Filed 4/2/26 CERTIFIED FOR PUBLICATION
IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA
SECOND APPELLATE DISTRICT
DIVISION ONE
In re ALONZO DEVON MELSON B336211 on Habeas Corpus. (Los Angeles County Super. Ct. No. TA143199)
ORIGINAL PROCEEDING; petition for writ of habeas corpus, Sean D. Coen, Judge. Petition granted. David Andreasen, under appointment by the Court of Appeal, for Petitioner. Rob Bonta, Attorney General, Charles C. Ragland, Chief Assistant Attorney General, Susan Sullivan Pithey, Assistant Attorney General, Noah P. Hill and Thomas C. Hsieh, Deputy Attorneys General, for Respondent.
________________________ In criminal prosecutions the People have a constitutional obligation to correct false testimony from their witnesses. (Glossip v. Oklahoma (2025) 604 U.S. 226 [145 S.Ct. 612, 221 L.Ed.2d 90] (Glossip); Napue v. Illinois (1959) 360 U.S. 264 [79 S.Ct. 1173, 3 L.Ed.2d 1217] (Napue).) They failed to comply with that obligation in this case. Petitioner Alonzo Devon Melson was charged with crimes arising out of a gang-related shooting. His first trial ended in a hung jury, with three jurors voting to acquit. When the People retried Melson, the only two eyewitnesses bolstered their identification of Melson with statements during direct examination that they had not made at the first trial. These new statements, about the witnesses’ purported identification of Melson to police soon after the shooting, were false. Although the prosecutor knew or should have known those statements were false, he did not correct them. Instead, he repeated one of the false statements in his closing argument as evidence of Melson’s guilt. This time, the jury convicted Melson. Although Glossip and Napue impose no constitutional obligation on a criminal defendant’s counsel to correct false testimony from prosecution witnesses, defense counsel has a constitutional obligation to render effective assistance. But Melson’s counsel during the retrial (who was not his counsel during the first trial) failed to impeach the two eyewitnesses with what they actually had said to the police or to otherwise cross- examine them about the false portions of their trial testimony concerning their police interviews. This resulted not from a strategic choice but a lack of fundamental preparation—counsel’s file contains no copies of the eyewitnesses’ police interviews and
2 no notes about their prior statements to police, indicating he did not adequately prepare to cross-examine these key witnesses. Melson requests that we issue a writ of habeas corpus vacating his conviction based on Napue error and ineffective assistance of counsel. The materiality of the Napue error that occurred here is presumed, and the People must rebut that presumption beyond a reasonable doubt. (Glossip, supra, 604 U.S. at p. 246.) Other evidence pointed to Melson’s guilt, but we do not assess whether the false testimony directly affected the trial’s outcome; we assess only whether it could have contributed to the verdict. (Id. at p. 253.) Under this standard, the People have not shown beyond a reasonable doubt that the false testimony from the two eyewitnesses did not contribute to Melson’s conviction. Accordingly, we must vacate that conviction. BACKGROUND A. The Charges Against Melson On January 13, 2017, Samuel Navarrete was shot and killed as he sat in his car just outside a housing project in Watts formerly called Hacienda Village. The People charged Melson as the shooter, based primarily on the testimony of two eyewitnesses who identified him from a photo lineup: Pedro Sabino, who was in the passenger seat of the car beside Navarrete, and Georgina Araiza, who lived five houses away from the scene of the shooting. Melson’s first trial resulted in a hung jury. The jury in a second trial convicted him of one count of first degree murder
3 (Pen. Code, 1 § 187, subd. (a)), two counts of attempted murder (one for Sabino and the other for Raul Garibay, who was in the back seat of the car) (§§ 187, subd. (a), 664)), one count of shooting at an occupied vehicle (§ 246), and one count of possession of a firearm by a felon (§ 29800, subd. (a)(1)). We reversed one of the attempted murder convictions on direct appeal because of a faulty kill zone instruction, but we otherwise affirmed the judgment. (People v. Melson (Apr. 1, 2020, B292679) [nonpub. opn.].) B. Evidence at the Retrial The prosecution’s theory was that Melson shot Sabino as part of a gang war between the Hacienda Village Bloods, a Black gang, and the Village Boys, a Hispanic gang. Melson is Black and was a member of the Hacienda Village Bloods. A member of the Village Boys, Juan Beltran, told police Melson had a reputation for “press[ing] the line . . . a lot.” At trial, Beltran testified that pressing the line meant “confront[ing] somebody.” Navarrete was an associate of the Village Boys, and Garibay, who was known by the nickname Little Bugsy and was in the backseat of the car during the shooting, was a member. Sabino was not associated with any gang, but his brother was a member of the Village Boys. 1. Sabino’s Testimony Sabino testified that, on the evening of the shooting, his friend Navarrete picked him up and drove to 106th Street, where Garibay joined them. Sabino did not know Garibay well. The
