Will v. Stephens

Filing 94

MEMORANDUM AND ORDER (Signed by Judge Keith P Ellison) Parties notified. (aar4)

United States District Court
Southern District of Texas

ENTERED
IN THE UNITED STATES DISTRICT COURT
FOR THE SOUTHERN DISTRICT OF TEXAS
HOUSTON DIVISION
§
§
Petitioner,
§
§
v.
§
§
BOBBY LUMPKIN, Director, Texas
§
Department of Criminal Justice, Correctional §
Institutions Division,
§
§
Respondent.
§

April 04, 2024
Nathan Ochsner, Clerk

ROBERT GENE WILL, II,

CIVIL ACTION NO. 15-CV-3474

MEMORANDUM AND ORDER
In this successive federal habeas corpus action, Robert Gene Will, II, seeks relief
from his Texas capital conviction and death sentence. After several years of litigation, the
parties have finally fully briefed the question of whether the Texas courts unreasonably
denied Will’s claim that the State suppressed material contrary to Brady v. Maryland, 373
U.S. 83 (1963). The Court finds that the state-court decision was contrary to federal law.
The litigation will proceed as discussed below.
Background
The matters raised in the instant lawsuit are the culmination of two decades of
litigation. Courts have previously described the underlying facts in great detail; there is no
need to recount it for purposes of the matter now before this Court. See In re Will, 970 F.3d
536, 544–48 (5th Cir. 2020); Will v. Thaler, No. H-07-CV-1000, 2010 WL 2179680, at
*1–4 (S.D. Tex. May 25, 2010); Will v. State, No. 74,306, 2004 WL 3093238, at *1 (Tex.
Crim. App. Apr. 21, 2004); Dkt. 57-74 at 155–66). Over the years, Will’s challenges to his
culpability have followed a constant theme: that his co-defendant Michael Rosario is the

1

man who shot Harris County Police Deputy Barrett Hill on the morning of December 4,
2000. Will’s overarching theory about the events of that morning has remained constant,
yet the evidence on which he bases his claims has shifted with time. The question has
always been whether, as two police officers chased Will and Rosario in different directions,
Rosario circled back around and shot Deputy Hill.
Previously, Will based his post-conviction challenges on a wide variety of issues:
forensics, timing, statements Will and Rosario made after the crime, and events that
transpired while the government held Rosario in the Harris County Jail. Now, the case
concerns evidence that the prosecution kept from Will’s trial attorneys.
The instant lawsuit arises out of a successive state habeas application Will filed in
2015. That application raised three claims: (1) Will is actually innocent; (2) the State
suppressed two documents from the defense; and (3) trial counsel should have uncovered
the documents allegedly suppressed by the State. (Dkt. 57, Ex. 71 at 2–3). Will summarizes
the allegedly suppressed material as follows:
The first was a Harris County Sheriff’s Office document, which reveals that
the county jail was holding Michael Rosario in a separate section of the
Harris County Jail because Rosario had asked a prison gang to kill Rob Will.
See Ex.1, Michael Rosario Administrative Separation Review Sheet (“Hit
Document”), at 1. Specifically, Rosario’s “Administrative Separation
Review Sheet” states that Rosario’s “Reason For Separation” was that
Rosario was “soliciting [the Texas Syndicate prison gang] to make [a] hit
on co-def. Robt. Will.” See id. The Hit Document also indicates that contact
was made “w[ith] [the Disruptive Group Unit] to visit w[ith] David Cruz
[Texas Syndicate].” Id. Copies of Mr. Cruz’s undisclosed jail records found
in the District Attorney’s case file show that the State knew of his
connection to Rob Will’s case before trial. Compare Ex. 6, David Cruz
Administrative Separation Review Sheet Updated 2-10-01, at 1, with Ex. 5,
David Cruz Administrative Separation Review Sheet Last Updated 4-2701, at 1. A second document never disclosed to Mr. Will’s defense counsel
was a report by Harris County Sheriff’s Deputy Patricia Schifani. See Ex.
7, Report of Deputy Patricia Schifani (“Schifani Report”). According to her
report, Deputy Schifani was returning Harris County jail inmates, including
2

Michael Rosario, from court on December 7, 2000, three days after the
murder.
Rosario looked directly at the mourning badge cover that Deputy Schifani
had been wearing in honor of Deputy Hill and said, “Do you know why you
are wearing that? . . . I am part of the reason you are wearing it, do you
know who I am?” Id. Rosario, the son of a Houston Police Department
officer, then “pointed at his armband caution text which indicated
‘*PROTECTION*’” and said, “I’m high-profile! Do you know who my
father is?” Id.
(Dkt. 37 at 3–4). Will’s successive habeas proceedings also raised a third item of
undisclosed evidence relating to trial witness Cassandra Simmons.1 Ms. Simmons is a
witness who testified at trial that Will told her “he had just shot a policeman.” Will claimed
that Ms. Simmons did not tell that to the police initially. Will claimed that the prosecution
did not divulge that, on the eve of trial, it had showed Ms. Simmons “very gruesome and
extremely graphic” photographs of the slain victim. (Dkt. 57, Ex. 71 at 25). Will claimed
that the photos “no doubt biased her testimony, causing her to recount a statement that, the
evidence shows, was not part of her actual recollection of the events.” (Dkt. 57, Ex. 71 at
25). Independent of the allegedly suppressed material, Will also relied on significant
evidence to show his innocence.

1

The state habeas record contains some confusion as to whether the allegations relating to Ms.
Simmons were properly before the state courts. Because of how Will had raised concerns about her testimony,
the lower state habeas court found that it was a “newly presented claim” which it was “without jurisdiction
to consider.” (Dkt. 57-74 at 182). The lower court also alternatively denied the claim on the merits. The Court
of Criminal Appeals did not comment on the procedural status of any allegations relating to Ms. Simmons.
The Court of Criminal Appeals, however, “den[ied] relief” which signals an adjudication on the merits. See
Ex parte Torres, 943 S.W.2d 469, 472 (Tex. Crim. App. 1997) (“In our writ jurisprudence, a ‘denial’ signifies
that we addressed and rejected the merits of a particular claim while a ‘dismissal’ means that we declined to
consider the claim for reasons unrelated to the claims merits.”). Earlier in this litigation, Respondent
questioned whether allegations about Ms. Simmons were properly before the Court. (Dkt. 73 at 29).
Respondent, however, does not now argue that a procedural bar prevents consideration of that issue.

