstances.” Id.

708

ARIZONA LAW REVIEW

[VOL. 47:655

neglect to account for the potentially innocent yet not currently imprisoned
defendant.331 As a consequence, a person who has already served his sentence
would have no recourse—aside from executive clemency332—in seeking to clear
his name.
Any jurisdiction open to the idea of directing new evidence claims to a
judge other than the original trial judge, irrespective of whether the recipient is
located in the county of conviction or the county of confinement, should consider
making its procedure compulsory.333 It might be tempting to give defendants the
ability to choose as between the original judge and a new, randomly assigned
judge, but providing that option would be fraught with peril. Such a choice would
promote forum-shopping, a much-dreaded inducement,334 and possibly overwhelm
those original trial judges perceived as “defense-friendly” with innocence claims
(or at least give them more than their fair share). Also, prisoners and especially
their lawyers might suffer artificial constraints in making this decision. Opting for
a new judge could send a negative and highly public signal about the original
judge’s objectivity—and this could discourage some defense lawyers from making
that choice for fear of potential retaliation during future appearances in that
courtroom.335
C. Appellate Review
Trial judges enjoy vast autonomy in considering claims of innocence
based on newly discovered evidence, whether in the form of new trial motions or
collateral petitions, and the appellate procedures governing denials of newly
discovered evidence claims embody this freedom. As mentioned above, not only
are decisions rejecting these claims occasionally appealable only as a matter of
permission,336 but the standard of review pertaining to those lower court decisions
331.
Litigants seeking redress through a procedure in the nature of coram nobis
are not generally required to be in custody at the time of filing. See supra note 99 and
accompanying text.
332.
For a discussion of some of the limitations of the executive clemency power,
see infra notes 382–84 and accompanying text.
333.
It should be noted that several scholars have proposed a completely different
method of handling post-conviction innocence petitions: the formation of an independent
commission. See, e.g., Lissa Griffin, The Correction of Wrongful Convictions: A
Comparative Perspective, 16 AM. U. INT’L L. REV. 1241, 1302–03 (2001). Griffin has
lauded the English treatment of innocence claims whereby the bipartisan Criminal Cases
Review Commission evaluates post-conviction innocence claims and directs the strongest
allegations to the Court of Appeal. Id. at 1275–78.
334.
See, e.g., Kevin M. Clermont & Theodore Eisenberg, Exorcising the Evil of
Forum-Shopping, 80 CORNELL L. REV. 1507 (1995) (arguing for the need to retain change
of venue rules to neutralize the danger of forum-shopping).
335.
Certain behavior by defense lawyers can conceivably spark retaliation—or at
least the fear of retaliation—against them or their clients by judges. See, e.g., Stephanos
Bibas, Plea Bargaining Outside the Shadow of Trial, 117 HARV. L. REV. 2463, 2486 (2004)
(discussing, for instance, how “[d]efendants whose lawyers take extensive discovery or file
many motions may suffer retaliation by judges and prosecutors” in the plea bargaining
context).
336.
See supra notes 178, 213 and accompanying text.

2005]