1 Subsequent unspecified statutory references are to the Penal Code.
4 three of them remained in the car waiting for someone else to join them when Sabino saw a figure walk in front of the car wearing a hoodie and appearing to be carrying a gun. A second figure, much taller, walked beside him. Sabino told Navarrete they should leave, but Navarrete said he wanted to see what the men wanted. The man with the gun knocked on the window, and Navarrete opened the car door. The man asked Navarrete where he was from. Navarrete gave his gang nickname (Fresh) and said he was from Village Boys. The man poked his head in the window and asked who was in the back seat. Garibay answered, “It’s me, Little Bugsy,” as if he knew the man. The man stepped back and started shooting. Sabino ducked down until the shooting stopped, then looked up and saw the men running away. Sabino moved Navarrete from the driver’s seat to the passenger’s side and drove him to the hospital, where he died of his wounds. In the early morning hours just after the shooting, the police interviewed Sabino, and he told them about the events of that evening. Sabino testified that he told the officers he needed a couple of days to clear his head so he could give a better description of the shooter. Two days later, on January 16, Sabino went to the police station, where detectives showed him a series of photographs. He chose three photos depicting people that resembled the shooter, and from those three, he selected Melson’s photo as the closest match. Sabino testified that this was in part because the photo showed Melson with a tattoo next to his eyebrow. Because the shooter was wearing a hoodie and a beanie, Sabino could not see much of the shooter’s face, but he saw a tattoo by the shooter’s
5 eyebrow when he ducked his head to look in the car. Sabino said he told the police that he had seen the tattoo. Transcripts of both police interviews, which were not introduced as evidence, show Sabino denied on both occasions that he saw any tattoo on the shooter. 2. Araiza’s Testimony Araiza lived five houses down the street from where the shooting took place. She testified that, about 15 to 30 minutes before the shooting, she was coming home from the store when she saw a man she knew as Chops in a heated argument with another man named Fatty. Fatty was inside the gate of the house where he lived, along with several other people, all of them Hispanic, and likely members of the Village Boys gang. Chops, who was Black, was outside the gate together with two other Black men, one of whom Araiza identified as Melson. Araiza had never spoken with Melson but she had seen him in the neighborhood in his red Honda and knew him by his gang name, Yank. Melson, who was wearing a red hoodie, was not participating in the argument. Araiza went inside her home, but shortly afterward, she heard shooting. She ran outside immediately because she thought her son was outside. She was outside by the time the last two gunshots were fired. She saw Chops 2 standing outside, and Melson was running away. Chops was standing near Fatty’s house with his hand outstretched and appeared to be holding a
2 Araiza testified that Chops was dead. A police officer confirmed this and said Chops left the state after Melson was arrested, and that he was murdered in Louisiana shortly thereafter.
6 gun. Araiza said, “you could see Chops with the gun, shooting.” The prosecutor asked, “Did Mr. Melson have a weapon in his hand?” Araiza replied, “I think he did have something. You could tell what’s a gun–and running [from] the scene.” Araiza was confident Melson was the man running away because of the red hoodie he had been wearing. Araiza did not see Melson shoot, but she did see him with a gun. Araiza testified she spoke with police after the shooting. They showed her a series of photographs, and she identified Melson in one of them as the man she knew as Yank or Tank. The transcript of Araiza’s police interview (which was not admitted at trial) shows that she in fact did not tell the police that she saw Melson (or Yank or Tank), or any Black men at all, after she heard gunfire. Instead, the only people she saw were Hispanic. 3. Other evidence The prosecution introduced other evidence implicating Melson. This included a recording of a jailhouse conversation involving Melson and two others, Juan Beltran and Miguel Aleman. 3 Officers placed Melson on a bench outside the cell where Beltran and Aleman were housed. During the conversation, Beltran and Aleman implied that Melson was involved in the murder of Sabino, and Melson did not object. Instead, Melson appeared concerned about whether anyone
3 Aleman, also known as Fatty, was the member of the Village Boys who, according to Araiza, argued with Chops prior to the shooting. In addition, the shooting happened in front of his house.
7 would snitch on him, and when he learned Sabino might talk to the police, asked where Sabino lived. The prosecution also introduced evidence of statements Melson made after his arrest. A detective who interviewed Melson at that time testified that he asked him where he was at the time of the murder, and Melson said he was in Anaheim with family. The detective was unable to locate any family of Melson’s that lived in Anaheim. An FBI agent testified that he reviewed the cell tower records from Melson’s phone, and that, from around 7:00 to 9:35 p.m. on the evening of the shooting, the phone communicated with towers within a few blocks of the area of the shooting. After that, the phone was tracked to an area several miles away in Long Beach, near two addresses associated with Melson. The shooting occurred at some point between 8:00 and 9:00 p.m. The agent testified that the phone could not have connected to the cell towers in question if it had been located in Anaheim. B. The Habeas Petition Melson filed a petition for habeas corpus in this court in conjunction with his direct appeal on September 6, 2019. The petition alleged that Sabino and Araiza had testified falsely, and that Melson’s trial attorney had been ineffective by failing to impeach those witnesses regarding those statements. The petition also alleged that Melson’s attorney rendered ineffective assistance in several other regards: by failing to reasonably cross- examine a detective regarding Melson’s statements about his whereabouts on the night of the shooting, by failing to object to the admission of portions of the jailhouse conversation with Aleman and Beltran, by failing to object to evidence regarding Melson’s violent character, and by failing to consult an