3

On February 5, 2014, the Court of Criminal Appeals entered an order remanding
the subsequent state habeas application. The trial-level habeas court considered Will’s
evidence without ordering an evidentiary hearing.
Both parties submitted proposed findings of fact and conclusions of law in early
December 2014. (Dkt. 57-74 at 2–77). For a reason that is not clear from the record, the
State submitted amended findings and conclusions on December 31, 2014. (Dkt. 57-74 at
153). On January 26, 2015, the trial court signed the State’s 49-page-long proposed
recommendation without alteration. (Dkt. 57-74 at 153–201). The Court of Criminal
Appeals adopted the lower court’s findings and conclusions, with several exceptions. Ex
parte Will, 2015 WL 13388366, at *1 (Tex. Crim. App. 2015).
Will then began to prosecute a successive habeas petition in federal court. Will filed
the instant federal petition in this Court in 2015. The parties’ briefing became ripe on
September 15, 2017. On September 25, 2017, the Court transferred this action to the Fifth
Circuit for a determination of whether the Anti-Terrorism and Effective Death Penalty Act
(AEDPA) would allow him to proceed with a successive habeas action. (Dkt. 20). The
question remained pending until the Fifth Circuit tentatively authorized successive
proceedings on August 5, 2020. In re Will, 970 F.3d 536 (5th Cir. 2020).
The parties returned to this Court and extensively briefed the question of whether
successive proceedings were appropriate. Once the issue became ripe, this Court entered a
Memorandum and Order allowing successive review on March 28, 2023. (Dkt. 79). The
briefing has since turned from the procedural question of whether the case should advance
to the question of whether Will’s claims merit relief.

4

The Court begins by acknowledging an underlying concern throughout this
litigation: The State has not denied that it has kept information from Will. On successive
state habeas review, the State did not dispute that the Hit Document and Schifani Report
had not been turned over to Will’s trial attorneys. (Dkt. 57-72 at 82–83). After returning to
federal court, the State turned over even more information that it had secreted away in the
prosecutor’s “work product folder.” Will argues that more material likely remains
undisclosed. Will’s action seeks additional factual development, in particular an in camera
review of what material the State is still maintaining in the work product folder.
The Court will first address the legal standards that govern this action and then turn
to the question of whether Will has filed this case in a timely manner. The Court will then
apply AEDPA to Will’s Brady claim and discuss the path forward in this case.
Standard Of Review
AEDPA governs this habeas proceeding. Honoring principles of comity and
federalism, Congress enacted AEDPA “to impose significant limits on the discretion of
federal courts to grant habeas relief.” Calderon v. Thompson, 523 U.S. 538, 554 (1998);
see also Danforth v. Minnesota, 552 U.S. 264, 278 (2008) (observing that the courts have
“adjust[ed] the scope of the writ in accordance with equitable and prudential
considerations”). AEDPA sets exacting procedural requirements over what a habeas court
may consider and establishes strict guidelines about how a court may review it.
If a habeas petitioner has presented his federal constitutional claims to the state
courts in a procedurally proper manner, and the state courts have adjudicated their merits,
AEDPA provides for a deferential federal review. Under AEDPA’s rigorous requirements,
a federal court reviews “[c]laims presenting questions of law” under Section 2254(d)(1).

5

Neal v. Vannoy, 78 F.4th 775, 783 (5th Cir. 2023). Section 2254(d)(1) is “divided into two
categories: the ‘contrary to’ standard, and the ‘unreasonable application’ standard.” Id. A
habeas petitioner may only secure relief after showing that the state court’s rejection of his
claim was “contrary to, or involved an unreasonable application of, clearly established
Federal law, as determined by the Supreme Court of the United States.” 28 U.S.C. §
2254(d)(1)–(2).
Courts review claims presenting questions of fact under two sections of AEDPA.
First, a federal habeas court presumes the underlying factual determinations of the state
court to be correct, unless the habeas petitioner “rebut[s] the presumption of correctness by
clear and convincing evidence.” 28 U.S.C. § 2254(e)(1); see also Miller-El v. Cockrell,
537 U.S. 322, 341 (2003); Young v. Dretke, 356 F.3d 616, 629 (5th Cir. 2004) (“As a
federal habeas court, we are bound by the state habeas court’s factual findings, both
implicit and explicit.”). Second, a petitioner must show that the state court’s ultimate
decision “was based on an unreasonable determination of the facts in light of the evidence
presented in the State court proceeding.” 28 U.S.C. § 2254(d)(2); see also Miller-El, 537
U.S. at 340. “[A] state-court factual determination is not unreasonable merely because the
federal habeas court would have reached a different conclusion in the first instance.” Wood
v. Allen, 558 U.S. 290, 301 (2010). “Claims presenting mixed questions of law and fact are
reviewed under a combination of these provisions; a state court’s ultimate legal conclusion
is reviewed under Section 2254(d)(1), while the underlying factual findings supporting that
conclusion are reviewed under Sections 2254(d)(2) and (e)(1).” Neal, 78 F.4th at 783.
In performing the AEDPA review, a federal court generally cannot “develop and
consider new evidence.” Shoop v. Twyford, 596 U.S. 811, 819 (2022). AEDPA limits

6

“review of factual determinations under § 2254(d)(2)” to “the evidence presented in the
State court proceeding,” and “review of legal claims under § 2254(d)(1) . . . ‘to the record
that was before the state court.’” Id. (quoting Cullen v. Pinholster, 563 U.S. 170, 181
(2011)). Thus, under AEDPA, “[a] federal court may admit new evidence only in two
limited situations: Either the claim must rely on a ‘new’ and ‘previously unavailable’ ‘rule
of constitutional law’ made retroactively applicable by this Court, or it must rely on ‘a
factual predicate that could not have been previously discovered through the exercise of
due diligence.’” Twyford, 596 U.S. at 812 (quoting 28 U.S.C. § 2254(e)(2)(A)).
Limitations Period
AEDPA provides that claims under 28 U.S.C. § 2254 are subject to a one-year
statute of limitations. See 28 U.S.C. § 2244(d)(1). The limitations period runs from the
latest of four possible events: (1) when “the judgment became final,” (2) when “the
impediment to filing an application created by the State action in violation of the
Constitution and laws of the United States is removed, if the applicant was prevented from
filing by such State action,” (3) when “the constitutional right asserted was initially
recognized by the Supreme Court . . . and made retroactively applicable to cases on
collateral review,” or (4) when “the factual predicate of the claim or claims presented could
have been discovered through the exercise of due diligence.” Id. § 2244(d)(1)(A)–(D).
The concerns in this lawsuit arose after Will had completed a full round of state
review. Will’s conviction became final on November 29, 2004, when the United States
Supreme Court denied his petition for a writ of certiorari. Will v. Texas, 543 U.S. 1004
(2004). Will filed his first state habeas application during the pendency of his direct appeal,
tolling any limitations period until its denial on March 29, 2006. See 28 U.S.C. §

7

2244(d)(2). Under the general timeline for filing a federal petition, the AEDPA limitations
period expired on March 29, 2007. See 28 U.S.C. § 2244(d)(2).
Will, however, argues that the Court should instead apply the more specific
AEDPA subsections which allow the limitations period to run from “the date on which [an]
impediment to filing an application created by State action . . . is removed” or “the date on
which the factual predicate of the claim or claims presented could have been discovered
through the exercise of due diligence.” 28 U.S.C. § 2244(d)(1)(B), (D). Relying on the
Fifth Circuit’s description of his diligence in this case, Will argues that his Brady claim
was either undiscovered or unactionable until he came across the material at issue in
September 2012. Will contends that AEDPA’s one-year clock started on that date, that his
state habeas action tolled the limitations period during its pendency from August 26, 2013,
to November 25, 2015, and that his federal petition filed on November 27, 2015, was
therefore timely.
Respondent, however, contends that “the clock under § 2254(d)(1)(D) does not start
when Will knew of the documents; it starts when the documents were available through
due diligence.” (Dkt. 86 at 31). Respondent, in essence, argues that “the evidence strongly
suggests the documents were available in the State’s file long before” September 2012.
(Dkt. 86 at 32).
Will, however, counters by pointing to language from the Fifth Circuit’s earlier
opinion in this case detailing his diligence. Albeit in the context of whether Will’s diligence
would allow for successive federal review, the Fifth Circuit recognized Will’s efforts to
uncover material:
Objectively, Will has demonstrated that the withheld records could not have
been discovered through due diligence. Here, the prosecutor “pledge[d] to
8