UP THE RIVER

709

chosen for appellate consideration is typically “abuse of discretion.”337 An
extraordinarily deferential standard of review, abuse of discretion acts, in the
words of one prominent scholar, as “a virtual shield from reversal.”338 In light of
these procedures, the chances of reversal on appeal are remote regardless of
whether a trial judge denies a newly discovered evidence claim summarily or after
an evidentiary hearing.339
Although it may be impracticable to depart from the routine of allowing
appeals only as a matter of permission in states where that is the norm, particularly
in jurisdictions with high caseloads, and instead provide appellate review for every
post-trial newly discovered evidence case, once authorization to appeal has been
granted the standard of review applicable to summary denials of these claims
should be less deferential than that afforded to dispositions rendered after a
hearing. To a large degree, the present system of appellate review of newly
discovered evidence claims offers few incentives—and arguably provides a
disincentive—for trial judges to do so much as hold an evidentiary hearing.340
337.
See supra notes 179–83, 215 and accompanying text.
338.
ROGER PARK ET AL., EVIDENCE LAW § 12.01, at 540–41 & n.6 (1998). For a
well-known case where the United States Supreme Court overturned a conviction upon a
finding of abuse of discretion, see generally Old Chief v. United States, 519 U.S. 172
(1997) (holding that a district court abused its discretion when it rejected the defendant’s
offer to concede the fact of a prior conviction, the existence of which was an element of the
crime at issue at trial, and instead admitted the full record of the prior offense).
339.
Even though many jurisdictions apply the abuse of discretion standard to
summary denials, see supra notes 182, 215 and accompanying text, some jurisdictions
provide differing standards of review in evaluating summary denials of newly discovered
evidence claims as opposed to post-hearing dispositions. For example, in evaluating
rejections of post-conviction relief petitions alleging newly discovered evidence, Florida
affords a more deferential standard of review to denials that occurred subsequent to a
hearing. See, e.g., FLA. R. APP. PROC. 9.141(b)(2)(D) (stating, with respect to appeals of
summary denials of motions for post-conviction relief in noncapital cases, that “[o]n appeal
from the denial of relief, unless the record shows conclusively that the appellant is entitled
to no relief, the order shall be reversed and the cause remanded for an evidentiary hearing or
other appropriate relief”); McLin v. State, 827 So. 2d 948, 955–56 (Fla. 2002) (holding that
the trial court erred in summarily denying post-conviction relief based on recantation
evidence); cf. Blanco v. State, 702 So. 2d 1250, 1252 (Fla. 1997) (stating that, in reviewing
a denial of a newly discovered evidence claim after an evidentiary hearing, assuming that
substantial competent evidence supported the trial court’s findings, “this Court will not
‘substitute its judgment for that of the trial court on questions of fact, likewise of the
credibility of the witnesses as well as the weight to be given to the evidence by the trial
court’” (quoting Demps v. State, 462 So. 2d 1074, 1075 (Fla. 1984))).
340.
A number of states allow courts significant freedom in determining whether
to hold an evidentiary hearing prior to ruling on a newly discovered evidence claim. See,
e.g., ALA. R. CRIM. PROC. 32.7(d) (“If the court determines that the [post-conviction relief]
petition is not sufficiently specific, or is precluded, or fails to state a claim, or that no
material issue of fact or law exists which would entitle the petitioner to relief under this rule
and that no purpose would be served by any further proceedings, the court may either
dismiss the petition or grant leave to file an amended petition. Leave to amend shall be
freely granted. Otherwise, the court shall direct that the proceedings continue and set a date
for hearing.”). To be fair, some jurisdictions provide a degree of encouragement for trial
judges to order evidentiary hearings or at least offer guidelines for judges in deciding

710

ARIZONA LAW REVIEW

[VOL. 47:655

Accordingly, many post-trial allegations of innocence are never heard in open
court, with the effect that litigants are prevented from developing a thorough
record of the new evidence.341 A vicious cycle then ensues; it may conceivably be
difficult for appellate courts to find an abuse of discretion based on a scanty
record, yet trial judges lack institutional incentives to order evidentiary hearings
and thereby produce a serviceable record for appeal.342 And, as the end result, the
system falls short of effectively assessing potentially viable innocence claims.343
To address this situation, appellate courts should retain a deferential
standard in evaluating denials of newly discovered evidence claims in the
aftermath of evidentiary hearings but undertake de novo review for claims rejected
summarily.344 This reform could achieve several beneficial policy aims. First,
bolstering the capacity of appellate courts to review summary denials anew would
make trial judges more accountable for their decisions and may, on the margins,
provoke judges to become more willing to conduct hearings where potentially
meritorious innocence claims hang in the balance.345 Second, spurring judges to
hold hearings in cases where the innocence claim may have validity would yield a
whether to hold hearings. See, e.g., N.Y. C.P.L § 440.30(2)–(5) (McKinney 2004) (offering
guidance regarding how trial courts should treat Article 440 motions, especially with regard
to when a court must grant, must deny, or has the discretion to deny the motion without a
hearing); Shockley v. State, 199 S.E.2d 791, 792 (Ga. 1973) (interpreting Georgia statutory
authority to signify that defendants are entitled to hearings on motions for a new trial); State
v. Allen, 744 P.2d 789, 792 (Haw. Ct. App. 1987) (discussing how evidentiary hearings
should be held on petitions for post-conviction relief where a petition puts forth a colorable
claim).
341.
Trial judges are often compelled to create a record of some fashion; that is,
even if a trial judge summarily disposes of a new trial motion, that judge may be required to
specify the grounds upon which she based her decision. See 58 AM. JUR. 2D New Trial
§ 436 (2004). Not every jurisdiction, though, imposes such an obligation. Id.
342.
On the one hand, holding an evidentiary hearing exposes the trial judge’s
decision to greater scrutiny; the appellate court has more to work with in evaluating the
decision. On the other hand, the fact that a judge held a hearing may suggest she did not
abuse her discretion in ultimately refusing to order a new trial, i.e., she took the time to
analyze the issues and thus her decision is entitled to deference.
343.
Indeed, appellate judges seem to take comfort in cases where an evidentiary
hearing was held by the lower court prior to the denial of a post-conviction petition. See,
e.g., Coleman v. Thompson, 501 U.S. 722, 755–56 (1991) (rejecting a defendant’s postconviction allegation of ineffective assistance of counsel, and pointedly mentioning that
“the Buchanan County Circuit Court, after a 2-day evidentiary hearing, addressed
Coleman’s claims of trial error, including his ineffective assistance of counsel claims”).
344.
This model is not unprecedented; Illinois, for example, has embraced a
comparable approach in regard to petitions denied pursuant to the principal post-conviction
remedy in the state. See Quinn & Hynes, supra note 204, at 641–42 (“The standard of
review for summary dismissals is de novo. The manifest error standard, however, is
applicable when reviewing orders granting or denying relief are made after an evidentiary
hearing.”); supra note 339 and accompanying text.
345.
Muskat has also insisted that state courts display greater willingness to hold
evidentiary hearings in cases revolving around newly discovered evidence of innocence,
maintaining that the legal standard applicable to the evaluation of bare innocence claims
should be altered to prompt state courts to order such hearings. Muskat, supra note 13, at
172–73.

