Hines, Chief Justice.This Court granted applications for certificates of probable cause from Mark Jason Jones, Kenneth Eric Gardiner, and Dominic Brian Lucci to appeal the denials of their petitions for writs of habeas corpus. The cases are consolidated, and for the reasons that follow, we reverse in each.Jones, Gardiner, and Lucci were tried and found guilty of malice murder in the shooting death of Stanley Jackson, as well as of possession of a firearm in the commission of a felony. See Gardiner v. State, 264 Ga. 329 (444 SE2d 300) (1994). The three defendants were Army servicemen stationed at Fort Stewart, near Savannah, and are Caucasian; Jackson was African-American. Evidence presented at the November, 1992 trial showed that, during the day of January 31, 1992, Jones sought to borrow some equipment from a fellow soldier, Sylvia Ann Wallace, after an inspection, and told Wallace that he was going to Savannah that night because “he had somebody that he was going to shoot.” When Wallace enquired who that would be, Jones replied, “I got a black guy up there I got to get.”Shortly after 10:00 p.m. that night, near the intersection of 33rd Street and East Broad Street, James White saw two men fire military-type automatic or semiautomatic rifles[1] through the window of a 1992 black Chevrolet Cavalier while a third man drove the car; Jackson was killed in the shooting. White met investigating officers at the scene and went with an officer to a police station. There, outside a topless bar named Club Asia, which was across the street from the station, he saw a car carrying Jones, Gardiner, and Lucci and told the officer accompanying him that the car appeared to be the one he saw at the scene of Jackson’s shooting.[2] The three men went into Club Asia, and were shortly removed by officers and caused to stand under a light outside the bar. From some distance away, White said that he could not be sure that they were the men from the car, but reiterated his identification of it.[3] Later, when White was subpoenaed to appear at a preliminary hearing, he viewed the men more closely and stated that he recognized Jones and Gardiner as the shooters. At the defendants’ trial, White identified Jones and Gardiner as the shooters.After production of records from the police file about the case in response to a 2010 open records request, the three men filed habeas petitions asserting that, contrary to Brady v. Maryland, 373 U.S. 83 (83 SCt 1194, 10 LE2d 215) (1963), exculpatory evidence was not provided to the defense team by the State, namely that: 1) prior to trial, White made statements to police officers that he could not identify the shooters, contradicting his positive identification of Jones and Gardiner at trial; 2) White was coerced into testifying at trial that he could, in fact, positively identify the men; and, 3) a police report of an incident that took place shortly after the shooting, namely that a man told a patrol officer that, at 1:00 a.m. on February 1, 1992, in the Yamacraw Village public housing complex, Caucasian males in a white Chevrolet pick-up truck and a silver Ford Thunderbird, with military style haircuts and semiautomatic weapons, were “threatening] to shoot blacks who hang out on street corners” (the “Yamacraw Report”); it is uncontroverted that at 1:00 a.m. on February 1, 1992, Jones, Gardiner, and Lucci were in police custody.As the habeas petitions of each of the three men presented the same assertions, the petitions were consolidated. After an evidentiary hearing, the habeas court determined that the claims were procedurally defaulted and denied relief. The three men applied to this Court for certificates of probable cause to appeal from the habeas court’s order. In 2014, this Court determined that the claims were not procedurally defaulted, granted the petitions, and remanded the case to the habeas court, directing it to perform a Brady analysis.Upon remand, the habeas court denied relief, determining that the Yamacraw Report would not have been admissible at trial and thus would not qualify as Brady material; the habeas court rejected the argument that, if the report had been produced to the defense at trial, “additional exculpatory evidence could have been gathered” as simply speculation, and ruled that there was no reasonable probability that producing the Yamacraw Report would have changed the outcome of the trial. As to White’s habeas testimony, which was contrary to his trial testimony as discussed below, the court noted that there was considerable evidence presented at trial regarding the level of certainty of White’s identification, and specifically discredited White’s habeas testimony. The petitioners again sought certificates of probable cause to appeal, and these appeals followed.“In reviewing the grant or denial of a petition for habeas corpus, this Court accepts the habeas court’s factual findings and credibility determinations unless they are clearly erroneous, but we independently apply the law to the facts.” [Cit.]“ Humphrey v. Lewis, 291 Ga. 202, 204 (II) (728 SE2d 603) (2012).To prevail on a Brady claim, appellant[s] must demonstrate that the prosecution wilfully or inadvertently suppressed evidence favorable to the accused, either because it is exculpatory or impeaching. Brady v. Maryland, 373 U.S. at 87. However, “the Constitution is not violated every time the government fails or chooses not to disclose evidence that might prove helpful to the defense.” Kyles v. Whitley, 514 U.S. 419, 436-437 (115 SC 1555, 131 LE2d 490) (1995). Brady comes into play only when the suppressed evidence is material, i.e., “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 ‘reasonable probability’ is a probability sufficient to undermine confidence in the outcome.” United States v. Bagley, 473 U.S. 667, 682 (105 SC 3375, 87 LE2d 481) (1985).Young v. State, 290 Ga. 441, 443 (2) (721 SE2d 839) (2012).1. During the habeas hearing, White testified that: he told the investigator outside Club Asia that the car the defendants were in “looked like” the shooters’ car, but that he did not recognize the three suspects and could not identify any