8 eyewitness identification expert. Melson also alleged the cumulative effect of the errors required reversal of his conviction. We summarily denied the petition, but the Supreme Court directed us to issue an order to show cause returnable before the superior court. After conducting an evidentiary hearing, the superior court denied the petition. At the conclusion of the hearing, the trial court “found [Sabino’s and Araiza’s] statements more to be prior inconsistent statements and not false testimony . . . . I didn’t find them to be false statements at all.” Melson again filed a petition in this court, which we again denied. On July 23, 2025, the Supreme Court directed us to vacate our order and issue an “order[] to show cause why [Melson] is not entitled to relief on the grounds he suffered prejudicial error pursuant to Napue . . . , or, in the alternative, that the cumulative effect of the errors alleged in the petition deprived him of due process.” STANDARD OF REVIEW “A habeas corpus petition is a collateral attack on a presumptively valid judgment, thus ‘ “the petitioner bears a heavy burden initially to plead sufficient grounds for relief, and then later to prove them.” ’ ” (In re Lewis (2018) 4 Cal.5th 1185, 1191.) “ ‘To do so, he or she must prove, by a preponderance of the evidence, facts that establish a basis for relief on habeas corpus.’ ” (In re Cudjo (1999) 20 Cal.4th 673, 687.) When a petition is referred for an evidentiary hearing, “ ‘we generally defer to the referee’s factual findings’ and give them great weight if supported by substantial evidence.” (In re Lewis, supra, 4 Cal.5th at p. 1191, fn. omitted.) This is “[b]ecause the referee observes the demeanor of the witnesses as they testify” (ibid.); if the referee’s findings are based solely on documentary
9 evidence, we do not accord them the same deference. (In re Long (2020) 10 Cal.5th 764, 774; In re Rosenkrantz (2002) 29 Cal.4th 616, 677.) We independently review mixed questions of law and fact, such as “ ‘[w]hether counsel’s performance was deficient, and whether any deficiency prejudiced the petitioner.’ ” (In re Long, supra, at p. 774.) We review pure questions of law de novo. (In re Hansen (2014) 227 Cal.App.4th 906, 914.) DISCUSSION “[A] conviction obtained through use of false evidence, known to be such by representatives of the [s]tate, must fall under the Fourteenth Amendment [citations]. The same result obtains when the [s]tate, although not soliciting false evidence, allows it to go uncorrected when it appears.” (Napue, supra, 360 U.S. at p. 269.) The introduction of false testimony requires reversing a defendant’s conviction if it “ ‘may have had an effect on the outcome of the trial,’ [(id. at p. 272)]—that is, if it ‘ “in any reasonable likelihood [could] have affected the judgment of the jury[]” ’ [citations]. In effect, this materiality standard requires ‘ “ ‘the beneficiary of [the] constitutional error to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained.’ ” ’ ” (Glossip, supra, 604 U.S. at p. 246.) Thus, “[t]o establish a Napue violation, a defendant must show: (1) that the testimony was actually false, (2) that the government knew or should have known that it was false, and (3) that the testimony was material, meaning there is a ‘reasonable likelihood that the false testimony could have affected the judgment of the jury.’ ” (U.S. v. Renzi (9th Cir. 2014) 769 F.3d 731, 751.) We begin by describing the facts Melson produced in support of his habeas claim. We then address the People’s
10 argument that Melson forfeited his claim of Napue error by failing to present it either before the trial court or in his direct appeal. We conclude he did not forfeit the claim. We proceed to analyze each of the three steps set forth in Renzi, and we conclude that Melson met his burden of demonstrating error. Lastly, we hold the People have not shown the error here to be harmless. A. Sabino’s Statements Sabino spoke with the police twice, first on January 14, 2017, hours after the shooting, and again two days later. In the first interview, he told the detectives about the events leading up to the shooting, giving a similar account to the one he later gave at trial. The detectives asked Sabino about the shooter’s appearance. Sabino said he saw the shooter’s face, but “I really didn’t see him ‘cause he had his hood and he had a beanie.” The detectives asked Sabino about height—the shooter was approximately 5’8”—and build—he was “kind of chubby.” He was wearing a black hoodie, a beanie, and cut-off jeans. He was young, perhaps 19 or 20, and of medium complexion. One of the detectives asked, “Any tattoos, did you see any tattoos?” Sabino answered, “No.” In the second interview, the detectives showed Sabino a series of 18 photographs to see if he could identify the shooter. Sabino set aside three photos that “looked like the guy that shot” Navarrete. From these three, Sabino said the photo of Melson was “the one that looks the most” like the shooter. The photo of Melson in the lineup depicts him with a tattoo of a thin black or gray line above his eyebrow. The detective asked, “Did you see any tattoos on the face or anything like that?” Sabino shook his head no. The detective
11 pointed to a photo of a suspect other than Melson and said, “And [did] you see any scars? This guy’s got tattoos on his cheek. Did you see that?” Sabino again shook his head no. The detective continued, pointing at the face of the man in the photo, “No? But you basically saw this area right in here?” Sabino answered, “Yeah.” The detective said, “So the eyes, nose, and the mouth?” Sabino said, “Yeah, by his mouth, I seen him.” At the evidentiary hearing on the habeas petition, the prosecutor testified that he immediately thought Sabino’s trial testimony about telling police he saw a tattoo was incorrect: “I was close to . . . 90 percent confident that I didn’t recall him mentioning [seeing the tattoo] in the interview.” He said that, while Melson’s attorney was cross-examining Sabino, he (the prosecutor) “lean[ed] over to the detective and ask[ed] him does he recall if Mr. Sabino made the statement about the tattoo, and he told me at that time he didn’t.” After confirming Sabino’s testimony was incorrect, the prosecutor elicited testimony on redirect from Sabino that he might not remember everything he told police because he was “kind of still in shock” from seeing his best friend murdered. At the habeas evidentiary hearing, the prosecutor explained that he asked this question “because I was trying to explain away in my mind why Mr. Sabino was providing this additional testimony. And from my experience, sometimes when a . . . witness experiences a very traumatic situation, you know, he just had [seen] his best friend get murdered, that . . . some witnesses and victims go into shock. And . . . they don’t remember what they told the police. So that’s why I asked him that question, is it possible . . . that . . . I forgot how I phrased it, but I brought that