[the] Court” that she would produce all Brady materials prior to trial;
Deputy Strickland, who prepared the Hit Document, appeared in court after
refusing to comply with a subpoena duces tecum and stated that he did not
have any documents pertaining to Will’s case; Will’s habeas counsel had
subpoenaed all inmate records concerning Rosario; an attorney in the DA’s
Office acknowledged that, between Brady and the DA’s open file policy,
the State would have been obligated to disclose the withheld documents
prior to trial; and Will had no exigent reason to know that the Hit Document
or Schifani Report existed. Accordingly, there was no reason for Will or his
counsel to suspect that documents were being withheld or to do more than
they did to uncover the withheld evidence.
Trial counsel need not assume the prosecution may be withholding
information in order to exercise diligence. The Supreme Court has stated
that its “decisions lend no support to the notion that defendants must
scavenge for hints of undisclosed Brady material when the prosecution
represents that all such material has been disclosed,” and trial counsel
should be able to reasonably rely on a prosecutor’s open file policy.
In re Will, 970 F.3d 536, 542–43 (5th Cir. 2020). Will also relies on the affidavit of
his trial prosecutor, Lynne Parsons, who in a 2014 affidavit claimed to “have no
recollection of seeing, or having knowledge of, Rosario’s Separation Review Sheet or
Deputy Shifani’s [sic] memorandum, and I cannot explain how these documents are in the
State’s files today.” (Dkt. 82-33). Ms. Parsons conceded that, “had [she] known of their
existence,” Brady would have obligated her to turn over the documents. (Dkt. 82-33). Ms.
Parsons’ affidavit does not explain when or how the documents came to be in the State’s
file, and it is unclear how Will could have known of their existence if the trial prosecutor
did not.
Respondent asks this Court to assume, contrary to the statements of a State
prosecutor, that Will could have uncovered this material far earlier. But the State of Texas
has been dilatory in fulfilling its constitutional obligation to turn over favorable material in
this case. Even after years of litigation in various courts, the State has only recently

9

divulged material which it had kept hidden in a “work product” folder. The State’s
reticence at disclosing information deprives it of any presumption of good faith.
The evidence shows that Will diligently pursued information in this case and
diligently acted when the evidence in this lawsuit came to light. The Court rejects
Respondent’s argument that AEDPA’s limitations period bars consideration of this
lawsuit.2
Analysis
The substantive issues in this case concern whether the State suppressed favorable
material and how that would have changed Will’s defense. The state court denied Will’s
successive state habeas claim based on the suppressed material. As discussed below, the
state court’s decision was contrary to federal law.
I.

The Constitutional Requirements of Brady
In Brady v. Maryland, 373 U.S. 83 (1963), the Supreme Court held that “the

suppression by the prosecution of evidence favorable to an accused upon request violates
due process where the evidence is material either to guilt or to punishment, irrespective of
the good faith or bad faith of the prosecution.” Id. at 87. “There are three components of a
true Brady violation: (1) the evidence at issue, whether exculpatory or impeaching, must
be favorable to the accused; (2) that evidence must have been suppressed by the State,
either willfully or inadvertently; and (3) prejudice must have ensued.” Canales v. Stephens,

2

The parties engage in significant discussion about whether Will has shown that he is actually
innocent, a factor which would allow for equitable tolling if he had not filed in a timely manner. See
McQuiggin v. Perkins, 569 U.S. 383, 386 (2013) (finding that a plea of actual innocence can overcome the
AEDPA statute of limitations). Because Will filed in a timely manner, the Court does not need to address his
actual-innocence arguments. Will previously tried to litigate a substantive actual-innocence argument in these
proceedings, but the Court found that it did not comply with AEDPA’s stringent limitations of successive
habeas review. (Dkt. 79 at 18).

10

765 F.3d 551, 574 (5th Cir. 2014) (quoting Strickler v. Greene, 527 U.S. 263, 281–82
(1999)).
This case turns on the prejudice component of the Brady analysis, generally framed
as the term “materiality.” The Fifth Circuit has observed that “[t]he Supreme Court has
imposed four criteria for determining whether evidence is material”:
First, materiality does not require the defendant to demonstrate by a
preponderance of the evidence that omitted evidence would have resulted
in acquittal. Second, he need not weigh the withheld evidence against the
disclosed evidence to show he would have been acquitted by the resulting
totality. Third, if evidence is found material, there is no need to conduct a
harmless error analysis. Fourth, the withheld evidence should be considered
as a whole, not item-by-item. Kyles v. Whitley, 514 U.S. 419, 434–37, 115
S.Ct. 1555, 131 L.Ed.2d 490 (1995). The sum of these four guideposts
means that to show a due process violation when the state withholds
evidence, a defendant need not prove that his trial necessarily would have
had a different outcome; a lack of faith in the result is sufficient.
DiLosa v. Cain, 279 F.3d 259, 263 (5th Cir. 2002). Evidence is material “if there is
a reasonable probability that, had the evidence been disclosed to the defense, the result of
the proceeding would have been different.” Kyles, 514 U.S. at 433. “A ‘reasonable
probability of a different result’ is shown when the suppression ‘undermines confidence in
the outcome of the trial.’” Graves v. Dretke, 442 F.3d 334, 340 (5th Cir. 2006) (quoting
Kyles, 514 U.S. at 434).
II.

Deference to the State Court Decision
Will argues that AEDPA’s relegation bar found in section 2254(d) does not

preclude de novo federal review of his Brady claim. “Because Brady claims involve mixed
questions of law and fact,” the Fifth Circuit has held that “§ 2254(d)(1), instead of subpart
(d)(2), is applied.” Floyd v. Vannoy, 894 F.3d 143, 161 (5th Cir. 2018). Will contends that
the state court decision was both contrary to and an unreasonable application of federal

11

law. As discussed below, the Court finds that Will has met his burden under section
2254(d)(1).
A.