2005]

UP THE RIVER

711

greater number of fully transcribed records of innocence claims for appellate
review, scholarly analysis, and public scrutiny. Third, preserving the abuse of
discretion standard (or a comparable test) with respect to post-hearing dispositions
correctly recognizes the fact-specific nature of those decisions, which explicitly
involve the critique of witness credibility and the weighing of conflicting
testimony and/or other evidence,346 and does not deter the ordering of hearings in
the first place. Lastly, this proposal, adopting a de novo standard of review for
summary dispositions while keeping a deferential approach to denials issued after
evidentiary hearings, comports with the core theories and principles underlying
appellate review and the interrelationship between trial and appellate courts.347
1.

Theoretical Justifications for Divergent Standards of Review

On appeal, labeling an issue as one of fact, law, or discretion is of critical
importance given that the categorization dictates the standard of review germane to
the issue.348 The standard of appellate review, in turn, affects the outcome of the
case: the more deferential the standard, the less likely the prospect of reversal.349
Issue characterization is by no means a simple endeavor,350 for judicial
decisionmaking often takes place in the murky waters between the islands of
clearly identifiable issues of fact and those of law, and may involve mixed
questions of fact and law or so-called “discretionary” decisions by the lower
court.351 Scholars have discussed how appellate courts should treat discretionary
decisions by trial courts, such as those regarding discovery, evidence, and jury-

346.
See Ronald R. Hofer, Standards of Review—Looking Beyond the Labels, 74
MARQ. L. REV. 231, 242 (1991) (suggesting that deference may be justified where the trial
court’s determination included the assessment of credibility, and the weighing of both
conflicting testimony and conflicting evidence).
347.
See infra notes 348–66 and accompanying text.
348.
See Maurice Rosenberg, Judicial Discretion of the Trial Court, Viewed From
Above, 22 SYRACUSE L. REV. 635, 645–46 (1971) (“[A]ll appellate Gaul is divided into
three parts for review purposes: questions of fact, of law and of discretion.”).
349.
See, e.g., MEADOR ET AL., supra note 179, at 154 (“Implicit in the effort to
formulate rational and workable standards of review is acceptance of the principle that the
reviewing court should subordinate its own view and defer to the trial court’s decision in
some situations and to some degree.”).
350.
See, e.g., Sch. Dist. v. Z.S. ex rel. Littlegeorge, 295 F.3d 671, 674 (7th Cir.
2002) (Judge Richard Posner observing that “the cognitive limitations that judges share with
other mortals may constitute an insuperable obstacle to making distinctions any finer than
that of plenary versus deferential review”); see also Chad M. Oldfather, Appellate Courts,
Historical Facts, and the Civil-Criminal Distinction, 57 VAND. L. REV. 437, 504–06 (2004)
(citing the aforementioned comment by Judge Posner and noting that “appellate courts often
remark on the difficulties involved in attempting to divine the difference between similar,
but purportedly distinct, standards of review”).
351.
See MEADOR ET AL., supra note 179, at 154 (mentioning “a four-way
classification of the nature of issues on appeal: questions of fact; questions of law; mixed
questions; and questions of discretion”); Ronald J. Allen & Michael S. Pardo, The Myth of
the Law-Fact Distinction, 97 NW. U. L. REV. 1769, 1769 (2003) (“The importance of the
law-fact distinction is surpassed only by its mysteriousness.”).