12 up to kind of explain away again, in my mind, why his testimony could be a little bit different.” The prosecutor believed he had no further duty to correct the record on Sabino’s testimony “[b]ecause I felt that the questions I had asked had clarified any confusion that the jury may have . . . [seen] with Mr. Sabino’s testimony. And from my recollection, [the] defense attorney got up and cross-examined him on his inconsistencies. So in my mind, the jury had the full story of what I offered as evidence and what [the defense attorney] offered on his cross-examination.” The defense attorney did cross-examine Sabino on his identification of the shooter, asking him, “When you spoke to the cops when you were interviewed, did you say you saw the tattoo, or was it something else?” Sabino answered, “I told them I seen the tattoo.” Defense counsel did not press Sabino further on what he told the police or impeach him with the transcripts showing Sabino had in fact denied having seen a tattoo. Counsel did ask about the inconsistencies in Sabino’s testimony about how he identified Melson: “So do you recall saying that you saw the shooter’s mouth, when you testified in court?” Sabino answered, “Yeah, I think so.” Defense counsel asked, “And did you also testify that you didn’t really see his eyes, when you were in court?” Sabino responded, “Yes.” During cross-examination, defense counsel also challenged Sabino’s claim that he chose Melson’s photo because of the tattoo, noting that the other two photos he chose as possible matches did not have tattoos like Melson’s. Later in the trial, defense counsel cross-examined the detective who interviewed Sabino about what the detective remembered Sabino saying about the shooter. The detective
13 responded, “[Sabino] remembered his height. He remembered his race. He remembered seeing his facial features from the nose down. He mentioned some of the eyes, but he mentioned that the suspect was wearing a beanie over his head.” Defense counsel asked, “So there was no mention of him saying a tattoo or anything of that nature?” The detective answered, “No, not at that time.” B. Araiza’s Statements Araiza’s testimony at trial differed significantly from what she told police, as well as from her testimony at the preliminary hearing and at the first trial. The police interviewed Araiza on January 26, 2017, about two weeks after the shooting. At the interview, Araiza said she saw three Black men, including one whom she knew as Chops, and another who was called Yank or Tank, near the scene of the shooting about 30 minutes beforehand. Araiza had not met Yank face to face, but had seen him in the neighborhood, and knew what car he drove and generally what he looked like. Chops and Yank were arguing with a man Araiza knew as Fatty outside of Fatty’s house. She went inside her house and remained there until she heard shooting outside and one of her kids screaming. She ran outside where she saw the car and Fatty standing nearby, along with several other people Araiza did not recognize. A detective asked Araiza, “Were they Hispanic or Black?” Araiza answered, “Hispanic.” The detective asked, “Did you see any Blacks out there?” Araiza answered, “No. When the shooting, no. After that, no.” The detective showed Araiza a series of photographs, and Araiza identified a photo of Melson as Tank. At the preliminary hearing, Araiza testified Chops was arguing with Fatty shortly before the shooting. Melson was
14 nearby but was not participating in the argument. Araiza was in her kitchen when she heard shooting, and she immediately ran to the front door because she thought her son was playing outside near Fatty’s house. Araiza saw three men wearing hoodies running away from the shooting. She recognized one of the men as Chops, but she did not see Melson. At the first trial, Araiza first testified she saw Black men running toward the projects after the shooting. On redirect, Araiza said she saw Chops and two others running away, but she did not identify the others. At the second trial, Araiza testified that after she heard gunfire she saw Melson running away with a gun in his hand. Araiza also testified that the police showed her a series of photographs after the shooting and she identified Melson in one of them as a person connected to Navarrete’s murder. At the evidentiary hearing, the prosecutor testified that he was aware “the statement that [Araiza] was giving in court was different than the statement she had provided to the police,” not merely with her testimony at the first trial. To address the inconsistencies, he asked, “Now, when you were interviewed back on January 26, 2017, do you remember everything that you told the police?” Araiza responded, “I can’t remember exactly. I’ve been through a lot since. . . . So it’s hard for me to remember a lot of things, you know?” The prosecutor asked, “Is it possible that you may not have told them that you saw Mr. Melson with a gun in his hand that night?” Araiza replied, “I can’t remember. I can hardly remember.” The prosecutor believed that Araiza’s new testimony was more accurate because “once she felt a little bit more secure as far as her safety and her family’s safety, I felt that she became a little bit more forthcoming about the facts.”
15 Even at the evidentiary hearing, however, the prosecutor appeared to believe that the primary inconsistency between Araiza’s testimony at the second trial and in her interview with police was as to whether she saw Melson with a gun, not whether she saw him at all after the shooting. The deputy district attorney conducting the evidentiary hearing read back Araiza’s testimony, where she described Chops shooting and Melson running away, and asked, “When you were listening to this testimony, did that appear in any way to be inconsistent with what she said to the police originally from what you recall?” The prosecutor answered, “No.” The prosecutor’s opening statement and closing argument reiterated Araiza’s false testimony about whether she had identified Melson to the police as involved in the shooting. In his opening statement, the prosecutor told the jury that Araiza was “going to tell you that she saw—this is what she told the detective when they interviewed her, that she saw Chops and someone that resembled [Melson] running from the scene.” In his closing argument, the prosecutor reiterated this: “We know on January 26, 2017, Georgina Araiza is interviewed. She I.D.’s Mr. Melson as the shooter.” During the evidentiary hearing, the prosecutor claimed he misspoke during his closing argument, and he was aware that Araiza identified Chops as the shooter. C. Forfeiture The People contend Melson forfeited any claim of Napue error “because he did not object in the trial court to the challenged testimony, [n]or did he request an admonition that the jury disregard any supposedly false testimony. (See People v. Fuiava (2012) 53 Cal.4th 622, 679.) And his Napue claims are procedurally barred because he could have, but did not, raise