The Lower Court Decision

When deciding a case under AEDPA’s deferential scheme, a federal court must
“train its attention” on the “last related state-court decision” that provides a “relevant
rationale” to a particular claim. Wilson v. Sellers, 138 S. Ct. 1188, 1191–92 (2018)
(quotation omitted). A federal court should focus “on the particular reasons—both legal
and factual—why state courts rejected a state prisoner’s federal claims and . . . give
appropriate deference to that decision.” Sheppard v. Davis, 967 F.3d 458, 467 (5th Cir.
2020) (quoting Wilson, 138 S. Ct. at 1192).
Federal review in this case centers on two different opinions. The lower court issued
a recommendation providing a detailed analysis of Will’s Brady claim. The Court of
Criminal Appeals adopted that recommendation—with significant exceptions—and then
denied relief. This Court’s AEDPA review must take into account the decisions rendered
by both the lower court and the Court of Criminal Appeals.
Under Texas procedure, the state district court does not make any final decision in
a capital habeas case. “On post-conviction review of habeas corpus applications, the
convicting court is the ‘original factfinder,’ and [the Court of Criminal Appeals] is the
‘ultimate factfinder.’” Ex parte Lane, 670 S.W.3d 662, 670 (Tex. Crim. App. 2023)
(quoting Ex parte Thuesen, 546 S.W.3d 145, 157 (Tex. Crim. App. 2017)). The lower
court’s role is to prepare findings of fact and conclusions of law for the Court of Criminal
Appeals’ review. The Court of Criminal Appeals will then “defer to and accept a trial
judge’s findings of fact and conclusions of law when they are supported by the record.” Ex

12

parte Storey, 584 S.W.3d 437, 439 (Tex. Crim. App. 2019). The Court of Criminal
Appeals, however, may exercise its authority “to make contrary or alternative findings and
conclusions.” Ex parte Reed, 271 S.W.3d 698, 727 (Tex. Crim. App. 2008).
Here, the lower habeas court adopted the State’s amended proposed findings and
conclusions. The lower habeas court’s recommendation separated Will’s Brady evidence
into three categories and analyzed each independently, making separate factual findings
and legal conclusions for each of the three groups of Brady evidence. The lower habeas
court discussed each of the three evidentiary categories—“Rosario’s Harris County Jail
administrative separation review sheet,” “Deputy Patricia Schifani’s memorandum,” and
“impeachment evidence regarding Cassandra Simmons”—under different headings and in
separate sections. (Dkt. 57-74 at 190, 192, 194). The lower court’s recommendation
followed a pattern. First, the lower court assumed that the State had suppressed or
inadvertently withheld each piece of Brady evidence. (Dkt. 57-74 at 190, 192, 194).
Second, the lower court found that “the State did not violate Brady because [Will] fails to
demonstrate, by a preponderance of the evidence that the document was favorable.” (Dkt.
57-74 at 190, 191–92, 194). Third, the lower court found that Will “fails to demonstrate,
by a preponderance of the evidence, that” each piece of evidence individually “was
material.” (Dkt. 57-74 at 190–91, 192–93, 194–95).
In short, the lower court’s analysis did not intermix its consideration of each claim.
The lower court considered each piece of evidence separately, assumed that each was
suppressed, but decided that all three were not favorable or material. The lower court did
not consider the effect of the three items of suppressed evidence as a whole.

13

But the lower court did not necessarily have to consider the collective effect of the
material. A habeas petitioner is entitled to relief only if he proves every element of the
Brady analysis. If a habeas petitioner fails to establish any element of Brady, a reviewing
court need not discuss the other elements. See United States v. Runyan, 290 F.3d 223, 245
(5th Cir. 2002). Because the lower court found that each piece of evidence was not
favorable, it did not need to continue to a full materiality analysis.
B.

The Court of Criminal Appeals Decision

The lower court forwarded the case to the Court of Criminal Appeals on January
26, 2015. (Dkt. 57-74 at 200–01). Without any additional briefing from the parties, the
Court of Criminal Appeals denied relief on November 25, 2015. The Court of Criminal
Appeals decided the case “based upon the trial court’s amended findings and conclusions
and [its] own review.” Id. at *1. In doing so, the Court of Criminal Appeals did not adopt
all of the lower court’s findings and conclusions. Specifically, the Court of Criminal
Appeals refused to adopt the findings that the jail review sheet and the Cassandra Simmons
evidence were not favorable. Ex parte Will, 2015 WL 13388366, at *1 (Tex. Crim. App.
2015). The Court of Criminal Appeals’ decision thus apparently left the status of Will’s
Brady claim as follows: the Schifani memorandum itself was possibly not favorable3 and
each piece of evidence was not material when considered individually.
This matter is not clear. The lower court’s finding of fact number 76 found, “based on the trial and
habeas records, that [Will] does not establish that the alleged newly discovered evidence—the 2000 Schifani
memorandum—is either favorable or material with respect to the primary case due to the inaccurate manner
in which [he] represents the content of the Schifani Memorandum and the fact that the memorandum is
cumulative of evidence previously rejected by the jury at trial.” (Dkt. 57-74 at 178). The Court of Criminal
Appeals “decline[d] to adopt Factual Findings Number[] 76 . . . to the extent [it] states that the evidence is
not favorable under Brady.” Will, 2015 WL 13388366, at *1. But the lower court had also issued a legal
conclusion—number 4—that Will “fail[ed] to demonstrate, by a preponderance of the evidence, that the
[Schifani memorandum] was favorable.” (Dkt. 57-74 at 191). The Court of Criminal Appeals decision left
conclusion number 4 intact. With the other two pieces of Brady material, the Court of Criminal Appeals had
declined to adopt both the finding of fact and the legal conclusion relating to favorableness. It is unclear why
the Court of Criminal Appeals would adopt the favorability legal conclusion relating to the Schifani
3

14

The Court of Criminal Appeals’ decision turned on the materiality discussion. As
previously discussed, the Court of Criminal Appeals could only comply with the
constitutional requirements set out by the Supreme Court in Kyles by assessing materiality
“collectively, not point by point.” Banks, 583 F.3d at 328. The issues presently under
consideration turn on whether the Court of Criminal Appeals engaged in the appropriate
analysis.
C.

Alleged Errors in the State Court Decision

The parties’ briefing assumes that Will has met Brady’s suppression and
favorability elements. Specifically, Respondent does not argue that the three pieces of
evidence considered by the state courts were not suppressed or not favorable. Relying on
section 2254(d)(1)’s contrary-to prong, Will first argues that the Texas courts made at least
three critical errors contrary to established Supreme Court precedent: (1) the state courts
applied the wrong standard of Brady materiality; (2) the state courts applied a ‘sufficiency
of the evidence’ test the Supreme Court has explicitly rejected; and (3) the state courts
failed to consider the cumulative effect of the Brady evidence Will presented. (Dkt. 83 at
6). The Court will address each of these arguments.
D.

The Standard Used by the State Court

Brady’s familiar materiality standard requires a habeas petitioner to show that
“there is a reasonable probability that, had the evidence been disclosed, the result of the
proceeding would have been different.” Cone v. Bell, 556 U.S. 449, 469–70 (2009). Will,

memorandum but decline to adopt the factual finding on which the lower court had based its conclusion. It
appears that this may be a drafting error, but the parties have not briefed that point. Respondent’s answer
does not argue that Will has failed to meet Brady’s favorability prong with regard to the Schifani
memorandum. The parties must provide briefing on the question of whether the lower court’s legal
conclusion number 4 has any effect in these proceedings.