712

ARIZONA LAW REVIEW

[VOL. 47:655

related matters,352 and the standard of review in assessing those decisions is
frequently articulated in terms of abuse of discretion.353 Under this standard,
appellate courts in general will only overturn discretionary decisions if the trial
court relied on an erroneous view of the law, misapplication of the law, mistaken
view of the evidence, or otherwise impermissible or irrelevant factors.354 Patrick
Brennan has stated that the abuse of discretion standard represents “a compromise
between the competing desires of affording power and flexibility to the trial court,
yet at the same time minimizing arbitrary decisions,”355 and it is a compromise in
which “[t]he scales are tipped in favor of trial court flexibility.” 356 Among the oftstated reasons for giving trial courts significant discretion with respect to certain
types of decisions are judicial economy,357 judicial comity,358 lower courts’
morale,359 and finality of judgments.360 Some observers also champion deference
on the basis that particular types of decisions are grist for a lower court’s daily
mill, which allows trial judges to gain vital experience in these areas that informs
their decisionmaking capability.361 Yet many commentators have found these
352.
See Patrick W. Brennan, Standards of Appellate Review, 33 DEF. L.J. 377,
412–14 (1984), reprinted in MEADOR ET AL., supra note 179, at 196 (observing that the
discretion of a trial court “is exercised in many situations: rulings on motions, discovery,
arguments of counsel, admissibility of evidence, instructions, and numerous other juryrelated matters”).
353.
See, e.g., 5 AM. JUR. 2D Appellate Review § 695 (2004) (“Various
formulations have been employed to elaborate on the practical effect and meaning of the
term ‘abuse of discretion.’”); MEADOR ET AL., supra note 179, at 196–221 (discussing the
abuse of discretion standard of review and including excerpts from articles on the topic).
354.
See Hofer, supra note 346, at 245 (describing the abuse of discretion
standard in Wisconsin). In the vernacular, as the Utah Supreme Court once declared, “[W]e
give the trial court a great deal of pasture” in reviewing certain discretionary decisions by
the lower court. State v. Pena, 869 P.2d 932, 938 (Utah 1994) (mentioning, at “the extreme
end of the discretion spectrum,” a trial court decision “to grant or deny a new trial based on
insufficiency of the evidence”).
355.
Brennan, supra note 352, at 412–14.
356.
Id.
357.
Id. (“An appellate court could not begin to handle all of the cases that would
be appealed if trial court rulings were a meaningless formality.”); see also Brooks v. State,
61 S.W.3d 916, 919 (Ark. Ct. App. 2001) (commenting that judicial economy would not
have been served by holding a hearing on the defendant's motion for a new trial where the
defendant failed to set out any new evidence).
358.
See Brennan, supra note 352, at 412–14 (“Appellate courts should and do
recognize the integrity and competence of the lower courts.”).
359.
See Hofer, supra note 346, at 241 (“Morale boosting . . . means that trial
judges would become demoralized if all their rulings were measured by a de novo
yardstick.”).
360.
See Brennan, supra note 352, at 412–14.
361.
See Oldfather, supra note 350, at 447–48 (describing one justification for a
trial court’s “supposed advantage” over an appellate court in reviewing facts as having “to
do with experience” and specifically the idea that trial judges are “better at disentangling
conflicting evidence simply because they have more practice”). Oldfather then proceeds to
engage in an interesting critique of the “institutional competence justification” for deference
to trial courts in the realm of facts: “To say, however, that juries and trial courts are superior
fact-finding instruments in most cases and with respect to most types of evidence is not to
say that they are superior in all cases and with respect to all types of evidence.” Id. at 449.

2005]

UP THE RIVER

713

justifications lacking, grounded as they are in general, macro-level policy concerns
rather than any issue-specific quality.362
Instead of relying on large-scale policy objectives, the strongest
justification for deferring to trial judges on appeal may stem from the fact that
those judges are often in a better position to render certain decisions than their
appellate peers.363 Specifically, deference has both practical utility and theoretical
legitimacy where a trial judge’s chance to view the witnesses “live” and receive
evidence proved instrumental to the decision.364 In those conditions, the lower
court arguably stood in a better decisionmaking posture than its appellate
counterpart ever could based solely on its interpretation of the cold, black-andwhite text of the record on appeal.365 Therefore, in ultimately ascertaining the
proper measure of deference to give the lower court, the key issue may lie in
determining whether the trial court was in a better position to consider the specific
question than an appellate tribunal would be.366