of them as the shooters; he asked investigators to provide him a line-up for identification purposes, but they would not; he was pressured by investigators, prosecutors, and members of the community – including clergymen – to identify the defendants as the shooters; he received anonymous telephone calls to his house that included threats to his family if he did not identify the defendants as the shooters; at a preliminary hearing, he identified Jones and Gardiner as the shooters; when he subsequently told an investigator that he could not truly identify them, he was threatened with being prosecuted for perjury if he did not testify at trial as he had at the preliminary hearing, and he was told that there would “be rioting in the city” if the defendants were not convicted; he then testified at trial, identifying Jones and Gardiner as the shooters;[4] in the years since the trial, he suffered significant mental and physical problems that he traced to his trial testimony; and, until approached in 2010 by a representative of a nonprofit organization that investigates claims of actual innocence of those incarcerated for life or under a death sentence, he told no one that he had testified falsely but his wife, who died before the habeas hearing.The habeas court specifically found that White’s habeas testimony “lacked credibility.” The court further noted that the defense had information that a detective noted that White’s identification was incomplete or uncertain, and the uncertainty of his identification was fully explored on cross-examination at trial. Although the petitioners contend that the result of the trial would have been different if the defense had known that White definitively informed investigators that he could not identify the defendants, and that he had been coerced by the State to make a definitive identification at trial, this ignores the factual findings of the habeas court that White’s testimony that these events occurred was not credible.[5] These contentions were rebutted by other evidence, the court’s credibility determination on these matters was not clearly erroneous, and the unestablished contentions do not form the basis for a Brady claim. See Humphrey, supra.2. At the time Jackson was killed, Officer B. J. Herron was assigned to patrol Yamacraw Village, a housing project in Savannah. The Yamacraw Report is handwritten, and was filled out on a Savannah Police Department form by Herron. The report shows Herron as the writer, with a preparation date of February 1, 1992, and the report is addressed to police precincts “1 & 2.” The report form’s field for “Reference” is filled in with: “Threats on citizen in Yamacraw.” The body of the report states:I was advised by a subject that on 2/1/92 [at 1:00 a.m.] two vehicles entered Yamacraw, silver 2 door 89-91 Ford Thunderbird and a white Chevy pick-up. Both vehicles were occupied by white males who were supposedly armed with semi-automatic weapons. All suspects appeared to have military style haircuts. Any contact with these suspects in any housing area, retain information. 10-0.[[6]] Suspects threaten to shoot blacks who hang out on corners.Herron testified during the habeas hearing that, while he was on foot patrol on February 1, 1992, a man he knew in the community came up to him and gave him the information he placed in the report; Herron could not remember exactly who gave him the information, but believed that, at the time Jackson’s killing was being investigated in 1992, he “probably could have found [him] again.” At the top of the report, in an unknown handwriting, was written, “Lt. Ragan”; at the time, Ragan was the lieutenant who oversaw the violent crime unit of the police department. In his testimony during the habeas hearing, Lt. Ragan said that he had no memory of the report being routed to him. The report was ultimately included in the police file about this case.The habeas court found that the defendants were not provided the Yamacraw Report “and could not have reasonably obtained it”; the court also stated in its order that there was no indication the State actively suppressed it.[7] However, such a finding does not eliminate the claim from our consideration; as noted, a Brady claim can rest upon a “demonstrat[ation] that the prosecution wilfully or inadvertently suppressed evidence favorable to the accused, either because it is exculpatory or impeaching. [Cit.]“ Young, supra. (Emphasis supplied.) Further, Brady’s disclosure requirement evenencompasses evidence “known only to police investigators and not to the prosecutor.” [Cit.] In order to comply with Brady, therefore, “the individual prosecutor has a duty to learn of any favorable evidence known to the others acting on the government’s behalf in this case, including the police.” [Cit.]Stickler v. Greene, 527 U.S. 263, 280 (II) (119 SCt 1936, 144 LE2d 286) (1999). See also Danforth v. Chapman, 297 Ga. 29, 29 (1) (771 SE2d 886) (2015) (“‘[T]he 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.’ [Cit.]“); Schofield v. Palmer, 279 Ga. 848, 853 (2) (621 SE2d 726) (2005) (“It is irrelevant that a police agency may have possessed the favorable evidence without the knowledge of the prosecutor; the law places the responsibility and ultimate burden on the prosecutor for the failure to provide the favorable evidence to the defendant if any part of the prosecution team possessed and suppressed the favorable evidence. [Cits.]“)The habeas court also stated in its order that, “in order to be considered Brady material, the Yamacraw Report would have to be admissible evidence.” However, this is a misstatement of the appropriate standard. The admissibility of the undisclosed material itself is not a prerequisite to finding a Brady violation; the question is whether, had the material “been disclosed to the defense, the result of the proceeding would have been different,” in reasonable probability. Young, supra. Thus, “inadmissible evidence may be material [under Brady] if it could have led to the discovery of [material] admissible evidence. [Cits.]