15

however, argues that the state court rendered a decision contrary to federal law because it
relied on a preponderance-of-the-evidence standard and employed a sufficiency-of-theevidence test. For the reasons set forth below, the Court agrees that the state decision was
contrary to federal law.
1. Preponderance-of-the-Evidence Standard
It is clearly established that evidence is material under Brady “if there is a
reasonable probability that, had the evidence been disclosed to the defense, the result of
the proceeding would have been different. Kyles v. Whitley, 514 U.S. 419, 433 (1995)
(quoting United States v. Bagely, 473 U.S. 667, 682 (1985) (opinion of Blackmun, J.)).
That is, Brady materiality is decisively not a preponderance-of-the-evidence test. Id. at 434
(“[A] showing of materiality does not require demonstration by a preponderance that
disclosure of the suppressed evidence would have resulted ultimately in the defendant's
acquittal.”). Yet, the lower court repeatedly included “preponderance-of-the-evidence”
language throughout its opinion. With regard to each category of evidence, the state court
held that Will “fail[ed] to demonstrate, by a preponderance of the evidence, that” each
piece of evidence “was material.” (Dkt. 57-74 at 190–91, 192–93, 194–95). At no point did
the state court identify the correct reasonable-probability standard mandated by Kyles.
(Dkt. 57-74).
Respondent argues that the inclusion of the preponderance-of-the-evidence
language is merely a reflection of traditional Texas habeas corpus law. Texas courts have
long held that “[a]n applicant for a post-conviction writ of habeas corpus bears the burden
of proving her claim by a preponderance of the evidence.” Diamond v. State, 613 S.W.3d
536, 545-46 (Tex. Crim. App. 2020); Ex parte Torres, 483 S.W.3d 35, 43 (Tex. Crim. App.

16

2016); Ex parte Richardson, 70 S.W.3d 865, 870 (Tex. Crim. App. 2002); Ex parte
Thomas, 906 S.W.2d 22, 24 (Tex. Crim. App. 1995). It is not improper for a state court to
apply this general preponderance of evidence standard to a petitioner’s factual contentions.
See Holland v. Jackson, 542 U.S. 649, 654 (2004). However, the legal issue of Brady
materiality must be decided under the reasonable-probability standard. Kyles, 514 U.S. at
434. Consequently, while many Texas habeas courts began their legal analyses with a
statement of the generally-applicable preponderance standard, where, as here, a more
specific legal standard applies to a portion of the habeas analysis, Texas habeas courts
identify and apply the more specific standard. See, e.g., Ex parte Reed, 670 S.W.3d 689,
763 (Tex. Crim. App. 2023); Diamond, 613 S.W.3d at 545–46; Ex parte Torres, 483
S.W.3d at 43; Ex parte Richardson, 70 S.W.3d at 870.
The Supreme Court endorsed this approach in Holland. There, the state court’s
opinion included a statement of the general preponderance-of-the-evidence standard, as
well as the more specific reasonable-probability standard. Holland, 542 U.S. at 654. The
habeas petitioner pointed to the state court’s preponderance-of-the-evidence language to
argue that it applied the wrong standard in assessing the prejudice prong of his ineffective
assistance of counsel claim. Id. The Supreme Court rejected the petitioner’s argument,
holding that the state court’s general statement of the preponderance standard did not
necessarily imply that it had applied the incorrect legal standard. See id. at 654–55.
The language in the state court opinion here is readily distinct from that in Holland.
In Holland, the only explicit invocation of the preponderance-of-the-evidence standard was
a statement that “[i]n a post-conviction proceeding, the defendant has the burden of proving

17

his allegations by a preponderance of the evidence.” Id. at 654 (internal citation omitted).4
In contrast, the state court here held that “the applicant’s Brady claim is meritless because
the applicant fails to demonstrate, by a preponderance of the evidence, that [the withheld
evidence] was material.” (Dkt. 57-74 at 190). Unlike the statement from Holland, this
statement cannot be “reasonably read as addressing the general burden of proof in
postconviction proceedings with regard to factual contentions.” Holland, 542 U.S. at 654.
Rather, it is a specific application of the preponderance standard to the materiality prong
of Will’s Brady claim. Further, the Supreme Court’s conclusion in Holland relied in large
part on the fact that the state court had elsewhere explicitly identified the correct
reasonable-probability standard for prejudice. Id. at 654–55; see also Woodford v. Visciotti,
537 U.S. 19, 23–25 (2002) (state court’s use of the word “probable” rather than “reasonably
probable” in three instances did not mean it erroneously applied preponderance-of-theevidence standard where it twice stated correct “reasonable probability” standard).
Meanwhile, the state court presiding over Will’s habeas case never identified the correct
standard for materiality.
While the Court “presum[es] that state courts know and follow the law,” Woodford,
537 U.S. at 24, the Court will not “assume a state court applied legal rules it did not,” Salts
v. Epps, 676 F.3d 468, 480 (5th Cir. 2012). The text of the state court decision in this case

4

The lower court in Holland also held that the following two statements implicitly evinced an application of
the preponderance standard: (1) “it is asking too much that we draw the inference that the jury would not
have believed Hughes at all had Melissa Gooch testified”; and (2) responded had “failed to carry his burden
of proving that the outcome of the trial would probably have been different but for those errors.” 542 U.S. at
654–55. The Supreme Court held that, to the contrary, these statements did not imply that the state court
applied a preponderance-of-the-evidence standard, explaining that (1) the first statement “does not imply any
particular standard of probability”; and (2) “use of the unadorned word ‘probably’ is permissible shorthand
when the complete Strickland standard is elsewhere recited.” Id. The Holland Court’s reasoning related to
these two additional statements is not instructive here; Will does not point to any analogous statements in the
state court’s opinion in his case.

18

clearly indicates that the court utilized the general preponderance-of-the-evidence habeas
standard instead of the reasonable-probability standard that Kyles requires.
Accordingly, the Court concludes that Will has shown that the state court’s
evaluation of Brady materiality involved an unreasonable application of clearly established
federal law. This reading comports with how others have interpreted similar language in
state habeas decisions. See, e.g., Long v. Hooks, 972 F.3d 442, 458–60 (4th Cir. 2020), as
amended (Aug. 26, 2020) (finding that a state court applied a standard contrary to clearly
established federal law where it analyzed Brady materiality under a preponderance-of-theevidence standard); Carusone v. Warden, N. Cent. Corr. Inst., 966 F.3d 474, 479–80 (6th
Cir. 2020) (granting habeas relief where the state court “effectively replaced the Kyles
standard with one more favorable to prosecution”); Bailey v. Rae, 339 F.3d 1107, 1118
(9th Cir. 2003) (determining that state court’s application of “more probable than not”
rather than “reasonable probability” standard in Brady analysis was contrary to clearly
established Supreme Court precedent).
The state court’s misapplication of the preponderance-of-the-evidence standard
means that, upon further briefing on the issues, this Court will assess the materiality of
Will’s Brady claims de novo. However, for completeness, it is appropriate to address
separately Will’s additional argument related to the state court’s Brady materiality analysis.
2. Sufficiency-of-the-Evidence Standard
Will contends that the state court considered whether the evidence was sufficient
to sustain the jury’s verdict and ignored the correct materiality standard. The Supreme
Court has instructed that Brady materiality “is not a sufficiency of evidence test.” Kyles,
514 U.S. at 434. As the Supreme Court “made clear in Kyles, the materiality inquiry is not