362.
See Hofer, supra note 346, at 241 (citing Maurice Rosenberg, Appellate
Review of Trial Court Discretion, 79 F.R.D. 173 (1975)). In Hofer’s view, many of the
arguments mentioned above to rationalize deference “fall to the wayside in a functional
analysis. They do not espouse deference because of any quality found in a particular issue.
Rather, they are ex post facto, arising from larger policy concerns; they arise from the
general rather than from the particular.” Id.
363.
See, e.g., State v. Bocharski, 22 P.3d 43, 56 (Ariz. 2001) (Martone, J.,
concurring) (noting that in assessing “whether the trial court abused its discretion in
weighing probative value against prejudicial effect” regarding the admissibility of
photographs “appellate courts are not in a very good position to second guess such
judgments”).
364.
See Henry Friendly, Indiscretion about Discretion, 31 EMORY L.J. 747, 784
(1982) (“An appellate court must carefully scrutinize the nature of the trial court’s
determination and decide whether that court’s superior opportunities of observation or other
reasons of policy require greater deference than would be accorded to its formulations of
law or its application of law to the facts.”); Hofer, supra note 346, at 239; cf. Oldfather,
supra note 350, at 447–49 (disputing the validity of the better position or “situational”
justification for deference to trial courts regarding all factual questions).
365.
See, e.g., State v. Pena, 869 P.2d 932, 936 (Utah 1994) (noting how,
regarding factual findings, trial courts are “considered to be in the best position to assess the
credibility of witnesses and to derive a sense of the proceeding as a whole, something an
appellate court cannot hope to garner from a cold record”). For the countervailing view,
Oldfather has insisted that transcripts, to some extent, have advantages over the rapid,
immediate processing of information by judges and juries at trial. Oldfather, supra note 350,
at 451–57. Also, as Oldfather points out, “Research consistently shows that people perform
poorly at using demeanor to determine whether a person is telling the truth . . . . Observers
tend to focus on facial expressions, which are highly manipulable and therefore unreliable,
rather than on speech patterns, which are better indicators of deception.” Id. at 457–58.
366.
See Miller v. Fenton, 474 U.S. 104, 114 (1985) (“At least in those instances
in which Congress has not spoken and in which the issue falls somewhere between a
pristine legal standard and a simple historical fact, the fact/law distinction at times has
turned on a determination that, as a matter of the sound administration of justice, one
judicial actor is better positioned than another to decide the issue in question.”); Hofer,
supra note 346, at 231 (advocating for “an analysis which affords deference to lower
tribunals where they were in a better position to address a question than the appellate court

714

ARIZONA LAW REVIEW

[VOL. 47:655

2. Altering the Standard of Review Applicable to Summary Denials of
Newly Discovered Evidence Claims
Application of the “better position” theory to newly discovered evidence
claims suggests deference to the trial judge’s determination is not always
warranted. By denying a new trial motion or post-conviction application premised
on newly discovered evidence of innocence without ordering an evidentiary
hearing, a trial judge has not taken advantage of any superior opportunities of
observation nor attempted to grasp a “feel” of the case.367 No evidence has been
received nor witnesses heard; the trial court has issued its ruling based only upon
the written submissions, a decision not unlike those usually made by appellate
courts, and concluded that the defendant’s allegations fail to merit so much as a
hearing. This decision, in essence, could be construed as a question of law or at the
very least a mixed question of law and fact: that the defendant’s assertions, even if
true, are legally insufficient to support further analysis.368 In reviewing mixed
questions of law and fact, appellate courts often afford almost total deference to a
trial court’s determination where such determination rested on assessments of
credibility and demeanor.369 In contrast, de novo review of mixed questions is
justified and occasionally utilized when the lower court’s decision did not entail
critical examinations of these variables.370 As a normative matter, therefore, de
novo review should apply to summary dismissals of state new trial motions and
post-conviction petitions grounded on newly discovered evidence, namely, cases
where no critical inspection and vetting of the new evidence occurred in open
court.
Admittedly, under the current regimes in which the judge who presided at
trial has jurisdiction over the subsequent presentation of newly discovered
evidence, that judge may try to lay claim to a superior decisionmaking platform
even in instances of summary dispositions. After all, she heard the witnesses and
received evidence at the original trial, making her all the more competent to
evaluate the strength of the defendant’s alleged newfound information and its

would be”); id. at 249 (“Finally, and certainly not least, some of the deference accorded to
discretionary acts again goes back to the ‘better position’ principle.”).
367.
See Brennan, supra note 352, at 412–14 (“It is said that one of the strongest
reasons behind according discretion to the trial court is the fact that the judge is in a better
position to see and hear the witnesses. This superior position allows a better ‘feel of the
case’ that examination of the appellate court record does not afford.”).
368.
Appellate courts tend to take a “functional approach” to mixed questions,
analyzing whether the issues are essentially factual or legal in nature. See, e.g., URSULA
BENTELE & EVE CARY, APPELLATE ADVOCACY: PRINCIPLES AND PRACTICE 206–12 (4th ed.
2004).
369.
See, e.g., Guzman v. State, 955 S.W.2d 85, 89 (Tex. Crim. App. 1997) (“The
appellate courts, including this Court, should afford [almost total deference] to trial courts’
rulings on ‘application of law to fact questions,’ also known as ‘mixed questions of law and
fact,’ if the resolution of those ultimate questions turns on an evaluation of credibility and
demeanor. The appellate courts may review de novo ‘mixed questions of law and fact’ not
falling within this category.” (internal citations omitted)).
370.
See id.