16 them on direct appeal. (See In re Dixon (1953) 41 Cal.2d 756, 759; In re Seaton (2004) 34 Cal.4th 193, 199-200.)” The cases the People cite on forfeiture, however, do not involve Napue error. Fuiava held that a defendant may not complain of prosecutorial misconduct on appeal unless he brought the claim to the trial court’s attention. (People v. Fuiava, supra, 53 Cal.4th at p. 679.) Dixon involved claims of illegal search and seizure and a coerced confession. The court stated, “The general rule . . . that habeas corpus cannot serve as a substitute for an appeal, and, in the absence of special circumstances constituting an excuse for failure to employ that remedy, the writ will not lie where the claimed errors could have been, but were not, raised upon a timely appeal from a judgment of conviction.” (In re Dixon, supra, 41 Cal.2d at p. 759.) In Seaton, the defendant sought to challenge his murder conviction and death sentence on the basis of racial bias by the prosecutor and by the county in manipulating the racial composition of its jurors (In re Seaton, supra, 34 Cal.4th at p. 197), and the court held on the basis of the principles described above that the defendant forfeited those claims by failing to raise them before the trial court or on direct appeal. (Id. at pp. 198-200.) We are aware of no cases applying these general rules of forfeiture to Napue error, and for good reason. As the United States Supreme Court explained in Glossip, it is “the prosecution’s duty to correct false testimony ‘when it appears.’ ” (Glossip, supra, 604 U.S. at p. 253, fn.10, quoting Napue, supra, 360 U.S. at p. 269.) Thus, “even if the defense had made a conscious choice not to raise the [issue of the false testimony], that would be irrelevant.” (Glossip, supra, at p. 253, fn.10.) To require Melson to object to the introduction of false testimony at
17 trial in order to preserve the issue would be incompatible with this rule. We also disagree with the People’s contention that Melson should have brought his Napue claim in his appeal. Raising a Napue claim in an appeal is often impractical because the ordinary appellate record may not include all information necessary to adjudicate Napue error. In particular, “[t]o establish a Napue violation, a defendant must show that the prosecution knowingly solicited false testimony or knowingly allowed it ‘to go uncorrected when it appear[ed].’ ” (Glossip, supra, 604 U.S. at p. 246, italics added, quoting Napue, supra, 360 U.S. at p. 269.) The question of what a prosecutor knew or should have known often cannot be answered without an evidentiary hearing like the one in this case. In this way, claims of Napue error are similar to “claims of ineffective assistance [of counsel, which] are often more appropriately litigated in a habeas corpus proceeding” and for which “the rules generally prohibiting raising an issue on habeas corpus that was, or could have been, raised on appeal” do not apply. (People v. Mendoza Tello (1997) 15 Cal.4th 264, 267.) Indeed, Melson first filed his habeas petition simultaneously with his direct appeal. He appears to have addressed Napue error in the habeas petition rather than the direct appeal specifically to facilitate the use of evidence from outside the appellate record. For these reasons, we conclude Melson did not forfeit his claims of Napue error. D. Was the Testimony False? To establish a Napue violation, a defendant must first show “that the testimony was actually false.” (U.S. v. Renzi, supra, 769 F.3d at p. 751.) Highly misleading statements may be deemed false for purposes of a Napue analysis, even if they are
18 not technically false. (In re Hill (2024) 104 Cal.App.5th 804, 830.) But “[m]ere inconsistencies between a witness’s testimony and her prior statements do not prove the falsity of the testimony.” (People v. Vines (2011) 51 Cal.4th 830, 874, overruled on another ground in People v. Hardy (2018) 5 Cal.5th 56, 104.) At the evidentiary hearing, the trial court rejected Melson’s Napue claim in part because it “found [Sabino’s and Araiza’s] statements more to be prior inconsistent statements and not false testimony . . . . I didn’t find them to be false statements at all.” This conclusion is correct as to many statements from both Sabino and Araiza. Their claims regarding the events of the shooting varied in particular aspects from their initial interviews with the police to the preliminary hearing to the first trial and the second trial, and we cannot say which version was accurate. (See People v. Vines, supra, 51 Cal.4th at pp. 874-875.) That analysis does not apply to statements Sabino and Araiza made regarding what they told the police. Sabino testified he told police that the shooter had a tattoo, but the transcripts of his police interviews show he said during both interviews that he had not seen a tattoo. Araiza testified that she told police she saw Melson fleeing from the scene of the shooting, but the transcript of her police interview shows she told police she did not see any Black men when she ran outside her house after the shooting. Even if the error resulted from nothing more than faulty memory, these portions of the witnesses’ testimony were untrue and thus meet the definition of false testimony. The People concede that Sabino’s testimony on the tattoo was “indisputably inaccurate,” but they argue it does not constitute false testimony under Napue because “Sabino’s testimony, taken as a whole, did not give the jury the false