19

just a matter of determining whether, after discounting the inculpatory evidence in light of
the undisclosed evidence, the remaining evidence is sufficient to support the jury’s
conclusions.” Strickler, 527 U.S. at 290 (citing Kyles, 514 U.S. at 434–35). The Fifth
Circuit has explained that “[i]t is not the role of a court applying Brady to weigh the existing
evidence against the excluded evidence” when determining materiality. DiLosa, 279 F.3d
at 264. Instead, a court must “ask whether the excluded evidence ‘could reasonably be
taken to put the whole case in such a different light as to undermine confidence in the
verdict.’” DiLosa v. Cain, 279 F.3d 259, 264 (5th Cir. 2002) (quoting Kyles, 514 U.S. at
435).
Will asserts that the state court’s decision committed that error by “improperly
weigh[ing] each piece of Mr. Will’s Brady evidence against the prosecution’s inculpatory
trial evidence to determine its materiality.” (Dkt. 83 at 18). The Brady materiality analysis
does not exist in isolation from the State’s trial evidence. The Supreme Court has indicated
that evidence “may not be material if the State’s other evidence is strong enough to sustain
confidence in the verdict.” Smith v. Cain, 565 U.S. 73, 76 (2012); see also United States v.
Agurs, 427 U.S. 97, 111–12 (1976). A court cannot ignore the relationship between the
Brady evidence and what happened at trial. “The materiality of Brady material depends
almost entirely on the value of the evidence relative to other evidence mustered by the
state.” United States v. Sipe, 388 F.3d 471, 478 (5th Cir. 2004) (quotation omitted). In fact,
“a thorough examination of the trial evidence . . . provides a starting point for assessing []
materiality.” Banks v. Thaler, 583 F.3d 295, 321 (5th Cir. 2009). What a court may not do,
however, is ask whether “after discounting the inculpatory evidence by the undisclosed
evidence” if “there would be enough evidence to sustain the conviction.” Graves v. Dretke,

20

442 F3d 334, 340 (5th Cir. 2006). The focus is on “whether in [the] absence [of the Brady
material] he received a fair trial.” Kyles, 514 U.S. at 434.
Will argues that the state court applied the materiality prong in a way that
functionally amounted to the disavowed review. The state-court decision, however, did not
assess whether, considering the Brady evidence, the trial evidence was still sufficient to
allow for his conviction. Instead, the state court placed the suppressed evidence into the
broader context of the trial record to determine its impact. The state court decision engaged
in a review permitted by the Supreme Court when it placed the Brady evidence into the
context of Will’s trial.
E.

Cumulative Effect

A court considering a Brady claim must look at the suppressed evidence
“collectively, not item by item,” Kyles, 514 U.S. at 436, and decide whether the cumulative
evidence “undermines confidence in the outcome of the trial.” Bagley, 473 U.S. at 678; see
also Wearry v. Cain, 577 U.S. 385, 394 (2016) (“[T]he state postconviction court
improperly evaluated the materiality of each piece of evidence in isolation rather than
cumulatively.”); Crawford v. Cain, 248 F. App’x 500, 504 (2007) (“[T]hough the court
may have to go over each piece of evidence item by item, it must ultimately evaluate the
cumulative effect of the evidence for purposes of materiality.”); United States v. Sipe, 388
F.3d 471, 478 (5th Cir. 2004) (“When there are a number of Brady violations, a court must
analyze whether the cumulative effect of all such evidence suppressed by the government
raises a reasonable probability that its disclosure would have produced a different result.”).
It is not uncommon for courts to “discuss the significance of each piece of evidence
in turn, then consider the importance of the evidence in combination.” Lesko v. Secretary

21

Pennsylvania Department of Corrections, 34 F.4th 211, 230 (3d Cir. 2022). Here, the lower
court reviewed each piece of evidence individually and found that it was not favorable and
not material. This initial item-by-item review itself “is not inconsistent with a cumulative
analysis. Indeed, the only way to evaluate the cumulative effect is to first examine each
piece standing alone.” Allen v. Sec’y, Fla. Dep’t of Corr., 611 F.3d 740, 748 (11th Cir.
2010); see also Kyles, 514 U.S. at 436 n.10 (“We evaluate the tendency and force of the
undisclosed evidence item by item; there is no other way. We evaluate its cumulative effect
. . . separately and at the end of the discussion.”); Maharaj v. Sec’y, Dep’t of Corr., 432
F.3d 1292, 1310 (11th Cir. 2005) (explaining that the “appropriate methodology [involves]
considering each Brady item individually, and only then making a determination about the
cumulative impact”); Miller v. Dretke, 431 F.3d 241, 251 (5th Cir. 2005) (discussing the
item-by-item and then cumulative review mentioned in Kyles).
But the item-by-item review alone is not enough. Because the Court of Criminal
Appeals found that at least two of the pieces of evidence were suppressed and material, it
had a constitutional obligation to consider the effect of that evidence as a whole. See
Crawford v. Cain, 248 F. App’x 500, 504 (5th Cir. 2007) (“[T]hough the court may have
to go over each piece of evidence item by item, it must ultimately evaluate the cumulative
effect of the evidence for purposes of materiality.”). The Court of Criminal Appeals’
analysis stopped short and never considered the effect of the Brady violations collectively.
Respondent admits that the Court of Criminal Appeals “did not expressly undertake
‘cumulative’ analysis of all the allegedly suppressed evidence” but nonetheless assumes
that it silently had the correct standard in mind and wordlessly applied it in this case. (Dkt.
86 at 81). Respondent relies heavily on Neal v. Puckett, 286 F.3d 230, 246 (5th Cir. 2002)

22

and its progeny for the proposition that the Court should analyze the ultimate decision of
the state court regardless of its specific analysis.
The matter at issue, however, is whether the state court decision was contrary to
federal law. “Neal does not speak to the standard of review where a state court applies
erroneous law.” Salts, 676 F.3d at 479. When considering a case under AEDPA’s “contrary
to” prong, a court does not “assume a state court applied legal rules it did not, and then ask
whether such rules could still reasonably support the result. That reading . . . would run
afoul of the Supreme Court’s command that, where a state court does not apply a legal test,
‘our review is not circumscribed by a state court conclusion.’” Salts, 676 F.3d at 479–80;
see also Wiggins v. Smith, 539 U.S. 510, 534 (2003) (holding that, where state court
analyzed only inadequate performance prong of Strickland, review of prejudice prong was
de novo); Rompilla v. Beard, 545 U.S. 374, 390 (2005) (“Because the state courts found
the representation adequate, they never reached the issue of prejudice . . . , and so we
examine this element of Strickland claim de novo.”).
Respondent relies on another Fifth Circuit case, Crawford v. Cain, 248 F. App’x
500, 506 (5th Cir. 2007), to argue that this Court should read adherence to constitutional
law in the Court of Criminal Appeals’ silence. Respondent argues:
In Crawford v. Cain, the Fifth Circuit addressed a substantially similar
claim that the state court “did not specifically quote the relevant language
from Kyles on the question of cumulative impact.” 248 F. App’x 500, 506
(5th Cir. 2007). The Court noted that “[t]he state appellate court cited the
relevant excerpts from Brady and quoted at length from Kyles and other
cumulative review cases.” Id. It thus concluded that, “[t]hough the court did
not cite the precise language we have to signal its cumulative review, it is
plain to us that the inquiry was conducted properly . . . Taken as a whole,
the evidence does not support Crawford’s contention that the state court was
objectively unreasonable in concluding that the suppressed evidence was
immaterial.” Id.