2005]

UP THE RIVER

715

comparative impact on the evidence adduced at trial.371 As stated in Part III.B of
this Article, however, the supposed benefits of having the original trial judge
entertain a new evidence claim are largely illusory in any circumstance given the
potential existence of cognitive biases.372 More to the point here, regardless of
whether the judge assigned to handle the post-trial innocence claim participated in
the initial trial, the main rationale for deference—the trial judge’s purportedly
superior position based on the ability to observe witnesses and take evidence—is
diminished, if not entirely absent, in the summary denial situation. Consequently,
where a trial court exercises its post-trial judgment regarding a newly discovered
evidence claim without holding an evidentiary hearing and viewing witnesses, an
appellate court should be entitled to look afresh at the case to provide a necessary
check on the lower court and a major precaution against injustice.

CONCLUSION
Any effort to reform post-trial procedures with an eye toward granting
prisoners greater access to state courts must address the wariness with which many
participants in the criminal justice system approach newly discovered evidence
claims. Most notably, this discomfort stems from fears about jeopardizing the
finality of judgments, undermining the trial process, and straining judicial
resources; doubts about the validity of most newly discovered non-DNA evidence
claims; and suspicions that guilty prisoners will capitalize on any “slackening” of
procedural safeguards.373 But reducing the rate of procedural default and providing
full-scale evidentiary hearings for a greater number of innocence claims would
mainly facilitate the exoneration of actually innocent prisoners, not to mention
inspire confidence in the accuracy of the criminal justice system.374 Moreover,
whereas the proposed easing of the procedural restrictions attendant to newly
discovered non-DNA evidence claims might hinder judicial economy to a degree,
the continued enforcement of demanding legal and evidentiary standards
371.
See, e.g., Whitley v. United States, 783 A.2d 629, 633–34 (D.C. 2001), reh’g
granted, opinion modified 796 A.2d 26 (D.C. 2002) (declaring that the trial judge who had a
chance to develop a “feel” for the case is in a superior position to appellate judges to
determine if the newly discovered evidence put forth as part of a motion for a new trial has
any potential for affecting the jury’s verdict). A Texas state appellate court applied the
abuse of discretion standard in reviewing the summary disposition of a habeas corpus
petition despite citing case law that “an appellate court must conduct a de novo review when
‘the trial judge is not in an appreciably better position than the reviewing court to make that
determination.’” Ex parte Carbajal, No. 08–03–00297–CR, 2004 WL 1772113, at *3 (Tex.
App., Aug. 5, 2004) (unpublished) (internal citations omitted). The court explained that
“[w]hile the trial court did not hear live testimony, the facts were contested, the affidavits
were conflicting, and the same trial judge presided at Appellant’s guilty plea and his petition
for writ of habeas corpus.” Id.
372.
See supra notes 288–335 and accompanying text.
373.
See, e.g., supra notes 62–64, 116–20 and accompanying text.
374.
At a fundamental level, holding evidentiary hearings with respect to
colorable innocence claims and subjecting the defendant’s witnesses to cross-examination
would advance the truth-seeking function of the system of litigation. See 6 JOHN HENRY
WIGMORE, EVIDENCE IN TRIALS AT COMMON LAW § 1838, at 463 (James H. Chadbourn ed.,
rev. ed. 1976) (describing cross-examination as the greatest legal engine ever invented for
the discovery of the truth).

