by David Kim

The Supreme Court of New Jersey unanimously held that the materiality standard governing a claim under Brady v. Maryland, 373 U.S. 83 (1963), is not the same as the materiality standard governing a motion for a new trial based on newly discovered evidence under State v. Carter, 426 A.2d 501 (N.J. 1981). Evidence that gives rise to a “reasonable probability” of changing a jury’s verdict satisfies a lower threshold than evidence that would “probably” change it, the Court explained, and it clarified that cross quotations in three of its own prior decisions had blurred the two standards. Applying Brady, the Court concluded that the defendant had not established materiality and affirmed the judgment of the Appellate Division as modified.

Background

Yusef B. Allen was tried for the October 15, 1997, murder of Lannie Silver, Jr. The State’s principal witness, Ruby Waller, testified that Allen and Ben McNeil came out of a house where she had bought crack cocaine and yelled at Silver to leave when he asked to buy drugs and that Allen went inside and returned carrying a gun. A jury convicted Allen of murder and two weapons offenses, and he was sentenced to life imprisonment with 30 years of parole ineligibility. His conviction was affirmed on direct appeal.

Waller also testified to cocaine possession convictions in 1990 and 1991 and to a June 21, 1998, shoplifting arrest, and she denied expecting any benefit on that pending matter for her testimony. No charge was pending when she gave police a formal statement on October 27, 1997, identifying Allen as the man she had seen holding a gun.

In 2013, Allen sought federal habeas relief, arguing that the prosecution violated Brady and Giglio v. United States, 405 U.S. 150 (1972), by failing to disclose Waller’s April 19, 1991, plea agreement, under which she received a three-year prison term for testifying truthfully at a co-defendant’s trial. The U.S. District Court denied Allen’s request for a stay to exhaust the Brady claim in state court, finding the claim “plainly meritless.” Allen then moved in state court for a new trial based on newly discovered evidence, and assigned counsel brought that motion under Carter. The motion court denied it, concluding that an agreement to testify against a different co-defendant eight years earlier had no relevance to Allen’s case.

The Appellate Division affirmed in part and remanded in part, holding that the motion court erred by proceeding under Carter rather than Brady but that the error was harmless because it saw “no practical difference between the materiality/harmless error elements set forth in the Brady and Carter formulations.” The Supreme Court granted certification limited to the proper measure of Brady materiality, whether the Carter standard applies, and whether this record meets it.

Analysis

The Court reviewed de novo the question of which legal standard governs a Brady claim. Whether evidence is material under Brady is a mixed question of law and fact. A reviewing court defers to supported trial-court factual findings but reviews de novo the application of the governing legal rule to those facts. State v. Marshall, 690 A.2d 1 (N.J. 1997); State v. Harris, 859 A.2d 364 (N.J. 2004). Brady 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,” and the rule was later extended to impeachment evidence, to evidence known to police investigators, and to cases in which the defense makes no request. Strickler v. Greene, 527 U.S. 263 (1999). The Court stated that a Brady claim thus has three elements: (1) the evidence must be “favorable to the accused, either because it is exculpatory, or because it is impeaching,” (2) it “must have been suppressed by the State, either willfully or inadvertently,” and (3) “prejudice must have ensued.” Strickler; State v. Martini, 734 A.2d 257 (N.J. 1999). That third element, prejudice, is what courts term materiality, the Court explained.

It then traced the evolution of that requirement. Brady announced no test for materiality, and United States v. Agurs, 427 U.S. 97 (1976), clarified only that a defendant whose favorable evidence the prosecutor withheld need not bear the severe burden applicable to newly discovered evidence. Strickland v. Washington, 466 U.S. 668 (1984), explained that the newly discovered evidence standard presupposes an otherwise accurate and fair proceeding and therefore does not fit Brady claims. Prejudice requires a reasonable probability of a different result, “a probability sufficient to undermine confidence in the outcome.” The U.S. Supreme Court applied that formulation to prosecutorial nondisclosure in United States v. Bagley, 473 U.S. 667 (1985) (evidence is material “only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different”), a standard New Jersey adopted in State v. Knight, 678 A.2d 642 (N.J. 1996). Materiality does not require a showing “by a preponderance” that disclosure would have produced an acquittal. Instead, the question is whether the defendant received “a trial resulting in a verdict worthy of confidence.” Kyles v. Whitley, 514 U.S. 419 (1995); Strickler.