23

Crawford is particularly instructive here. The CCA’s written decision
clearly recited and applied the correct prongs of Brady. See SHCR-03 Supp.
At 718 ¶ 57 (listing the three Brady elements). It also correctly identified
materiality as the correct standard for measuring harm. Id. at 718 ¶ 57, 726
¶ 76, 729–30 ¶ 89, 738 ¶ 3, 740–41 ¶ 5, 743–44 ¶ 7. Similarly, “though the
[CCA] did not cite the precise language” to “signal its cumulative review”
it is plain that “the inquiry was conducted properly.” Crawford, 248 F.
App’x at 506.
(Dkt. 86 at 92–93). Respondent’s argument, however, highlights the reasons for
which an assumption that the Court of Criminal Appeals engaged in a cumulative review
is unwarranted. The lower court’s decision cited the three Brady prongs, but did not once
cite Kyles or mention the relevant cumulative-review cases. The lower court then sectioned
off its analysis of each piece of Brady evidence individually. This itself was not an error.
A state court properly “evaluates the tendency and force of the undisclosed evidence item
by item,” but it then must follow by “evaluat[ing] its cumulative effect . . . separately and
at the end of the discussion.” Kyles, 514 U.S. at 436 n.10. Nothing in the lower court
decision suggested that it took the final step and considered the material as a whole.
Unlike in Crawford, the Court of Criminal Appeals did not cite any relevant excerpt
from Brady, did not quote Kyles, and did not mention any cumulative review cases. The
lower court had expressly considered each piece of Brady material individually, and
nothing in the Court of Criminal Appeals’ decision suggested any other type of review.
The state court’s failure to complete the Brady materiality analysis was contrary to federal
law.
In conclusion, this is not a case where a court merely used imprecise or incomplete
language when rendering its decision. This is a case where the state court omitted a crucial
component of the review required by Supreme Court precedent. The state court’s decision
that was based only on individual materiality determinations was “contrary to” clearly
24

established federal law. See Woods v. Smith, 660 F. App’x 414, 432 (6th Cir. 2016) (“[T]he
state court’s assumption that Brady requires only individual materiality assessments was
‘contrary to’ clearly established federal law.”); Barker v. Fleming, 423 F.3d 1085, 1094
(9th Cir. 2005) (finding a decision contrary to federal law when “[d]espite the clarity of the
Supreme Court’s directive that a materiality analysis include an assessment of the
cumulative effect of undisclosed evidence, the Washington Supreme Court did not conduct
such an analysis”); Castleberry v. Brigano, 349 F.3d 286, 291–92 (6th Cir. 2003).
F.

Conclusion of AEDPA review

Because the state court’s decision was contrary to federal law, the Court need not
decide whether it was also an unreasonable application of that law. See Salts, 676
F.3d at 479–80 (where the “state court applies erroneous law,” the petitioner “need not also
show that the state appeals court’s decision involved an ‘unreasonable application’ of such
law”). When a habeas petitioner shows that the state-court decision was contrary to federal
law, “a federal court will be unconstrained by § 2254(d)(1).” Williams v. Taylor, 529 U.S.
362, 406 (2000). Will is entitled to de novo review of his Brady claim.
III.

Factual Development and Lingering Questions
Will asks the Court to allow factual development before proceeding to adjudication

of his Brady claim, seeking “an opportunity to specifically set out what items of discovery
and testimony of witnesses will further development of the merits of his Brady claim.”
(Dkt. 92 at 46). Consistent with Will’s request, the Court directs Will to submit, within
thirty (30) days of the entry of this Order, a motion for discovery or evidentiary hearing
which lists, with specificity, what material he wishes to discover and what testimony he

25

seeks to present in any evidentiary hearing. The Court will defer ruling on the merits of
Will’s Brady claim until the parties brief whether factual development is necessary.
In the briefing which follows, the Court also expects the parties to address the
below-described issues.
A.

Issue 1: What Remains from the State Habeas Court’s Decision?

Will has shown that the state habeas court’s decision relating to materiality was
contrary to federal law. But a federal court does not necessarily toss the baby out with the
bathwater when the reasoning in a state court decision is incomplete. The state habeas court
made extensive factual findings which may support a materiality analysis. The parties will
brief, with specificity, what remnants of the state habeas court’s findings and conclusions
require deference under AEDPA. The parties’ briefs should include a discussion of whether
the lower court’s legal conclusion number four (i.e., that Will failed to demonstrate that
the Schifani memorandum was favorable) (Dkt. 57-74 at 191) has any effect in these
proceedings. See supra note 3.
B.

Issue 2: What May the Court Consider in the Brady Materiality

Review?
With the Court’s holding that the state-court decision was contrary to federal law,
federal review will consider the question of Brady materiality de novo. Will has presented
a factually rich narrative in support of his Brady claim which includes extensive evidence
and argument that far exceeds that presented at trial. Will links some of the evidence
directly to the State’s suppression of evidence. For example, Will argues that “[d]iscovery
of the Hit Document led to the new testimony of Mr. Rosario’s would-be hitman, David
Cruz.” (Dkt. 82 at 52). Other evidence—such as Will’s reassessment of the forensic

26

evidence, reinterpretation of the timetable of events, and reliance a K-9 report—does not
appear to flow directly from the suppressed evidence.
Will’s successive proceedings have sought to advance two separate claims: his
Brady claim and an actual-innocence claim. Actual-innocence claims cover broad ground.
Actual innocence encompasses an all-inclusive view of all evidence amassed over time.
See House v. Bell, 547 U.S. 518, 538 (2006) (“Schlup makes plain that the habeas court
must consider all the evidence, old and new, incriminating and exculpatory, without regard
to whether it would necessarily be admitted under rules of admissibility that would govern
at trial.”). The Fifth Circuit did not authorize Will to proceed on his actual-innocence claim.
Echoes of the sweeping actual-innocence review, however, permeate Will’s briefing on the
Brady arguments. Will argues that, when conducting the materiality assessment of the
claims he exhausted in state court, the Court should take into account significant evidence
he has developed after trial (and which he hopes yet to develop in federal court). The
parties’ briefing does not neatly separate the breadth of Brady’s materiality review from
the sweeping review in actual-innocence claims. The parties must define the parameters of
Brady’s materiality review.
The Brady materiality question has a sharper focus than the actual-innocence
inquiry. The Supreme Court has described the Brady materiality issue as “legally simple
but factually complex.” Turner v. United States, 582 U.S. 313, 324–25 (2017). The
Supreme Court has instructed that a court “must examine the trial record, ‘evaluat[e]’ the
withheld evidence ‘in the context of the entire record,’ and determine in light of that
examination whether ‘there is a reasonable probability that, had the evidence been