19 impression that (1) during his police interviews, he actually told the police the shooter had a tattoo over his eyebrow and (2) there was no evidence that Sabino had not actually mentioned the tattoos. The parties presented evidence showing that Sabino’s initial testimony that he told the police about the tattoos was not accurate.” The People point to additional evidence that called into question the accuracy of Sabino’s claim, including that he had told police he identified Melson primarily by his mouth, because he had been wearing a hat and a hoodie, and that his memory of the interview might be compromised because he was in shock at the time. In addition, the People note that defense counsel asked one of the detectives who interviewed Sabino about the tattoo, and the detective testified that Sabino had not mentioned the tattoo. The People likewise concede that Araiza’s testimony was “incorrect,” but argue it was not false for purposes of Napue “because her testimony, taken as [a] whole, did not give the jury the false impression that she actually had made these prior statements.” On further questioning by the prosecutor, Araiza stated that she could not remember everything she told the police, and she might not have told the police she saw Melson fleeing the scene with a gun. In support of their position, the People cite case law indicating that a violation of due process occurs if a witness’s “testimony, taken as a whole, gave the jury [a] false impression.” (Alcorta v. Texas (1957) 355 U.S. 28, 31 [78 S.Ct. 103, 2 L.Ed.2d 9]; accord, In re Hill, supra, 104 Cal.App.5th at p. 831.) But these cases involve situations in which a witness’s testimony is merely misleading. In such a case, “outright falsity need not be shown if the testimony taken as a whole gave the jury a false
20 impression.” (People v. Westmoreland (1976) 58 Cal.App.3d 32, 42, citing Alcorta, supra, at p. 31.) We are aware of no case in which a court held that literally false testimony was not false for purposes of Napue because the testimony did not give the jury a false impression. This case presents no reason to depart from the general rule. Although both Sabino and Araiza expressed uncertainty about what they told the police, they did not disavow their testimony. Sabino reaffirmed on cross-examination that he told police about the tattoo. As to Araiza, the prosecutor reinforced the testimony by stating during his opening statement that Araiza would testify she told police she saw Melson after the shooting, and by repeating in his closing argument that Araiza identified Melson as the shooter during her police interview. The witnesses’ testimony qualifying or expressing uncertainty about what they previously said is more directly relevant to the questions of materiality and to whether the prosecutor corrected the false testimony. We analyze those issues further in those sections below. E. Did the Prosecution Know or Have Reason to Know of the Falsity? The second step in a Napue analysis is to determine whether “the government knew or should have known that [the testimony] was false.” (U.S. v. Renzi, supra, 769 F.3d at p. 751.) At this stage, “the governing federal decisions establish that the trial prosecutor’s lack of personal knowledge of the false and misleading nature of a prosecution witness’s testimony is not controlling. The United States Supreme Court has held that the state’s duty to correct false or misleading testimony by prosecution witnesses applies to testimony which the prosecution knows, or should know, is false or misleading (see United States
21 v. Agurs [(1976)] 427 U.S. [97,] 103 [96 S.Ct. 2392, 49 L.Ed.2d 342]), and has concluded this obligation applies to testimony whose false or misleading character would be evident in light of information known to other prosecutors, to the police, or to other investigative agencies involved in the criminal prosecution. (See, e.g., Giglio v. United States [(1972)] 405 U.S. 150, 154 [92 S.Ct. 763, 31 L.Ed.2d 104] [information known to prior prosecutor]; United States v. Bagley [(1985)] 473 U.S. 667, 670-672 & fn. 4 [105 S.Ct. 3375, 87 L.Ed.2d 481] [information known to federal investigators]; Barbee v. Warden, Maryland Penitentiary (4th Cir.1964) 331 F.2d 842, 846 [information known to investigating police officers].” (In re Jackson (1992) 3 Cal.4th 578, 595.) Under this standard, the government knew Sabino and Araiza testified falsely. The People argue this is not the case because the prosecutor did “not intend[] for Sabino to testify that he had told the police about the tattoo.” The People further note that the trial court, at the conclusion of the evidentiary hearing, found “there’s nothing done in bad faith” with respect to the false testimony from Sabino and Araiza. That may be the case, and indeed, nothing in this opinion should be interpreted as implying the prosecution (or Sabino or Araiza) acted in bad faith. But the prosecutor’s subjective knowledge or intent is not decisive. (See In re Jackson, supra, 3 Cal.4th at p. 595.) A Napue violation does not require bad faith or the solicitation of false evidence, but merely that “the [s]tate . . . allows it to go uncorrected when it appears.” (Napue, supra, 360 U.S. at p. 269.) The prosecutor had access to the recordings of Sabino and Araiza’s police interviews, as well as to the detectives who conducted them. Indeed, in the evidentiary hearing, the prosecutor testified that when Sabino testified about what he had told the police, the prosecutor
22 conferred with the detective regarding whether Sabino had claimed to have seen a tattoo during his police interviews. The People also argue the prosecutor corrected Sabino’s false testimony by “[taking] steps to clarify it,” including “elicit[ing] Sabino’s testimony that he might not have remembered everything he told the police during his two interviews” and “elicit[ing] Detective Flaherty’s testimony that Sabino had not been fully cooperative during the interviews and appeared to be in shock.” As to Araiza, the People contend the prosecutor corrected the testimony by questioning her as to whether she actually saw Melson fleeing with a gun and getting her to admit that she did not remember clearly what she told the police. The People conclude that “Napue does not require that a prosecutor present specific evidence to correct a witness’s false testimony.” Although there is a paucity of case law regarding what steps a prosecutor must take to correct a witness’s false testimony, Glossip makes clear that impeaching a witness’s credibility in general is insufficient. (See Glossip, supra, 604 U.S. at pp. 249-250 [rejecting the argument that there was no Napue error because the false testimony had been impeached on other grounds].) A general, non-specific question from the prosecutor to a witness about whether he or she remembers precisely what the witness told the police does not correct a specific false statement from that witness, not least of all because such a question gives no indication to the jury which part of the witness’s testimony is known to be false. Without an effort by the prosecutor to inform the jury more specifically what testimony was false, or to confront the witness about the specific falsity, the prosecutor did not correct the false testimony.