716

ARIZONA LAW REVIEW

[VOL. 47:655

minimizes the risk that factually guilty inmates will succeed in taking advantage of
the relaxation of procedural constraints.375 If states were to adopt the procedural
reforms set forth in Part III of this Article, in the end, the cost to administrative
efficiency would be a proper one to absorb—assuming the failure to adequately
review claims by potentially innocent prisoners runs counter to the goals of our
society in this age of DNA exonerations.376
The rise of DNA testing since 1989 has lent scientific credence to the
long-suspected fear that actually innocent prisoners are not always acquitted at
trial, and has provoked many people to reconsider their views of the criminal
justice system’s effectiveness.377 Indeed, the extensive efforts by many state
legislatures in the past decade to improve post-conviction access to DNA testing of
biological evidence deserve acclaim.378 State treatment of newly discovered nonDNA evidence, however, need not mirror the traits of modern post-conviction
DNA testing legislation; non-DNA evidence is often less reliable than DNA
evidence and may earn more rigorous evidentiary and legal requirements to ensure
reliability.379 Rather than pressing for wholesale changes to the legal and
evidentiary rules governing the treatment of newly discovered non-DNA evidence
claims, my position is that the lessons from the DNA revolution justify
modification of the procedural attributes of most state regimes. Ideally, these
changes would provide more meaningful access to state courts for those potentially
innocent defendants whose greatest misfortune lies in the happenstance that the
375.
See supra notes 225–29 and accompanying text.
376.
Several scholars suggest that the focus on factual innocence in the recent
debate on reforming the criminal justice system may be misplaced. Some of these critiques
have taken the form of attacks on the overall number of actually innocent prisoners. See,
e.g., Joshua Marquis, The Myth of Innocence, 95 J. CRIM. L. & CRIMINOLOGY 501 (2005)
(arguing that the emphasis on innocence is misleading given that only a few wrongful
convictions occur and that such mistakes are inevitable as well as acceptable). Others have
contended that the focus on innocence may partially serve to obscure the problem of more
mundane and pervasive errors in the criminal justice system, “including arbitrary and
unequal treatment of offenders as well as disproportionate punishment of the ‘guilty.’” See
Carol S. Steiker & Jordan M. Steiker, The Seduction of Innocence: The Attraction and
Limitations of the Focus on Innocence in Capital Punishment Law and Advocacy, 95 J.
CRIM. L. & CRIMINOLOGY 587, 597 (2005).
377.
The spate of DNA exonerations also famously contributed to the decision by
Governor George Ryan of Illinois to study the death penalty in that state, a decision that
culminated in the pardoning of four death row inmates and the commutation of 167 death
sentences to life in prison. See, e.g., Jodi Wilgoren, Citing Issue of Fairness, Governor
Clears Out Death Row in Illinois, N.Y. TIMES, Jan 12, 2003, at 1.
378.
See supra notes 3–4 and accompanying text.
379.
See White, supra note 25, at 10–11 (arguing that evidence that becomes
“newly available” due to technological advances and is highly reliable, such as DNA
evidence, should be treated differently from most forms of newly discovered evidence). To
a degree, the emergence of DNA testing has possibly raised the bar for proving innocence in
other cases. See, e.g., Kirk Makin, The Reliance on Science as a Cure for Injustice, GLOBE
& MAIL (Toronto, Can.), Nov. 22, 2004, at A1 (quoting a veteran defense attorney as
claiming that “[b]ecause DNA is now the best-known and surest means of exoneration, it
has come to be seen as a sort of benchmark for testing wrongful convictions . . . . I have had
prosecutors say to me, ‘After all, this isn’t a case where innocence can be shown by
DNA’”).

2005]