27

disclosed, the result of the proceeding would have been different.’” Id. (quoting United
States v. Agurs, 427 U.S. 97, 112 (1976) and Cone v. Bell, 556 U.S. 449, 470 (2009)).
The Brady materiality analysis is not a retrial with new evidence, new testimony,
and new defensive theories. Even so, the Fifth Circuit has previously considered whether
the disclosure of suppressed evidence would have led to other exculpatory evidence. See,
e.g., United State v. Sipe, 388 F.3d 471, 485 (5th Cir. 2004); Felder v. Johnson, 180 F.3d
206, 212 (5th Cir. 1999). The relationship between the suppressed material and other
evidence, however, cannot be “mere speculation.” See Wood v. Bartholomew, 516 U.S. 1,
7 (1995). Some courts ask if the Brady evidence would “lead directly” to the other evidence
on which a habeas petitioner relies. Barton v. Warden, Southern Ohio Correctional
Facility, 786 F.3d 450, 465 (6th Cir. 2015) (quotation omitted); see also East v. Scott, 55
F.3d 996, 1004 (5th Cir. 1995) (asking whether the Brady material “would have led to” the
petitioner’s other evidence).
The Supreme Court has instructed courts to evaluate materiality “in the context of
the entire record.” United States v. Agurs, 427 U.S. 97, 112 (1976). The parties must
explain what “the entire record” means in this case. The parties’ briefing has put forward
a factually dense narrative that, in many ways beyond just the material which the State
suppressed, differs from the case put before jurors. The parties must brief whether, under
a de novo materiality review, the Court may do more than plug the suppressed material into
the testimony, evidence, and defensive theories that came before jurors. The parties must
specify what the Court may consider when adjudicating Brady’s materiality prong,
particularly in contrast to the parties’ extensive briefing on Will’s actual-innocence
arguments.

28

C.

Issue 3: May the Court Consider Newly Turned-Over Material in this

Action?
The State of Texas turned over material from the prosecutor’s work product folder
after the Fifth Circuit had authorized successive proceedings on discrete issues. The
extremely late disclosure of information has placed Will in an awkward procedural posture.
Will likely cannot insert a new claim into these proceedings.5 Will argues that the longsuppressed material merely supplements his federal claim. Yet, as discussed above, it is
not clear what role evidence beyond the trial record and specific evidence underlying a
claim plays into Brady’s materiality review.
This is not the first case which has made apparent that the Harris County District
Attorney’s Office has used a “work product folder” as a means of preventing criminal
defendants from litigating on a level playing field. See, e.g., Prible v. Davis, 2020 WL
2563544, at *25 n.21 (S.D. Tex. 2020) (finding that “the work-product doctrine cannot
excuse [the Harris County District Attorney’s Office] efforts to hide relevant and
exculpatory information from the defense”), vacated sub nom. Prible v. Lumpkin, 43 F.4th
501 (5th Cir. 2022). The State of Texas withheld evidence from Will for more than two
decades throughout extensive litigation. Had the State complied with its constitutional
obligations, Will could have used the new evidence at trial, relied on it in his first two state
habeas applications, presented it in his initial federal petition, and included it in his motion
for leave to file a successive petition. Respondent gives no explanation for withholding the
The mandate rule generally restricts this Court’s review to the matters sent down from the circuit
court. See United States v. Lee, 358 F.3d 315, 321 (5th Cir. 2004) (“[T]he mandate rule compels compliance
on remand with the dictates of a superior court and forecloses relitigation of issues expressly or impliedly
decided by the appellate court.”). Yet, it does not bar claims based on newly discovered evidence. See Webb
v. Davis, 940 F.3d 892, 897 (5th Cir. 2019). The more-specific AEDPA successive-petition provisions bar
the introduction of new habeas claims into these proceedings. See 28 U.S.C. § 2244(b)(3)(A).
5

29

material until this late stage of the proceedings. The State of Texas’ failure to honor the
constitutional requirements of the Brady decision is deeply troubling.
The State of Texas revealed material only when the procedural posture was such
that Will could not make use of it in his initial or successive state habeas proceedings.
Respondent blames the current procedural entanglement on Will. Respondent says that
“Will himself has chosen to bring the new evidence in a posture that renders it
unreviewable.” (Dkt. 86 at 110 n.29). Respondent goes on to assert that “[t]here was no
rule stopping [Will] from filing a third federal petition that would run concurrently with
his second petition,” even though “there would undoubtedly be procedural hurdles, such as
seeking authorization to file another successive petition.” (Dkt. 86 at 110, n.29).
Respondent proposes a procedure that further complicates the proceedings and wastes
judicial resources, particularly when the newly disclosed material covers the same themes
as the evidence properly before the Court. 6
The briefing to date does not provide sufficient clarity as to what role the newly
disclosed evidence may have in these proceedings. The parties must provide additional
briefing, supported by legal precedent, describing what role newly turned over or newly
developed evidence could have in the Brady materiality analysis. With that briefing, the
parties should discuss whether it may be appropriate to stay any de novo adjudication of
the properly presented Brady claim to allow the state courts to consider the newly divulged
evidence in the first instance.
D.

Issue 4: Do the Circumstances Warrant In Camera Review of the Work
Product Folder?

Respondent “agrees that work-product evidence first turned over by the State in June 2022 would
not be barred by § 2254(e)(2).” (Dkt. 86 at 106).
6

30

When the State recently turned over material which had recently been withheld, it
still did not disclose the full contents of the work product folder. The State has turned over
some redacted material during these proceedings which Will uses to supplement his federal
claim. Will has asked for discovery of the work product folder, possibly in camera. If Will
resumes his request for disclosure of material remaining in the prosecutor’s work product
folder, he should supply the Court with a proposed discovery order which will provide for
in camera review. In addition, the parties will provide the Court with a detailed explanation
of what material has already been turned over from that folder, to preclude redundancy.
Conclusion
For the reasons discussed above, the Court finds that the state court’s decision
regarding Will’s Brady claim was contrary to federal law. Will is entitled to de novo review
of the materiality of his Brady evidence.
The Court expresses concern about the time it has taken to arrive at this point in the
litigation. Over two decades have passed since Will’s conviction. It took the State over ten
years following Will’s conviction to turn over the relevant Brady material. The Court
expresses concern about any future delay caused by the government’s failure to comply
with its constitutional obligations.
Will shall file any motion for factual development within thirty (30) days from the
entry of this Order. Respondent may file any opposition within thirty days (30) thereafter.
Will may file a reply within fourteen (14) days. The Court expects that the parties’ briefing
will address the concerns raised in his Order. The Court may extend these deadlines upon
a sufficient showing of good cause.
IT IS SO ORDERED.

31

SIGNED at Houston, Texas on this the 3rd of April, 2024.

___________________________________
KEITH P. ELLISON
UNITED STATES DISTRICT JUDGE

32