23 F. Was the False Testimony Material? The final step in showing a Napue violation is whether the false testimony was material. (U.S. v. Renzi, supra, 769 F.3d at p. 751.) “[M]ateriality . . . requires courts to assess whether ‘the error complained of’ could have contributed to the verdict.” (Glossip, supra, 604 U.S. at p. 253.) Courts have frequently emphasized that this standard is relatively “lenient.” (Dickey v. Davis (9th Cir. 2023) 69 F.4th 624, 637.) It “is considerably less demanding than the standard for Brady [v. Maryland (1963) 373 U.S. 83 [83 S.Ct. 1194, 10 L.Ed.2d 215]] claims, which requires that a petitioner show ‘there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.’ [Citations.] The Supreme Court has explained that Napue’s materiality threshold is lower ‘not just because [Napue cases] involve prosecutorial misconduct, but more importantly because they involve a corruption of the truth-seeking function of the trial process.’ ” (Dickey, supra, at p. 637, quoting United States v. Agurs, supra, 427 U.S. at p. 104.) “This standard is equivalent to the harmless beyond a reasonable doubt standard for determining whether constitutional error is prejudicial.” (In re Hill, supra, 104 Cal.App.5th at p. 835.) Thus, to establish materiality, it is not necessary to show that false testimony “directly affected the trial’s outcome.” (Glossip, supra, 604 U.S. at p. 253.) Glossip illustrates how the materiality standard applies in a case where, as here, the false testimony is relevant primarily to a witness’s credibility. In that case, the prosecution accused the defendant, Richard Glossip, of hiring Justin Sneed to kill Glossip’s employer. The primary evidence of Glossip’s involvement was testimony from Sneed himself, who agreed to
24 testify against Glossip in a plea deal to avoid the death penalty. (Glossip, supra, 604 U.S. at p. 233.) The prosecution theorized that Sneed “had no propensity to violence except at Glossip’s direction.” (Id. at p. 236.) On direct examination, Sneed acknowledged he was prescribed lithium after his arrest. He explained, “When I was arrested I asked for some Sudafed because I had a cold, but then shortly after that somehow they ended up giving me [l]ithium for some reason, I don’t know why. I never seen no psychiatrist or anything.” (Id. at p. 235.) Years later, the state turned over eight boxes of previously undisclosed prosecution documents pertaining to the case. In one of the boxes “ ‘was a page of notes handwritten by [the head prosecutor] during a pretrial interview with Sneed,’ indicating ‘that Sneed had told [the prosecutor] that he was “on lithium” not by mistake, but in connection with a “Dr. Trumpet.” ’ ” (Glossip, supra, 604 U.S. at p. 239.) Glossip deduced this was a reference to Dr. Larry Trombka, the only psychiatrist who worked in the Oklahoma County jail while Sneed was in custody there. Sneed’s previously withheld medical records “showed that Sneed had received lithium to treat his undisclosed bipolar disorder.” (Ibid.) The Supreme Court held that the prosecutor committed Napue error by failing to correct Sneed’s false testimony regarding the reason for his lithium prescription. The court reasoned that the testimony was material: “Because Sneed’s testimony was the only direct evidence of Glossip’s guilt of capital murder, the jury’s assessment of Sneed’s credibility was necessarily determinative here. Besides Sneed, no other witness and no physical evidence established that Glossip orchestrated [the] murder. Thus, the jury could convict Glossip only if it believed Sneed. [¶] Had the prosecution corrected Sneed on the
25 stand, his credibility plainly would have suffered. That correction would have revealed to the jury not just that Sneed was untrustworthy (as amicus points out, the jury already knew he repeatedly lied to the police), but also that Sneed was willing to lie to them under oath.” (Glossip, supra, 604 U.S. at pp. 248- 249.) The facts here are not as egregious as in Glossip, but the false testimony in this case had a similar effect. 1. Sabino Sabino was the prosecution’s primary witness. He saw the shooting up close and was the only witness who testified that Melson fired the gun. Even better for the prosecution, he had no apparent bias against Melson. He was not a gang member and claimed he had never met Melson before. The primary weakness in his testimony was the reliability of his identification of Melson. Because the shooter was wearing both a hoodie and a beanie, Sabino primarily saw the bottom part of the shooter’s face, and he saw it only briefly, in a highly stressful and traumatic situation. As he acknowledged during his second interview with the police, his identification of Melson was only a “best guess.” The claim that he saw Melson’s tattoo shored up this weakness considerably. It provided a concrete basis for the identification—Sabino did not choose Melson’s photo because he generally resembled the shooter, but because he had a specific tattoo Sabino remembered seeing. This claim would be stronger still if Sabino told police about the tattoo within days of the shooting. In that case, it would mean the tattoo was a part of Sabino’s identification from the beginning. The revelation that Sabino had not actually told the police about the tattoo, but instead affirmatively denied seeing any such
26 tattoo, would have had the opposite effect. It would mean Sabino either misremembered an important part of his identification of Melson, or that he was lying. Either way, the correction of the false testimony might have created a doubt in the jurors’ minds about the credibility of Sabino’s identification, especially in combination with Sabino’s admission that he primarily saw the lower part of the shooter’s face. The People argue Sabino’s false testimony was not material because the jury heard other testimony indicating that Sabino did not tell the police about the tattoo. 4 In particular, they point to the cross-examination of one of the detectives who interviewed Sabino. As noted above, after the detective testified about what Sabino said he remembered about the shooter, Melson’s attorney asked, “So there was no mention of—him saying a tattoo or anything of that nature?” The detective replied, “No, not at that time.” As an initial matter, the detective’s testimony did not satisfy the prosecutor’s responsibility to correct Sabino’s testimony as it was elicited by Melson’s attorney on cross- examination. (See Glossip, supra, 604 U.S. at p. 252 [“the Due Process Clause imposes ‘ “the responsibility and duty to correct” ’
4 The People also claim that a “recording of [Sabino’s second police] interview was played at trial and a transcript was provided to the jury.” If the jury had seen the entire recording, it might have significantly reduced the materiality of Sabino’s false testimony, because the jury would have seen for itself that Sabino shook his head when asked if the shooter had a tattoo. But as Melson notes in his traverse, the record shows that the jury saw only a small portion of the interview, which did not contain any discussion of tattoos.
27 false testimony on ‘representatives of the [s]tate,’ not on defense counsel”], quoting Napue, supra, 360 U.S. at pp. 269-270.) Nor did the defense counsel’s cross-examination of the detective sufficiently deflect the materiality of Sabino’s false testimony. 5 First, the statement that Sabino did not mention a tattoo “at that time” is not the same as a statement that Sabino affirmatively denied seeing a tattoo during both police interviews. Second, the prosecutor made no attempt to confront Sabino with his statements to police that he did not see a tattoo so that the jury could evaluate Sabino’s demeanor to that revelation in real time. Third, the prosecutor relied on Sabino’s trial testimony about the tattoo during closing argument (albeit not the portion saying he told detectives about it), but Melson’s attorney made no mention of the detective’s statement during closing arguments that Sabino’s memory of the tattoo was a recent development. Finally, as we explain in more detail below, “prejudice analysis requires a ‘cumulative evaluation’ of all the evidence.” (Glossip, supra, 604 U.S. at p. 251.) The materiality of Sabino’s false testimony must be considered in conjunction with the other errors in the case, including that of Araiza’s false testimony.