Carter Standard for Newly Discovered Evidence

In contrast, Carter governs the situation in which evidence is discovered after trial from a source other than the prosecution. A new trial under Carter requires evidence that is “material to the issue and not merely cumulative or impeaching or contradictory,” was “discovered since the trial and not discoverable by reasonable diligence beforehand,” and is “of the sort that would probably change the jury’s verdict if a new trial were granted,” which is a test Carter itself characterized as “more stringent” than the “Brady analysis,” the Court explained. Its third prong is based on authority more than a century old. State v. Bunk, 73 A.2d 245 (N.J. 1950); Christie v. Petrullo, 128 A. 853 (N.J. Sup. Ct. 1925); Paradise v. Great Eastern Stages, Inc., 176 A. 711 (E. & A. 1935) (new evidence must be such that it “more than likely would change the result”).

Two Standards Are
Not Equivalent

The Court agreed with the parties that the Appellate Division erred in equating the two standards. The difference lies in the modifier, the Court explained. As Kyles instructed, where an adjective modifies “probability,” “the adjective is important.” For Brady claims, that modifier is “reasonable,” and a reasonable probability of a different result imposes no obligation to establish that the verdict “more likely than not” would have changed or to “demonstrat[e] by a preponderance” that it would have. In contrast, motions for a new trial premised on newly discovered evidence are generally assessed under a standard demanding proof that the evidence “more likely than not” would, or “by a preponderance of the evidence” would, change the outcome. Strickland. Courts commonly use “probably” to mean more likely than not, the Court observed. Schlup v. Delo, 513 U.S. 298 (1995); Sawyer v. Whitley, 505 U.S. 333 (1992) (Blackmun, J., concurring); DCPP v. J.R.-R., 258 A.3d 1094 (N.J. 2021). Carter’s formulation has that meaning, having been taken from Bunk, which in turn relied on Paradise. The Court concluded that as a measure of materiality, “probably” is the more demanding standard.

The Court acknowledged that some of the state Supreme Court’s own decisions may have introduced confusion but stated that the Supreme Court had never held the two tests identical. State v. Ways, 850 A.2d 440 (N.J. 2004), applied Carter yet framed its holding in terms of a “probability” of a different verdict. State v. Allegro, 939 A.2d 754 (N.J. 2008), correctly quoted Strickland’s prejudice prong, which matches Brady’s materiality test, but then stated that it was guided in part by Carter’s “would probably change the jury’s verdict” language, and State v. Gideon, 242 A.3d 855 (N.J. 2021), relied on Allegro’s recitation. Resolving the confusion those cross quotations created, the Court held that “would probably change the jury’s verdict” is a more stringent standard than “a reasonable probability” that the jury’s verdict would change.

Application to Allen’s Claim

Turning to the present case, the Court assumed, as the Appellate Division had, that Allen satisfied Brady’s first two prongs, and it determined that he could not show a reasonable probability that disclosure of the 1991 plea form would have changed the verdict. Waller identified Allen to police roughly eight months before her shoplifting arrest, when she had no charge pending and no leniency to seek. When she testified in January 1999, that charge was pending in municipal court, but she was unrepresented, had never discussed it with the municipal prosecutor, and had never sought any benefit.

The Court reasoned that Waller had no rational incentive to commit perjury, a third-degree crime carrying up to five years’ imprisonment, to reduce a $500 maximum fine on what was then a disorderly persons shoplifting offense. The Court concluded that it was “confident that even if Waller’s 1991 plea form had been disclosed to the defense, the jury’s verdict would have been the same.”

Conclusion

Accordingly, the Court affirmed the judgment of the Appellate Division as modified. See: State v. Allen, 2026 N.J. LEXIS 702 (2026).  

Subscribe today

Already a subscriber? Login