UP THE RIVER

717

actual perpetrators of the crimes for which they were wrongfully convicted did not
leave biological evidence at the crime scene.
The need to amend state court procedures has reached a critical stage
partly because of the dearth of alternative venues for potentially innocent
prisoners. In particular, the rising number of procedural hurdles that prisoners must
overcome to obtain relief in federal court through a writ of habeas corpus has made
that option effectively unavailable.380 Nor is parole a feasible solution for most
innocent state prisoners. Parole boards customarily frown upon an unwillingness to
convey remorse, which puts actually innocent prisoners in a Catch-22: continue to
proclaim innocence or boost the chances for parole by “admitting” guilt and
showing regret.381 Furthermore, the executive clemency power—an oft-cited,
purported panacea for the ills of wrongful convictions382—is seldom exercised by
government officials.383 Even when used, clemency may be aimed chiefly toward
attaining political objectives, with any correction of injustice as a side effect.384
380.
Inmates may pursue innocence claims in federal court via habeas corpus
petitions, but the United States Supreme Court has clarified that freestanding claims of
actual innocence based on newly discovered evidence do not provide an independent
ground for habeas relief absent compelling circumstances. See generally Herrera v. Collins,
506 U.S. 390 (1993). Moreover, the Antiterrorism and Effective Death Penalty Act
(AEDPA) of 1996 provided additional restrictions on federal habeas corpus review. See
generally Stephen B. Bright, Is Fairness Irrelevant?: The Evisceration of Federal Habeas
Corpus Review and Limits on the Ability of State Courts to Protect Fundamental Rights, 54
WASH. & LEE L. REV. 1 (1997). If passed, a bill recently introduced in Congress—the
Streamlined Procedures Act of 2005—would further restrict federal habeas corpus review of
capital cases. See Editorial, Stop This Bill, WASH. POST, July 10, 2005, at B6; Alan Elsner,
Bill Would Cap Appeals Over Death Sentences, STAR-LEDGER (NEWARK, N.J.), July 7,
2005, at 6. Notably, the Supreme Court is slated to examine the issue of DNA evidence of
innocence and federal habeas corpus during the 2005–06 Term in House v. Bell, No. 04–
8990. See Linda Greenhouse, Justices to Review Rules for Death Case Appeals, N.Y. TIMES,
June 29, 2005, at A18; Charles Lane, Court May Revise Rule on Death Row Appeals,
WASH. POST, June 29, 2005, at A3.
381.
See, e.g., Stanley Z. Fisher, Convictions of Innocent Persons in
Massachusetts: An Overview, 12 B.U. PUB. INT. L.J. 1, 19 (2002) (recounting how a
prisoner in Massachusetts, later exonerated through DNA testing, “was denied parole
‘because he proclaimed his innocence and refused to enter treatment for sexual deviance’”);
Martin, supra note 5, at 98–100 (quoting one former inmate as having been told by a
California parole commissioner that “‘[i]f you don’t admit that you did this crime, you’ll
never get out’”).
382.
See, e.g., Herrera, 506 U.S. at 411–12 (terming clemency the “historic
remedy for preventing miscarriages of justice where judicial process has been exhausted”);
State v. Mixon, 983 S.W.2d 661, 671 (Tenn. 1999) (noting that “convicted defendants who
discover new non-scientific evidence of actual innocence too late to file a motion for new
trial or petition for writ of error coram nobis may always seek executive clemency”).
383.
See HOWARD ABADINSKY, DISCRETIONARY JUSTICE: AN INTRODUCTION TO
DISCRETION IN CRIMINAL JUSTICE 147–48 (1984) (“All states and the federal government
have provisions for clemency. . . . The basis for a pardon varies from state to state, and it is
not used extensively anywhere.”); Heise, supra note 319, at 241 (noting that “an increase in
the number of death sentences has coincided with a decrease in the number of defendants
removed from death row through clemency”); cf. Beau Breslin & John J.P. Howley,
Defending the Politics of Clemency, 81 OR. L. REV. 231, 231 (2002) (“In the past quarter

718

ARIZONA LAW REVIEW

[VOL. 47:655

More than ever, then, state post-conviction procedures comprise the most
appropriate vehicle to rectify wrongful convictions and a subset of those
procedures, the rules concerning newly discovered evidence, have the potential to
operate as the principal engine driving cases toward fair resolutions.385 Ultimately,
it may be impossible to determine for certain whether prisoners are actually
innocent when their cases rest solely upon non-DNA evidence.386 That is a poor
excuse, though, for depriving defendants like Stephen Schulz, the inmate
convicted of robbery in New York, of the opportunity to present their claims in a
manner that both justice and common sense require. The lone dissenter to the New
York Court of Appeals’ recent decision affirming Schulz’s conviction criticized
the trial court’s denial of the post-conviction innocence claim without holding an
evidentiary hearing and summarized his views in a sentence that aptly serves as a
wrap-up to this Article: “The interests of finality count for a great deal, and may be
alluring, but they are not always consistent with the higher ends of justice.”387

century more than forty-five prisoners have been removed from death row because of
executive orders, while countless others have not been so successful.”). Even if governors
want to use their clemency powers, they may lack funding to investigate clemency petitions
sufficiently. See Hart & Dudley, supra note 27, at 641.
384.
See KATHLEEN DEAN MOORE, PARDONS: JUSTICE, MERCY, AND THE PUBLIC
INTEREST 203 (1989) (“It is not overly cynical to suggest, however, that politicians weigh
public opinion in a pardon decision.”); Hugo A. Bedau, The Decline of Executive Clemency
in Capital Cases, 18 N.Y.U. REV. L. & SOC. CHANGE 255, 257 (1990–91) (“Clemency
decisions—even in death penalty cases—are standardless in procedure, discretionary in
exercise, and unreviewable in result.”); cf. Breslin & Howley, supra note 383, at 232
(arguing that “clemency is, and should remain, a political process”).
385.
See Muskat, supra note 12, at 161 (“The best way to formulate an adequate
process for the resolution of bare innocence claims is not to force review on unenthusiastic
federal courts, but instead to encourage the states to review such claims through effective
state postconviction remedies.”).
386.
See supra note 14 and accompanying text.
387.
People v. Schulz, 829 N.E.2d 1192, 1201 (N.Y. 2005) (Rosenblatt, J.,
dissenting).