5 We do not mean to imply the detective’s testimony on cross-examination has no impact on the materiality analysis. To the contrary, the testimony was particularly useful in impeaching Sabino because the detective was a prosecution witness and was part of the People’s prosecution team who had no reason to testify favorably on behalf of Melson. We merely conclude that, although the cross-examination of the detective on this point was helpful to Melson’s case, it did not fully blunt the force of Sabino’s earlier false testimony.
28 2. Araiza Araiza’s false testimony had a similar effect of bolstering her credibility and usefulness to the prosecution. Araiza told three different versions of what she saw when she came out of her house after the shooting. In her police interview, she said she did not see any Black men. At the preliminary hearing and again in the first trial, she testified she saw Chops and two other unidentified Black men fleeing the scene, but she did not say she saw Melson. Finally, at the second trial, she testified she saw Melson holding what appeared to be a gun and fleeing along with Chops. Testifying that she told the police she saw Melson after the shooting made it appear as if she had been telling a more consistent story all along. Araiza’s false claim about what she said in her police interview went essentially unchallenged. As noted above, the prosecutor never corrected it. Instead, the prosecutor reinforced the error by saying during his opening statement before Araiza testified that she was “going to tell you that she saw—this is what she told the detective when they interviewed her, that she saw Chops and someone that resembled [Melson] running from the scene,” (italics added), testimony that the prosecutor should have known would be false. Defense counsel cross-examined Araiza about how often she had seen Melson before and how well she could see the aftermath of the shooting given lighting conditions and her distance from the scene, but he never impeached her with her statement to police that she not only did not see Melson but saw no Black men at all right after the shooting. Then, at the end of the trial, the prosecutor repeated Araiza’s false testimony during closing argument and relied on it
29 to argue for a conviction, saying that when Araiza was interviewed by the police “[s]he I.D.’s Mr. Melson as the shooter.” Araiza was much farther away at the time of the shooting, she disagreed with Sabino as to the color of Melson’s hoodie, and, contrary to the prosecution’s theory, she testified that Chops was the shooter rather than Melson. But she was still one of only two witnesses who directly connected Melson to the crime. Her false testimony thus contributed to the verdict. 3. Defense Counsel’s Lack of Preparation as a Contributing Factor Because we do not analyze the materiality of a Napue violation in isolation (Glossip, supra, 604 U.S. at p. 251), we consider not only the effect of Sabino’s and Araiza’s false testimony, but also the related issue of Melson’s attorney’s preparation to cross-examine them. Melson raised the issue of insufficient trial preparation in his habeas petition, noting that there was no indication that his attorney had taken any notes prior to the retrial, nor that he had reviewed the recording of Araiza’s police interview or Melson’s first police interview. At the evidentiary hearing, Melson’s attorney claimed he took notes, but also claimed he turned over all his trial materials. The materials the attorney turned over included no such notes, nor transcripts of the two police interviews. A lack of adequate preparation can alone demonstrate ineffective assistance of counsel. (See In re Edward S. (2009) 173 Cal.App.4th 387, 407 [“a defense attorney who fails to investigate potentially exculpatory evidence, including evidence that might be used to impeach key prosecution witnesses, renders deficient representation”].) In addition, “[f]acts set forth in the return that are not disputed in the traverse are deemed true.” (People v.
30 Duvall (1995) 9 Cal.4th 464, 477.) This is because courts must be able to examine the return and traverse in order “to determine whether there are facts legitimately in dispute that may require holding an evidentiary hearing.” (Id. at p. 485.) In this habeas proceeding, the People have never contested Melson’s allegations regarding his attorney’s preparation to cross-examine Sabino and Araiza. 6 This is presumably because they drew the same obvious inference we do from the record— that defense counsel did not take notes and did not have a copy of the police interview transcripts despite their production to Melson’s prior trial counsel. 7 This would explain why Melson’s attorney did not raise the inconsistencies between the witnesses’ statements in those interviews with their trial testimony on cross-examination. We need not decide whether this lack of preparation alone constituted ineffective assistance of counsel under the standard established in Strickland v. Washington (1984) 466 U.S. 668 [104 S.Ct. 2052, 80 L.Ed.2d 674]. Instead, we view it as one part of the analysis of the materiality of the Napue violations. We infer that Melson’s attorney’s lack of preparation left him unable to counter Sabino’s and Araiza’s false testimony, and the cross- examination of both witnesses amply supports that inference.
6 Nor did the trial court make any factual findings at the conclusion of the habeas evidentiary hearing regarding defense counsel’s preparation to cross-examine Sabino and Araiza about their prior statements to police. 7 There is one exception: the record suggests that at some point Melson’s attorney obtained a copy of the transcript of Sabino’s second police interview.
31 4. The People Fail to Show a Lack of Prejudice Melson’s first trial ended in a hung jury, with nine jurors voting to convict, and three to acquit. At the second trial, both of the prosecution’s key witnesses changed their stories to claim falsely that they had told the police certain key identifying facts about Melson from the get-go. Melson’s attorney, who had failed to prepare adequately to cross-examine those witnesses, challenged the false testimony only in part and only as to Sabino, and then failed in his closing argument even to mention the detective’s contradiction of Sabino’s claim. The prosecutor did not correct the false testimony, and in fact repeated Araiza’s false testimony during closing argument. The result was a conviction. Take away those errors, and the evidence at the retrial was basically the same as it was the first time around when three jurors voted to acquit. In such circumstances, the People have not shown beyond a reasonable doubt that Sabino’s and Araiza’s false testimony did not contribute to the guilty verdicts. (Glossip, supra, 604 U.S. at p. 246.) We therefore grant Melson’s petition and vacate the judgment of conviction.
32 DISPOSITION The writ of habeas corpus is granted and Melson’s judgment of conviction is vacated. The matter is remanded to the superior court for further proceedings, including for the People to elect whether to retry Melson. CERTIFIED FOR PUBLICATION
WEINGART, J.
We concur:
ROTHSCHILD, P. J.
BENDIX, J.