“ Johnson v. Folino, 705 F3d 117, 130 (3rd Cir. 2013). See also Bradley v. Nagle, 212 F3d 559, 567 (II) (11th Cir. 2000) (“[I]n order to find that actual prejudice occurred—that our confidence in the outcome of the trial has been undermined—we must find that the evidence in question, although inadmissible, would have led the defense to some admissible material exculpatory evidence [Cit.]“); Young, supra (“Because the report was hearsay and inadmissible, and appellant has not shown how its disclosure would have led to admissible evidence, it did not constitute Brady material. [Cit.]“) (Emphasis supplied.)As to whether the Yamacraw Report is material within the meaning of Brady, the question is whether in the absence of the production of the report, the defendantsreceived a fair trial, understood as a trial resulting in a verdict worthy of confidence. A “reasonable probability” of a different result is accordingly shown when the government’s evidentiary suppression “undermines confidence in the outcome of the trial.” [Cit.]Brownlow v. Schofield, 277 Ga. 237, 239 (2) (587 SE2d 647) (2003). Accord Young, supra.During the habeas hearing, investigating detectives recognized that the incident reported in the Yamacraw Report presented certain similarities to the allegations against the defendants, and would have warranted further investigation as to that incident’s relation to Jackson’s slaying. The prosecutor who conducted the trial testified that, had the Yamacraw Report been in his file, he would have either disclosed it to the defense, “or been specific about why I was withholding it.” And, the Yamacraw Report clearly would have been helpful to the defense; it was evidence that others similar in appearance were threatening a racial attack similar to that alleged to have been suffered by Jackson, but three hours after his slaying,[8] when the defendants were already in custody. Compare Upton v. Parks, 284 Ga. 254, 255-256 (1) (664 SE2d 196) (2008). Herron’s habeas testimony indicated that, at the relevant time, he could have identified the subject who gave him the information in the Yamacraw Report. The attorneys who represented the defendants at trial testified that, if they had been provided the report, they would have followed up with Herron, would have sought other residents of the area who might have witnessed or received the threats, argued to the jury that the fact that the defendants were in custody at the time meant that they could not have been those who committed the acts reported, and thus there were other potential assailants who the police had not sought. Thus, the report would have also enabled the defense attorneys to further their attack on the thoroughness of the police investigation, see Kyles v. Whitley, 514 U.S. 419, 446 (IV) (B) (115 SCt 1555, 131 LE2d 490) (1995), and allowed them to present an alternative theory regarding the actors responsible for the shooting. See Walker v. Johnson, 282 Ga. 168, 170 (2) (646 SE2d 44) (2007).It is certainly true that, due to the passage of time, the man who provided the information in the Yamacraw Report was not located so as to enable him to testify at the habeas hearing, and thus the information that he might have given if his existence had been known to the defense at the time of trial – and what further information this might have produced – is unknown. However, even if he had merely testified at trial to a repetition of that which he had told Herron, the information would have been impactful, given how the case would have proceeded at trial.Although White’s subsequent claims about his trial testimony do not constitute Brady violations, and thus are not grounds for habeas relief, we must consider the importance of identity testimony on the course of the trial. The State’s case was heavily dependent on White’s testimony and his eyewitness identification of the defendants. That testimony was attacked as unreliable at trial, as was the quality of the police investigation into White’s identification; the defense attorneys elicited testimony that the investigators failed to ask on the night of the shooting whether White could identify the suspects by their faces, and that the investigators conducted no identification line-up, then or later, either in person or by photograph, even though the detective who coordinated White’s view of the suspects the night of the shooting testified that the preferred method of establishing an identification of a suspect at that point in an investigation was to do a photographic line-up.Jackson was killed shortly after 10:00 p.m. on January 31, 1992. There was trial testimony from several witnesses that, until 9:15 or 9:30 p.m., the petitioners were at the rehearsal of Jones’s wedding,[9] which was to take place the next day, and a dinner afterward, which took place in a town that was over a 50- minute drive away from the relevant areas of Savannah. No murder weapon was ever recovered; no firearm was found in the defendants’ car, no casings from an automatic weapon were found there, and the forensic scientist who vacuumed the interior of the car looking for gunshot residue found none.[10]There were also significant racial overtones to the trial; in addition to those factors mentioned in our prior opinion, see Gardiner, supra at 330 (1), 332 (2), 333 (4), and 334 (9), witnesses were regularly asked about the defendants’ attitudes toward members of other races; had the jury been presented with information that other persons, not the defendants, were in the area that same night, apparently ready to engage in racially motivated violence, the outcome of the trial might well have been different. Thus, in light of the totality of the circumstances, confidence in the outcome of the trial was undermined by the State’s failure to provide the Yamacraw Report to the defense. See Walker, supra. Certainly, in the face of the Yamacraw Report, the jury “could have voted to convict [the defendants], [but] we have ‘no confidence that it would have done so.’ [Cit.]“ Wearry v. Cain, __ U.S. __, __ (II) (136 SCt 1002, 194 LE2d 78) (2016). Accordingly, the habeas court’s denial of the petitions for writs of habeas corpus must be reversed.Judgments reversed. All the Justices concur.
